The Hague Trusts Convention: Scope, Application and Preliminary Issues 9781474200462, 9781841131108

Trusts cross borders. When they do,real difficulties may arise. Will the understanding of what a trust is be different i

233 18 3MB

English Pages [555] Year 2002

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Hague Trusts Convention: Scope, Application and Preliminary Issues
 9781474200462, 9781841131108

Citation preview

Preface RUSTS CROSS BORDERS.

Today, trusts are crossing borders ever more frequently. When they do, real difficulties may arise. Will the understanding of what a trust is be different in a foreign state? Will the rights, powers and duties of the trustee and settlor be the same? What rights will the beneficiary be able to assert? To what extent will the trust assets be safe from the claims of creditors, forced heirs, or third parties? Which legal system will be applied to the trust? Within what limits? What if the trust needs to be recognised in a state which does not have the institution of the trust in its domestic law? The importance of these questions is self-evident. The divergence in attitudes to the trust world-wide makes the prospect of a conflict of laws inevitable. Indeed, it is not easy to think of too many areas of law more likely to give rise to a conflict of laws. Yet, (save for the enormous contribution made to the advancement of this subject by Professor David Hayton, who headed the United Kingdom delegation to the Hague Conference where the Trusts Convention was concluded, and whose writings have proved invaluable to an understanding of the Convention in England), detailed literature on the topic by English commentators is very sparse. The Hague Trusts Convention, enacted into English law by the Recognition of Trusts Act 1987, seeks to ameliorate the situation by providing harmonised choice of law rules for “trusts created voluntarily and evidenced in writing”. It also provides for the recognition of trusts complying with the requirements of the Convention in contracting states. Those contracting states should recognise the trust, even if they do not have the institution in their domestic law. Of course, the problems raised by the recognition of a trust in non-trust states for private international law purposes can be varied and complex. This monograph is the first published in England to devote itself entirely to a detailed analysis of the Hague Trusts Convention. It is aimed at academics and practitioners; at private international lawyers and at trust lawyers. The perspective is that of an English private international lawyer, though it is very much hoped that it will be of interest to lawyers in other contracting (and, indeed, non-contracting) states to the Convention and frequent reference is made to the position in civil law states and in other trust states, both offshore and onshore. The comparative law problems created by the recognition of a trust in non-trust states are a key backdrop to understanding the aims, scope and effectiveness of the Convention and are considered throughout the book as they become relevant. Nevertheless, this is not intended to be a work on the comparative law of trusts and does not purport to present a detailed comparison of the English trust

T

vi Preface with its counterparts in other trust states and with homologues and analogues in civil law states. Some comment on the structure of the book may assist the reader. It is a book of two parts, of which the second (the analysis of the Hague Trusts Convention itself) is by far the larger. The author decided that, for ease of reference, it would be most convenient to the reader for the analysis of the Hague Trusts Convention to be set out in Article-by-Article format. Of course, monographic treatment does not always suit such an approach and some topics inevitably straddle more than one Article. Such topics are considered under the Article under which questions concerning them might most naturally arise. The Hague Trusts Convention deals with the operation of the trust itself. It does not deal with the preliminary steps needed to create a trust and it would certainly have been possible to limit discussion in this book to the Convention itself. However (and having taking the advice of several academics and practitioners), the author felt that it would be undesirable to consider the operation of the Hague Trusts Convention without considering how such trusts are created in the first place. Creation and operation of Hague Convention trusts are inextricably inter-related in practice, when contemplating the validity of a transnational trust. This book affords in Part One detailed treatment to these preliminary matters, without which a Hague Convention trust can never come into effect and considers which choice of law rules an English court would apply. Many of the choice of law rules applicable to these preliminary steps raise some of the most complex and uncharted choice of law questions, e.g. the capacity of the settlor, to which detailed consideration is given. Of course, other preliminary matters are more general, e.g. the need for a valid will in order for an express testamentary trust to take effect, or the potential impact of matrimonial property regimes on a trust. These topics could each be the subject of a book in their own right. The law applicable in these areas is concisely summarised. Although these issues may not be trusts specific, their practical importance to the creation of Hague Convention trusts is paramount and the author felt that it might prove useful to some readers briefly to explain the basic choice of law rules for those areas in this part. Indeed, it did not seem desirable to discuss some preliminary issues relating to the creation of Hague Convention trusts, but to omit others. Often far more developed choice of law rules on preliminary issues exist in offshore trust states and close attention is paid to their legislation in this area. Indeed, frequent reference is made to the choice of law rules in offshore jurisdictions throughout the book. It is intended by the author that this book will eventually be followed by a second volume considering other private international law issues affecting trusts: in particular, the law of jurisdiction and enforcement of foreign judgments in transnational trusts litigation and the choice of law rules affecting resulting and constructive trusts (insofar as they fall outside the Hague Trusts Convention).

Preface vii I owe debts of gratitude to numerous people and shall only mention a few. Richard Hart and all those at Hart Publishing have been extremely helpful and approachable and have greatly assisted me in my work. John Glasson, consultant to Eversheds and editor of International Trust Laws is largely responsible for fuelling my interest in the private international law of trusts over the years. He has also provided me on numerous occasions with materials that I was struggling to track down, as well as frequently providing me with invaluable advice on this book and on the concerns of practitioners in this area. The Official Report to the Convention is reproduced with the kind permission of the Permanent Bureau of the Hague Conference on Private International Law and Professor von Overbeck. At the University of Birmingham, I would like to thank Professor David Feldman (for encouraging me to pursue this project), Imogen Moore (for her friendly advice and for discussing with me esoteric points of property law), Frank Meisel (likewise) and Sophie Boyron (for listening to my rants about my (lack of) progress on the book). At the University of Nottingham, I would like to thank Professor James Fawcett, who assiduously read a draft of the entire book and provided numerous invaluable comments. I would also like to thank Professor Andrew Simester for his advice and encouragement; for similar reasons I would like to thank Professor Michael Bridge, University College, London and Professor David Hayton, King’s College, London. At the University of Oxford, I would like to thank Sir Peter North for his advice. I would also like to thank Adrian Briggs, both for advising me on this project and, more generally, for his enormous support and encouragement of me in my work over the years. At the University of Durham, I would like to thank Panos Koutrakos, who, whilst making clear that he could not imagine how anyone could possibly want to write a book on international trusts, was nonetheless very supportive and willing to advise me. I would especially like to thank John Stevens (University of Birmingham) for reading a draft of the entire book and for providing invaluable comments on it. John has encouraged and assisted me greatly during the writing of this book and had, on innumerable occasions, to face a barrage of the most obscure property questions imaginable from me (normally, whilst he was trying to eat his lunch or to drink his coffee). Finally, I would like to thank my parents for their unfailing support and encouragement and I can be sure that this book will find a home on their coffee table. Jonathan Harris August 2001

Table of Cases ABN—Amro Bank Luxembourg SA v. Trustees of the C (Jersey) Foundation (Case no 15430) 27 April 1994, Luxembourg CA ...........328, 334 Achillopoulos, Re [1928] Ch 433.............................................................43, 45 Adams v. Clutterbuck (1883) 10 QBD 403 ....................................................26 Afanoor’s Trusts, Re (1895) 64 LJ Ch 521...................................................199 Akai Pty Ltd v. The People’s Insurance Co Ltd (1996) 141 ALR 374, High Ct of Aus .......................................................................................195 Aldrich v. A-G [1968] P 281..........................................................................44 Amin Rasheed Shipping Corp v. Kuwait Insurance Co [1983] AC 50.....................................................................185, 193, 195, 228 Annesley, Re [1926] Ch 692......................................27, 28, 47, 51, 55, 366, 392 Arab Monetary Fund v. Hashim (No 9) [1996] 1 Lloyd’s Rep 589, CA; affirming in part [1993] 1 Lloyd’s Rep 543 .........................140, 239, 250, 324 Arab Monetary Fund v. Hashim (No 11) The Times, 11 October 1994................................................................246, 247, 250, 377 Ardagh, In the Estate of [1914] 1 IR 5 .........................................................236 Armar Shipping Co Ltd v. Caisse Algérienne [1981] 1 WLR 207 ..................298 Armitage v. Nurse [1998] Ch 241................................................................395 Ashby, Re [1892] 1 QB 873 .........................................................................376 Astor’s Settlement Trusts, Re [1952] Ch 534 ...............................................295 Attenborough v. Solomon [1913] AC 76 .......................................................54 A-G v. Bellilos [1928] 1 KB 798.....................................................................74 A-G v. Campbell (1872) LR 5 HL 524....................................200, 210, 211, 220 A-G v. Jewish Colonization Association [1901] 1 KB 123...............................................203–5, 207, 209, 210, 221, 348 A-G v. Johnson [1907] 2 KB 885 ...............................................................34–7 A-G v. Lepine (1818) 2 Swan 181 ................................................................293 A-G v. Mill (1831) 2 Dow & Cl 393 ............................................................292 A-G v. Stephens (1834) 3 My & K 347 ........................................................294 A-G v. Sturge (1854) 19 Beav 594; (1854) 18 Beav 597..................................293 A-G for Hong Kong v. Reid [1994] 1 All ER 1.............................................272 A-G of New Zealand v. Ortiz [1984] AC 1..................................................398 A-G of the United Kingdom v. Heinemann Publishers Australia Pty Ltd (1988) 165 CLR 30; (1987) 19 NSWLR 86 .....................250, 251, 324 A-G’s Reference (No 1 of 1985) [1986] 2 All ER 219 ....................................272 Augustus v. Permanent Trustee Co (Canberra) Ltd (1971) 124 CLR 245 .................................................24, 161, 169, 208, 211, 227, 285

xxii Table of Cases Bank of Africa v. Cohen [1909] 2 Ch 129.................................................15, 16 Bankes, Re [1902] 2 Ch 333 ..........................................................................12 Bannister v. Bannister [1948] 2 All ER 133 ..................................................146 Baring Bros & Co Ltd v. Cunninghame District Council [1997] CLC 108 ......................................................................................126 Barnett’s Trusts, Re [1902] 1 Ch 846.............................................................51 Barrie’s Estate, Re 240 Iowa 43, 35 NW 2d 658 (1949)...................................50 Batthyanay v. Walford (1887) 36 Ch D 269 .................................................176 Beatty’s Will Trust, Re (No 2) judgment of 28 February 1991, Ch D, Trust Law International 77 ............................................236, 299, 308 Beaudoin v. Trudel [1937] 1 DLR 216...........................................................64 Berchtold, Re [1923] 1 Ch 192...............................................9, 34, 37, 172, 173 Betts Brown Trust Fund Trustees, Re 1968 SC 170......................................284 Billson v. Crofts (1873) LR 15 Eq 314..........................................................376 Black v. Black’s Trustees 1950 SLT (Notes) 32 ........................................15, 17 Bodley Head Ltd v. Flegon [1972] 1 WLR 680 ...........................................8, 14 Bonacina, Re [1912] 2 Ch 394 .....................................................................117 Bonython v. Commonwealth of Australia [1951] AC 201 ............................169 Boston Safe Deposit & Trust Co v. Alfred University 339 Mass 82 ..............204 Boys v. Chaplin [1971] AC 356, HL; affirming [1968] 2 QB 1, CA ........253, 254 Branco v. Veira (1995) 8 ETR (2d) 49..........................................................285 Brassard v. Smith [1925] AC 371 ..................................................................30 Bridge Trust Co Ltd and Slatter v. A-G, Wahr Hansen and Compass Trust Co Ltd, Cayman Islands..........................................299–301 Buchanan v. McVey [1955] AC 516.............................................................399 Bullen v. Wisconsin 240 US 625, 36 Sup Ct 473 .............................................55 Bundesgerichtshof 4 April 1968, Juristenzeitung 1968, 428 ..........................375 Burroughs-Fowler, Re [1916] 2 Ch 251 .......................................................376 Caithness, Re (1891) 7 TLR 354 ...................................................................50 Camille and Henry Dreyfus Foundation Inc v. IRC [1956] AC 39, HL; affirming [1954] Ch 672 ............................................................292, 293 Cammell v. Cammell [1965] P 467 ..............................................................257 Cammell v. Sewell (1858) H & N 617 ...........................................................28 Caron v. Odell (1986) 75 Revue Critique de Droit International Privé 66...........................................................................................335, 368 Casani v. Mattei (1998/99) 1 ITELR 925, Tribunale di Lucca ....350–2, 370, 371 Casio Computer Co Ltd v. Sayo (No 3) The Times, 6 February 2001......................................................................................144 Casio Computer Co Ltd v. Kaiser, judgment of 11 April 2001, [2001] EWCA Civ 661, CA ..............................................................144, 256 Charron v. Montreal Trusts Co (1958) DLR (2d) 240, Ontario CA ...............14 Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd [1981] Ch 105................................136, 146, 253, 255, 324, 377

Table of Cases xxiii Chelleram v. Chelleram [1985] Ch 409............................206, 207, 210, 218–21, 228, 234–6, 241, 253, 257, 258, 285, 286, 384 Chilwell v. Carylon (1897) 14 SC 61, South Africa ........................................63 Church of Scientology California v. Commissioner of Police (1976) 120 SJ 690 ....................................................................................240 Cigala’s Settlement Trusts, Re (1878) 7 Ch D 351 .........................................35 Clark, In Re Estate of 21 NY 2d 478, 288 NYS 2d 993...................................55 Cloncurry’s Case [1932] IR 687 ...........................................................202, 203 Clunet 1896, 628, judgment of 9 March 1895 ..............................................371 Clunet 1935, 651 ........................................................................................131 Coast Lines Ltd v. Hudig & Veder Chartering NV [1972] 2 QB 34 ..............194 Cohn, Re [1945] Ch 5 .............................................................................9, 156 Collens, Re [1986] Ch 505.......................................................................50, 51 Colonial Bishoprics Fund (1841), Re [1935] 1 Ch 148 ..................................293 Commissioner of Stamp Duties (Queensland) v. Livingstone [1965] AC 694, PC....................................................................................35 Cooke’s Trusts, Re (1877) 56 LJ Ch 637 .......................................................12 Cooke’s Trusts, Re (1887) 56 LT 737............................................................64 Cooper v. Cooper (1888) 14 App Cas 88 ............................................12, 13, 64 Corte di Cassazione, sez 1 civile, 14 ottobre 1997 n 10031 ...........................316 Cour d’appel Paris (18 April 1929) [1935] Revue Critique de Droit International Privé 149 ...........................................................................314 Cour d’appel Paris (10 January 1970) [1971] Revue Critique de Droit International Privé 518 ...........................................................................168 Cour de Cassation: judgment of 20 February 1996 [1996] Dalloz-Sirey Jurisprudence 390...................................................................................331 Courtois v. De Ganay, decision of 10 January 1970 [1971] Revue Critique de Droit International Privé 518, Paris CA.......................332 Craven’s Estate, Re [1937] 1 Ch 423 ........................................................42, 43 Custom Made Commercial v. Stawa Metallbau (Case C–288/93) [1994] ECR I–2913 ..........................................................................224, 288 D’Alemida Araujo Ltd v. Becker & Co Ltd [1953] 2 QB 329........................254 De Nicols, Re [1900] AC 639........................................................................63 De Nicols, Re (No 2) [1900] 2 Ch 410 .........................................................364 De Nicols v. Curlier [1900] AC 21 ................................................................63 Denley’s Trust Deed, Re [1969] 1 Ch 373 ....................................................295 Devos v. Devos (1970) 10 DLR (3d) 603 ......................................................228 Dexter Ltd (In Administration) v. Harley, The Times, 8 March 2001, 2 April 2001 ....................................................................................144, 256 Dr Barnado’s Homes National Incorporated Association v. Commissioner for Special Purposes of the Income Tax Acts [1921] 2 AC 1 .......................35 Dubai Aluminium v. Salaam [2000] 2 Lloyd’s Rep 168, CA; reversing [1999] Lloyd’s Rep 415 ...........................................................................247

xxiv Table of Cases Duke of Marlborough v. A-G (No 1) [1945] Ch 78, CA ..........63, 205, 211, 228, 297–9, 304, 305, 364 Duke of Wellington, Re [1947] Ch 506 .......................................27, 28, 51, 388 Duncan v. Lawson (1889) 41 Ch D 394..................................................49, 174 Duyvewaardt v. Barber (1992) 43 RFL (Can) (3rd) 139................................297 Dynamit Actien-Gesellschaft v. Rio Tinto Co Ltd [1918] AC 260 ................390 Egerton’s Will Trusts, Re [1956] 2 All ER 817...............................................62 Egon Oldendorff v. Liberia Corp (No 1) [1995] 2 Lloyd’s Rep 64..............................................................................193, 279 Egon Oldendorff v. Liberia Corp (No 2) [1996] 1 Lloyd’s Rep 380 ...................................................................................193 El Ajou v. Dollar Land Holdings plc [1993] 3 All ER 717; reversed [1994] 2 All ER 685 .......................................22, 140, 246, 253, 255, 324, 377 El Du Pont de Nemours & Co v. Agnew and Kerr [1987] 2 Lloyd’s Rep 585.........................................................................298 Elefanten Schuh v. Jacqmain (Case 150/80) [1981] ECR 1671 ......................273 Emery v. Clough 63 NH 552, 4 A 796 (1885) .................................................42 Emery v. Hill (1826) 1 Russ 112 ..................................................................293 Emery’s Investment Trusts, Re [1959] Ch 410 .............................................398 Eric Beach Co v. A-G for Ontario [1930] AC 161 ..........................................30 Euro-Diam v. Bathurst [1987] 2 All ER 113.................................................383 Eves v. Eves [1975] 1 WLR 1338 .................................................................132 Faillité Four Seasons Overseas NV v. SA Finimtrust, decision of 21 January 1971 [1972] Revue Critique de Droit International Privé 51, Trib d’arrondissement de Luxembourg ...................................................334 Favorke v. Steinkopff [1922] 1 Ch 174 ...............................................34, 35, 37 Federal Trade Commission v. Affordable Media LLC 179 F 2d 13130 (1999).....................................................................................................309 First National Bank in Mitchell v. Daggett 242 Neb 734, 497 NW 2d 358 (1993) ............................................................................169 Fitzgerald, Re [1904] 1 Ch 573..................................63, 155, 168, 176, 244, 391 Fletcher v. Fletcher (1844) 4 Hare 67.............................................................43 Forbes v. Cochrane (1824) 2 B & C 448 ......................................................390 Ford v. Newman 77 Ill 2d 235, 396 NE 2d 539 (1979) ..................................198 Fordyce v. Bridges (1848) 2 Ph 497..............................................................200 Foster v. Driscoll [1929] 1 KB 470.................................................382, 390, 393 Forsyth v. Forsyth [1891] P 363 ...........................................................259, 262 Foskett v. McKeown [2001] AC 102 ...........................................................324 Fraser, Re (1883) 22 Ch D 527 ....................................................................294 Freke v. Lord Carbery (1873) LR 16 Eq 461 ................................9, 49, 154, 174 Fry, Re [1946] Ch 312 .............................................................................37, 40 Funk v. Krombach (2001) 15 Trust Law International 103 .....................55, 368

Table of Cases xxv Garner v. Bermuda Trust Co Ltd (1992) J Int Corp P 133 (Note), Bermuda CA.....................................................................................70, 369 Gartside v. IRC [1968] AC 553 .....................................................................39 Genoese Commercial Court, 24 March 1997 ...............................................352 Georges-Picot v. Henderson [1963] NZLR 950............................................299 Godard v. Gray (1870) LR 6 QB 139 ................................................67, 94, 101 Goff v. Goff [1934] P 107 ....................................................................259, 262 Golden Acres Ltd v. Queensland Estates Pty Ltd [1969] Qd R 378 .....................................................................................180 Government of India v. Taylor [1955] AC 491 .....................................397, 398 Grainge v. Wilberforce (1889) 5 TLR 436 .....................................................41 Grant v. Edwards [1986] 2 All ER 426.........................................................132 Grey v. IRC [1960] AC 1 ..............................................................................41 Groos, Re [1915] 1 Ch 572 ...........................................................................47 Groupe Concorde v. Suhadiwarno Panjan (Case C–440/97) [1999] ECR 1–6307..........................................................................224, 288 Grupo Torras v. Al-Sabah [2001] Lloyd’s Rep Bank 36 ..............................................................140, 246, 247, 255, 377 Guépratte v. Young (1851) 4 De G & Sm 217 ........................................63, 275 Habershon v. Vardon (1851) 4 De G & Sun 467...................................291, 292 Halley, The (1968) LR 2 PC 193 .................................................................240 Hamlin v. Hamlin [1985] 2 All ER 1037 ......................................................367 Harbour Assurance Ltd v. Kansas Ltd [1993] QB 701..................................277 Harris Investments Ltd v. Smith [1934] 1 DLR 748 ................203, 211, 221, 252 Harrison, Re [1891] 2 Ch 349 .....................................................................239 Harrison v. Crédit Suisse ATF 96.II.79 JT 197.I.329, Swiss Federal Tribunal........................................................................331–3 Hayward (Dec’d), Re [1996] 3 WLR 674 .....................................................358 Helbert Wagg & Co Ltd, Re [1956] 1 Ch 323 ........................................20, 307 Hellenic Steel Co v. Svolamar Shipping Co Ltd (‘The Komminos S’) [1991] 1 Lloyd’s Rep 370.........................................................................192 Hellman’s Will, Re (1866) LR 2 Eq 363.........................................................48 Hernando, Re (1884) 27 Ch D 284 ............................................50, 60, 168, 364 Hewitt’s Settlement, Re [1915] 1 Ch 228 ..............................................284, 364 Holzberg v. Sasson (1986) 75 Revue Critique de Droit International Privé 685 ....................................................................368, 369 Hope v. Brewer 136 NY 126 .......................................................................292 Hoyles, Re [1911] 1 Ch 179..............................................................9, 174, 291 Hutchinson v. Ross 262 NY 381, 187 NE 65 (1933) .......................169, 198, 219 Income Tax Special Purposes Commissioners v. Pemsel [1891] AC 531 .......292 Industrie Tessili Italiana Como v. Dunlop AG (Case 12/76) [1976] ECR 1473 .............................................................................224, 288

xxvi Table of Cases IRC v. Holmden [1968] AC 685 ..................................................................271 Irani Finance Ltd v. Singh [1971] Ch 59 ......................................................172 Irving v. Snow 1956 SC 257 ........................................................................244 Iveagh v. IRC [1954] Ch 364 .........................................168, 205, 206, 210, 211, 213, 220, 224, 225, 228 Jacobs, Re (1970) 114 Sol Jo 515.................................................................292 Jakob Handte GmbH v. Traitements Mecano-Chimiques des Surfaces (Case C–26/91) [1992] ECR I–3967 ..................................................121, 125 Janred Properties Ltd v. Ente Nazionale Italiano per il Turismo [1989] 2 All ER 444 ..................................................................................20 Jewish National Fund Inc v. Royal Trust Co (1965) 53 DLR (2d) 577 Canadian SC..................................................200, 221, 229 Joseph’s Will Trusts, Re (1907) 26 NZLR 504.............................................294 Kahler v. Midland Bank [1950] AC 24 ........................................................190 Kapetan Georgis, The [1988] 1 Lloyd’s Rep 352 ..........................................239 Kaufman, In the Goods of [1952] P 325 ........................................................44 Kehr (Dec’d), Re [1952] Ch 26 ........................................46, 234, 236, 237, 241, 267, 268, 289, 373, 384 Ker’s Settlement Trusts, Re [1963] Ch 553 .....................241, 256, 261, 262, 384 Khan’s Settlement, Re [1966] Ch 567 ............................................................60 King’s Will Trusts, Re [1964] Ch 542............................................................54 Korvine’s Trust, Re [1921] 1 Ch 343 .............................................................42 Kuwait Oil Tanker v. Al Bader [2000] 2 All ER (Comm) 271 ..............................................................140, 246, 247, 255, 377 Kwok Chi Leung Karl v. Estate Duty Commissioners [1988] 1 WLR 1035...................................................................................30 Lashley v. Hog (1804) 4 Pat 581....................................................................63 Lashmar, Re [1891] 1 Ch 258........................................................................41 Lee v. Abdy (1886) 17 QBD 309....................................................................16 Lemenda Trading Co Ltd v. African Middle East Petroleum Ltd [1988] QB 448 ........................................................................................390 Lemos v. Coutts & Co (1992–3) CILR 460, Cayman Islands .........................67 Levick’s Will Trusts, Re [1963] 1 WLR 311.................................................233 Levy Estate, Re (1989) 58 DLR (4th) 375.....................................................223 Lewal’s Settlement, Re [1918] 2 Ch 391 ........................................................60 Lewis v. Hanson 36 Del CH 235, 128 A 2d 819 (1957) .................................198 Liddiard, Re (1880) 14 Ch D 310.................................................................236 Lightning v. Lightning Electrical Contractors Ltd [1998] NPC 71 ...........................................................127, 149, 176, 177, 187 Lindsay v. Miller (No 1) [1949] VR 13........................205, 209–11, 218–20, 288 Lister & Co v. Stubbs (1890) 45 Ch D 1 ......................................................272

Table of Cases xxvii Liverpool Marine Credit Co v. Hunter (1868) LR 3 Ch App 479..................253 Lord Cable, Re [1977] 1 WLR 7 .............................................39, 199, 201, 205, 209–11, 381, 383, 397, 398 Lord Sudeley v. A-G [1897] AC 11.....................................................34, 35, 37 Lorillard, Re [1922] Ch 638 ..........................................................................45 Lyons v. New Bonaker (1867) LR 4 Eq 655 .................................................294 McGovern v. A-G [1982] Ch 321 ................................................................292 McKain v. RW Miller & Co (SA) (1991) 174 CLR 1....................................253 Macmillan Inc v. Bishopsgate Trust plc (No 3) [1996] 1 WLR 387, CA ....................................................8, 30, 141, 324, 325 McMorran, Re [1958] Ch 624 ......................................................................60 Maimann v. Maimann, judgment of 27 June 2001 [2001] EWCA Civ 1132...........................................................................251 Maldonado’s Estate, Re [1954] P 223 ........................................................9, 51 Male v. Roberts (1800) 3 Esp 163..................................................................13 Manisty’s Settlement, Re [1974] Ch 17........................................................271 Martin, Re [1900] P 211..........................................................................49, 62 Martin v. Paxton (1826) 1 Russ 115 ............................................................293 Matthews (Dec’d), Re (1980) JJ 139............................................................198 Mayor of Canterbury v. Wyburn [1895] AC 89 ...........................................292 Meatyard, In the Goods of [1903] P 125........................................................44 Mégret, Re [1901] 1 Ch 547 .............................................................55, 60, 369 Metall und Rohstoff AG v. Donaldson Lufkin & Jenrette Inc [1990] 1 QB 391......................................................................................142 Micallef’s Estate, Re [1977] 2 NSWLR 929 ...................................................50 Milan Commercial Court, 27 December 1996 ......................................352, 370 Miller, Re [1914] 1 Ch 511 ...........................................................................49 Minet v. Vulliamy (1826) 1 Russ 113...........................................................293 Mirrlees’ Charity, Re [1910] 1 Ch 163.........................................................294 Mitchner, Re [1922] St R Qd 39..................................................................200 Mitchner, Re (No 2) [1922] St R Qd 252 .....................................................200 Montgomery v. Zarifi 1918 SC 128 .............................................................264 Moore (Dec’d), Re; Moore v. His Holiness Pope Benedict XV [1919] IR 316..........................................................................................292 Moore v. Livingstone 148 Ind App 275, 265 NE 2d 251 (1970) .....................173 Morgan Guaranty Trust Co v. Huntingdon 140 Conn 331, 179 A 2d 604 (1962) ..................................................................................60 Morice v. Bishop of Durham (1804) 9 Ves 399 ............................................295 Municipality of Canterbury v. Wyburn [1895] AC 89..................................152 Murphy v. Deichler [1909] AC 446 ...............................................................61 NV Handel Maatschappij J Smiths Import-Export v. English Exporters Ltd [1955] 2 Lloyd’s Rep 69 ....................................................254

xxviii Table of Cases Nanton Estate, Re [1948] WWR 113...........................................................284 National Commercial Bank v. Wimborne (1978) 5 BPR [97323] ..................247 National Shawmut Bank v. Cumming 325 Mass 457 ...................................227 Nelson v. Bridport (1846) Beav 547 ..................................................26, 49, 169 Netherlands v. Rüffer (Case 814/79) [1980] ECR 3807.................................116 New York Breweries Co Ltd v. A-G [1899] AC 62 ...................................43, 44 New York Life Insurance Co v. Public Trustee [1924] 2 Ch 101 ........................................................................................30 Nunneley v. Nunneley and Marrian (1890) 15 PD 186...................256, 259, 262 OD-Bank in Liquidation v. Bankrupt’s Estate WKR (Case No. FB920075) 98 Blätter für Zürcherishe Rechtssprechung (1999) No. 52, Zurich DC ...............................................................................100, 332, 333 Ogden v. Ogden [1908] P 46...........................................................................8 O’Keefe, Re [1940] Ch 124 ...........................................................................51 Oldfield (No 2), Re [1949] 2 DLR 175 .........................................................293 Oliphant v. Hendrie (1784) 1 Bro CC 571 ...................................................292 Ontvanger v. Hamm qq, unreported, The Netherlands........................328, 399 Oppenheimer v. Cattermole [1976] AC 249 ..................................................20 O’Rourke v. Darbishire [1820] AC 581 .......................................................271 Paget’s Settlement, Re [1965] 1 WLR 1046 ................................256, 261–3, 268 Paramasivam v. Flynn (1998–99) 160 ALR 203 .................................246–8, 251 Pearce v. Ove Arup Partnership Ltd [2000] Ch 403, CA; reversing [1997] Ch 293 ...................................................................194, 263 Pearse v. Green (1819) 1 Jac & W 135.........................................................271 Peillon v. Brooking (1858) 25 Beav 218 .....................................39, 40, 199, 210 Penn v. Baltimore (1750) 1 Ves Sen 444.........................................................22 Pepin v. Bruyere [1900] 2 Ch 504 ................................................................174 Permanent Trustee Co (Canberra) Ltd v. Permanent Trustee Co of New South Wales Ltd (1969) 14 FLR 246 ............................................310 Perpetual Executors & Trustees Association of Australia Ltd v. Roberts [1970] VR 732 .................................................................................206, 233 Peters Bauunternehmung GmbH v. Zuid Nederlandse Aannemers Vereiniging (Case 34/82) [1983] ECR 987 ................................................116 Petroleo Brasiliero SA v. Mellitus Shipping Inc [2001] 2 Lloyd’s Rep 203..................................................................................................239 Philipson-Stow v. IRC [1961] AC 727..................................34, 37, 49, 173, 233 Phrantzes v. Argenti [1960] 2 QB 19............................................................253 Piercy, Re [1895] 1 Ch 83 ......................................................................37, 173 Piercy v. EFTAS, judgment of 15 March 1956, Tribunal of Oristano ...........316 Pilkington’s Will Trusts, Re [1937] Ch 574 ..........................................233, 275 Pipon v. Pipon (1744) Amb 799 ....................................................................50 Pollak’s Estate, Re [1937] Transvaal Provincial Decision 91 .................207, 284

Table of Cases xxix Pollard v. Ashurst [2001] 2 WLR 722, CA; affirming [2000] 2 All ER 772 ..................................................................171, 172, 358 Pouey v. Hordern [1900] 1 Ch 492 ...................................................18, 59, 369 Powell Duffryn v. Petereit (Case C–214/89) [1992] ECR I–1745 ...................255 Power Curber International Ltd v. National Bank of Kuwait SAK [1981] 3 All ER 607 ..................................................................................30 Preston v. Melville (1841) 8 Cl & Fin 1 .........................................................45 Price, Re [1900] 1 Ch 442 .............................................................................60 Private Trust Corp v. Grupo Torras SA [1997/98] 1 OFLR 443, Bahamian CA ..........................................................................110, 309, 333 Provost of Edinburgh v. Aubery (1754) Amb 256;(1753) Amb 236 ...............293 Pryce, Re [1911] 2 Ch 286.............................................................................60 Public Curator of Queensland v. The Union Trustee Co of Australia Ltd (1922) 32 CLR 66....................................................................................200 Pullan v. Koe [1913] 1 Ch 9 ..........................................................................62 QRS 1 v. Fransden [1999] 1 WLR 2169, CA ................................................399 R v. Williams [1942] AC 541 ........................................................................30 Rabaiotti 1989 Settlement, Re [2000] ITELR 763, Royal Ct of Jersey ...260, 261 Rahman v. Chase Bank (CI) Trust Co Ltd [1991] Jersey Law Reports 113..............................................................73, 110, 333 Raiffeisen Zentralbank Osterreich AG v. Five Star General Trading LLC (2001) 2 WLR 1344, CA; affirming (2000) CLC 1359 .................................32 Ralli Bros v. Cia Naviera Sota y Aznar [1920] 2 KB 287 .........223, 291, 383, 393 Ralli’s Will Trusts, Re [1964] Ch 288............................................................53 Rawluk v. Rawluk (1990) 65 DLR (4th) 161, Canadian SC..........................146 Regazzoni v. KC Sethia (1944) Ltd [1958] AC 301, HL; affirming [1956] 2 QB 490, CA ...........................................................20, 382, 393, 394 Renard, In re Estate of 56 NY 2d 973, 453 NYS 3d 625 .................................55 Renshaw’s Trust Deed, Re [1928] NZLR 460..............................................299 Republica de Guatemala v. Nunez [1927] 1 KB 669.......................................16 Revenue Commissioners v. Pelly [1940] IR 122.................................202–4, 210 Richard v. Mackay, judgment of 4 March 1987, Ch D, (1997) 11 Trust Law International 22 ................................236, 297, 299, 308 Roberts, Re [1983] 70 FLR 158 ...................................................................299 Robertson v. Howden (1892) 10 NZLR 609 ................................................252 Robinson, Re [1931] 2 Ch 122 .............................................................223, 292 Robson v. Flight (1865) 4 De GJ & Sm 608.....................................................6 Rochefoucald v. Boustead [1987] 1 Ch 196..................................................273 Rose v. St Louis Union Trust Co 43 Ill 2d 312, 253 NE 2d 417 (1969)...........220 Ross, Re [1930] 1 Ch 377 ......................................................27, 47, 49, 51, 388 Rossano v. Manufacturers’ Life Assurance Co Ltd [1953] 2 QB 352......................................................................................398

xxx Table of Cases Royal Boskalis Westminster NV v. Mountain [1999] QB 674; [1998] 2 WLR 538 ...........................................................................390, 393 Royal Trust Co v. Tucker [1982] 1 SCR 250, Canadian SC .........................107 Saliba v. Falzon [1998] NSWSC 302 ...........94, 155, 200, 201, 210, 244, 361, 406 Saunders v. Vautier (1841) 4 Beav 115 ...................................................22, 270 Schrebsman, Re [1944] Ch 83 .......................................................................43 Seaconsar Far East Ltd v. Bank Markazi Jombouri Islami Iran [1994] 1 AC 438........................................................................................82 Seale’s Marriage Settlement, Re [1961] Ch 574............................................299 Shannon v. Irving Trust Co 275 NY 95, 9 NE 2d 792 (1937)..........169, 198, 282 Sharpe (A Bankrupt), Re [1980] 1 WLR 219 ................................................146 Smith, Re; Lawrence v. Kitson [1916] 2 Ch 206.............................................26 Smyth, Re [1898] 1 Ch 89 ..................................................................34–6, 310 Snook v. London & West Riding Investments Ltd [1967] 1 All ER 518.................................................................................110 Sottomayor v. De Barros (No 1) (1877) 3 PD 1..............................................12 Spiliada Maritime Corp v. Cansulex [1987] AC 460 ......................................82 Standard Chartered Bank Ltd v. IRC [1978] 3 All ER 644 .............................30 State of Norway’s Application, Re (Nos 1 and 2) [1990] 1 AC 723...............398 Stevens v. Head (1993) 176 CLR 433...........................................................254 Strong v. Bird (1874) LR 18 Eq 315 ...............................................................42 Swiss Federal Court, judgment of 3 September 1999 (2000) I Semaine Judiciare, p. 269 .....................................................................................333 Tallack v. Tallack [1927] P 211 ...........................................................257, 267 Tatry, The (Case C–406/92) [1994] ECR I–5439..........................................412 Tempest v. Lord Camoys (1882) 21 Ch D 571 .............................................271 Thahir v. Pertamina [1993] 1 SLR 735 ........................................................324 Thompson v. Thompson (1844) 1 Coll 381 .................................................292 Treasurer of Ontario v. Blonde [1948] AC 24 ...............................................30 Trendtex v. Crédit Suisse [1980] QB 629, CA ..........................................31, 32 Tribunale di Roma, judgment of 6 December 1989 .....................................316 Trustees Executors and Agency Co Ltd v. Margottini [1960] VR 417 .................................................................................168, 233 Trustees of the C (Jersey) Foundation v. Internationale Nederlanden Bank (Luxembourg) SA (Case no 17370) 22 May 1996, Luxembourg CA..........334 Trustor AB v. Smallbone, judgment of 9 May 2000, CA ........140, 246, 255, 377 Tyndall, Re [1913] SALR 39................................................................284, 309 United Services Fund v. Richardson Greenshields of Canada Ltd (1987) 40 DLR (4th) 94 ...........................................................................344 United States Surgical Corp v. Hospital Products International Pty Ltd [1982] 2 NSWLR 766 ........................................................................................251

Table of Cases xxxi Vagliano, Re [1905] WN 179 ......................................................................294 Van Grutten v. Digby (1862) 31 Beav 561 ..............................................63, 275 Vandervell v. IRC (No 1) [1967] 2 AC 291 ....................................................40 Vanquelin v. Bouard (1863) 15 CBNS 341.....................................................44 Vansant v. Roberts 3 Md 119 (1852) ...........................................................208 Velasco v. Coney [1934] P 143 ......................................................................61 Viditz v. O’Hagan [1900] 2 Ch 87 ................................................12, 13, 18, 64 Vita Food Products Inc v. Unus Shipping Co Ltd [1939] AC 277 ..........................................................................169, 180, 344 Waite’s Settlement, Re [1958] Ch 100 ...........................................................60 Wardlow-Milne (Dec’d), Re (1970) JJ 1539.................................................198 Watson’s Settlement Trusts, Re [1959] 1 WLR 732......................................306 Waziristam, The [1953] 1 WLR 1446 ..........................................................240 Webb v. Webb [1991] 1 WLR 1410 .............................................................177 Webb v. Webb (Case C–294/92) [1994] ECR I–1717 ................22, 171, 177, 221 Webb (Dec’d), In the Estate of (1992) 57 SASR 193 ..............................285, 410 Welch v. Tennent [1891] AC 639 ..................................................................63 Weston’s Settlements, Re [1969] 1 Ch 223 ...................................................299 Whitehead’s Trusts, Re [1971] 1 WLR 833 ...................................236, 308, 309 Whitworth Street Estates (Manchester) Ltd v. James Miller and Partners Ltd [1970] AC 583, HL..............................................................213 Wilks, Re [1935] Ch 645 ..............................45, 46, 234, 236, 241, 289, 373, 384 Williams and Humbert Ltd v. W and H Trade Marks (Jersey) Ltd [1986] AC 368 .................................................................................396, 398 Wilmington Trust Co v. Wilmington Trust Co 26 Del Ch 397, 24 A 2d 309 (1942)...........................................................................304, 309 Wilson v. Moore (1833) 1 My & K 126 .......................................................239 Winkworth v. Christie, Manson & Woods Ltd [1980] Ch 496 .................................................................................28, 29, 375, 388 Zivonestenska Bank v. Freeman [1950] AC 57 ............................................383

Table of Legislation INTERNATIONAL Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters................................82, 85, 89, 116, 135–7, 144, 256, 263, 287, 294, 358, 424 Art. 1 .....................................................................................................116 Art. 1(1) .................................................................................................238 Art. 1(2) .................................................................................................358 Art. 2 ..............................................................................................100, 232 Art. 5(1).....................................................................116, 121, 125, 223, 288 Art. 5(6).................................................................82, 86, 232, 313, 330, 460 Art. 16(1) ...............................................................................................171 Art. 16(1)(a) ...........................................................................................358 Art. 17 ..............................................................................................86, 273 Art. 25....................................................................................................312 Art. 29..................................................................................67, 94, 101, 135 Art. 57....................................................................................................412 Art. 57(1) ...............................................................................................412 Council Regulation (EC) No 44/2001 of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters: OJ 2001 L 12/1 ..............................................67, 82 Art. 5(1)(b) .............................................................................................224 Art. 5(6)............................................................................................86, 313 Art. 7(1) .................................................................................................412 Art. 22(1) ...............................................................................................171 Art. 23 .....................................................................................................86 Art. 33(1) ...............................................................................................312 Art. 36.........................................................................................67, 94, 135 Art. 45(2) ...............................................................................................101 Hague Convention on the Celebration and Recognition of the Validity of Marriages (14 March 1978) .................................................................476 Hague Convention on the Conflict of Laws Relating to the Form of Testamentary Dispositions (5 October 1961)..................................48, 490 Hague Convention on International Access to Justice (25 October 1980) Art. 33....................................................................................................495 Hague Convention on the International Administration of the Estates of Deceased Persons 1973 ...........................................................43, 412, 469

xxxiv Table of Legislation Hague Convention on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (Worldwide Judgements Convention) .......................................................................................412 Hague Convention on the Law Applicable to Agency 1978..........................412 Art. 16....................................................................................................481 Art. 19 ............................................................................................409, 492 Art. 22....................................................................................................492 Hague Convention on the Law Applicable to Contracts for the International Sale of Goods 1955 ...................................182, 485, 492, 493 Art. 1(4) .................................................................................................183 Art. 2 .....................................................................................................191 Hague Convention on the Law Applicable to Contracts for the International Sale of Goods 1986 .............................................................................191 Art. 1(b) .................................................................................................183 Art. 7(1) ..........................................................................................191, 464 Hague Convention on the Law Applicable to Maintenance Obligations (2 October 1973) Art. 16....................................................................................................491 Art. 21....................................................................................................494 Hague Convention on the Law Applicable to Matrimonial Property Regimes 1978 ................................................................................62, 412 Hague Convention on the Law Applicable to Succession to the Estates of Deceased Persons 1989 ...........................................56, 57, 59, 87, 412, 413 Art. 1(2)(d) .........................................................................................57, 58 Art. 3(1) ...................................................................................................56 Art. 3(2), (3) .............................................................................................57 Art. 5(1) ...................................................................................................56 Art. 6 .......................................................................................................56 Art. 7 .......................................................................................................58 Art. 7(2) ...................................................................................................58 Art. 7(2)(d)...............................................................................................57 Art. 18 .....................................................................................................58 Hague Convention on the Law Applicable to Trusts and on their Recognition ................................iii–v, 3, 5, 24, 38, 39, 40, 46, 47, 54, 56, 66, 70, 72, 74, 76–8, 81, 83–97, 431–3 Preamble ........................................99, 104, 112, 115, 318, 341, 352, 435, 454 Art. 1 .................................................100–2, 239, 435, 444, 452, 454, 456–60 Art. 2 .....................................6, 8, 86, 92, 93, 99, 103–22, 129, 139, 151, 159, 184, 186, 197, 232, 255, 278, 287, 311, 313, 314, 318, 341, 343, 352, 358, 395, 406, 412, 420, 422, 423, 435, 443, 444, 455, 458, 461, 462, 467 Art. 2(1)............................................................................105, 458, 459, 461 Art. 2(2) ..........................................................................................458, 459 Art. 2(2)(a)........................................................................106, 107, 114, 459

Table of Legislation xxxv Art. 2(2)(b) .......................................................................106, 108, 459, 476 Art. 2(2)(c) ......................................................................................109, 459 Art. 2(3) ..........................................................................................110, 459 Art. 3 ................................................70, 96, 122, 123–50, 163, 165, 213, 214, 229, 232, 248, 254, 272, 273, 279, 311, 326, 327, 354, 402, 403, 406, 411, 422, 436, 443, 444, 454, 458–60, 463, 481 Art. 4 .........................................................5–7, 33, 46, 54, 55, 106, 121, 124, 151–7, 180, 200, 244, 272, 318, 320, 356, 360, 362, 366, 391, 422, 436, 444, 460–4, 467, 468 Art. 4(2) .................................................................................................459 Art. 5.............................................113, 158–65, 166, 192, 209, 313, 328, 344, 345, 351, 352, 432, 444, 462, 465, 488 Ch II (Arts 6–10) ..................................158, 311, 437, 444, 473, 474, 476, 490 Art. 6 ..............................................13, 15, 113, 124, 143, 155, 160, 166–214, 221–3, 225, 227–31, 233, 249, 259, 275, 277, 278, 281, 289, 302, 303, 307, 312, 313, 319, 344–7, 356, 361, 365, 373, 374, 389, 409, 410, 436, 437, 445, 463, 465, 467, 469, 470, 472, 473, 490 Art. 6(1) .............................................................166, 196, 463, 464, 465, 476 Art. 6(2)....................................................................158, 162, 166, 209, 211, 215, 290, 462, 463, 465, 471 Art. 6(3) ..........................................................................................469, 470 Art. 7.............................................13, 15, 26, 76, 158, 160–63, 166, 176, 178, 179, 184–6, 190–2, 201, 205, 206, 210, 211, 213, 215–32, 233, 252, 259, 276, 278, 280, 281, 287, 289, 298, 312, 313, 347–9, 388, 409, 426, 436, 437, 445, 462, 463, 465–7, 469–70, 490, 491 Art. 7(1) ...................................................................................462, 465, 471 Art. 7(2) ...................................................................................465, 467, 491 Art. 7(2)(a) ................................................................219, 220, 222, 465, 476 Art. 7(2)(b)......................................................................................219, 465 Art. 7(2)(c).................................................................219, 220, 222, 465, 476 Art. 7(2)(d) .......................................................................219, 225, 465, 466 Art. 7(3) ..........................................................................................469, 470 Art. 8.............................................124, 153, 159, 164, 189, 190, 229, 232–80, 309, 314, 437, 445, 456, 460, 461, 467, 469, 472 Art. 8(1).....................................................................190, 233, 276, 279, 280 Art. 8(2) .................................................................234, 235, 280, 467–9, 472 Art. 8(2)(a) .............................................6, 21, 46, 54, 69, 152, 237, 238, 243, 268, 285, 286, 308, 364, 373, 384, 462, 468 Art. 8(2)(b) .............................................................................................239 Art. 8(2)(c) ......................................................................................245, 468 Art. 8(2)(d) ............................................46, 241, 258, 268, 271, 286, 373, 384 Art. 8(2)(f).......................................................................................245, 468

xxxvi Table of Legislation Art. 8(2)(g)............................................................24, 96, 105, 132, 240, 245, 248, 253, 254, 272, 325, 468 Art. 8(2)(h) ....................................................64, 244, 258, 259–61, 263, 265, 266, 270, 286, 299, 309, 361, 362 Art. 9 .........................................................159, 185, 216, 223, 234, 235, 238, 241, 246, 270, 281–96, 307, 308, 343, 348, 389, 391, 395, 396, 437, 445, 457, 463, 465, 467–9, 471–473 Art. 10 .........................................................68, 220, 224, 228, 252, 257, 280, 297–310, 405, 437, 445, 467–9, 471, 472, 474, 490 Ch III (Arts 11–14) ..........................................93, 404, 454, 456, 489–91, 495 Art. 11 ...........................................78, 100, 101, 103–105, 107, 108, 136, 159, 234, 294, 311–36, 337, 338, 341, 343, 344, 363, 373, 376, 379, 404, 424, 437, 446, 473, 476, 482 Art. 11(1) .................................................................................456, 473, 490 Art. 11(2) ....................................................312, 313, 317, 319, 459, 473, 474 Art. 11(3) ..........................................................................312, 314, 319, 474 Art. 11(3)(a), (b).......................................................................313, 314, 474 Art. 11(3)(c) ....................................................................................314, 474 Art. 11(3)(d) ......................................................................40, 107, 110, 129, 141, 234, 254, 294, 321–4, 326–9, 360, 373, 375–8, 406, 422, 426, 473–5, 481 Art. 12..................................................................337–40, 404, 438, 446, 476 Art. 13.............................................89, 92, 101, 113, 115, 159, 179, 181–184, 192, 328, 329, 336, 341–53, 354, 370, 381, 382, 404, 414, 422, 424, 438, 463, 473, 476, 477, 480, 481, 484, 485 Art. 14 ...............93, 182, 215, 312, 344, 354, 379, 404, 438, 446, 477, 478, 480 Ch IV (Arts 15–25) .................................................................................454 Art. 15....................................................46, 54, 70, 74, 77, 97, 153, 181, 190, 244, 256, 259, 260, 264–70, 282, 291, 308, 315, 317, 318, 322, 328, 329, 336, 338, 339, 343, 351, 355–79, 380, 381, 384–6, 389, 391, 392, 406, 413, 421, 423, 425, 438, 443, 446, 454, 456, 461, 473, 475, 476, 480–2, 484, 485 Art. 15(1) ..........................................................................461, 480, 481, 483 Art. 15(1)(a)......................................................................363, 364, 482, 485 Art. 15(1)(b)...........................................................264–6, 364, 365, 482, 485 Art. 15(1)(c)...................................................55, 66, 74–6, 83, 155, 350, 366, 371, 372, 374, 392, 412, 482, 485 Art. 15(1)(d).................................................58, 264, 291, 317, 318, 320, 321, 323, 325, 360, 373–6, 378, 422, 482, 485 Art. 15(1)(e)...........................318, 320, 321, 373, 374, 376, 377, 482, 483, 485 Art. 15(1)(f) ...........................321, 323, 373, 374, 377, 378, 422, 482, 483, 485 Art. 15(1)(g) ...........................................................................................485 Art. 15(2)...................................................................378, 379, 480, 481, 483

Table of Legislation xxxvii Art. 16..................................46, 74, 77, 97, 181, 190, 237, 239, 242, 256, 259, 264–70, 274, 282, 308, 315, 336, 338, 339, 354, 357, 362, 363, 380–6, 391, 413, 414, 421, 423, 439, 440, 443, 447, 454, 473, 476, 481, 483, 484, 493 Art. 16(1)............................................................382, 385, 386, 393, 484, 486 Art. 16(2) ...........................181, 291, 381–3, 385, 393, 406, 415, 484, 486, 487 Art. 16(3) ..........................................................................291, 382, 385, 484 Art. 17 .........................................................27, 186, 190, 242, 269, 289, 307, 387–9, 411, 439, 443, 447, 454, 463, 487 Art. 18 ................................................20, 34, 74, 77, 190, 237, 242, 245, 259, 266, 268, 269, 279, 295, 308, 336, 343, 353, 354, 356, 368, 373, 376, 381, 390–6, 421, 423, 439, 447, 454, 488 Art. 19 .................................................397–401, 414, 439, 454, 473, 475, 488 Art. 20........................................................129, 139, 145, 147, 149, 311, 354, 402–3, 420, 439, 442, 454, 460, 488, 489, 495 Art. 20(1) ...............................................................................................139 Art. 21.............................................................93, 342, 404–5, 407, 408, 414, 416, 440, 454, 458, 489–91, 493, 495 Art. 22............................94, 97, 225, 406–8, 414, 440, 443, 447, 454, 491, 493 Art. 22(1) ..........................................................................406, 407, 414, 490 Art. 22(2) .................................................................................407, 408, 490 Art. 23.........................................................184, 307, 409, 440, 454, 491, 492 Art. 23(1) ...............................................................................................413 Art. 24.........................................................................410–11, 440, 454, 492 Art. 25..................................................................412–13, 414, 440, 454, 492 Ch V (Arts 26–32) ...........................................................................454, 492 Art. 26 ...................................383, 414, 417, 420, 440, 442, 487, 489, 490, 493 Art. 26(2) ........................................................................................414, 494 Art. 27 .................................................414, 415, 418, 420, 440, 441, 442, 494 Art. 28 .................................................414, 415, 416, 418, 420, 441, 442, 494 Art. 28(3) ........................................................................................494, 495 Art. 29 .................................................414, 417, 418, 420, 440, 441, 442, 495 Art. 29(3) ...............................................................................................496 Art. 30................................................................414, 418, 420, 441, 442, 495 Art. 30(1) ........................................................................................416, 441 Art. 30(2)(b) ....................................................................................418, 495 Art. 31 ..........................................402, 414, 419, 420, 439, 442, 492, 493, 496 Art. 32 .......................................................................414, 416, 420, 442, 496 Lugano Convention ..............................................................................85, 116 Art. 5(6) ...................................................................................................86 Mexico Convention (Inter-American Convention of the Law Applicable to International Contracts 1994) Art. 9(2) .................................................................................................185 Art. 10....................................................................................................185

xxxviii Table of Legislation OECD Model Convention on the taxation of international trusts 1977 .....................................................................397 Rome Convention on the Law Applicable to Contractual Obligations ..........................................16, 120–2, 139, 182, 185–7, 191–4, 196, 213, 228, 275, 464, 483, 485, 486 Art. 1 .....................................................................................................117 Art. 1(1) .................................................................................................122 Art. 1(2)(a)........................................................................................11, 238 Art. 1(2)(b)...............................................................................................63 Art. 1(2)(e) ...............................................................................................19 Art. 1(g).....................................................................................31, 121, 332 Art. 3(1) ...................................................................................191, 281, 282 Art. 3(2) ..........................................................................................298, 301 Art. 3(3).....................................................................................179–81, 345 Art. 3(4) ...................................................................................188, 212, 302 Art. 4 .......................................................................................................15 Art. 4(2) ...................................................................................216, 218, 332 Art. 5(2) .................................................................................................481 Art. 6 .......................................................................................................10 Art. 6(1) .................................................................................................481 Art. 7..................................................................................10, 481, 484, 485 Art. 7(1) ..........................................................................................381, 383 Art. 7(2) ...................................................................................................10 Art. 8 ................................................................................................10, 188 Art. 8(1) ...................................................................................160, 188, 224 Art. 8(2) .................................................................................................279 Art. 9 .......................................................................................188, 273, 275 Art. 9(1) ...................................................................................................32 Art. 9(2) ......................................................................................26, 32, 275 Art. 9(6) ...................................................................................................26 Art. 10 .....................................................................................................10 Art. 10(1)(c) .............................................................................121, 253, 254 Art. 10(1)(e) ....................................................................................122, 126 Art. 11.........................................................................................11, 12, 188 Art. 12(1) .................................................................................................32 Art. 12(2) .................................................................................31, 32, 40, 41 Art. 15 ............................................................................................190, 387 Art. 16 .....................................................................................................10 Art. 21....................................................................................................412 Art. 22....................................................................................................486 Art. 22(1)(a) ....................................................................................383, 485 Art. 25....................................................................................................121

Table of Legislation xxxix

NATIONAL AND OFFSHORE JURISDICTIONS

Australia Trusts (Hague Convention) Act 1991 ...........................................................88 s. 7(1), (2) ...............................................................................................410

Bahamas Trustee Act 1998..........................................................................................67 Trusts (Choice of Governing Law) Act 1989 ...........................................67, 68 s. 4 .........................................................................................................175 s. 5 .........................................................................................................300 s. 7 .........................................................................................................175 s. 7(2)(a)(ii) ............................................................................................175

Barbados International Trusts Act 1995 s. 15 .........................................................................................................71 ss. 16, 17 ..................................................................................................68

Bermuda Trusts (Special Provisions) Act 1989 .............................................................71 Pt I...........................................................................................................70 ss. 3, 5–7 ..................................................................................................70 s. 9 ...........................................................................................................72 s. 10(1), (2) ...............................................................................................71 s. 11 ..................................................................................................70, 372

British Virgin Islands Trustee (Amendment) Act 1993 s. 81 .......................................................................................................302 s. 83(1) .....................................................................................................72

xl Table of Legislation s. 83(1)(a), (b)...........................................................................................73 s. 86(1) ...................................................................................................110

Canada Alberta International Convention Implementation Act, SA 1990 s. 1(2) .....................................................................................................410 s. 1(4) .....................................................................................................383 s. 1(6) .....................................................................................................139 British Columbia International Trusts Act, RSBC 1996 s. 3(1) .....................................................................................................139 Manitoba International Trusts Act, SM 1993 s. 4(1) .....................................................................................................139 New Brunswick International Trusts Act, SNB 1988 s. 2 .........................................................................................................410 s. 6(1) .....................................................................................................139 Newfoundland International Trusts Act, RSN 1989 s. 5(1) .....................................................................................................139 Prince Edward Island International Trusts Act, RSPEI 1988 s. 3(1) .....................................................................................................139 Saskatchewan Trusts Convention Implementation Act, SS 1994 s. 5(1) .....................................................................................................139

Table of Legislation xli Cayman Islands Special Trusts (Alternative Regime) Law 1997.......................295, 391, 395, 396 Trusts (Foreign Element) Law 1987 ...................................................66, 75, 76 s. 4(1) .....................................................................................................175 s. 4(4) .....................................................................................................300 s. 5 .........................................................................................................175 s. 5(e) .....................................................................................................175 Trusts Law 1967 (1998 Revision)..................................................................69 Pt VII................................................................................................66, 175 s. 89 ..................................................................................................66, 175 s. 89(3) .....................................................................................................68 s. 89(4).......................................................................................68, 300, 302 s. 90 ..............................................................................................67–9, 175 s. 91.............................................................................................67, 73, 372

Cook Islands International Trusts Act 1984 s. 13(C) ..................................................................................................110 International Trusts Amendment Act 1989 ss. 6, 11 ..................................................................................................110 International Trusts Amendment Act 1995–6 s. 12 .......................................................................................................110

Germany Bankruptcy Act 1877 Art. 43....................................................................................................377 Bundesgesetz über das Internationale Privatrecht Art. 150....................................................................................................85 EGBGB Art. 25....................................................................................................365

Guernsey Trusts (Guernsey) (Amendment) Law 1990 s. 11A.......................................................................................................55 s. 11(A)(1) ................................................................................................73 s. 11(A)(1)(b) ......................................................................................71, 72

xlii Table of Legislation Trusts (Guernsey) Law 1989 s. 1 .........................................................................................................113

Isle of Man Judgments (Reciprocal Enforcement) (Isle of Man) Act 1968 ........................75 Matrimonial Proceedings Act 1986 s. 17 .......................................................................................................269 Recognition of Trusts Act 1988.............................................................76, 372 s. 5 .........................................................................................................371 Trusts Act 1995 ......................................................................................75, 76 s. 1(1) .................................................................................................76, 77 s. 1(1)(b)(i), (ii).........................................................................................77 s. 3 ...........................................................................................................75 s. 4......................................................................................................68, 75 s. 4(1).......................................................................................................76 s. 4(1)(b) ..................................................................................................76 s. 4(2)(a), (d) ............................................................................................76 s. 4(3).......................................................................................................76 s. 5 ...........................................................................................................76 s. 5(b)(i) ...................................................................................................75 s. 5(b)(ii) .......................................................................................67, 75, 76

Italy Civil Code ..........................................................................................315, 316 Art. 602 ..................................................................................................370 Art. 1362 ................................................................................................214 Art. 2643 ................................................................................................339

Japan Horei (Law on Private International Law) Art. 26....................................................................................................365 Art. 28(3) ...............................................................................................365 Law no 62, 21 April 1922............................................................................111 Trust Act 1922 ...........................................................................................108 Art. 16(2) ...............................................................................................377 Art. 31....................................................................................................375

Table of Legislation xliii Jersey Age of Majority (Jersey) Law 1999 ...............................................................73 Trusts (Amendment) (Jersey) Law 1989........................................................72 Art. 8(A) ..................................................................................................72 Art. 8(A)(2).........................................................................................73, 74 Art. 8(A)(2)(a) ..........................................................................................72 Art. 8(A)(2)(b)..........................................................................................73 Trusts (Amendment No 2) (Jersey) Law 1991..........................................72, 89 Trusts (Jersey) Law 1984 ........................................................................72, 89 Pt II..........................................................................................................72 Art. 1 .......................................................................................................72 Art. 6 .......................................................................................................72 Art. 20(1) ...............................................................................................104 Art. 45 .....................................................................................................72 Art. 45(2) .................................................................................170, 372, 374 Art. 45(2)(a), (b) .......................................................................................74 Art. 45(3) ........................................................................................170, 372 Wills and Succession (Jersey) Law 1993 ........................................................73

Liechtenstein Personen- und Gesellschaftsrecht Art. 897 ..................................................................................................111

Netherlands Civil Code Book 3, Art. 84(3)..........................................................319, 320, 328, 374 Wet Conflichtenrecht Trusts 1995 .......................................................319, 320 Art. 3 .....................................................................................................338 Art. 4 .......................................................................................320, 327, 374

Peru Ley general de Bancos Art. 314 ..................................................................................................111

xliv Table of Legislation Switzerland Loi Federale sur le Droit International Privé (Private International Law Statute) (1 January 1989) ....................................................................332 Art. 91(1) .................................................................................................51 Art. 150...........................................................................................332, 333 Art. 154(1)..............................................................................................333

United Kingdom Administration of Estates Act 1925 ................................................46, 241, 384 s. 33 .....................................................................................45, 46, 133, 141 s. 33(1) ............................................................................................241, 384 s. 36(4) .....................................................................................................54 s. 42 .......................................................................................................236 Administration of Estates Act 1971 s. 1 ...........................................................................................................45 s. 2(1).......................................................................................................45 s. 3(1).......................................................................................................45 Administration of Justice Act 1932 s. 2(1).......................................................................................................44 Arbitration Act 1996 s. 7 ..................................................................................................276, 277 s. 46(1)(b) ...............................................................................................185 Civil Jurisdiction and Judgments Act 1982 ...................................................82 s. 3(3).......................................................................................................89 s.45(3) ....................................................................................................232 Civil Liability (Contribution) Act 1978 s. 1(1) .....................................................................................................239 Colonial Act 1892 ........................................................................................45 Companies Act 1985 s. 360 ..................................................................................................337–9 Contracts (Applicable Law) Act 1990 ....................................................11, 120 s. 2(2).......................................................................................122, 126, 383 Contracts (Rights of Third Parties) Act 1999 s. 1 ...........................................................................................................43 Domicile and Matrimonial Proceedings Act 1973 s. 1(1).......................................................................................................62 Family Law Act 1996 s. 15 .......................................................................................................257 Sched. 2..................................................................................................257 Foreign Judgments (Reciprocal Enforcement) Act 1933.................................67 s. 4 ...........................................................................................................75

Table of Legislation xlv Foreign Limitation Periods Act 1984 s. 1(1) .....................................................................................................245 s. 2(1) .....................................................................................................245 Inheritance (Provision for Family and Dependants) Act 1975 ..................................................................................57, 62, 367 s. 10 ................................................................................................368, 369 ss. 12, 13.................................................................................................368 s. 25(1) .....................................................................................................60 Insolvency Act 1986 ............................................................................318, 374 Law of Property Act 1925 s. 1(6).......................................................................................................21 s. 2 ..................................................................................................338, 339 s. 20 .........................................................................................................21 s. 27 ................................................................................................338, 339 ss. 34, 36.................................................................................................141 s. 53(1)(b) ..................................................................................41, 273, 274 s. 53(1)(c)............................................................................................40, 41 s. 164..................................................................................33, 244, 361, 391 s. 164(1) .................................................................................................154 ss. 165, 166 .............................................................................................361 Matrimonial and Family Proceedings Act 1984 Pt III.......................................................................................................269 s. 17 ..................................................................................................64, 257 Matrimonial Causes Act 1859.....................................................................262 Matrimonial Causes Act 1973 .............................................64, 256–9, 264, 267 s. 24.........................................................................259, 260, 264–6, 268–70 s. 24(1) .....................................................................................................64 s. 24(1)(b) ...............................................................................................133 s. 24(1)(c), (d) .........................................................................................257 s. 25 .......................................................................................................258 Mortmain Acts ..........................................................................................292 Perpetuities and Accumulations Act 1964................................33, 154, 245, 361 Private International Law (Miscellaneous Provisions) Act 1995 Pt III ...............................................................................................296, 387 s. 9(5) .....................................................................................................387 s. 11 .......................................................................................................296 Recognition of Trusts Act 1987 .............................................69, 89, 95–7, 127, 139, 142–4, 146, 147, 149, 150, 177, 179, 241, 343, 383, 392, 397, 404, 408, 411, 413, 414 s. 1 .........................................................................................................443 s. 1(1)................................................................................................89, 117 s. 1(2)................................70, 96, 122, 125, 137, 139, 141–5, 147–9, 165, 197, 210, 214, 248, 273, 279, 311, 326, 354, 402, 403, 411 s. 1(3)..............................................................................................266, 362

xlvi Table of Legislation s. 1(4) .....................................................................................................411 s. 1(5) ........................................................................................94, 407, 414 s. 2....................................................................................................89, 443 s. 3 ......................................................................................................., 444 Sched.........................................................................89, 99, 117, 342, 444–7 Supreme Court Act 1981 s. 25(1) .....................................................................................................44 Trustee Act 1906 ........................................................................................284 Trustee Act 1925 ss. 23, 30.................................................................................................208 ss. 31, 32.............................................................................46, 236, 241, 267 s. 33(1) ...................................................................................................376 s. 40 .........................................................................................................54 Trustee Act 2000 ...........................................................................95, 208, 243 ss. 1, 3–5 ..................................................................................................95 s. 6(1)(b) ..................................................................................................95 s. 7(1).......................................................................................................95 Sched 1, para. 1(a) ....................................................................................95 Trustee Investments Act 1961.......................................................208, 243, 244 s. 2(1) .....................................................................................................244 Trusts of Land and Appointment of Trustees Act 1996 ........................170, 171 s. 1(1).......................................................................................................21 s. 3 .........................................................................................................173 ss. 10, 11.................................................................................................171 ss. 12, 13 .........................................................................................172, 356 s. 14 .......................................................................................................172 s. 14(2)(b) ...............................................................................................172 s. 19 .......................................................................................................171 Sched. 1 ...................................................................................................20 Sched. 2, paras 3–5 .................................................................................141 Unfair Contract Terms Act 1977 s. 27(2)..............................................................................................14, 357 Variation of Trusts Act 1958 .......................125, 256, 257, 262–4, 268, 299, 384 s. 1 .........................................................................................................261 s. 1(1) .................................................................................258, 261, 268–70 s. 1(3) .....................................................................................................261 s. 2 .........................................................................................................262 Wills Act 1837 s. 18 .........................................................................................................49 s. 18(2) .....................................................................................................61 s. 27 .........................................................................................................60 Wills Act 1963 .............................................................................................50 s. 1 ..............................................................................................48, 61, 273 s. 2 ...........................................................................................................61

Table of Legislation xlvii s. 2(1)(b) ..................................................................................................50 s. 2(1)(c) .............................................................................................49, 61 s. 2(1)(d) ..................................................................................................61 s. 4 ...........................................................................................................48 s. 6(1) .................................................................................................48, 51 s. 6(3).......................................................................................................48 Statutory Instruments Civil Procedure Rules 1998, SI 1998/3132 ....................................................263 r. 6.20(11)–(15).........................................................................................82 Non-Contentious Probate Rules 1987, SI 1987/2024 r. 30(1)(a)–(c) ...........................................................................................44 r. 30(3)(a)(i) .............................................................................................44 r. 30(3)(b).................................................................................................45 Recognition of Trusts Act 1987 (Overseas Territories) Order 1989, SI 1989/673........................................................................69 Trustee Investments (Division of Trust Fund) Order 1996, SI 1996/845 ......................................................................244

United States of America Probate Code (Cal) ......................................................................................64 Uniform Disposition of Community Property at Death Act...........................64

1

The Creation of Transnational Trusts: Launching the Rocket 1. INTRODUCTION 1

(A) General The image has been formed in relation to transnational trusts of the rocketlauncher and the rocket.2 The former concerns those preliminary matters necessary for creation of the trust, such as the capacity of the settlor to dispose of his property on trust, the vesting of the property in the trustee on trust and the validity of the formal instrument creating the trust; the latter governs matters affecting the trust once it has come into existence, such as the role of the trustee, the administration of the trust and the relationship between the trustee and the beneficiary. Whilst the latter is largely the province of the Hague Trusts Convention, it will not come into consideration unless the pre-conditions for the formation of a trust are satisfied. As von Overbeck states “the law designated by the Convention applies only to the establishment of the trust itself, and not to the validity of the act by which the transfer of assets is carried out”.3 Each contracting state should apply its pre-existing national conflict of law rules to determine “whether the rocket was successfully launched into space or whether it merely fizzled like a damp squib”.4 This Part considers the preliminary steps to the creation of express transnational trusts.5 It will examine the creation of both inter vivos and testamentary 1 This Part is an updated and very substantially expanded version of my article, “Launching the Rocket: Capacity and the Creation of Inter Vivos Transnational Trusts”, in J Glasson (ed.) International Trust Laws (Bristol, Jordans, loose-leaf), ch. C2. 2 See the Explanatory Report on the Convention on the Law Applicable to Trusts and on their Recognition by Professor M Alfred von Overbeck (hereafter “the von Overbeck Report”), Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition 370, para. 53, p. 381; D Hayton, “The Hague Convention on the Law Applicable to Trusts and on their Recognition” (1987) 36 ICLQ 260, 268; D Hayton, Underhill and Hayton, The Law Relating to Trusts and Trustees, 15th edn. (London, Butterworths, 1995), 946; L Collins (ed.), Dicey and Morris, The Conflict of Laws, 13th edn. (London, Sweet and Maxwell, 2000), 1088–9. 3 Von Overbeck Report, para. 54, p. 381. 4 A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, p. 61. 5 See also J Mowbray, L Tucker, N Le Poidevin and E Simpson, Lewin on Trusts, 17th edn. (London, Sweet and Maxwell, 2000), 277–84.

4 Introduction trusts. The discussion is principally of the English choice of law rules.6 However, it goes on to consider the position in selected offshore states. Those states very frequently have more developed “rocket-launching” rules than exist in England; rules which have pragmatism and the protection of local trusts from claims arising by foreign law very much at the forefront. Inconvenient though it might be that all issues that arise in relation to a transnational trust are not governed by one law, the law applicable to many issues of the rocket-launcher process is relatively settled. There is one notable exception to this, on which there is a dearth of authority and on which the proffering of advice is notoriously difficult: namely the law which governs capacity to create a transnational trust. Accordingly, this part will give particular attention to this issue.

(B) Creation and Transfer Lupoi tells us that the distinction between the rocket-launcher and the rocket “makes me shiver, because it embraces two different legal negotia within the ‘launch pad’: creation and transfer.”7 It does not make this author shiver. One can readily agree with Lupoi’s view that the rocket-launcher idea fuses two distinct elements: the creation of the trust and the transfer on trust. They may not occur, and (outside the context of testamentary trusts), usually will not occur at the same time. But Lupoi goes further, and argues that creation is the preliminary question, not transfer. He points out that a trust remains valid even if the trustee chosen by the settlor does not accept, or dies before accepting, the position. “None of this would be possible if the negotium of creation were not autonomous, and did not prevail over the negotium of transfer.”8 This writer would agree only in part. It is certainly true that creation of a trust is perfected even in the absence of a trustee; but that trust still needs to be constituted by the vesting of legal title in a willing9 trustee before it will come into effect and the powers and duties arising under the trust instrument can be exercised. In other words, until creation and transfer take place, the trust is not fully in effect and it seems to this author quite sensible and helpful to speak of this stage as “preliminary” and “rocket-launching.” What is, however, important is that the creation of the trust and the transfer on trust are both considered when the “rocket-launching” process is examined. That is not to say that it will be undesirable for the same law to be applied to both matters; indeed, it may be thought more coherent for a single law to determine whether a trust has come into operation. It will be seen that this author 6

And relates to what Lupoi would call the “English-model” trust. M Lupoi, Trusts: A Comparative Study, translated by S Dix, (Cambridge, Cambridge University Press, 2001), 346. 8 Ibid., 347. 9 In the case of express trusts. 7

Inter Vivos Trusts 5 favours, for choice of law purposes, a somewhat different distinction between (i) questions going to whether any property right may be transferred; and (ii) questions going to whether the trust structure has been created and equitable title duly vested in the beneficiary.

(C) Inter Vivos Trusts—Matters to be Considered We will look first at inter vivos trusts. A variety of issues will arise in the creation of an inter vivos trust. The most important ones are: (i) does the would-be settlor have capacity to transfer property on trust; and (ii) has legal title passed from the would-be settlor to the would-be trustee and equitable title passed to the would-be beneficiary?

2. TO WHAT EXTENT ARE MATTERS OF CAPACITY AND ESSENTIAL VALIDITY EXCLUDED FROM THE HAGUE TRUSTS CONVENTION ?

The exact extent to which the two matters listed above form part of the rocketlaunching process must now be considered. Article 4, Hague Convention states that: “the Convention shall not apply to preliminary issues relating to the validity of wills or of other acts by virtue of which assets are transferred to the trustee”.

(A) Capacity (i) Whose capacity? The capacity of a settlor validly to transfer assets to the trustee seems to be an act “by virtue of which assets are transferred”. Von Overbeck states that the views of civil law and common law states were largely opposed to each other in the negotiation of the Convention, the former regarding capacity as requiring a distinct choice of law rule, the latter tending to subject capacity to the law governing essential validity. This latter view is highly suspect in relation to the English conflict of laws.10 In any event, it appears that the settlor’s capacity is outside the Hague Convention: “in the end the Commission did not expressly exclude general capacity, although a consensus emerged that this was not to be governed by the Convention, while the specific capacity to become a trustee was to be so governed”.11 10 11

But see M Lupoi, Trusts: A Comparative Study, 242. Von Overbeck Report, para. 59, p. 382.

6 Capacity and Essential Validity Exclusions The latter rule is expressly stated in Article 8(2)(a) of the Convention.12 It is not immediately self-evident that the trustee’s capacity should be a part of the “rocket” itself.13 The Report explains that whilst a question e.g. as to the age required to act as trustee was not obviously a question relating to the person’s fitness to be a trustee, with which Article 8(2)(a) is largely concerned, there may well be even stronger protection available under the applicable law of the trust itself. This is scarcely convincing and does not really seem to address the point. A better answer would be that if a person has “general” capacity to receive property at all, it is more defensible for his “specific” capacity to act as trustee to be determined by the law chosen to govern the trust.14 Moreover, a trust will not fail for want of a trustee,15 so that it is easier to see the capacity of the trustee as part of the “rocket” itself. (ii) Declarations of oneself as trustee More critically, it leaves the situation where the settlor declares himself to be trustee of property, rather than transferring it to another person on trust, in a profoundly complex position. For on one view, the settlor still must have capacity to create the trust; but on another view, there is no transfer of legal title. Accordingly, this might be said to be a question relating simply to his capacity to act as trustee. The von Overbeck Report states that declarations of trust raise certain issues which are excluded from the Convention by Article 4.16 This seems correct in principle, in that the key “rocket-launcher” questions are not really focused on whether legal title has been transferred, but on whether the settlor may alienate any property right at all.17 Clearly, in the case of the declaration of trust, the equitable interest will be alienated. Indeed, were this not correct, a settlor who is forbidden from alienating property to X could evade that restriction simply by declaring a trust for Y. Y could then terminate the trust and claim the property absolutely, if the governing law of the trust so permitted.

12

Discussed in Part Two of this book. But see the discussion of “the capacity of the trustee” below, where it is argued that he must still possess “general” capacity to receive property at all and that this is a “rocket-launching” matter. 14 See the discussion of “the capacity of the trustee” in section 5, below and of Art. 8(2)(a) in Part Two of this book. 15 Robson v. Flight (1865) 4 De GJ & Sm 608. 16 “. . . The acts by which . . . [the] change in the capacity in which the assets were held was effectuated must also be envisaged by Article 4 and therefore excluded from the Convention’s scope”: von Overbeck Report, para. 57, p. 382. 17 However, it would be wholly unsatisfactory if trusts which are duly declared according to the rocket-launching rules did not then fall within the Hague Trusts Convention. This conclusion is suggested by Art. 2 of the Convention, which speaks of assets being placed “under the control of a trustee”. Nevertheless, once the trust is “launched”, it should be wholly immaterial for Convention purposes whether the launch was by transfer to a trustee or by the settlor declaring himself to be a trustee. See further the discussion of Art. 2 in Part Two of this book. 13

Constitution of Trusts 7

(B) Constitution of Trusts It is clear from the wording of Article 4 that the vesting of the property in the trustee on trust is also excluded from the Convention. The Report states that: “Article 4 is intended to exclude from the Convention’s scope of application both the substantive validity and the formal validity of the transfers which are preliminary to the creation of the trust.”18

Again, this must include the passing of equitable title where a settlor simply declares himself trustee, for the reason outlined above.

3. CAPACITY

(A) What is Capacity? (i) Status and capacity A person’s status refers to his connection with a particular country (such as his domicile, nationality or habitual residence) or to his membership of a particular class of persons (e.g. his race, marital status, minority, religion or bankruptcy). The law of the country from which the individual derives his status does not, however, necessarily determine whether he has capacity to enter into a transnational transaction. (ii) Essential validity and capacity (1) General The most critical distinction to be drawn is between essential validity and capacity. Capacity defines a particular class of persons who by a given law cannot enter into a transaction into which other people generally may enter e.g. a provision that no person aged under 16 may enter into a contract to buy cigarettes. Essential validity, on the other hand, concerns provisions which state that no person may validly enter into the transaction in question. Yet this seemingly straightforward distinction can lead to great complexity. Imagine a provision of the law of state A that no person may enter into a given transaction. Prima facie, this looks like an issue of essential validity. Yet if this provision is only intended to apply to persons who are domiciled in state A, then it is arguably correctly to be considered as a rule relating to capacity, limited as it is to a particular class of persons. But it is not unreasonable to assume that many countries have particular restrictions which are intended to apply either 18

Von Overbeck Report, para. 55, p. 382.

8 Capacity and Essential Validity Exclusions only to purely domestic transactions, or, if they also apply to transnational transactions, only to domiciliaries of that state; in which case, the potential scope of capacity rules could be very wide. The matter is complicated by the decision in Bodley Head Ltd v. Flegon,19 where Brightman J held that an alleged incapacity by Russian law to trade abroad was to be treated as a rule concerning essential validity, since it was a law applicable to all Russian citizens. Yet if, in a dispute governed by Russian law, a particular provision is not applicable to non-Russian citizens, it should arguably be treated as a question of capacity. In contrast, if the Russian law were that no person could enter into a given transaction, but Russian domestic law was not designed to apply to transnational matters and the Russian courts themselves would not apply it to them, but would instead apply a different country’s law, the rule is better viewed as a matter of essential validity, since when Russian law does apply its effects are not confined to a particular class of persons. Of course, this distinction is often acutely difficult to make. Were it the case that the same law applied to all questions of capacity and essential validity, the distinction would be relatively unimportant. However, given the uncertainty of the law on capacity, that cannot be taken for granted. Accordingly, it is vital to attempt to distinguish between the two. (2) Characterisation Unfortunately, matters are still more complex. Imagine that, adopting the English classification, a matter of capacity is raised, to which we are minded to apply the law of state A. What if state A would classify this rule, which we would think of as relating to capacity, as one relating to essential validity? How can an English court then decide, when confronted with transnational trusts litigation, into which category an issue falls? This is the process of characterisation. It is a doctrine of considerable complexity, the details of which are beyond the scope of this book.20 Put very simply, however, three obvious methods of characterisation could be used. (α) Characterisation could be done entirely by English domestic law categories.21 So if, for argument’s sake, capacity were to be governed by the law of 19

[1972] 1 WLR 680, 688. See Dicey and Morris, ch. 2; P North and J Fawcett (eds.), Cheshire and North’s Private International Law, 13th edn. (London, Butterworths, 1999), ch. 3. For recent consideration of the doctrine, see Macmillan Inc v. Bishopsgate Trust (No 3) [1996] 1 WLR 387; (on which, see A Briggs, “From Complexity to Anticlimax: Restitution and Choice of Law” (1996) 4 RLR 88; J Stevens, “Restitution or Property? Priority and Title to Shares in the Conflict of Laws” (1996) 59 MLR 741; R Stevens, “The Law Applicable to Priority in Shares” (1996) 112 LQR 198; J Bird, “Choice of Law Rule of Priority Disputes in Relation to Shares” [1996] LMCLQ 57; W Swadling, “A Claim in Restitution?” [1996] LMCLQ 63; C Forsyth, “Characterisation Revisited: an Essay in the Theory and Practice of the English Conflict of Laws” (1998) 114 LQR 141). See also the discussion of Art. 2, Hague Convention in Part Two of this book. 21 See Ogden v. Ogden [1908] P 46. 20

What is Capacity? 9 the settlor’s domicile (lex domicilii), one would simply ask if any rules existed in that legal system which restricted the class of persons who may enter into such a transaction, ignoring how state A would classify that rule. This is the approach most favoured in English law, but has a distinct drawback: for if state A would not regard this as a matter of capacity, it could lead to the application of state A’s laws in a situation where even the courts of state A would not apply that law if the dispute were litigated in their courts. Accordingly, it can lead to distortion of the foreign law. (β) One could characterise according to the categories of the governing law.22 In other words, one would ask whether according to the lex domicilii a matter was regarded as an issue of capacity, and if so apply, that law. But this is circular reasoning. For one would only consider the lex domicilii at all if one had already taken the view that the issue was one of capacity to be referred to that law; yet whether it is a capacity issue is precisely the point in question. Moreover, if the lex domicilii does not think that e.g. its law prohibiting under 16s from creating a trust is a capacity issue, the result could be that foreign restrictions are not applied in situations where an English court would feel it appropriate to apply them. (χ) An English court could characterise prima facie according to English law, but, in deciding whether finally to characterise a matter as relating to capacity or essential validity, it would have regard to how the matter is treated in the state whose law might stand to be applied. Whilst this has the attraction of avoiding logistical difficulties and at the same time introducing an element of discretion to produce what might be considered a “fair” characterisation, it could be so uncertain as to prove unworkable. For this reason, characterisation is still essentially done by English categories alone,23 and the fact that state A does not consider a particular rule to be one of capacity will largely be irrelevant if it is so treated by English law. (3) A meaningful distinction? Sykes and Pryles argue that the distinction between capacity and essential validity is not meaningful in any event, since, they say, capacity is but an aspect of essential validity.24 They argue that “it is difficult to see why capacity should be different from say essential validity”.25 It may indeed be that, ultimately, the same choice of law rule will be applied to both. Yet the distinction is still highly important. An example will illustrate the point. 22

See Re Maldonado’s Estate [1954] P 223; Re Cohn [1945] Ch 5. A notable exception is the classification of property for private international law purposes as movable or immovable (rather than as real or personal property). The classification is determined by the law of the place where the property is situated, rather than by the law of the forum: Freke v. Carbery (1873) LR 16 Eq 461; re Hoyles [1911] 1 Ch 179; re Berchtold [1923] 1 Ch 192. 24 E Sykes and M Pryles, Australian Private International Law, 3rd edn. (Sydney, Law Book Co, 1991), 614 and 715. 25 Ibid., 614. 23

10 Choice of Law Rule for Capacity Questions The rule that a contract governed by English law must be for consideration is a law which is considered appropriate for all contracts governed by English law; however, it is not to be used when the contract is governed by a different law.26 It is a rule of essential validity. By contrast, a rule that no person under 16 may buy cigarettes has a more mandatory character, and it is reasonable to assume that the policy of that legislator requires that we at least treat with some reservation the view that an English domiciliary should be able to evade that rule by choosing another law to govern the contract, including his capacity to make that contract. Accordingly, it is suggested that the Sykes and Pryles view cannot be supported. It is arguable that the freedom given to a party to choose the law governing a contract or trust cannot be extended to his capacity to enter into that transaction. As to capacity itself, Matthews and Sowden distinguish between two aspects thereof: (i) qualities relating to the individual settlor which prevent him from passing property (e.g. a minor); and (ii) whether the particular transaction can be carried out by the settlor e.g. because he may not alienate a particular piece of property.27 It is true that if the alleged reason for a restriction of the second type is that the settlor has e.g. entered into a marriage settlement, or there is community of property by the law of his domicile, this question must be answered by the relevant choice of law rules for these areas,28 since without the alleged restriction on property disposition, the settlor would be competent to alienate the property. But this is no different from e.g. the obvious point that a testamentary trust cannot be created unless there is first a valid will according to the law governing essential validity of the will.29 Generally, however, the two categories of capacity do not require separate treatment. An alleged restriction on distribution of property imposed upon a particular person of full age and sound mind of which he is sole owner is no less a question of capacity than an alleged restriction imposed on minors and should be governed by the same law.

4. THE CHOICE OF LAW RULE FOR CAPACITY QUESTIONS

At first sight, it might seem that, in terms of practical convenience, it would be desirable if the capacity of the settlor could be governed by the applicable law of the trust, as determined by the Hague Trusts Convention.30 Clearly, this would offer the most certain and easily applicable choice of law rule. However, a glance at other relevant areas of the conflict of laws where capacity is in question reveals a reluctance to adopt this approach. 26 See Arts. 8 and 10, Rome Convention; note, however, the limited exceptions to be found in Arts. 7(2) and 16. 27 P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn. (London, Key Haven Publications, 1993), 53. 28 On which see below; see further Dicey and Morris, ch. 28, Cheshire and North, ch. 34. 29 On which see below; see further Dicey and Morris, ch. 27, Cheshire and North, ch. 33. 30 See, especially, Arts. 6 and 7 (considered in Part Two of this book).

Capacity and Conflict of Laws 11

(A) Capacity and other Branches of the Conflict of Laws A detailed examination of the treatment of capacity throughout the conflict of laws would be inappropriate here. So, whilst in matters relating e.g. to wills of movable property or to marriage settlements, which are closely bound up with personal status, the lex domicilii has a strong claim to govern capacity questions, the principles applied in those areas are less obviously of use in relation to inter vivos trusts. Accordingly, this section is confined to an examination of the law of capacity in areas where commercial considerations might be thought broadly analogous to those of the law of trusts, from which there may be important lessons to be drawn. As Sykes and Pryles note: “The topic of trust partakes of the character of contract . . . and of assignment.”31 For this reason, it is upon contracts and property transfers that we shall focus. (i) Contracts It is worth reflecting at some length upon the approach taken to capacity when a transnational contract is purported to be made, since if freedom of choice is not permitted in this area, it is unlikely to be permitted in trusts law either. Anton comments that “capacity to create an inter vivos trust of moveables is thought to depend upon principles similar to the common law rules regulating general capacity to contract”.32 Article 1(2)(a), Rome Convention33 excludes from its ambit “questions involving the status or legal capacity of natural persons, without prejudice to Article 11.” Article 11, Rome Convention provides that: “in a contract concluded between persons who are in the same country, a natural person who would have capacity under the law of that country may invoke his incapacity resulting from another law only if the other party to the contract was aware of this incapacity at the time of the conclusion of the contract or was not aware thereof as a result of negligence.”

This accords a limited role to the law of the place of conclusion of the contract (lex loci contractus), designed to protect a person who acts reasonably, believing that he is contracting with a person of full capacity. But it certainly does not support the general application of the law of the place of conclusion of a contract to questions of capacity, which may, particularly where the parties are in different countries, be wholly arbitrary. Article 11 is a negative provision. It limits the application of the law which otherwise governs a natural person’s capacity; but 31

E Sykes and M Pryles, Australian Private International Law, 3rd edn., 711. A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn. (Edinburgh, W Green, 1990), 636. 33 Enacted by the Contracts (Applicable Law) Act 1990. 32

12 Choice of Law Rule for Capacity Questions it does not state what the choice of law rule for capacity is.34 In England, this general capacity rule will be determined by the common law. Unfortunately, English common law is largely short of authority on this question.35 The Rule in Dicey and Morris, Rule 179(1) reads thus:36 “The capacity of an individual to enter into a contract is governed by the law of the country with which the contract is most closely connected or by the law of his domicile or residence: (a) If he has capacity to contract by the law with which the contract is most closely connected, the contract will (semble) be valid so far as capacity is concerned. (b) If he has capacity to contract by the law of his domicile and residence, the contract will (semble) be valid so far as capacity is concerned”.

Given that questions of status are intrinsically personal37 it is tempting to argue that capacity should be determined by the law of the domicile of the individual alleged to lack capacity. Dicey and Morris point out that: “An individual’s capacity to contract can be looked at as an emanation of his status and therefore governed by the law of his domicile. Alternatively, it can be considered as a factor determining the validity of the contract and therefore determined by its governing law”.38

They do not go on to commit themselves as to which characterisation is correct and their rule contains aspects of both classifications. However, they rightly point out that in the modern commercial world it would be impractical for all questions of capacity to be governed by the law of the domicile. If X, domiciled in Brazil, enters into a contract with an English domiciliary which is governed by English law, it would be very inconvenient if X could escape liability for an alleged breach of contract on the grounds that he lacked capacity by the law of his domicile. It is true that there is some judicial authority for the application of the law of the domicile;39 but this is largely in the context of cases concerned with matrimonial capacity or marriage settlements. One case relied upon by Lewin on Trusts is Viditz v. O’Hagan.40 In that case, an 18 year old Irishwoman purported 34 Furthermore, Art. 11, Rome Convention limits the role of the law governing capacity insofar as that law states that capacity is lacking. Art. 11 is a provision which mitigates in favour of the validity of contracts. It cannot be used by a person to avoid a contract, where that person does have capacity according to the law governing that issue. Moreover, Art. 11 may only be used by a party with respect to his own capacity; it cannot be used by one contracting party to demonstrate the capacity or incapacity of the other contracting party. 35 See H Fischer, “The Law Governing Capacity with Regard to Bills of Exchange” (1951) 14 MLR 144; J Smith, “Capacity in the Conflict of Laws—a Comparative Study” (1952) 1 ICLQ 446. 36 Dicey and Morris, 1271–2. 37 Although it has been stated above that capacity and status are not conceptually the same, there is an obvious link between the two. 38 Dicey and Morris, 1272. 39 See Sottomayor v. De Barros (No 1) (1877) 3 PD 1, 5; Re Cooke’s Trusts (1877) 56 LJ Ch 637, 639; Cooper v. Cooper (1888) 14 App Cas 88, 99–100. 40 [1900] 2 Ch 87; Lewin on Trusts, 17th edn., 303. See also Re Bankes [1902] 2 Ch 333.

Capacity and Conflict of Laws 13 to make a marriage settlement upon her marriage to an Austrian man. Irish law provided that the settlement was voidable unless ratified when she obtained the age of 21; however, if she did nothing within a reasonable period after reaching 21, she would be deemed to have ratified the agreement. Upon her marriage she had acquired an Austrian domicile and Austrian law stated that a wife could not ratify the settlement. Twenty-nine years later the husband purported to revoke the settlement. Lindley MR ruled that the agreement was invalid, since by the laws of the woman’s successive domiciles, she lacked capacity to make the settlement.41 He essentially applied Irish law to determine whether the agreement was initially binding and found that it could be revoked; and Austrian law to determine whether it had been revoked; in other words, the laws of the prematrimonial and post-matrimonial domiciles. Nevertheless, it is apparent that, in a modern commercial setting, the convenience of applying the applicable law of the contract is far greater and it is likely that the state of the contracting party’s domicile will have less interest in being applied to a commercial contract than to matters such as matrimonial settlements. More importantly for present purposes, the lesser interest of the law of the domicile is likely also to extend to the capacity of the settlor to create inter vivos trusts.42 In Male v. Roberts43 Lord Eldon, in an unconvincing judgment, appeared to suggest that the relevant governing law is the law of the place of contracting. Yet there appears to be no reason for that law’s application. The place of contracting may be entirely fortuitous, and, in the trust context, so equally may the place of creation of a trust. Both are capable of manipulation to avoid the clutches of any potentially applicable incapacity. Moreover, the lex loci contractus was at the time of the Male decision presumed to be the proper law, so is consistent with the argument that the proper law governs capacity. The law chosen by the parties is favoured by some Australian writers. Nygh argues that “In principle the capacity to enter into a settlement inter vivos should, by analogy with contracts, be regarded as part of the question of validity and thus governed by the proper law of the trust determined in accordance with Article 6 or 7 [of the Hague Convention]”.44

Sykes and Pryles, who view capacity as an aspect of essential validity, argue that parties should be able deliberately to: “choose the laws of a state which upholds the validity of the transaction as opposed to the law of a state which does not. . . The problems of scope of party autonomy in 41 However, Dicey and Morris, 1077, relying on a passage of Lindley MR ibid., 98, argue that this case, along with Cooper v. Cooper, rather provide support for application of the objective proper law. 42 At least outside the context of inter vivos family settlements. 43 (1800) 3 Esp 163. 44 P Nygh, Conflict of Laws in Australia, 6th edn. (Sydney, Butterworths, 1995), 518.

14 Choice of Law Rule for Capacity Questions cases of essential validity and capacity are analogous and it is not immediately apparent why a different rule should be employed for each issue”.45

Yet there is an obvious retort to that. Many rules of English law are designed with the purely domestic factual situation in mind. They are not necessarily designed to apply automatically to transnational matters.46 Rules which are restricted to a particular class of person are likely to take on more of a mandatory character, in the sense that they are specifically directed at certain parties, and should not be freely avoidable by those parties, at least not if the transaction in question has no material connection with another state. The law of closest connection has some support from the Canadian case of Charron v. Montreal Trusts Co., in which the Court of Appeal of Ontario accepted that capacity is governed by the proper law of the contract, “that is the law of the country with which the contract is most substantially connected”.47 However, it has been pointed out that the case is weak authority, since it was primarily concerned with an initial question of essential validity, not capacity.48 Brightman J also stated obiter in Bodley Head Ltd v. Flegon 49 that the proper law of the contract governs capacity, by which he is thought to have meant the proper law objectively ascertained.50 As seen above, the application of this law is supported by Dicey and Morris, on the at least partly persuasive grounds that otherwise e.g. a 14 year-old boy could confer capacity on himself to buy cigarettes simply by choosing a law under which he had no incapacity. Of course, this assumes that some legal system’s rules of capacity ought to be applied on a mandatory basis to the contracting party and that freedom of choice is unacceptable; however, it is not hard to see the force in the argument that capacity rules are not supposed to be “voluntary.” The objective proper law approach attempts to reconcile the undesirability of a potential lifelong burden being imposed on a person by the law of his domicile with the need to prevent evasion of capacity rules. However, the Dicey and Morris fear of giving autonomy of choice free rein does not mean that the law of closest connection must be used instead. Whilst their rule tends to mitigate in favour of capacity, in that capacity can be obtained under either of two laws, it lacks the very commercial convenience which the authors purport to provide. For one must look at the law of the domicile of the allegedly incapacitated party; then, if he has no capacity according to that law, one must work out with which legal system the alleged contract as a whole is most closely connected. This will be a very uncertain process, not least 45

E Sykes and M Pryles, Australian Private International Law, 3rd edn., 614. Although there are some “overriding” rules which are intended to be applied in an English court in certain circumstances whatever the governing law of the contract. An example is s. 27(2), Unfair Contract Terms Act 1977. 47 (1958) DLR (2d) 240, 244–5. 48 C Clarkson and J Hill, Jaffey on the Conflict of Laws, 2nd edn. (London, Butterworths, 1997), 234. 49 [1972] 1 WLR 680. 50 Cheshire and North, 593–4. 46

Capacity and Conflict of Laws 15 because the common law rules on the objectively applicable law will have to be used, rather than those found under Article 4, Rome Convention for determining the applicable law in the absence of choice. Such matters as a choice of court clause, the place of conclusion of the contract, the place of performance and the currency in which payment is to be made may all need consideration. Nor is it reasonable to expect the objectively applicable law to be any easier to determine in relation to transnational trusts. It would not be far-fetched to imagine a case where a settlor in state A purports to transfer assets which he has in state B to a trustee in state C to hold for a beneficiary in state D. The proffering of advice in such situations would be immensely difficult and fraught with danger.51 Uncertain though the authorities are, it is tolerably clear that, in English law, unfettered freedom of choice is not permitted in relation to contractual capacity. Indeed, Cheshire and North state that the choice lies between the law of the domicile, the place of contracting and the proper law objectively ascertained, not even considering the law chosen by the parties as a viable option.52 This is particularly material in formulating a choice of law rule for trusts. For if the chosen law will not be allowed to govern contractual capacity, one cannot expect that it will be allowed to govern all issues relating to the capacity of the settlor to dispose of property on trust. The entire premise of the conflict of laws property rules is that property rights are sufficiently powerful and connected to the particular state in which assets are situated that freedom of choice should not generally be allowed, even on questions of essential validity. This should apply a fortiori to the rules on capacity. Since the creation of a trust involves the alienation of property,53 it would be most surprising if the law chosen by the settlor could alone54 determine his capacity to create a trust. (ii) Capacity to transfer property (1) Land Capacity to convey land is governed, unsurprisingly, by the law of the place where the property is situated (lex situs). In Bank of Africa v. Cohen,55 a married 51 Although assistance might be gleaned from Art. 7, Hague Convention, this is not inevitably so. This is because we are looking for what the common law considers to be the objective proper law of the trust here. In any event, even if for common law purposes it was decided to align the method of determining the law of closest connection with the Convention approach, Art. 7 still leaves considerable room for flexibility, and uncertainty, as to which law will be found to be objectively applicable. For further consideration of the common law, see the discussion of Arts. 6 and 7 in Part Two of this book. 52 Cheshire and North, 592. 53 It is argued in this Part that even where the settlor declares himself to be trustee of his own property, there is an alienation of a property right (equitable title) and he must have capacity to alienate that right according to the private international law property rules affecting capacity. 54 Although, as will be seen below, it may be applied to the issue of whether he may create the trust structure with property of which he otherwise has capacity to dispose. 55 [1909] 2 Ch 129. Capacity to create a trust of immovable property was subjected to the lex situs in the Scottish case of Black v. Black’s Trustees 1950 SLT (Notes) 32, discussed by A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 636.

16 Choice of Law Rule for Capacity Questions woman domiciled in England, by deed executed in England, purported to agree to mortgage to a bank in England her land in South Africa, in order to secure debts of her husband. Under South African law she had no legal capacity to do so. The South African rule was one intended to protect married women. It was not clearly designed to protect non-South African domiciliaries. Nonetheless, the lex situs was applied and the woman was held to lack capacity. (2) Tangible movables The case for the application of the lex situs is greater for immovable property than other species of property, since the situs cannot be manipulated so as to confer capacity on the settlor. Yet property rights are so powerful that for conflict of laws purposes the lex situs is also generally regarded as having the strongest interest in being applied to the essential validity of inter vivos transfers of movables, especially in relation to tangibles.56 Given the “mandatory” characteristics of capacity rules, one might expect that the lex situs has an even stronger claim to be applied in relation to capacity to transfer tangible movables than it does to the essential validity of such transfers. (3) Intangible movables The Rome Convention does not address the capacity of a natural person to assign, or to receive an assignment of, an intangible. Unfortunately, the common law authorities are of limited use. Lee v. Abdy57 and Republica de Guatemala v. Nunez58 support the application of the proper law, which is thought to be either the law place of the transferor’s domicile or of the conclusion of the transaction. Both are rightly criticised by Cheshire and North.59 The former is commercially inconvenient; the latter is capable of easy manipulation. Given that the situs of an intangible is actually likely to be more enduring than that of a tangible movable,60 and that the lex situs can sensibly be applied to capacity to transfer the latter, it is difficult to see in principle why the lex situs ought not also to be applied to the transfer of intangibles.

(B) Trusts and Capacity Very little authority is to be found on the choice of law rule for capacity to create a trust.61 Sykes and Pryles lend express support to the proper law, which, 56

See below, sections 7(C) and (D). (1886) 17 QBD 309. 58 [1927] 1 KB 669. 59 Cheshire and North, 1059–60. 60 See below, sections 7(C) and (D). 61 At least outside the context of marriage settlements. Some discussion can be found in Lewin on Trusts, 17th edn., 302–6 and in M Lupoi, Trusts: A Comparative Study, 153. See also A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1. Curiously, K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative 57

Trusts and Capacity 17 where the settlor has made a choice, will be the law chosen by the parties.62 Yet this is largely based on their view that capacity should be treated as an aspect of essential validity, a view which has been criticised above. Nor is it easy to believe, on the basis of the above analysis of other commercial areas, that unfettered freedom will be given here when it is not permitted in those fields. (i) One question or two? It is contended that most of the problems in this area have arisen from a failure, common to both English writers and offshore legislators, to appreciate that there are in fact two distinct aspects to the capacity to create a trust. Matthews and Sowden hint at this when they state that whilst the use of the proper law is unsuitable in relation to the settlor’s capacity to transfer property, it might be used in relation to the specific capacity to create the trust structure rather than some other structure.63 (ii) The transfer of property This dichotomy is critical. For a settlor who wishes to transfer property to a trustee is disposing of his property no less than an owner who simply gives away or sells his property to another person absolutely. He needs capacity to alienate that property every bit as much as the outright donor and should not be able to avoid the otherwise applicable law for property transfers by creating a trust structure, especially as the beneficiary may well have the right to terminate the trust and have the property vested in him absolutely. In which case the rules which determine whether he may alienate his property must logically be the same as those advocated in relation to capacity to transfer property above. It is suggested that, whatever the form of property involved, the law of the place where the property is situated should be applied to that question.64 Support for this view is also to be found in Underhill and Hayton,65 though their comments are not wholly clear and, on one view, appear to be confined to cases where the situs is England. “Inter vivos trusts of English movables and immovables transferred validly under the lex situs to trustees of an English or other trust by a settlor of foreign domicile, habitual residence or nationality are valid and unimpeachable. . .” Law, ch. 23, at 10 observes that “the choice of law rules governing capacity to create a trust and the required formalities are those which apply generally to transactions involving a foreign element.” It is far from clear that this is correct, if, indeed, its meaning can be understood at all. 62 E Sykes and M Pryles, Australian Private International Law, 3rd edn., 614 and 715 63 P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 45. 64 At least in the case of inter vivos trusts. But see Black v. Black’s Trustees 1950 SLT (Notes) 32, where the lex situs was applied only in respect of immovable trust property and not in respect of movable property. 65 D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 241.

18 Choice of Law Rule for Capacity Questions Earlier, they had stated that: “It is clear that questions of capacity, form and essential validity affecting English land are governed by the English lex situs irrespective of the settlor’s domicile, habitual residence or nationality. . . The position seems to be the same for gratuitous transfers to trustees by foreign settlors of English movables, whether tangible or intangible”.66

An application of the lex situs at the time of transfer might be thought to secure the optimal compromise between certainty and convenience on the one hand and the application of the law which might legitimately be said to have the strongest interest in its law being applied to capacity questions, on the other. It might be objected that this will mean that capacity to transfer property will be governed by one law, but that the trust itself will be governed by another and that this is inconvenient. However, it should be remembered that another aspect of the “rocket-launching” process is the actual vesting of the property in the trustee. That will be determined by the lex situs.67 An advantage of the approach advocated here is that both questions68 relating to the initial transfer of the property to be held on trust to the trustee will be governed by the same, easily ascertainable, law. (iii) The creation of the trust structure There will then be a second question as to whether the settlor, who has capacity to alienate property according to the lex situs, has capacity to create the trust structure with that property. On this issue, it is suggested that there is a much stronger case for settlor autonomy. For the interest of the lex situs can sensibly be confined to the question whether the settlor can dispose of his property at all. Once he is able to do so, it should not be that law’s concern whether he makes an outright transfer of the property or transfers on trust. Other potential choice of law rules do not seem appropriate. Application of the law of the domicile should be rejected:69 “It would amount to allowing the itinerant domiciliary to carry around the disabling incapacity in his baggage and use it to escape liability for it in other jurisdictions, even those which do not recognise the status at all”.70 66 Ibid., 203, citing Pouey v. Hordern [1900] 1 Ch 492. Note that the 14th edn. (1987) simply states that “for a trust of movables questions of capacity and validity are governed by the proper law of the contract or trust, subject to the possible disapplication of that law on public policy grounds,” though there is no real analysis of the matter and the (now amended) views of Dicey and Morris (10th edn.), Rule 120, are relied upon. 67 See below, section 7. 68 I.e. the essential validity of the transfer to the trustee and the capacity of the settlor to transfer property to the trustee. In turn, problems of characterisation are minimised if the same law is applied to essential validity and to capacity. 69 Although, in relation to movables, it is favoured by P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 44. Theirs seems to be something of a lone voice in advocating this rule. 70 E Sykes and M Pryles, Australian Private International Law, 3rd edn., 344; cf. Viditz v. O’Hagan [1900] 2 Ch 87.

Trusts and Capacity 19 Another candidate, the law with which the trust has its closest connection, suffers from grave uncertainty and impracticality. Nor does it seem necessary to apply that law to the second question. There is no obvious policy reason for not applying the law chosen by the settlor to this issue. The restrictions upon a settlor seem legitimately targeted at the question whether he may dispose of property at all, not whether he may dispose thereof on trust. On the second question at least, this would accord with the views of Sykes and Pryles, and also with those of Nygh.71 (iv) Declarations of oneself as trustee This leaves one residual and complex question. What if the settlor simply declares himself trustee of his property? Arguably, he has not alienated his legal title, so the first question above does not apply; he has merely created the trust structure with the property, which, on the view advocated here, would be a second question issue, to be resolved by the chosen law. Some support for this view is provided by Lewin on Trusts. The authors argue that: “. . . the references in the Report to a transfer to the trustee as a condition of the creation of a trust must be taken to relate only to cases where the settlor intended to create the trust by making such a transfer”.72

With respect, it is suggested that this view should not be followed. If the initial step of asking whether the settlor has capacity to alienate his property is not taken where declarations of trust are concerned, the effect might be the equitable alienation of property of which the settlor may not dispose at all according to the lex situs. That cannot be right in principle, not least because the beneficiary might be able to call for the property to be vested in him absolutely and to terminate the trust altogether, thereby by-passing the lex situs rule. Since it has been suggested that the lex situs is likely to have an interest in whether any interest in the property can be disposed of by the settlor, it is suggested that, in the case of the declaration of trust, one should ask (i) does the settlor have power to alienate any interest in his property by the lex situs; (ii) if so, does he have power to create a trust according to the proper law of the trust? (v) Corporate trustees It is sometimes suggested that a different choice of law rule might apply to the capacity of a corporation to create a trust. So, for example, capacity of corporations to contract is excluded altogether from the Rome Convention.73 Certainly, a company may not enter into a transaction which is prohibited by the rules of 71 72 73

P Nygh, Conflict of Laws in Australia, 6th edn., 518. Lewin on Trusts, 17th edn., 288. Art. 1(2)(e), Rome Convention.

20 Choice of Law Rule for Capacity Questions its constitution. But in addition to being within its constitution, the company must also have capacity according to the relevant choice of law rule.74 On one view, which seems to have an effect on some of the offshore legislators,75 capacity should be governed by the law of the place of incorporation.76 However, that cannot be correct in principle, at least in relation to trusts, because it is in substance an application of the lex domicilii to companies. Yet a company is likely to have a much weaker “binding” connection with the state of incorporation than an individual has with his state of domicile, because companies are set up entirely by law and can choose where to incorporate.77 And we have already seen that the lex domicilii is not the appropriate choice of law for capacity of individuals. There is no reason why the lex situs should have any less interest in the capacity of a corporation than an individual. Both affect the alienation of property within its jurisdiction. Accordingly, it is suggested that exactly the same choice of law rules should be applied to corporate settlors as to individual settlors. (vi) Foreign penal rules and rules which infringe English public policy Finally, it is important to remember that, as with any choice of law rule, a foreign law applicable to capacity issues78 will not be enforced if it is penal,79 or applied to the extent that it infringes English public policy.80

(C) Conclusion In sum then, the following choice of law rule is suggested: (a) The capacity of any settlor to dispose of any property right inter vivos is governed by the lex situs at the time of the purported transfer. (b) The capacity of any settlor to create a trust of any property of which he may, according to rule (a), dispose, is governed by the proper law of the trust.

74

See Dicey and Morris, 1109–11, Rule 156. See the discussion in section 21, below. 76 See Janred Properties Ltd v. Ente Nazionale Italiano per il Turismo [1989] 2 All ER 444. 77 Although an individual may subsequently change his domicile, he is born with a given domicile (whereas a company’s place of incorporation is chosen) and most people, in practice, do not change their domicile of origin. 78 Including the issues of the capacity of the trustee and beneficiary, considered below. 79 E.g. Regazzoni v. KC Sethia (1944) Ltd [1956] 2 QB 490; Re Helbert Wagg & Co Ltd [1956] Ch 323; Oppenheimer v. Cattermole [1976] AC 249. See also Lewin on Trusts, 17th edn., 306. 80 In relation to “rocket” issues and public policy, see the discussion of Art. 18, Hague Convention in Part Two of this book. 75

The Capacity of the Trustee 21

5. THE CAPACITY OF THE TRUSTEE

We shall see in the discussion of Article 8(2)(a), Hague Convention that the “specific” capacity of the intended trustee is governed by the law applicable to the trust.81 However, it is argued there that the intended trustee must also have “general” capacity to receive property at all.82 It would be unsatisfactory if this “general” capacity requirement could be bypassed simply because the recipient is to hold the property on trust, since the recipient will hold legal title to the property and may alienate it to a third party.83 What then should be the law which regulates the “general” capacity to receive property at all? It would be most unfortunate if the capacity of the settlor to alienate property at all were governed by one law, but the capacity of the trustee to receive property were governed by a different law.84 It is suggested that the lex situs should likewise govern the capacity of the trustee to receive property at all. Accordingly, the following choice of law rule is suggested: (a) The capacity of a “trustee”85 to receive any property right inter vivos is governed by the lex situs at the time of the purported transfer. (b) The capacity of a “trustee” to act as trustee of any property which he may, according to rule (a), receive, is governed by the proper law of the trust.

6. THE CAPACITY OF THE BENEFICIARY

A putative beneficiary likewise requires capacity to receive a benefit at all,86 and specifically to receive a benefit under a trust. Where a beneficiary lacks capacity in English domestic law, he will not normally be able to give a good receipt for trust capital and/or income and alternative arrangements will need to be made.87 Lewin on Trusts notes that: 81

See also von Overbeck, para. 59, p. 382. An example of a “general” property rule regulating capacity in England is s. 1(1), schedule 1, Trusts of Land and Appointment of Trustees Act 1996, which provides that a purported transfer of land to a minor is ineffective to pass legal title, but operates as a declaration that the land is to be held in trust for the minor. 83 In England, s. 1(6), Law of Property Act 1925 prevents a child from holding a legal estate in land. If a settlor sought to create a trust governed by Utopian law of land situated in England, it would be most unsatisfactory if a minor could act as trustee thereof on the basis that Utopian law so permits. See also s. 20, which states that “The appointment of an infant to be a trustee in relation to any settlement or trust shall be void. . .” 84 Even though the settlor alienates absolute title to the property, whereas the trustee receives legal title only to the property. 85 I.e. a putative trustee. 86 We are concerned here with restrictions which the law imposes on the receipt of a benefit by a person or class of persons, not with further restrictions which the trust instrument might expressly impose on a minor as to when he may become entitled to an interest in possession. 87 Although there are exceptions to this rule. See Lewin on Trusts, 17th edn., 690–1. 82

22 The Capacity of the Beneficiary “. . . If a beneficiary domiciled abroad is mentally incapable and a curator has been properly appointed by the court of the domicile to look after the beneficiary’s affairs, with authority to receive the money, it should be paid to the curator”.88

It goes on to state that: “The authority of the parent or guardian of a foreign domiciled minor to give a good receipt for trust income or capital payable to the minor is governed by the law of the domicile”.89

However, it is no simple process to draw up suitable choice of laws rules for determining whether the beneficiary does lack capacity. For one thing, the nature of his interest under a trust is a complex issue, particularly where discretionary trusts are concerned.90 He may not be seen in some legal systems as enjoying a right in the trust property itself, but simply in ensuring that the trust is duly administered according to its terms.91 Furthermore, there has been a tendency to treat equity as acting in personam for private international law purposes, even where the assertion of beneficial title is concerned.92 The fact that states may differ in their views of the beneficiary’s position makes it unwise to seek authoritatively and autonomously to determine the matter for private international law purposes. It is far more helpful to ask what the key issue(s) concerning the beneficiary’s capacity are and to determine to which law they ought most naturally to be subjected. It might appear that a beneficiary’s capacity is a trusts specific issue, and should be subjected to the law governing the trust. He will seek to uphold the terms of the trust, and, even if he is regarded as receiving a title under the trust, the title is a beneficial one and not a “general” receipt of property. Against that, however, the trust cannot be “launched” unless there is a beneficiary93 (or valid purpose). Moreover, although it might be said that “The prime entitlement of the beneficiaries of a trust is that the trustees carry out the terms of the trust”,94 it might also be said that the matter in which any one law is likely to have the greatest interest is the ability of the beneficiaries to bring a trust to an end in accordance with the rule in Saunders v. Vautier,95 if they possess capacity to do so. In other words, a beneficiary under a trust is, putatively at least, an absolute owner of property; and on the question of absolute ownership of property 88

See Lewin on Trusts, 17th edn., 305. Ibid, 305–6. 90 See R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn. (London, Butterworths, 1998), 418–20. 91 G Gretton, “Trusts without Equity” (2000) 49 ICLQ 599. 92 Most notably in Case C–294/92 Webb v. Webb [1994] I ECR 1717. See also Penn v. Baltimore (1750) 1 Ves. Sen. 444; El Ajou v. Dollar Land Holdings Plc [1993] 3 All ER 717. See further, G Panagopoulos, Restitution in Private International Law (Oxford, Hart Publishing, 2000), 70–6. 93 Although, of course, the fact that the beneficiary is a minor and cannot immediately claim his interest in possession will not prevent the trust from taking effect if he is otherwise competent to receive a beneficial interest. 94 R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 115. 95 (1841) 4 Beav 115; or its equivalent in a foreign legal system. 89

Introduction 23 acquired inter vivos, one would expect the lex situs to hold sway, as the question is one of the “general” capacity to receive property. On balance, it is submitted that the “general” capacity of a beneficiary to receive any form of benefit under the trust should be determined by the lex situs, as it is for the settlor and for the trustee. This law will determine whether the beneficiary is precluded from taking any interest or benefit in the property in question. However, in accordance with the line of argument advanced above, it would be unsatisfactory for that law to determine the specific question of whether he may benefit under a trust, and objectionable for the lex situs to deny him a benefit on the ground that it does not know the trust concept. Accordingly, it is suggested that the “specific” question whether a person in principle entitled to take a benefit may do so in the capacity of a beneficiary under a trust (including the question of whether he is presently entitled to take a beneficial interest in possession under the trust because he is e.g. a minor) may be better viewed as a trusts specific matter for the law applicable to the trust. Accordingly, the following choice of law rule is suggested: (a) The capacity of a person to receive a benefit at all inter vivos96 is determined by the lex situs at the time of the purported transfer of property to the trustee.97 (b) The capacity of a person who may, according to rule (a), obtain a benefit, to assert a beneficial interest under a trust, including the question of whether he is presently entitled to that benefit, is governed by the proper law of the trust.

7. THE VESTING OF PROPERTY IN THE TRUSTEE INTER VIVOS

(A) Introduction Once it has been established that the settlor has capacity to dispose of his property on trust, the next question concerns the constitution of the trust. This involves consideration of the law that determines whether property may be, and has successfully been, vested in the trustees. This is a question of the essential validity of a purported transfer of property. In principle, there is no reason why this should not be governed by the general conflict of laws rules for the transfer of property, which will be outlined below. Put very simply, inter vivos property 96 In the case of a testamentary disposition of movables, the law of the deceased’s last domicile would be applicable to this matter. 97 Although the beneficiary may subsequently seek to terminate the trust and claim the property absolutely and the situs may have changed since the time of the trust’s creation, it would be unfortunate if the entitlement of the beneficiary to receive property could not be determined at the outset. It would also be regrettable if a different law were applied to the beneficiary’s capacity to receive property at all to that which is applicable to determine whether the settlor may alienate property at all and whether the trustee may receive property at all.

24 Vesting of the Property in the Trustee Inter Vivos transfers are generally governed by the lex situs, at least if the transfer is of immovable or tangible movable property. (i) One question or two? Technically, one could argue that two questions arise as to the passing of property: (i) the transfer of legal title to the trustee (ii) the transfer of equitable title to the beneficiary.98 The latter question arises because, in the words of Sykes and Pryles, “the trust document may therefore be viewed as a type of equitable conveyance.”99 Thus, if A purports to transfer property to B to hold on trust for C, the lex situs might permit the transfer of property to B, but not accept the validity of the equitable transfer. Accordingly, one could argue that the lex situs should apply to both questions. Yet this cannot be correct in principle. Otherwise, it would seem to preclude the creation in England of a trust of property situated in any country where the passing of equitable ownership is not permitted or recognised, even if English law is stated to be the governing law of the trust. Indeed, it would seem to prevent the creation of a trust, the assets of which are held in a country which does not have any analogous concept to that of equitable ownership. Rather, it is suggested that once legal title has validly been transferred to B, the “rocket-launching” process is at an end.100 The particular interest of the beneficiary might be considered as an aspect of the relationship between trustee and beneficiary,101 and hence governed by the law determined by the rules of the Hague Trusts Convention. This view appears to be supported by Nygh.102 He postulates an example of a trust of land situated in New Caledonia which is governed by the law of New South Wales. If New Caledonian law treats the conveyance as an absolute conveyance without restriction, he argues that “This will not affect the validity of the trust from the New South Wales point of view”.103 It follows that the sole question with which the lex situs should concern itself is whether legal title to an interest may be alienated. It should not be a reason to 98 It is argued here that the lex situs should determine solely whether legal title has been alienated. The law applicable to the trust should determine whether equitable title has vested in the beneficiary. Inevitably, bootstraps objections may be levelled at this approach, because until equitable title has vested in the beneficiary, it may be objected that no trust has been created, and that until a trust has been created, it has no applicable law. However, a degree of pragmatism is essential. Moreover, that pragmatism is kept firmly in check by the requirement that lex situs permits the property of the settlor to be alienated at all. 99 E Sykes and M Pryles, Australian Private International Law, 3rd edn., 712. 100 The Australian decision in Augustus v. Permanent Trustee Co. (Canberra) Ltd (1971) 124 CLR 245 may lend some support to this view. In that case the application of the rule against perpetuities, which would obviously affect the creation of an equitable interest in a would-be beneficiary, was held to be governed by the proper law of the trust. It is argued below (section 8) that perpetuity rules which affect the alienation of property at all should be governed by the lex situs; whereas provisions which specifically affect the creation of the trust structure should be governed by the law applicable to the trust. 101 Which Art. 8(2)(g) states to be within the Convention’s scope. 102 Although he does not expressly adopt it, it is consistent with the example given by him. 103 P Nygh, Conflict of Laws in Australia, 6th edn. (Sydney, Butterworths, 1995).

Introduction 25 invalidate the “rocket-launching” process that the situs does not know the trust concept. For this reason, this author would disagree with the view of KoppenolLaforce that: “Without specific new rules the creation of a trust will not be possible for property that is situated in a non-trust country in which the creation of a trust was not possible before the Trusts Convention”.104

Such an approach to the “rocket-launching” process would fatally undermine the Convention’s philosophy, which very clearly sanctions the location of trust assets in non-trust states. It follows that we are concerned in this section solely with the principles which will govern the vesting of legal ownership in the trustee, not the vesting of equitable title in the beneficiary. (ii) Declarations of oneself as trustee Once again, the case of the declaration of trust creates problems. Lewin on Trusts contends that: “. . . where trusts are created by a mere declaration by the owner of the original trust assets there is no preliminary issue to be decided about the validity of any transfer, there being no transfer”.105

It is suggested that this view should not be followed. Whilst there will be no transfer of legal title, there will still be the passing of a property right from the wouldbe settlor. Consistent with the argument made above in relation to capacity,106 the question whether any property right has passed from the settlor should be governed by the lex situs, so as not to evade the application of that law to property transfers; but the question whether the title which the would-be beneficiary has received is equitable ownership should be determined by the proper law of the trust; otherwise, no valid trust could be made whenever assets were situated in a country in which the concept of equitable ownership was unknown. (iii) Intention to create the trust Before we proceed, it should be noted that there is one issue which should not be subject to the choice of law rules affecting the creation and transfer of property rights: namely the determination of whether the settlor manifested sufficient certainty of intention to create the trust. This is a question of resolving what the settlor meant by the words used. As with wills,107 the law intended by the settlor to govern the matter should prevail. 104 M Koppenol-Laforce, Het Haagse Trustverdrag (Deventer, Kluwer, 1997), 266. Provided that the lex situs says that a settlor may alienate his property, that should suffice. 105 Lewin on Trusts, 17th edn., 288. 106 See von Overbeck Report, para. 57, p. 382; and section 4(B)(iv) above. 107 See section 14, below.

26 Vesting of the Property in the Trustee Inter Vivos Where the trust is created inter vivos, this will almost always (subject to contrary indication) be the law intended by the settlor to govern the trust itself. Where this law is not expressly stipulated and the implied intention of the settlor cannot be ascertained, the assumption that the putative applicable law of the trust should apply is still good and the law which would be applicable in the absence of choice under Article 7, Hague Trusts Convention should be used. Testamentary trusts may be more complex, since the law applicable to the will in which the “trust” was declared and the law putatively applicable to the trust itself may differ. Which should determine whether the settlor intended to create a trust? A hard and fast rule cannot be stated, because the answer remains that whichever law was intended by the settlor to govern should prevail. But where no such law can be ascertained, it may be presumed that a settlor who made no provision as to which law would govern the testamentary “trust” intended that the law of his domicile at the time of making the will, which would determine the construction of the will itself,108 should prevail.

(B) Immovable Property (i) The general rule The lex situs clearly holds sway in relation to the essential validity and formalities109 necessary for transfers of immovable property.110 So, in Nelson v. Bridport,111 Lord Nelson had sought to devise his Sicilian estate in a way which infringed Sicilian law. It was held that he could not do so. The rationale of the decision, the correctness of which is beyond doubt, is that “At the end of the day, only the law of the situs can control the way in which land, which constitutes part of the situs itself, is transferred.”112 (ii) Renvoi It is for the same reason that the doctrine of renvoi will also be applied here, since it is important to apply the law of the situs precisely as it would be applied

108

See section 14, below. Adams v. Clutterbuck (1883) 10 QBD 403. 110 However, a contract to transfer an interest in land need not satisfy the formal requirements of the lex situs—Re Smith, Lawrence v. Kitson [1916] 2 Ch 206. Art. 9(2), Rome Convention states that a contract is formally valid if it satisfies the formal requirements of either the applicable law of the contract or the law of the place where either party was at the time of conclusion of the contract. Art. 9(6) states that in the case of contracts for the sale of land, these formality rules still apply but are subject to “the mandatory requirements of form of the law of the country where the property is situated if by that law those requirements are imposed irrespective of the country where the contract is concluded and irrespective of the law governing the contract.” 111 (1846) Beav 547. 112 Cheshire and North, 929. 109

Immovable Property 27 by a judge in the courts of that state.113 Detailed discussion of the highly complex doctrine of renvoi is outside the scope of this work.114 However, some basic consideration is provided, given the importance of the doctrine in the areas of property law and succession. In areas of the law where renvoi is inapplicable, an English court simply uses its choice of law rules and applies the domestic law of the state identified. Renvoi essentially entails the English court seeking to decide a case using the substantive law that a judge in the courts of the situs would use. To do this, it applies not merely the domestic law of the state identified by its choice of law rules, but the private international law of that state. So, to take a very simplified example, if a dispute about succession to land situated in Brazil arose, an English court would apply Brazilian law. However, it might find that a Brazilian judge, applying Brazilian choice of law rules, would apply the law of the testator’s nationality, Argentinean law, in a Brazilian court. If an English court applies Argentinean law too, it has invoked the doctrine of renvoi. Application of renvoi promotes unity of outcome and may be justified on the basis that since the situs has total control over land within its jurisdiction, it would be futile to decide the case in a manner inconsistent with how it would be resolved in that legal system. However, matters are more complex for two reasons. First, the foreign law, when applying its choice of law rules, may refer the matter back to English law, creating something of a logical paradox. Secondly, it becomes apparent that there are degrees of renvoi. Single renvoi: In Re Duke of Wellington,115 an issue arose as to who succeeded to land situated in Spain. English choice of law rules lead to the application of the lex situs, Spanish law. However Spanish private international law referred the dispute to the law of the deceased’s nationality, which was taken to mean English law. This being an area where renvoi was applicable, the English court would have regard to Spanish private international law, and not merely apply the domestic law of Spain. Spanish law would not itself apply a doctrine of renvoi and its choice of law rules pointed to the application of English domestic law only. Accordingly, the English court applied a “single” renvoi and used English law to determine the dispute. Double renvoi: In Re Annesley,116 an Englishwoman died domiciled in France according to English law. English choice of law rules in respect of succession to movable property led to the application of French law. However, by French law, the deceased was considered to have died domiciled in England and, as a matter of French private international law, English law should be applied. But, of 113 See Re Ross [1930] 1 Ch 377; Re Duke of Wellington [1947] Ch 506. See also the discussion of Art. 17, Hague Convention in Part Two of this book. 114 See Dicey and Morris, ch. 4; Cheshire and North, ch. 5. 115 [1947] Ch 506. 116 [1926] Ch 692.

28 Vesting of the Property in the Trustee Inter Vivos course, English private international law stated that French law should be applied. This time (in contrast to re Duke of Wellington), French law itself applied a doctrine of renvoi and would seek to apply English private international law and not merely English domestic law. French law would accept this remission back to its own law and a French court would apply French law. It followed that an English court should do the same and accept the “double” renvoi. Accordingly, in cases where renvoi is applicable, the law which is ultimately applied (and whether single or double renvoi is applied) depends solely on whether the private international law of the state identified itself applies a doctrine of renvoi on the facts.

(C) Tangible Movable Property117 (i) What does “lex situs” mean? One obvious problem in relation to movable property is that its situs can change. Potentially, this could mean that if A purports to transfer property to B in state X, by which the transfer is valid, but B moves the property to state Y, by which the transfer is invalid, ownership of the property would change. This would clearly be a most unsatisfactory and uncertain state of affairs. However, the conflict of laws rule is that title is governed by the law of the place where the most recent purported transfer took place. So, in our example, the fact that property was subsequently taken to state Y would not affect title, and the law of state Y would be irrelevant, unless a purported transaction took place in that state. If such a transaction did occur, the law of state Y would determine ownership, even if its findings would be inconsistent with those of state X. So, in Cammell v. Sewell,118 Cammell owned timber, title having been acquired by Russian law when in Russia. The timber was shipped from Russia to England on a Prussian ship which was wrecked off the coast of Norway. The ship master sold the timber to Sewell in Norway, which gave Sewell title under Norwegian, but not English, law. Sewell brought the timber to England and Cammell sued him. It was held that Sewell’s title prevailed. Whilst this might seem harsh on the previous owner, who could lose title without knowing that his goods were being dealt with in another country, the case shows just how much power the lex situs exerts in this area. In the converse case of Winkworth v. Christie,119 pictures were stolen from England and taken to Italy, where they were bought by D2 and were later 117 See generally P Lalive, The Transfer of Chattels in the Conflict of Laws: A Comparative Study (Oxford, Clarendon Press, 1955); G Zaphiriou, The Transfer of Chattels in Private International Law: A Comparative Study (London, Athlone Press/University of London, 1956). 118 (1858) H&N 617. 119 [1980] Ch 496.

Tangible Movable Property 29 returned to England to be put up for auction by D1; but since no transaction in them had taken place in England, Italian law still determined title. So D2 acquired good title against the world. The effect was that the original owner found goods which he had previously owned validly on sale by D1 in England. Slade J said that the rule is necessary for commercial certainty and because the innocent purchaser overseas is just as worthy of protection as the original owner.120 Moreover, as North notes,121 to apply the lex situs will generally accord with the expectations of parties to a transfer; and even with movables, a party may need to invoke the authorities of the situs to ensure transfer of title is completed. In multiple assignment cases, buyers would otherwise have to investigate title of the vendor not just by the current lex situs but also to trace the entire history of the goods. (ii) Exceptions There are a limited number of exceptions to the lex situs rule, most of which are stated by Slade J in Winkworth v. Christie.122 These include: (a) where goods are in transit, with a casual or unknown presence;123 (b) where the application of the lex situs is contrary to English public policy; (c) where an English statute prescribes the mandatory application of English law; (d) general assignments of movables on bankruptcy or succession are governed by the law of the domicile of the bankrupt or the deceased. (iii) Renvoi There is some support for the application of this doctrine in relation to movable property in Winkworth, where Slade J stated that he “supposes” it would be open to P to plead renvoi.124 Application of the doctrine is also supported by Dicey and Morris.125 Although the need to decide a case in exactly the same manner as would the courts of the situs is obviously not as great where property can be removed from that jurisdiction, nonetheless, if the lex situs is considered to be the most appropriate law to be applied to tangible movables, there is a strong argument for applying the same substantive law as would be applied in that jurisdiction, and hence for applying renvoi. Lewin on Trusts argues for the non-application of renvoi to testamentary trusts of movables. The authors contend that, since the relevant authorities are 120 It may be that an exception will operate where the subsequent purchaser did not act bona fides. Slade J suggested this exception in Winkworth, but its existence is dubious, and rejected by Cheshire and North, 944–5. 121 P North, Private International Law Problems in Common Law Jurisdiction (The Hague, Martinus Nijhoff, 1993). 122 [1980] Ch 496. 123 Ibid, 501. For the applicable law in such a case, see Cheshire and North, 951–2. 124 Ibid, 514. 125 Dicey and Morris, 966.

30 Vesting of the Property in the Trustee Inter Vivos at first instance, the Court of Appeal should restrict the application of renvoi to immovables.126 However, in relation to inter vivos trusts of chattels, they state that: “Though the chattel could subsequently be moved to a different territory, there is much to be said for the view that if the situs is foreign then. . . the lex situs should be taken to be the domestic law that the courts of the situs would apply, not necessarily its own”.127

No further reason is given by the authors for advocating renvoi in this case. One reason that might have been given is that the lex situs is not applied to succession to movables,128 whereas it is applied to determine the effects of a purported inter vivos transfer of movables.129

(D) Intangible Movable Property (i) The situs The law still purports to ascribe a situs to intangible property. Shares have their situs in the country where they may be dealt with between shareholder and company. So, if shares are only transferable by entry on the register, they are situated in the country where the register is kept.130 If the company has a register in two countries, and the transfer may be registered in either, the situs is the place where in the ordinary course of business the transfer would be registered.131 The greatest complexity is caused by the debt due under a contract. A debt is generally regarded as having its situs where the debtor resides,132 because that is where it generally needs to be enforced. If he has more than one residence, it is where the creditor stipulates for the debt to be paid133 and in the absence of such a stipulation, where it is payable in the normal course of business.134

126

Lewin on Trusts, 17th edn., 284. Ibid, 284. Emphasis added. 128 See section 14, below. 129 See also Macmillan Inc v. Bishopsgate Trust plc (No 3) [1996] 1 WLR 387 (on which, see A Briggs, “From Complexity to Anticlimax: Restitution and Choice of Law” (1996) 4 RLR 88; J Stevens, “Restitution or Property? Priority and Title to Shares in the Conflict of Laws” (1996) 59 MLR 741; R Stevens, “The Law Applicable to Priority in Shares” (1996) 112 LQR 198; J Bird, “Choice of Law Rule of Priority Disputes in Relation to Shares” [1996] LMCLQ 57; W Swadling, “A Claim in Restitution?” [1996] LMCLQ 63; C Forsyth, “Characterisation Revisited: an Essay in the Theory and Practice of the English Conflict of Laws” (1998) 114 LQR 141). 130 See e.g. Brassard v. Smith [1925] AC 371; Eric Beach Co v. A-G for Ontario [1930] AC 161; R v. Williams [1942] AC 541; Macmillan Inc v. Bishopsgate Trust plc (No 3), ibid. 131 Treasurer of Ontario v. Blonde [1948] AC 24; Standard Chartered Bank Ltd v. IRC [1978] 3 All ER 644. 132 New York Life Insurance Co v. Public Trustee [1924] 2 Ch 101. 133 Kwok Chi Leung Karl v. Estate Duty Comrs [1988] 1 WLR 1035. 134 Power Curber International Ltd v. National Bank of Kuwait SAK [1981] 3 All ER 607. 127

Intangible Movable Property 31 It could be objected that it is artificial to attribute a situs to intangible property, and especially to debts.135 However, the rules for ascertaining the situs of property provide a ready way to locate intangible property and hence provide a straightforward choice of law rule. Indeed, the situs of a debt, contingent as it is upon the residence of the debtor, may well be less likely to change than the situs of tangible movables. (ii) Assignment of debts: the issues raised Suppose that A, a debtor resident in Ruritania, owes B £10,000 under a contract governed by English law and B purports to assign the right to that £10,000 to C to hold on trust for D under a contract governed by Utopian law. Which law determines (i) whether the right can be assigned at all; (ii) if so, whether a fully constituted trust exists? (iii) Assignability of debts For matters within the scope of the Rome Convention, an apparently straightforward answer is given. Article 12(2) states that questions relating to the assignability of a debt and to the relationship between that debtor and assignee are to be governed by the law governing the right to which the assignment relates. In our example, that would be the law of the original contract, i.e. English law. Article 1(2)(g), Rome Convention excludes from the scope of the Convention the constitution of trusts. Accordingly, it might appear necessary to have regard to common law principles. It appears that the assignability of an intangible will still, however, be governed by the law from which the right originates, which will, of course, in the case of a debt, be the original law of the contract. In Trendtex v. Crédit Suisse,136 a Swiss company sued a Nigerian bank in England. The Swiss company then assigned this right of action in the English court to a Swiss bank in Geneva. The Swiss company alleged that the assignment was void as contrary to English law. The Court of Appeal agreed, since the assignability of the English cause of action was governed by English law, not by the proper law of the assignment itself. The characteristics of a cause of action could only sensibly be answered by the law of its origin.

135 For criticism, see P Rogerson, “The Situs of Debts in the Conflict of Laws—Illogical, Artificial and Misleading” [1990] CLJ 441. See also M Moshinsky, “The Assignment of Debts in the Conflict of Laws” (1992) 108 LQR 591. 136 [1980] QB 629.

32 Vesting of the Property in the Trustee Inter Vivos (iv) The assignment of the debt Whether a debt which can be assigned actually has been assigned is a matter for the law applicable to the contract of assignment.137 In our example above, this would be Utopian law.138 (v) Title to the chose in action More complex is the question whether the assignee has legal title to the debt purported to be transferred. The assignee may not seek to claim a contractual right thereto, but rather may assert that he is already the owner of the chose in action. Jaffey states that, “from a policy point of view what is required is a choice of law rule which promotes stability in proprietary relations. . . Essentially there are two options: the lex situs or the law under which the interest was created.”139

Since the lex situs holds sway in relation to proprietary questions in relation to tangible movables and immovables, there is an obvious attraction in applying that law here as well. In our earlier example, this would point to Ruritanian law, the residence of the debtor. This argument has, however, recently been rejected by the Court of Appeal in Raiffeisen Zentralbank Osterreich AG v. Five Star General Trading LLC.140 Mance LJ ruled that the assignment of the benefit of a marine insurance policy must be resolved by application of contractual, and not proprietary choice of law rules. A similar rule should apply to assignment of debts. The debt is a form of property whose existence cannot easily be divorced from the contract which creates it. For matters within the scope of the Convention, Articles 12(1) and 12(2) exclusively determine the validity and effects of an assignment. Even for matters outside the Convention’s scope, the view that the debt and its contractual origin are inextricably intertwined should today lead to rejection of the lex situs with respect to assertions of “title” to the chose in action. Nevertheless, the law of the contract of assignment cannot be used where the assignment is by gift. In that scenario, the assignability of the debt may be governed by the law under which it was created; but whether it has been assigned should be a matter for the lex situs.141

137 Presumably, the formal validity of the assignment itself should be determined as for any other contract by Arts. 9(1) and 9(2) of the Rome Convention: i.e. it will be formally valid if it satisfies the requirements of the applicable law or that of the state where either party was at the time of conclusion of the contract. 138 See Art. 12(1), Rome Convention; Trendtex v. Crédit Suisse, [1980] QB 629. 139 Jaffey on the Conflict of Laws, 2nd edn., 453. 140 (2001) 2 WLR 1344. On the first instance decision (reported at (2000) CLC 1359), see P Rogerson, “Choice of Law—Notice of Assignment” [2001] CLJ 14. 141 A view shared by D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3, 7.

Conclusion 33

(E) Conclusion The following choice of law rule is suggested to determine the essential validity of purported inter vivos transfers of property on trust: (a) (i) Whether any proprietary interest has been transferred inter vivos from a would-be settlor to a would-be trustee is determined by the lex situs at the time of transfer.142 (ii) However, in the case of debts, the assignability of the debt will be determined by the law under which the debt arose; and the question whether the debt has been assigned to the trustee will be determined by the law governing the assignment itself. (b) Whether the disposition is effective to transfer equitable title to the wouldbe beneficiary, and thus to form a completely constituted trust, is determined by the proper law of the trust.

8. PERPETUITIES AND ACCUMULATIONS

It is argued in the discussion of Article 4, Hague Convention that perpetuity rules should fall within the Convention only insofar as they appertain specifically to the ability to create the trust structure with a disposition of property.143 Insofar as they affect any disposition of property, however, they should be regarded as a preliminary, “rocket-launching” matter and subjected to the “ordinary” transfer of title rules discussed in this Part. As Morris and Leach observe: “The Rule against Perpetuities is a rule of policy. It is not a mere canon of construction, still less a mere formality. Hence, in general, the law which governs the material or essential validity of instruments should be the governing law”.144

It is argued that section 164, Law of Property Act 1925,145 which deals with accumulations, should be regarded as a “rocket” provision. As such, it should be applied only where English law governs a trust, but should not routinely be 142 This rule should be qualified in the case of some forms of intangible property. In the context of the assignment of a right to sue, the approach is likely to be similar to that applicable to debts. Accordingly, the effectiveness of the transfer is likely to be determined by a court asking (i) can the right be assigned according to the law under which the right arises; and (ii) has the right been assigned according to the law governing the assignment itself? 143 See the discussion of “distinguishing between the ‘rocket’ and the ‘rocket-launcher’” in respect of Art. 4 in Part Two of this book. 144 J Morris and W Leach, “The Rule Against Perpetuities”, 2nd edn. (London, Stevens, 1962), 20. 145 See also the Perpetuities and Accumulations Act 1964. See further Law Commission Report No 251 on the Rules against Perpetuities and Excessive Accumulations (London, HMSO, 1998).

34 The Situs of Equitable Interests applied to foreign trusts of property located in England.146 Although it appears to apply to all dispositions of property, it is essentially concerned with the creation of trusts.147

9. THE SITUS OF EQUITABLE INTERESTS AND DEALINGS WITH SUBSISTING EQUITABLE INTERESTS

(A) General It may become important to determine where a beneficial interest under a trust is located. Dispositions of an equitable interest within a certain jurisdiction may attract taxation liability within that state. It may also be important to the extent that the lex situs falls to be applied to any dealings with a subsisting equitable interest. Unfortunately, it is not always straightforward to determine the situs of a beneficial interest. Lewin on Trusts states that: “A beneficial interest under a trust may be settled so as itself to be held on the trusts of another settlement. Such an interest is situated where the trustees of the original trust can be sued to enforce it, unless it amounts to equitable ownership of a trust asset, when it is situated where the trust asset is”.148

Distinguishing between these two scenarios may not be easy in practice and it is instructive to look at the cases more closely.

(B) Trusts and Unadministered Estates Certainly, the right of a legatee to a residuary beneficial interest in an unadministered estate is not considered to be located in the place where the assets themselves are situated. The right of the legatee is to compel in personam the performance of the executor’s duties. So much is clear from Lord Sudeley v. A-G.149 In that case, a testator died domiciled in England, bequeathing one quarter of the residue of his real and personal estate to his wife. The will was proved in England by executors domiciled in England. Included in the estate were mortgages of land situated in New Zealand. The wife died before the estate of her husband became fully administered and the residue ascertained and no appropriation of the New Zealand mortgages had been made. The question 146 Save insofar as application of the governing law of the trust infringes English public policy under Art. 18, Hague Convention. 147 E Burn, Cheshire and Burn’s Real Property, 16th edn. (London, Butterworths, 2000), 309. 148 Lewin on Trusts, 17th edn., 282; relying on Re Smyth [1898] 1 Ch 89; A-G v. Johnson [1907] 2 KB 885; Favorke v. Steinkopf [1922] 1 Ch 174; Re Berchtold [1923] 1 Ch 192; Philipson-Stow v. IRC [1961] AC 727, 762. 149 [1897] AC 11.

Beneficial Interests and Rights to Enforce a Trust 35 of the wife’s executors’ rights in relation to the New Zealand mortgages arose in connection with liability for probate duty. This, in turn, raised the issue of whether those rights were situated in England or New Zealand. Lord Halsbury LC made clear that the right of the wife had not constituted a beneficial interest in the property, since until administration, the extent of any residuary interest could not be determined and the trust was not constituted. Accordingly, it was: “. . . impossible to maintain that the character of any part of that estate can be ascertained so as to make it possess a specific locality until that has happened. . . . [T]he thing that the legatee is entitled to, call it a debt, call it something that must be administered either by trustee or executor, the character of that, the local character, is fixed by the persons, call them debtors or call them trustees. . .. Under these circumstances it appears to me there can be but one answer to the question, and that is that the debtors are here and have to administer here. . . and if it is a debt and the debtor is here, that is the character of the asset as fixed by the residence of the debtor, and the asset is English”.150

(C) Beneficial Interests and Rights to Enforce a Trust Dicey and Morris state that:151 “The location of an interest under a trust is dependent upon whether, under the law governing the trust, the beneficiary has a beneficial interest in the trust property or whether under that law he has merely a right of resort to a court in order to compel the trustees to discharge the task imposed upon them. If the beneficiary is given a beneficial interest in the trust property then his interest under the trust is located in the country where the trust property is situated. If the beneficiary is given merely a right of action against the trustees then his interest under the trust is located where the action may be brought, i.e. at the trustees’ place of residence”.152

It follows that, where a beneficiary has an entitlement under a fixed trust to a particular asset,153 his equitable interest is located where the property itself is located. Where he has only an entitlement to receive a sum of money from the trust assets,154 his interest is located where the trustee resides. If a beneficiary has a right to receive £1,000 from the trust, to be taken from the trust assets which are situated around the world, it would not be possible to apply the law 150 Ibid., 16. The decision was applied in Dr Barnado’s Homes National Incorporated Association v. Commr for Special Purposes of the Income Tax Acts [1921] 2 AC 1; Commissioner of Stamp Duties (Queensland) v. Livingstone [1965] AC 694. Lord Shand in Lord Sudeley (at 20) expressly left open the position where a legatee had been left an entire estate, tacitly suggesting that he might have found that the right of the legatee was located in the situs of the property in question. 151 Dicey and Morris, 934. 152 For this proposition, the authors cite Re Smyth [1898] 1 Ch 89; A-G v. Johnson [1907] 2 KB; Favorke v. Smith [1922] 1 Ch 174; Re Cigala’s Settlement Trusts (1878) 7 Ch D 351; Lord Sudeley v. A-G [1897] AC 11; Commissioners of Stamp Duty (Queensland) v. Livingstone [1965] AC 694 (PC). 153 The law applicable to the trust should determine the nature of the interest which the beneficiary has under that trust. 154 Or a share of an undivided bulk, such as identical shares in a company.

36 The Situs of Equitable Interests of the place where the assets themselves are located, since there is no particular asset which is his.155 However, application of this rule has not always proved straightforward in the case law. This has not been helped by the fact that the bulk of cases have concerned the rather specialised environment of testamentary trusts for sale of land. In Re Smyth,156 a testator domiciled in England made an English will bequeathing a plantation in Jamaica to trustees domiciled in the United Kingdom on trust for certain persons and their issue for life, thereafter on trust for sale. The trust for sale took effect and the proceeds became divisible amongst certain persons. One of the persons entitled to the ultimate proceeds had died whilst the trust for life was still in existence and their liability to probate duty in respect of that interest came into issue. This turned on whether the interest was situated in Jamaica or England. Romer J ruled that the right in issue was solely to compel the trust for sale to be carried into effect at the due time i.e. an equitable chose in action, and was located in England, where the trustee could be sued to enforce the right. Yet one could argue that what was in issue was the ultimate proprietary entitlement to the proceeds of sale, whatever they might turn out to be. In A-G v. Johnson,157 a trust for sale created by a testator domiciled in England with English resident trustees included a tea estate in Assam. The income was to be used to pay life annuities, with surplus income to be paid to the surviving annuitants. The trustees postponed sale of the estate and, whilst still unsold, two persons entitled to the surplus income died. The increase in the survivors’ share in the surplus income of the tea estate was held to be located in England. Bray J stated that: “I think it is clear that, but for the intervention of the trustees and the special directions and powers given to them, the property would have been situated out of the United Kingdom. It consisted of real estate in Upper Assam. . . .”

However, even although the estate remained unsold, the testator had intended the land to be sold and “it has become property in England because the testator wished that it should be so.”158 Yet it is not easy to see how the legatees had a mere right of action against the executor. They possessed interests under an administered estate, albeit that their interests under a trust for sale would be deemed “converted” into personalty. True, they could not lay claim to specific assets until the extent of the surplus was known. However, the fact that the property had not yet been sold pursuant to the trust for sale did not mean that the legatees possessed mere personal rights of action; they had an interest prior 155 But against that, the residence of the trustees may differ, which will necessitate a further choice of law rule (such as the law of the place where the majority reside, or the law of the place of administration of the trust). 156 [1898] 1 Ch 89. 157 [1907] 2 KB 885, 893 and 895. 158 Ibid., 893 and 895.

Beneficial Interests and Rights to Enforce a Trust 37 to sale, and it is not easy to see how their beneficial interests could sensibly be located other than in Assam prior to sale.159 A similar difficulty emerges in Favorke v. Steinkopff.160 The rights of beneficiaries under a will to income from funds invested in the Hamburg State Loan was described by Russell J as “undoubtedly . . . a property right and interest in this trust fund.” But he found that the beneficiary was entitled only to say “You, the English trustees, must pay me so many Marks out of this particular fund which the testator designated for that purpose”;161 in other words, an equitable chose in action located where the trustees were resident, namely England. This represents something of an extreme view of equity acting in personam. A claimant may not be able to say “I (beneficially) own that particular sum of money; but he can say “I (beneficially) own a right to a share in the money.” Why should that property right not be located where the money itself is located? May it not equally need to be enforced in the situs as would a right to a defined sum of money? The beneficiary’s right is far more tangible than that of the residuary legatee in Lord Sudeley and should arguably be located where the assets themselves are situated. In contrast, in Re Berchtold,162 a testator left an interest in the proceeds of sale of English land, which had not yet been sold, to a legatee. This was a defined share and Russell J this time could: “find no real foundation in the various authorities . . . for altering the view that the beneficial interest in relation to the Birmingham freeholds and the proceeds of sale thereof . . . was immovable property according to English law [and that] . . . the persons . . . beneficially entitled are [those] . . . who would be entitled thereto according to the law of England”.163

One can only remark that the difference in reasoning between this case, and, in particular, A-G v. Johnson, illustrates the difficulty in attempting to delimit the distinction drawn by Dicey and Morris between a right to enforce a trust and a beneficial interest in the property itself.164

159 Indeed, the correctness of the case is doubted by Dicey and Morris, 934. Viscount Simonds in Philipson-Stow v. Inland Revenue Commrs [1961] AC 727, 743 expressly left open the possibility that once property held on trust for sale has been sold, the law governing the interest in the proceeds would become that of the trustee’s residence. Prior to sale, however, it was located where the land itself was situated. A similar view was taken by North J in Re Piercy [1895] 1 Ch 83, 88. 160 [1922] 1 Ch 174. 161 Ibid., 177. 162 [1923] 1 Ch 192. 163 Ibid., 209. 164 The decision in A-G v. Johnson [1907] 2 KN 885 was described as “unsatisfactory” in Philipson-Stow v. Inland Revenue Commrs [1961] AC 727, 749 per Lord Reid, in finding that the beneficial interest in unsold land in South Africa subject to a trust for sale was situated in South Africa. He pointed out that, “any sale to be effective must comply with the law of . . . [the situs].” See also Re Fry ]1946] Ch 312.

38 The Situs of Equitable Interests

(D) A Principled Answer Given the difficulty that a court may have in distinguishing between a right to enforce a trust and equitable ownership, it is worth asking whether, in principle, a beneficial interest’s situs should follow that of the legal interest. After all, legal rights are easily identifiable and may be thought to have some form of enduring connection with their locality. Equitable rights are not so and do not obviously have a physical relationship with any country. They may also be enforced by in personam action against the trustee and might be considered to have as close a, if not a closer, relationship with the law of the trustee’s residence, or with the place of administration of the trust. Moreover, the spirit of the Hague Trusts Convention might suggest that little importance should be attached to the physical location of property where equitable interests are in issue. Attribution of the same situs to the equitable interest as the legal interest is also a recipe for difficulty where the property is situated in a non-trust state. Against that, however, a beneficial interest is just that: it is a right carved from the legal property itself. It cannot in that sense be “enjoyed” outside the state where the “legal” property is located. Any purported transfer of the equitable interest which infringes the rules of the situs of the “legal” property may prove ineffective when it is sought to assert equitable rights in respect of that property. On balance, it is submitted that the situs of an equitable interest should be the same as that of a legal interest and not that of the trustee’s residence, whenever the claim is to a proprietary interest in the trust property. It might otherwise present an easy way of avoiding taxation e.g. by appointing trustees resident in Utopia (a tax haven) of property situated in Ruritania. More pertinently still, the beneficial interest may not always be enforceable by in personam order against the trustees. It may be necessary to demonstrate e.g. that a third party transferee in Ruritania took subject to the equitable interest of the beneficiary. The rule should apply wherever the claimant asserts a proprietary entitlement to the trust assets. This should also apply to fixed trusts where the claimant cannot isolate specific assets as belonging beneficially to him (e.g. where he has been left a share of the trust’s income), provided that the assets from which a share is claimed are all located in a single state.165 Many of the “right to enforce” cases considered above are still ones where a property right is, or will ultimately be, asserted and it would be preferable if the physical situs of the assets were adopted, where it is practical to do so. The “trustee residence” rule should be restricted to: (i) residuary interests in unadministered estates; (ii) claims under a fixed trust to a share of the trust assets, where no particular asset can be identified as belonging to the beneficiary and where the physical assets in question are 165 Of course, where he can show entitlement to specific trust assets, it would not matter that those assets were located in different states: their situs would simply differ.

Transfers of Subsisting Equitable Interests 39 not all located in a single state; (iii) the right of an object of a discretionary trust “to be considered as a potential recipient from the fund by the trustees.”166 Where the trustees are resident in different states, it may be preferable to apply the law of the place of administration of the trust to these three matters.167 So much for ascertaining the situs of an equitable interest. That is not, of course, to say that the lex situs is the appropriate choice of law rule to apply where dealings with a subsisting equitable interest are in issue. That question will now be considered.

10. TRANSFERS OF SUBSISTING EQUITABLE INTERESTS

A question left unanswered by the Hague Trusts Convention is what happens where B, the beneficiary of a trust governed by Utopian law, seeks to dispose of his beneficial interest to X, a third party, so that the trustee, T, holds the assets henceforth on trust for X. Two main questions may arise: (i) can B alienate his beneficial interest; and (ii) has title passed to X? The answer to the first question must depend upon the alleged restriction on B’s right to alienate. It may be that the trust instrument provides that B’s interest cannot be transferred. In Peillon v. Brooking,168 the testator precluded his daughter beneficiary from assigning her interest under a trust. When she subsequently purported to do just that, the validity of the transfer was raised. Though not stated in these terms, there are four clear candidate laws which might today be used to determine this issue: (i) the law governing the succession under the will itself, on the basis that the restriction is a “rocket-launching” issue; (ii) the law governing the trust pursuant to which B acquired a beneficial interest; (iii) the law governing the transfer pursuant to which the interest is alienated; (iv) the situs of the assets at the time of transfer. Option (i) should be rejected. “Rocket-launching” questions are concerned with whether the settlor may vest property in the trustee and whether he may create the trust structure with that property. Once a trust has been validly constituted and entered into force, questions concerning the interests of the parties are matters arising under the trust itself. Option (iii) has a certain appeal, given that the law which normally applies to “ordinary” property transfer is that of the last transaction in the property. X might legitimately argue that he should not be concerned with restrictions on alienation imposed on B in the original settlement. However, such an approach would seriously compromise the settlor’s freedom to dispose of his estate as he wished. The beneficiary would be able to evade any restrictions on alienation simply by subjecting the transfer of 166 R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 420; citing Lord Wilberforce in Gartside v. IRC [1968] AC 553, 606. 167 N.B. not the law governing the administration of the trust, since the concern is to apply the law of a place where a beneficiary might naturally sue to enforce his rights. 168 (1858) 25 Beav 218, 218; see also Re Lord Cable [1977] 1 WLR 7.

40 Transfers of Subsisting Equitable Interests his interest to a different law. Option (iv) might be supported on the basis that a transfer of an equitable interest is clearly a property matter and an equitable conveyance demands the application of the lex situs as much as a legal conveyance. However, the lex situs approach suffers from three main defects: first, it also allows a beneficiary to evade a restriction imposed on him concerning alienation of his beneficial interest; secondly, the interest which X is seeking to acquire is B’s interest under a subsisting trust, with a subsisting choice of law; and thirdly, application of the lex situs may be problematic if the trust property is situated in a non-trust state.169 It is suggested that option (ii) should be preferred, as it was in Peillon v. Brooking and the beneficiary’s ability to alienate a subsisting equitable interest should be subjected to the law of the trust under which the interest arose. A similar approach is taken in the conflict of laws to contractual assignments of debts.170 Where a creditor under a contract governed by Utopian law purports to assign a debt to X under a contract governed by Ruritanian law, Article 12(2), Rome Convention states that “the law governing the right to which the assignment relates shall determine its assignability . . .”; in other words, the right to transfer the debt is (in the example given above) governed by Utopian law. As to the second question of whether title has been transferred to X, two candidates emerge: (i) the law governing the original trust; or (ii) the lex situs. A case for the lex situs can be made on the basis that a transfer of a property right is in issue. Indeed, it is argued in this book that the property rights of third party purchasers acting in good faith should normally be governed by the lex situs and not by the law governing the trust.171 However, that argument is made in the case of purported outright transfers of property to the third party. In this case, B and X both have in mind the transfer of B’s subsisting equitable interest under the trust only.172 Since the whole philosophy of the Hague Trusts Convention is not to subject title to equitable interests to the lex situs, the law governing the trust ought to determine whether equitable title passes to the third party. This view also attracts the support of Lewin on Trusts, whose authors state that “The validity and effect of an assignment of an interest under a trust is governed by the governing law of the trust.”173 Likewise, the law governing the formal validity of the original trust will determine whether the transfer of a subsisting equitable interest need be in writing.174 169

Assuming that the equitable situs is the same as the legal situs. See section 7(D), above. 171 See the discussion of Art. 11(3)(d), Hague Convention in Part Two of this book. 172 The position would differ if the beneficiary were to direct the trustee to transfer legal title to X, and at the same time himself transferred equitable title to X (as in Vandervell v. IRC (No 1) [1967] 2 AC 291). In that situation, the trust would terminate. Since the transfer is an outright one, the lex situs should govern it. 173 Lewin on Trusts, 17th edn., 284; relying on Re Fry [1946] Ch 311. It is far from clear, however, that the case cited by the authors lends any real support to Lewin’s proposition. 174 In England, s. 53(1)(c), Law of Property Act 1925 states that; “a disposition of an equitable interest or trust subsisting at the time of the disposition, must be in writing signed by the person disposing of the same, or by his agent thereunto lawfully authorised or by will.” 170

Sub-Trusts 41

11. SUB - TRUSTS

A related question concerns where T holds property for B pursuant to a trust governed by Utopian law. B declares himself in writing to be trustee of his interest for X and states that the sub-trust shall be governed by Ruritanian law. In principle, such a sub-trust is valid, if permitted by Ruritanian law and if satisfying the requisite formalities.175 However, a question may arise as to the relationship between T and X. If, for example, it is alleged that T inappropriately invested the trust fund and thereby caused loss to the trust, X might wish to sue T directly, especially if B has since been declared bankrupt. He might claim that his rights derive from Ruritanian law. However, it is suggested that the relationship between T and X must be subjected to the law governing the “main” trust i.e. Utopian law. Otherwise, T could find that, although he has acted in accordance with the standards of care of Utopian law, he is in breach of trust according to Ruritanian law and that his position has been adversely affected by an arrangement to which he was not party. The position is once again analogous to the assignment of debts. Article 12(2), Rome Convention subjects the relationship between debtor and assignee to the law governing the right to which the assignment relates, and not to the law governing the contract of assignment.176 A further question arises if it is alleged that what appears to be a declaration of sub-trust is in substance a disposition of a subsisting equitable interest. This may be the case if B purports to create a sub-trust, but has no active duties to perform as trustee, so that in reality he drops out of the picture altogether.177 It may be that the two forms of transfer attract different formality rules.178 The law governing the main trust and the law putatively applicable to the sub-trust may take different views on the nature of the transaction. Since, on either view, the transaction involves the carving of an interest from the initial trust, it is suggested that the law governing the main trust should determine whether the transaction is a purported declaration or disposition. If it is the former, it will then be for the law governing the sub-trust to determine its validity.

12. INCOMPLETELY CONSTITUTED TRUSTS AND TRUSTS OF THE BENEFIT OF A COVENANT

Where a trust is incompletely constituted, a would-be beneficiary may wish to rely upon one of the exceptions to the rule that equity will not assist a 175 For example, in England, s. 53(1)(b), Law of Property Act 1925 states that “a declaration of trust respecting any land or any interest therein must be manifested and proved by some writing signed by some person who is able to declare such trust or by his will.” 176 See further section 7(D), above. 177 Grainge v. Wilberforce (1889) 5 TLR 436; re Lashmar [1891] 1 Ch 258; Grey v. IRC [1960] AC 1. 178 In English domestic law, see ss. 53(1)(b) and (c), Law of Property Act 1925. The former applies to declarations of trusts of land; the latter to dispositions of subsisting equitable interests.

42 Incompletely Constituted Trusts/Trusts of the Benefit of a Coventant volunteer.179 Suppose that S promises T in writing that he will make an inter vivos transfer to him of movable property located in England on trust, to hold for B, a volunteer. S expressly states that the trust will be governed by Utopian law. S never constitutes the trust by vesting legal title to the property in T. It is submitted that, if B tries to invoke one of the exceptions to the volunteer rule in order to constitute the trust, the availability of those exceptions should be subjected to the law applicable to the putative transfer of property, i.e. the lex situs in the case of inter vivos transfers, English law. The matter principally180 concerns the vesting of title in T on trust. It is a “rocket-launching” matter and accordingly it would not be appropriate to apply Utopian law, the law applicable to the putative trust. However, if it is alleged that the incompletely constituted inter vivos trust is constituted upon S’s death,181 it is less clear whether the law applicable to inter vivos transfers or the law applicable to succession (lex successionis) should determine whether an exception to the rule that “equity will not assist a volunteer” is satisfied. In re Korvine’s Trust,182 in the context of donatio mortis causa, Eve J remarked that: “No doubt the gift is consummated by the death of the donor, but it is not a legacy or a testamentary act—it is I think a gift, though an incomplete gift, inter vivos—a gift subject to revocation by the act of the donor himself or by his recovery from his illness. If he dies without revoking the gift, the donee’s title is derived from the act of the donor in his lifetime and relates back to the date of that act. On these grounds, and notwithstanding that the subject matter of the donatio is liable to the donor’s debts upon a deficiency of assets and is also subject to legacy and estate duty, I think the law to be applied is that applicable to gifts inter vivos and not that applicable to testamentary dispositions, and I so hold.”

This view is supported by Miller, who argues that whilst donatio mortis causa is sui generis, it is a form of inter vivos gift.183 Although this might be said to be the predominant view, it is suggested, with respect, that it should not be followed and that the lex successionis ought rather to be applied. True, the act of constitution refers back to the trust purported to be created by the settlor in his lifetime. However, it is upon the settlor’s death that the donatio principle may take effect and imposes an obligation on the executors or administrators to perfect the donee’s title. The transfer takes place upon death and the law with the greatest legitimate interest in whether property can and has been alienated on 179 On which see R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn., 176–87. 180 Of course, a valid trust will also vest equitable title in the beneficiary. However, this is contingent on the constitution of the trust by the transfer of legal title to the trustee. 181 As with e.g. the rule in Strong v. Bird (1874) LR 18 Eq 315 or donatio mortis causa (on which, see J McGhee, Snell’s Equity, 30th edn. (London, Sweet and Maxwell, 2000), ch. 20). 182 [1921] 1 Ch 343, 348. See also Re Craven’s Estate [1937] 1 Ch 423; Emery v. Clough 63 NH 552, 4 A 796 (1885). 183 G Miller, International Aspects of Succession (Aldershot, Ashgate, 2000), 232. He cites the view of Smith J in Emery v. Clough, ibid., in support.

The Need for an English Grant 43 death is the lex successionis. In the case of movables, the law of the deceased’s domicile at death should be applied.184 Returning to our example above, the law applicable to the putative property transfer, English law should also determine in the case of incompletely constituted trusts, whether a fully constituted trust of the chose in action, i.e. the benefit of the promise, has been created.185 However, if it rules that such a trust has arisen, the law applicable to that trust should be Utopian law, the law putatively applicable to the trust of the assets themselves. It would be most unfortunate if a trust of the benefit of a promise and a trust of the assets themselves were governed by different laws. 13. TESTAMENTARY TRUSTS : ADMINISTRATION OF ESTATES

The law which governs the administration of an estate is a broad subject. It is not proposed to examine this area in detail here. However, since it may impact upon the creation of a testamentary trust, a short summary of the basic choice of law rules is provided.186 (A) The Need for an English Grant English law requires an executor appointed under a will to obtain a grant of probate before he may deal with property. Where the deceased died intestate a grant of letters of administration must be sought by a family member or other suitably interested person. Only once the estate’s liabilities have been cleared by the administration process may the administration of the estate according to the terms of the will, or the rules of intestacy, occur. Civil law states do not necessarily require the intermediate stage of administration of the estate and dealing with outstanding liabilities before distribution of the testator’s assets to those entitled under the will or upon intestacy may take place.187 Nonetheless, a grant must still be obtained from an English court. The position is likewise even if an administrator has already been appointed in the state of the deceased’s domicile.188 Unless and until a foreign administrator 184 On one view, this is what Farwell J decided in Re Craven’s Estate. However, the case is also consistent with application of English law qua lex fori: see G Miller, ibid, 230. See also section 14, below. 185 See Fletcher v. Fletcher (1844) 4 Hare 67; re Schebsman [1944] Ch 83; R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 185–7. This area of law is today of reduced importance in English Law: see s. 1, Contracts (Rights of Third Parties) Act 1999. 186 See Cheshire and North, ch., 32; Dicey and Morris, ch. 26; G Miller, International Aspects of Succession, ch. 9. The United Kingdom signed (on 2 October 1973), but has not ratified, the Hague Convention on the International Administration of the Estates of Deceased Persons 1973 (in force in the Czech Republic, Portugal and Slovakia). 187 Re Achillopoulos [1928] Ch 433. 188 New York Breweries Co Ltd v. A-G [1899] AC 62. A foreign administrator will, however, be personally liable if he obtains assets situated in England without an English grant as executor de son tort.

44 Testamentary Trusts: Administration of Estates obtains an English grant, his authority to deal with the estate will not be recognised in England.189 In practice, this means that an English grant must be sought whenever there are assets located in England.190

(B) The Jurisdiction of the English Courts The English court’s jurisdiction to make a grant is not limited to where there are assets located within the jurisdiction.191 However, where the deceased leaves no English assets and does not die domiciled in England, the English court is likely to decline to make the grant.192 Where the testator leaves two separate wills in respect of his English assets and his overseas assets, the English court will normally grant probate only in respect of the former. Where the deceased dies domiciled in England, an English court will normally make a grant to the named executors, or, upon intestacy, grant letters of administration to a person having a beneficial entitlement to the estate. Where the deceased dies domiciled abroad, the English court attaches great store to the law of the deceased’s last domicile. It will normally make a grant to the person entrusted with the estate’s administration in the courts of the deceased’s last domicile.193 Where no such grant has been made, an English grant will normally be made to a person beneficially entitled to the estate according to the law of the deceased’s last domicile.194 However, the matter is in the court’s discretion and it may appoint someone different if it sees fit.195 These rules may be thought to achieve some uniformity of outcome with the lex successionis. However, Cheshire and North point out that the position is less readily justifiable where the estate contains movable and immovable property, since succession to the latter will be governed by the lex situs and not by the law of the deceased’s last domicile.196 It appears that the English court will continue to follow the law of the deceased’s last domicile,197 however, unless the estate 189 He will not be allowed to sue in England in his capacity as foreign administrator. He may, however, sue in his personal capacity to enforce a judgment obtained against a debtor to the estate: Vanquelin v. Bouard (1863) 15 CBNS 341. 190 New York Breweries Co v. AG [1899] AC 62. 191 S. 25(1), Supreme Court Act 1981 (replacing s. 2(1), Administration of Justice Act 1932). The administrator appointed by an English grant will be accountable in an English court even in respect of property situated overseas; whether the courts where the property is situated will recognise the English grant is, of course, a quite separate matter. 192 Aldrich v. AG [1968] P 281, 295. 193 Non-Contentious Probate Rules 1987, r. 30(1)(a). It does not matter upon which basis the foreign court appointed the administrator, provided that the administrator has capacity by English law. 194 Non-Contentious Probate Rules 1987, r. 30(1)(b). However, if a testator domiciled overseas writes a will in English naming an executor, the English court will normally grant probate to that person: Non-Contentious Probate Rules 1987, r. 30(3)(a)(i). 195 Non-Contentious Probate Rules 1987, r. 30(1)(c); In the Goods of Kaufman [1952] P 325. 196 Cheshire and North, 978. See s. 14, below. 197 In the Goods of Meatyard [1903] P 125.

Grants in Scotland, Northern Ireland and the Commonwealth 45 consists principally of immovable property in England, when it will appoint an administrator in accordance with the rules of the law of the forum (lex fori) and not those of the deceased’s last domicile.198

(C) Grants obtained in Scotland, Northern Ireland and the Commonwealth Where a grant is obtained in Scotland or Northern Ireland in respect of a person domiciled in that state, the grant will be effective in England without the need for resealing.199 Commonwealth grants made in states to which the Colonial Act 1892 has been extended by Order in Council may be sealed by the English probate registry and become effective in respect of English assets. This applies even if the Commonwealth state in question was not that of the deceased’s last domicile. However, the English court’s decision whether to seal is discretionary.200

(D) Applicable Law The administration of the estate is governed by the law of the state where the administrator obtained his grant.201 Where an English grant is made, the administrator must settle all liabilities existing according to English law, whether those liabilities arise in England or overseas. This must be done even if the English grant is ancillary to a foreign grant. Once the liabilities have been settled, however, the English administrator may, at its discretion, remit the assets to the principal administrator in the state of the deceased’s last domicile.202

(E) Delimiting the Administration of Estates and the Trust It may sometimes be difficult to draw a line at a precise point where the administration of an estate ceases and the operation of the trust begins. This is illustrated by the judgment of Farwell J in re Wilks.203 He decided that an English administrator could exercise the power contained in section 33, Administration of Estates Act 1925 to postpone a sale of trust assets to which infants were entitled, notwithstanding that the ensuing trust would not be governed by English law. He ruled that: 198

Non-Contentious Probate Rules 1987, r. 30(3)(b). S. 1, Administration of Estates Act 1971. The converse is true for an English grant’s effect in Scotland and Northern Ireland (ss. 2(1) and 3(1)). 200 The court will normally only seal the order where it is granted to the person who would have been appointed according to the law of the testator’s last domicile. 201 Preston v. Melville (1841) 8 Cl & Fin 1, 12. 202 Re Achillopoulos [1928] Ch 433; Re Lorillard [1922] Ch 638. 203 [1935] Ch 645. 199

46 Testamentary Trusts: Succession “The question, therefore, is at what point the administrators cease to administer the estate in this country and become trustees for the purpose of distributing the estate amongst the persons interested. So long as they are performing the duties of administrators their powers and rights are to be found in the Administration of Estates Act 1925. . . it appears to me that this is a matter of administration and that the administrators are entitled to exercise the power of postponement which is given them by the Act. They have in their hands money which, no doubt, is the property of the infants, but in respect of which they can get no receipt from the infants. . . and until they can distribute the estate by handing over the proceeds or the property to persons who can give them a good receipt for it they are trustees, but holding in their character as administrators and having, therefore, the powers which are given to administrators under s. 33 of the Administration of Estates Act 1925”.204

Such a broad interpretation may seriously impact on the operation of the Hague Trusts Convention. If the administration of an estate does not cease until receipt can be given, then the period throughout which English powers of administration might operate on a foreign trust is unexpectedly long. Moreover, in re Kehr,205 Danckwerts J ruled that an English-appointed administrator of an estate could appoint trustees of a foreign trust and that the trustees, once appointed by English law, had the statutory powers of maintenance and advancement contained in sections 31 and 32, Trustee Act 1925. It is hoped that these cases would be decided differently today. Articles 8(2)(a) and (d), Hague Convention clearly indicate that the appointment of trustees and their powers of administration are matters for the applicable law of the trust. Nor is there any obvious reason why the administration of estate provisions should be treated as mandatory rules to be superimposed onto the applicable law.206

14. TESTAMENTARY TRUSTS : THE LAW APPLICABLE TO SUCCESSION

It goes without saying that the law applicable to succession is a complex and large subject and that a full discussion of the matter is not possible or appropriate here. However, it is also apparent that it is one of the most important preliminary issues which may arise in connection with testamentary trusts. For this reason, a summary of the law applicable to succession claims is provided here.207 204

[1935] Ch, 648–51. [1952] Ch 26. 206 One would normally expect that issues relating to the administration of estates would be preliminary matters necessary to launch the trust and that they would fall outside the scope of the Convention by Art. 4. Where they affect the operation of the trust itself, however, they would be within the Convention and Arts. 15 and 16 could be applicable. 207 See Cheshire and North, ch. 33; Dicey and Morris, ch. 27; D Hayton, “The Problems of Diversity”, in D Hayton (ed.), European Succession Laws (Bristol, Jordans, 1998), ch. 1; J Denker, “The British Isles: England”, ibid. ch. 4, esp. 79–80; G Miller, International Aspects of Succession, esp. chs. 5 and 6. 205

Testate Succession 47

(A) Testate Succession (i) Movables (1) Essential validity Whilst the lex situs rule holds sway for inter vivos transfers of property, the essential validity of a testamentary disposition of movables is governed by the law of the deceased’s last domicile.208 To the extent that rules of forced heirship209 and perpetuities210 fall outside the Convention, it will be this law that will apply to such matters. In Re Annesley,211 a British national who died domiciled in France had made an English will. The testator failed to leave two-thirds of her estate to her children, as was required by French law. By French law, the will was valid only as to the remaining one-third share, so that the respective shares of those entitled under the will would be reduced proportionately. The English court applied the law of the deceased’s last domicile, French law, to determine the succession rights of the forced heirs and legatees to the testator’s movable property.212 Once again, it should be stressed (consistently with the approach advocated for inter vivos dispositions) that the lex successionis should answer the question whether the will is valid, and whether legal title to assets may be alienated by will. It should not determine whether equitable title may pass to the beneficiary. Otherwise, a lex successionis which does not recognise the trust concept could undermine the Hague Trusts Convention, even though the Convention makes it plain that trusts may be validly created over assets located in non-trust states. For this reason, this author would reject the view of Koppenol-Laforce that where Dutch law is the lex successionis “The crucial question is whether Dutch succession law has a closed system of possible testamentary dispositions on death. If so, then a trust is not possible”.213 Rather, the crucial question is simply whether Dutch law allows the testator to dispose of his property at all. (2) Construction The importance generally attached to the law of the testator’s domicile is inappropriate where the meaning of words contained in the will is concerned. This 208

Re Groos [1915] 1 Ch 572; Re Annesley [1926] Ch 692. Re Ross [1930] 1 Ch 377. See section 17, below. 210 See section 8, above. 211 [1926] Ch 692. 212 Complex problems of renvoi were raised; but the important point for present purposes is that the choice of law rule applied by the English court was that of the deceased’s domicile at death. 213 M Koppenol-Laforce, Het Haagse Trustverdrag, 272. Fortunately, Koppenol-Laforce concludes that Dutch succession law does permit the creation of a testamentary trust, (subject to any mandatory succession rules of Dutch law): ibid., 275. However, the author also notes that since Dutch law requires the heir to exist at the time of the testator’s death, there may be difficulties if the “heir” is considered to be the beneficiary for these purposes. A beneficiary not born at the time of the testator’s death could not benefit under the will. A better view may be that the trustee is the “heir” for the purposes of Dutch succession law: ibid, 275. 209

48 Testamentary Trusts: Succession is a question of discerning what the testator intended the will to mean. Construction should accordingly be governed by whichever law the testator intended to govern it: and in the absence of stipulation, this is likely to be the law of his domicile at the time of making the will. Certainly, the law of his domicile at death does not hold sway. Section 4, Wills Act 1963 expressly provides that the construction of a will shall be unaffected by a change in the testator’s domicile after the date of execution of the will. (3) Capacity The capacity of a testator to make a will is governed by the law of his domicile. However, it is uncertain whether this refers to his domicile at the time of making the will or to his domicile at the time of death. The former may seem preferable from the standpoint of legal certainty; otherwise a will valid at its time of creation may be invalid at the time of death due to a change in the testator’s capacity. On the other hand, the same is true of the essential validity of a will, which is clearly determined by the deceased’s domicile at the time of death. The will is ambulatory until death and, upon death, is likely to be administered in the place where the testator is then domiciled. On balance, however, the domicile at the time of creation of the will is preferred as a choice of law rule for capacity. As Cheshire and North point out,214 a testator requires capacity at the time of creation of the will; a will purported to be created by a minor will not be validated by subsequent events. A legatee must have capacity to receive a testamentary disposition. It would appear that he need only have capacity either by the law of his domicile at the point when he claims the legacy or the law of the testator’s domicile at death.215 (4) Formal validity The Wills Act 1963 gives effect to the Hague Convention on the Conflict of Laws Relating to the Form of Testamentary Dispositions 1961. It is extremely liberal in the formality rules which it lays down. Section 1 provides no fewer than seven alternative laws, compliance with any one of which will be sufficient. These seven are: the law of the place where the will was made; or the law of the testator’s domicile, nationality or habitual residence either at the time when he made the will or upon his death.216 Any alteration in the relevant law after the date of making the will should be given effect if it validates a will, but not if it invalidates it.217 (5) Revocation Since a will may be revoked in a number of ways, it is difficult to lay down a single choice of law rule to determine whether this has occurred. 214

Cheshire and North, 986–7. Re Hellman’s Will (1866) LR 2 Eq 363. 216 These are the domestic laws of the states identified. The application of renvoi is excluded: s. 6(1). 217 S. 6(3), Wills Act 1963. 215

Testate Succession 49 (α) Subsequent will or codicil A later will may revoke an earlier one. Section 2(1)(c), Wills Act 1963, provides that the second will is effective to revoke the first if the new will satisfies any of the laws specified in the Act as applicable to its formal validity, or any of the laws applicable to the original will’s formal validity. (β) Act of revocation Suppose that a testator makes a will and subsequently destroys it when domiciled in state A. He then acquires a domicile in state B, which remains his domicile at the time of death. Is the effectiveness of the purported revocation determined by the law of the testator’s domicile at the time of the act (state A) or at the time of death (state B)? There is no clear authority, but the former view is much to be preferred. As Cheshire and North state, “The legal effect of the act of destruction fell to be determined at the time of its performance. . . The mind recoils from the suggestion that the effect of such an act might be nullified or changed merely because at some later period in his life he happened to be subjected to a different legal system”.218 (γ) Subsequent marriage In English law, a will is revoked by marriage unless demonstrated to have been made in contemplation of marriage.219 For choice of law purposes, revocation in this manner could be regarded as part of the law of succession or of matrimonial law. In Re Martin,220 it was held to be the latter and accordingly governed by the law of the testator’s domicile at the time of marriage. (ii) Immovable Property (1) Essential validity Succession to immovables is governed by the lex situs, the state with exclusive control over the property.221 (2) Construction Since construction is essentially a matter of determining the settlor’s intentions, the lex situs need not be applied. It will be presumed that the testator intended construction to be referred to the law of his domicile at the time of making the will.222 However, if the mooted construction is illegal by the lex situs, that latter law should prevail, it being somewhat pointless for an English court to rule otherwise.223 218

Cheshire and North, 997–8. S. 18, Wills Act 1837, as amended. [1900] P 211. 221 Nelson v. Bridport (1846) 8 Beav 547; Freke v. Carbery (1873) LR 16 Eq 461; Duncan v. Lawson (1889) 41 Ch D 394; Re Ross [1930] 1 Ch 377. 222 Philipson Stow v. IRC [1961] AC 727. 223 Re Miller [1914] 1 Ch 511. 219 220

50 Testamentary Trusts: Succession (3) Capacity This should be determined by the lex situs, since it is again rather futile for an English court to determine that a disposition is valid by the law of the testator’s domicile if the lex situs regards the testator as lacking capacity and refuses to recognise the title of the legatee.224 (4) Formal validity The same liberal rules apply under the Wills Act 1963 as for movables, but yet another alternative is provided: namely that compliance with the formal requirements of the lex situs will suffice.225 (5) Revocation (α) Subsequent will or codicil It would appear that the position is the same as for wills of movables and governed by the Wills Act 1963. (β) Act of revocation Binding authority is lacking, but the effectiveness of such a purported revocation is probably governed by the lex situs.226 (γ) Subsequent marriage The matrimonial characterisation of this issue in relation to movables might suggest that the law of the testator’s domicile at the time of marriage should apply. Although the authorities are not clear cut,227 Cheshire and North favour application of this law.228

(B) Intestate Succession (i) Movables In the event of whole or partial intestacy, the law of the deceased’s domicile at the time of his death determines the devolution of his estate.229 Indeed, it would be curious if any other law were to govern, since an individual may have decided 224 Re Hernando (1884) 27 Ch D 284. But see Jaffey on the Conflict of Laws, 2nd edn., who suggests (at 476) that capacity to dispose of land in England might rather be governed by the law of the testator’s domicile. The authors readily admit that their view is not shared by Dicey and Morris (at 1030) or by Cheshire and North (at 1001). 225 S. 2(1)(b); this is a reference to the domestic law of the situs. 226 See the discussion of the American case of Re Barrie’s Estate 240 Iowa 43, 35 NW 2d 658 (1949) in Cheshire and North, 1007. 227 Re Caithness (1891) 7 TLR 354; cf. the Australian decision in Re Micallef’s Estate [1977] 2 NSWLR 929. 228 Cheshire and North, 1007–8. 229 Pipon v. Pipon (1744) Amb 799; Re Collens [1986] Ch 505.

Renvoi and Succession 51 not to make a will on the basis that his estate would devolve according to the law of his domicile. It should be noted that this rule is one of succession. Where property is found to be bona vacantia, title to it will be determined by the lex situs and not by the law of the deceased’s last domicile.230 But if the Crown takes property as the ultimate heir, rather than on the basis that it is ownerless property, that is a rule of succession and the law of the deceased’s last domicile will apply.231 Accordingly, fine questions as to whether a state’s claim to property is based on bona vacantia or succession must be made, in order to determine the applicable law. (ii) Immovables Intestate succession to immovables will be governed by the lex situs.232

(C) Renvoi and Succession The doctrine of renvoi is applicable to succession both to movable233 and immovable property.234 The notable exception is formal validity, where the application of renvoi is excluded by section 6(1), Wills Act 1963. This may be highly significant where the lex successionis would apply a different choice of law rule to that applied in an English court. This may be especially likely where succession to immovables is concerned, which might be determined by personal connecting factors in some states, rather than by the law of the situs. If a testator, domiciled and habitually resident in England and a British national, leaves property situated in Utopia, a Utopian judge might refer the matter back to English law. The decisive question then becomes whether a 230

Re Barnett’s Trusts [1902] 1 Ch 846. Re Maldonado’s Estate [1954] P 223. 232 Re Collens [1986] Ch 505. But see Jaffey on the Conflict of Laws, 2nd edn., 479, where the authors argue that the law of the deceased’s last domicile would be a better rule where property located in England is concerned, on the basis that English law has little interest in precisely how property is carved up between the intestate’s relatives. 233 Re Annesley [1926] Ch 692; Re Ross [1930] 1 Ch 377; Re O’Keefe [1940] Ch 124. Lewin on Trusts, 17th edn., 284 advocates the elimination of the doctrine in respect of succession to movables. It might be argued that there is no obvious need to apply the law of a testator’s last domicile exactly as it would be applied in his home courts. However, this author would favour the preservation of renvoi, which promotes unity of outcome. Without it an English court might apply a different law to the courts of the testator’s domicile, creating a recipe for a conflict of laws and inviting the claimant to forum shop. It is interesting also to note that Switzerland applies a doctrine of renvoi pursuant to Art. 91(1) of its Private International Law Statute (which came into force on 1 January 1989 and is known as the “Loi Federale sur le Droit International Privé”) in respect of deceased persons not domiciled in Switzerland and refers to the law designated by the private international rules of the state of his domicile: see A Burnand, “Laws of Succession and Testamentary DispositionsConflicts of Laws and Recognition of Foreign Trust—a Swiss Overview” (1995) 1(10) Trusts and Trustees 15, 16. 234 Re Ross [1930] 1 Ch 377; Re Duke of Wellington [1947] Ch 506. 231

52 Capacity to Create a Testamentary Trust Utopian judge would accept the renvoi back from English law to Utopian law. If yes, then an English court should apply Utopian law; if not, the English court should apply English law. The difference may be critical, especially as regards forced heirship provision. It follows that: “Properly to assist a client with his estate planning this requires ascertaining the client’s likely last domicile or habitual residence or nationality, discovering which States apply domicile or habitual residence or nationality, and ascertaining the situs of the client’s assets. It is then important to discover which States are ‘schismatic’ (like England, France and Belgium) in having one law governing succession to movables and another governing succession to immovables, and which States are ‘unitarians’ in having one lex successionis for movables and immovables”.235

15. CAPACITY TO CREATE A TESTAMENTARY TRUST

The analysis of capacity in relation to inter vivos trusts above subjects the settlor’s capacity to alienate a property right to the law applicable to “ordinary” rules applicable to capacity to dispose of property. By analogy, in the case of testamentary transfers, this would lead to the application of the law of the testator’s domicile (probably at the time of creation of the will), in the case of movables, and to the application of the lex situs in the case of immovables, to determine whether the settlor has capacity to alienate his property by will at all. If he does, then by analogy to the inter vivos transfer, the question whether the testator has capacity to create the trust structure by will should be determined by the proper law of the trust. Hence, the following choice of law rule is suggested: (a) The capacity of a testator to dispose of any property right by will is governed: (i) in the case of movable property, by the law of that party’s domicile at the time of making the will; (ii) in the case of immovable property, by the lex situs. (b) The capacity of a testator to create a trust of any property of which he may, according to rule (a), dispose, is governed by the proper law of the trust.

16. ESSENTIAL VALIDITY AND TESTAMENTARY TRUSTS

(A) General Rule Where a trust is created by will, the executor will have to pass legal title to the trustee. As argued throughout this chapter, the “ordinary” choice of law rules 235 D Hayton, “Cross Border Estates: Part 1: an Introduction” [1994] Private Client Business 168, 168–9. See also “Cross Border Estates: Part 2: the Problems of Diversity”, 250, where the application of the renvoi doctrine is considered in more detail.

Executor Trustees 53 should apply to determine the essential validity of a property transfer. Although the transfer to the trustee does not take place immediately upon the deceased’s death, it is transferred pursuant to a will, so should be subject to the property rules on testamentary transfers. Hence the law of the testator’s last domicile (for movables) or the lex situs (for immovables) should be applied. That law will determine whether legal title236 vests in the trustee. However, that law should not be applied to the question whether equitable title has vested in the beneficiary and a testamentary trust should not be rendered invalid merely because the lex successionis does not know the trust concept. The choice of law rule to determine the essential validity of the vesting of equitable title in the beneficiary should be determined by the proper law of the trust (which may, of course, be chosen by the settlor). Hence, the following choice of law rule is suggested. (a) Whether any proprietary interest has been transferred by will from a would-be settlor to a would-be trustee is governed: (i) in the case of movable property, by the law of the testator’s domicile at the time of death; (ii) in the case of immovable property, by the lex situs. (b) Whether the disposition is effective to transfer equitable title to the wouldbe beneficiary, and thus to form a completely constituted trust, is determined by the proper law of the trust.

(B) Executor Trustees A particular problem may arise where the executor of a will is also named as trustee. Lewin on Trusts suggests that: “If the will makes the executors as such into trustees, it is not thought that there is any transfer or act by which the assets are transferred to them other than the will itself, whose validity is conclusively established by the probate, even though the executors hold the assets (after they have cleared the estate and assented to the gift) in the capacity of trustees rather than executors.”237

It is not certain whether Lewin’s view should be followed. The assets will have been duly transferred to the executor in his capacity as executor and there will be an “alienation” of property to him in the role of trustee. It is a form of alienation 236 Or, more accurately, whether the subject-matter of the trust vests in the trustee (since the subject-matter of the trust may be a subsisting equitable interest). 237 Lewin on Trusts, 17th edn., 278. There might be an analogy to be drawn with the comments of Buckley J in Re Ralli’s Will Trusts, [1964] Ch 288, 301, who remarked that “In my judgment the circumstances that the plaintiff holds the fund because he was appointed a trustee of the will is irrelevant. He is at law the owner of the fund and the means by which he became so have no effect on the quality of his legal ownership.” However, that case concerned the constitution of a marriage settlement by the vesting of the property in the trustee of a will. In the instant case, the executor is yet to become trustee.

54 Forced Heirship of a legal estate; and where legal estates are concerned, it is attractive to require compliance with the “ordinary” choice of law rules applicable to property transfers (especially in the case of immovable property). Hanbury and Martin comment on the position in English law thus: “The principle is that the transition from executors to trustees occurs automatically after completion of the administration, but as far as powers of disposition of property are concerned, there must also have been a sufficient assent by the executors in their own favour as trustees”.238

In the case of land, an executor must vest title in himself qua trustee by means of a written assent.239 On balance, it is suggested that Lewin’s view should be followed in part. The vesting of the legal estate in the trustee qua trustee should not be subjected to a further “rocket-launching” choice of law rule. It has been argued in this Part that only issues which affect the alienation of property at all should be subjected to such rules. In this case, property will already have vested in the executor and the vesting in him of legal title qua trustee is a change in the specific capacity in which he holds it. However, that specific capacity in which the property is now held is a matter in which the proper law of the trust has a legitimate interest.240 Accordingly, it is suggested that the proper law of the trust should determine whether legal title to the property held by the executor has been duly vested in the trustee qua trustee.

17. FORCED HEIRSHIP

The characterisation of forced heirship claims is controversial.241 The key issue is whether a claim affects the transfer of legal title to the trustee, or whether it affects the trust itself. Insofar as it affects the ability to create the trust structure only (because it does not render the alienation of legal title to the trustee void),242 238 J Martin, Hanbury and Martin, Modern Equity, 16th edn. (London, Sweet & Maxwell, 2001), 57, considering Attenborough v. Solomon [1913] AC 76. See also s. 36(4), Administration of Estates Act 1925; s. 40, Trustee Act 1925. 239 Re King’s Will Trusts [1964] Ch 542; A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn. (London, Sweet & Maxwell, 1998), 30–1. 240 Compare Art. 8(2)(a), Hague Convention. 241 See the discussion in Part Two of this book of Arts. 4 (“distinguishing between the ‘rocket’ and the ‘rocket-launcher’”) and 15 (“Article 15 and the rocket-launcher” and “forced heirship”). 242 Hayton suggests that if it renders the transfer of legal title to the trustee voidable, it may be said that, in principle, the rocket was launched and it may be better to treat a forced heirship claim as a “rocket” matter, (at least, one might possibly add, insofar as the claim renders voidable only part of, and not the entirety of, the disposition). It should be borne in mind that, in the context of inter vivos trusts, there will probably be no scope for applying forced heirship rules under the Hague Trusts Convention which render a trust voidable; this is because if forced heirship rules are characterised as rules of succession, an English court will apply such succession rules only to assets which it regards as forming part of the testator’s estate on death; and where is considers those assets to have been validly transferred on inter vivos trust, it will not apply the lex successionis to those assets.

Forced Heirship 55 it falls within the Hague Trusts Convention and the mandatory rules of the law applicable to succession may be applied pursuant to Article 15(1)(c).243 But insofar as it restricts the alienation of property at all to the trustee, it should be treated as concerning the “rocket-launcher” and outside the Convention. In the latter case, since forced heirship rules are most naturally to be classified as rules of succession,244 the law governing succession to the deceased’s estate245 should determine whether, and to what extent, a testamentary disposition should be restricted by forced heirship rules.246 As Scoles and Hay note, rules of forced heirship and family protection: “. . . reflect social policies reflective of presumed contribution, probable needs for support, and concepts of family fairness, all policies centering on the family, and its maintenance as a social and economic unit. In the conflict-of laws setting, these policies strongly suggest the predominant relationship to, and concern of, the state of the decedent’s domicile, the usual center of family life”.247

Hence, forced heirship claims to movables will be governed by the law of the testator’s last domicile. Where forced heirship claims relate to immovables, the lex situs should be applied.248 It should be stressed that a validly created inter vivos trust will be safe, even if the lex successionis decrees that it is subject to retroactive claw-back.248a This is because English law applies the law governing succession only to property which it regards as forming part of the settlor’s estate on death. Where property

243

See the discussion of Arts. 4 and 15(1)(c) of the Convention. See A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, 32, who considers other possible classifications, but concludes that the succession classification is the most likely. However it is not easy to reconcile this with his discussion of the direct enforcement of the heir’s rights against the trustee or beneficiary where he argues (at 36), that “The rights asserted by the heir might have arisen in consequence of the settlor’s death and in that respect have the characteristics of succession rights, but they could hardly be described as rights in respect of the settlor’s property.” In this author’s view, however, such a claim should clearly be classified as a succession claim. The forced heir may be claiming against the trustee, but he is in substance asserting that he has a better claim to the testator’s assets on death than the legatee and that the legatee does not succeed to that portion of the estate. 245 And not just the mandatory rules of the law of succession. 246 A view supported by Re Annesley [1926] Ch 692. See also E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1018–21; Bullen v. Wisconsin 240 US 625, 36 Sup Ct 473; In Re Estate of Clark 21 NY 2d 478, 288 NYS 2d 993; contrast In Re Estate of Renard 56 NY 2d 973, 453 NYS 3d 625 (considered by M Lupoi, Trusts: A Comparative Study, 152). 247 Ibid., 1021. 248 For the possibility of claims by forced heirs in tort and unjust enrichment, see A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, 21, 36–9. For discussion of the choice of law rules applicable to succession in France, see M Revillard in D Hayton (ed.), European Succession Laws, ch. 7, 185. 248a Although, upon the settlor’s death, the law applicable to succession may decree that any dispositions made to a person during the deceased’s lifetime should be taken into account in determining the extent of their entitlement to a forced share of the deceased’s estate. Indeed, there is a presumption that lifetime payments should be brought into account for these purposes in Guernsey law: see the Royal Court of Guernsey’s decision in Funk v. Krombach (2001) 15 Trust Law International 103, (considering s. 11A, Trusts (Guernsey) Law 1989, as amended). 244

56 Hague Succession Convention 1989 has been validly disposed of inter vivos, it will not be regarded as part of the estate and the lex successionis will have no say in the matter.249 The same may not be said if proceedings take place in the courts of a nontrust state with claw-back provisions: “The practical lesson is that trustees must not allow assets that can be traced to the trust to be found in Nontrusto . . . nor must they themselves become subject to the jurisdiction of Nontrusto if such jurisdiction could make orders compelling them to bring trust assets within the jurisdiction”.250

18. THE POTENTIAL IMPACT OF THE HAGUE CONVENTION ON THE LAW APPLICABLE TO SUCCESSION TO THE ESTATES OF DECEASED PERSONS OF

1 AUGUST 1989

(A) General This Convention was intended to complement the Hague Trusts Convention and to provide a degree of harmonisation in an area of law where the connecting factor or factors used vary from state to state.251 However, it has not been ratified by the United Kingdom and has not entered into force.252 Largely, this is due to the ambitious scope of the project and the major departure that it brings from the common law.253 It essentially provides for uniformity of choice of law, whether the property be immovable or movable, and allows the settlor to choose between the law of his habitual residence254 and nationality at the date of making the will.255 There is a prevailing choice of law, but this is subject 249 See further the discussion in section 18, below and of Art. 15(1)(c) in Part Two of this book. See also Re Mégret [1901] 1 Ch 547. 250 D Hayton, “Cross Border Estates: Part 3: the Hague Conventions” [1994] Private Client Business 329, 334. 251 See J Schoenblum, “Choice of Law and Succession to Wealth: a Critical Analysis of the Ramifications of the Hague Convention on Succession to Decedent’s Estates” (1991) 32 Vanderbilt Journal of International Law 84; D Hayton in E Pluribus Unum, 121; D Hayton, [1994] Private Client Business 329, 329–30; C Robertson, “International Succession Law” (1995) 2 Trusts and Trustees 20; G Miller, International Aspects of Succession, 285–90. On the succession laws of national states in Europe, see D Hayton (ed.), European Succession Laws. See also M Oldham, “Financial Obligations within the Family-Aspects of Intergenerational Maintenance and Succession in England and France” [2001] CLJ 128. 252 See the Lord Chancellor’s Department Consultation Paper, 1990. The Convention has been signed by Argentina, Luxembourg and Switzerland, but only signed and ratified in the Netherlands (but has not entered into force therein). 253 D Hayton describes this Convention as “much ado about a lot” in E Pluribus Unum, 121, 128, in contrast to the Hague Trusts Convention which has been described as “much ado about very little” by J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 1 J Int Corp P 5. 254 A term which may need clarification before ratification might occur in England: see D Hayton, “Cross Border Estates: Part 2: the Problems of Diversity” [1994] Private Client Business 250. 255 Art. 5(1).

Forced Heirship and Claw-back 57 to the possibility of the settlor selecting another law to deal with assets in a particular jurisdiction.256 In the absence of choice, the objectively applicable law is that of the settlor’s habitual residence and nationality at the time of death;257 if different, it will be his habitual residence, if he lived there for at least five years prior to his death;258 and if this is not satisfied, the law of the settlor’s nationality applies.259 It has been pointed out that ratification of this Convention in the United Kingdom would necessitate not just a change in choice of law rules, but e.g. modification of the English Inheritance (Provisions for Family and Dependants) Act 1975, which is stated to apply to persons dying domiciled in England.260 (B) Forced Heirship, Claw-back and the Hague Trusts and Succession Conventions One aspect of the Succession Convention might be thought to be of particular concern for trusts lawyers: namely the impact of forced heirship claims on inter vivos trusts. Article 7(2)(d), Hague Succession Convention includes as a matter for the applicable law: “the disposable part of the estate, indefeasible interests and other restrictions on dispositions of property on death.” The concern for trusts lawyers is that an inter vivos trust of assets deposited in an offshore state by a settlor precisely to escape the clutches of forced heirship rules might be struck down, if the assets are still considered to be part of his estate by the lex successionis. Indeed, Duckworth goes as far as to say that “For the planner of offshore trusts the darkest cloud on the horizon is the Hague Convention on Succession”.261 The Official Report by Waters states that the effect of Article 7(2)(d): “. . . would include a situation where the deceased in his lifetime responds to an advertisement from a tax haven (a non-contracting state) that he deposit assets in an investment device in the haven, and thereby avoid family protection legislation in his home jurisdiction”.262

The fear arises that the trust assets will be clawed-back by the lex successionis.263 However, Hayton argues that such concern is misplaced, since Article 256

Art. 6. Art. 3(1). 258 Art. 3(2). This is a presumption which can be rebutted where the state of closest connection is nonetheless overwhelmingly that of the settlor’s nationality. 259 Art. 3(3). Unless he was more closely connected with another state. 260 D Hayton in E Pluribus Unum, 121, 131. See also D Hayton, “Cross Border Estates, Part 3: the Hague Conventions” [1994] Private Client Business 329. 261 A Duckworth, “An Offshore View of Forced Heirship—Global Conflict and its Planning Implications: Part 3” [1995] Private Client Business 408. 262 Explanatory report on the Convention on the Law Applicable to Succession to the Estates of Deceased Persons by Professor D Waters, (hereafter “the Waters Report”), Proceedings of the Sixteenth Session of the Hague conference on Private International Law 1990, Book II—Succession to Estates—Applicable Law, 526, para 79. 263 A Duckworth, in J Glasson (ed.) International Trust Laws, ch. B1, 62–3. 257

58 Hague Succession Convention 1989 1(2)(d) excludes from the Succession Convention “property rights, interests or assets created or transferred other than by succession, such as in joint ownership with right of survival, pension plans, insurance contracts, or arrangements of similar nature”.264 So much is not self-evident, however, since it is not certain whether claw-back claims arise “other than by succession” within the meaning of Article 1(2)(d).265 Duckworth argues that the exclusion applies only to: “a lifetime disposition as to which the question or complaint is that it is in effect a willsubstitute. But if the question or complaint is that the lifetime disposition has encroached on the entitlements of forced heirs, that is a different matter; Article 7(2) applies and Article 1(2)(d) has no effect”.266

It is submitted that this interpretation is quite legitimate. True, Article 7 cannot subject matters to the applicable law which are ex hypothesi outside the Succession Convention’s scope. However, since the word “succession” is not defined in the Succession Convention, it is natural to look at Article 7 to determine those matters which might fall within it. Moreover, Duckworth points out that it is stated to be a vital goal of the Convention that “family laws against disinheritance of the surviving spouse or children be honoured”.267 Such a goal can scarcely be met if the Convention does not apply to lifetime dispositions of property which are made in order to evade otherwise applicable forced heirship provisions. In any event, Hayton argues that the United Kingdom has little to fear from the Succession Convention’s claw-back potential. If the property were transferred to England: “it is Article 15(1)(d) of the Trusts Convention that is relevant, because the English forum will apply English law as the lex situs to the transfer of English assets to English trustees . . . unimpeachably removing the assets from D’s estate. . .”268 264 D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3, 8–9. Hayton’s view is shared by M Koppenol-Laforce, Het Haagse Trustverdrag, 266, 275. However, KoppenolLaforce points out that the position may be different in respect of revocable trusts declared inter vivos: “If the settlor is also the trustee and during his life the only beneficiary, it looks much like a testamentary agreement in contemplation of death. The applicable succession law will have to set the marks” (ibid., 266–7). 265 D Hayton himself comments that the meaning of this phrase is “fudged by the Convention.” However, he argues that ratifying legislation should clarify the matter, and that, failing that, clawback claims can always be resisted on public policy grounds under Art. 18 of the Succession Convention: ibid., 8–9; see also D Hayton, “Cross-Border Estates: Part 3: the Hague Conventions” [1994] Private Client Business 329. 266 A Duckworth, in J Glasson (ed.) International Trust Laws, ch. B1, 63. 267 Waters Report, para. 26; quoted by A Duckworth, in J Glasson (ed.) International Trust Laws, ch. B1, 64. 268 D Hayton in E Pluribus Unum, 121, 132. See also D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3. In this author’s view, the transfer to the trustee is a “rocketlaunching” matter and the lex situs is applied to the inter vivos transfer according to the common law rules. However, the effect is still that assets validly disposed of inter vivos by English choice of law rules do not form part of the deceased’s estate on death and the lex successionis will not be applied to them.

General 59 In other words, application of the Hague Trusts Convention will allow an English court to treat it as a mandatory rule of English law that assets transferred to England under an inter vivos trust are beyond the reach of forced heirship claims. This, of course, will not assist in respect of assets outside the jurisdiction if claims are brought in the courts of another state which does permit claw-back. On balance, it suggested that it cannot unequivocally be stated that ratification of the Hague Succession Convention will have no potential claw-back implications for inter vivos trusts, where those trusts are created in order to defeat the forced heirship rights arising under the lex successionis. Nevertheless, the most likely outcome is that an English court would rule that where property is alienated pursuant to a validly created inter vivos trust, it is ex hypothesi not within the estate of a testator upon his death. Moreover, an English court could always fall back on its public policy to prevent claw-back in respect of assets within the jurisdiction. That does not, however, remove the risk that the English inter vivos trust might be “attacked” in another contracting state to the Succession Convention.269 19. TESTAMENTARY POWERS OF APPOINTMENT 270

(A) General Separate choice of law rules are applicable where an instrument confers on a person the power to appoint those who shall succeed to property. Such a power may be exercisable in the will of the donee or inter vivos. It may be general (in which case the donee may appoint any person or persons, including himself, as objects of the power) or special (i.e. limited to a specified class of persons).271 Since ownership of the property remains with the donor, it is less obviously appropriate that a testamentary exercise of the power by the donee should be governed by the “ordinary” rules of succession.272 (B) Essential Validity The applicable law depends on whether the power is general or special. The matter may be crucial if e.g. the law of the donee’s domicile on death treats the exercise of the power as essentially invalid for failure to comply with forced heirship rules. 269 Then again, this is a risk which persists even in the absence of an international succession Convention. 270 See G Miller, International Aspects of Succession, ch. 7; K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 32–7. 271 The question whether a power is general or special ought, in cases of doubt, to be governed by the law under which the instrument creating the power was made: Dicey and Morris, 1062. 272 E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1038–47, esp. 1039; A Scott and W Fratcher, The Law of Trusts, 4th edn. (Boston, Little, 1989), §629– 642.

60 Testamentary Powers of Appointment With special powers, the property is in no sense that of the donee and it would be inappropriate for the essential validity of a testamentary exercise of that power to be subjected to the law of his domicile. Rather, the law under which the instrument creating the power was created will apply.273 General powers are more complex, since the donee may mix the property with his own assets and treat them effectively as one.274 If so, the essential validity of a testamentary exercise of the power by the donee should be governed by the lex successionis.275 In the case of movables, this means the law of the donee’s last domicile; in the case of immovables, the lex situs.276 If, however, the donee treats the assets subject to the power as a separate fund, the assets cannot be regarded as forming part of his estate on death and the essential validity of a testamentary exercise of the power should be determined by the law governing the instrument creating the power.277 (C) Construction It may be unclear whether words used in the donee’s will are intended to constitute an exercise of the power.278 This is a matter of the donee’s intentions and, unless he provides to the contrary,279 it will be assumed that he intended matters of construction to be determined by the law of his domicile at the time of creation of the will.280 (D) Capacity The donee’s capacity to exercise the power is determined by the law of his domicile at the time of the will’s creation. It matters not that the donee lacked capacity by the law of the instrument creating the power.281 273

Pouey v. Hordern [1900] 1 Ch 492. In England, property over which the deceased retained a general power of appointment is included in his estate for the purposes of discretionary judicial assessment under s. 25(1), Inheritance (Provision for Family and Dependants) Act 1975. Such property may also fall within his estate for the purpose of determining mandatory forced heirship rights under a foreign law: D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3, 5–6. 275 Re Pryce [1911] 2 Ch 286; Re Khan’s Settlement [1966] Ch 567. Cf. Re Waite’s Settlement [1958] Ch 100. 276 Re Hernando (1884) 27 Ch D 284. 277 Re Mégret [1901] 1 Ch 547. 278 Dicey and Morris, 1057–8 explain that different legal systems require different levels of certainty of intention. See also s. 27, Wills Act 1837. 279 See e.g. Re Price [1900] 1 Ch 442; Re Waite’s Settlement [1958] Ch 100. 280 Re McMorran [1958] Ch 624. The same view is taken of American law by E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1047; and see Morgan Guaranty Trust Co v. Huntingdon 140 Conn 331, 179 A 2d 604 (1962). 281 Re Lewal’s Settlement [1918] 2 Ch 391. E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1043 state that capacity may be established “either by the law applicable to the transaction by which the donor created the power, or that of the donee’s domicile”. 274

Formal Validity 61

(E) Formal Validity As with “ordinary” testamentary disposition, a testamentary appointment under a power will be formally valid if it satisfies one of the laws specified in section 1, Wills Act 1963.282 However, section 2(1)(d) provides another validating law. It states that: “a will so far as it exercises a power of appointment [is formally valid] if the execution of the will conformed to the law governing the essential validity of the power”.283

(F) Revocation (i) Subsequent will or codicil A subsequent will duly executed under the Wills Act 1963 may revoke a power exercised over movable or immovable property.284 (ii) Act of revocation The effectiveness of an act of revocation of a testamentary power of appointment over movables is governed by the law of the donee’s domicile.285 (iii) Subsequent marriage The effect of marriage on the exercise of a power is a matrimonial law question and should be referred to the law of the donee’s domicile at the time of marriage.286

20. MATRIMONIAL PROPERTY REGIMES

(A) General Some legal systems provide for the community of property of spouses. This may affect property acquired during the marriage, but may in some regimes also 282

Or one of the other validating laws under s. 2. See also Murphy v. Deichler [1909] AC 446. 284 Although s. 2(1)(d) will not apply if the subsequent will does not “exercise a [substitute] power of appointment” but only revokes the initial exercise of the power: Dicey and Morris, 1064. See also s. 2(1)(c), Wills Act 1963. 285 Velasco v. Coney [1934] P 143. Cheshire and North comment (at 1013) that “presumably compliance with the law of the situs is required in the case of immovables.” 286 See also s. 18(2), Wills Act 1837 which provides that, as a matter of English law, marriage does not ordinarily revoke a power. 283

62 Matrimonial Property Regimes affect property acquired before marriage.287 Clearly, such regimes may affect trust property owned beneficially by a spouse prior to marriage. Community of property may also have a “rocket-launching” effect on trusts purported to be created by one spouse during the marriage or upon death. Indeed, a matrimonial property regime could be regarded as analogous to a specifically enforceable contract between the parties to settle after acquired property.288 It may accordingly be necessary for one spouse to gain the consent of the other before subjecting matrimonial property to a trust. A matrimonial property scheme may also restrict the need for forced heirship provision on death in states where there is a mere discretionary provision available on death, as in England under the Inheritance (Provision for Family and Dependants) Act 1975.

(B) Choice of Law Rules The law applicable to matrimonial property regimes is a large and complex subject. Only a short summary of the relevant choice of law rules will be provided here.289 (i) No ante-nuptial contract (1) Movable property Where there is no ante-nuptial contract, the effect of marriage on movable property is governed by the law of the husband’s domicile at the time of marriage.290 However, this is no more than a presumption and may give way to the law of the intended matrimonial domicile.291 Moreover, the rule was established at a time when the wife acquired the domicile of her husband on marriage. This is no longer the law292 and there is no reason in principle to favour the husband’s domicile.293 287 See M Rheinstein and M Glendon, International Encyclopaedia of Comparative Law Vol. IV, Persons and Family (Mohr (Dordrecht, Paul Siebeck), Tübingen and Martinus Nijhoff, 1980), ch. 4. 288 Compare Pullan v. Koe [1913] 1 Ch 9. This analogy was suggested to me by Professor David Hayton, to whom thanks are due. 289 See further Cheshire and North ch. 34; Dicey and Morris, ch. 28; G Miller, International Aspects of Succession, ch. 3. The United Kingdom has not signed or ratified the Hague Convention on the Law Applicable to Matrimonial Property Regimes 1978 (in force in France, Luxembourg and the Netherlands). 290 Re Martin [1900] P 211. 291 Re Egerton’s Will Trusts [1956] 2 All ER 817. 292 S. 1(1), Domicile and Matrimonial Proceedings Act 1973. 293 There may, accordingly, be a greater willingness to apply the law of the intended matrimonial home where the domiciles of the spouses differ at the time of marriage. Dicey and Morris (at 1069) advocate that where the spouses are not domiciled in the same state, the law of domicile should be abandoned altogether and the “law of the country with which the parties and the marriage have the closest connection [should be applied], equal weight being given to connections with each party.”

Choice of Law Rules 63 It does not appear that a change in domicile will affect the community of property.294 This is certainly true of property acquired prior to the change of domicile; where property is subsequently acquired, however, a spouse’s rights cannot be retrospectively prejudiced and a much stronger case for applying the law of the new domicile exists.295 (2) Immovable property It may be that the lex situs should be applied in this scenario, at least where the land is situated overseas, in order to ensure the enforceability of any order made by an English court.296 The cases appear to indicate, however, that where land is situated in England, English law does not need to be applied and that the question whether the land is subject to community of property may be governed by the law of the husband’s domicile at the time of marriage.297 (ii) Ante-nuptial contracts Some legal systems permit the parties to determine for themselves the effects of marriage on property. The validity and effects of such contracts are excluded from the Rome Convention.298 Essential validity is governed by the proper law of the settlement. This law may be chosen.299 In the absence of an express or implied choice, the law of closest connection will govern. In ascertaining this law, the matrimonial domicile will be especially important,300 although not determinative.301 It appears that the same rules apply to immovable property as to movable property.302 Ante-nuptial contracts will be formally valid if they satisfy the proper law or that of the place of execution of the settlement.303 The law which governs the parties’ capacity to make an ante-nuptial agreement is not clear-cut. The leading cases are criticised by Dicey and Morris, and are consistent with application either of the law of the domicile at the time of 294 De Nicols v. Curlier [1900] AC 21 (but the case may have been one where there was an antenuptial agreement: see Jaffey on the Conflict of Laws, 2nd edn., 460–1). See also Lashley v. Hog (1804) 4 Pat 581 (although this appears to be a case on succession and not on matrimonial property). 295 See further Cheshire and North, 1020–1. 296 Welch v. Tennent [1891] AC 639. 297 Re De Nicols [1900] 2 Ch 410; Chilwell v. Carylon (1897) 14 SC 61 (South Africa); Jaffey on the Conflict of Laws, 2nd edn., 462–3. These cases may, however, be explained on the basis that there was a tacit ante-nuptial agreement. 298 Art. 1(2)(b) excludes “rights in property arising out of a matrimonial relationship.” 299 Although some objective connection with the settlement may be required: Re Fitzgerald [1904] 1 Ch 573. 300 Ibid.; Duke of Marlborough v. AG [1945] Ch 78. 301 Dicey and Morris point out (at 1074–5) that other factors may be significant, such as: the domicile of the wife, where this differs from that of the husband; the language of the settlement; the place of execution; the place of administration of a trust; the residence of the trustees; the situs of property; and invalidity by the law of the matrimonial domicile. 302 Re De Nicols [1900] 2 Ch 410; Chilwell v. Carylon (1897) 14 SC 61 (South Africa). 303 Van Grutten v. Digby (1862) 31 Beav 561; Guépratte v. Young (1851) 4 De G & Sm 217.

64 Matrimonial Property Regimes contracting or the objective proper law.304 Cheshire and North argue that the matter should be resolved by the objective proper law of the agreement, which will prima facie be that of the matrimonial domicile.305 In the case of an agreement which, according to the law of the domicile at the time of marriage, must be ratified upon reaching the age of majority, the domicile of the parties once they are married will determine whether ratification or revocation has occurred.306

(C) Matrimonial Property Regimes and the Law Applicable to Succession: Double Recovery and No Recovery Hayton points out that since the law applicable to matrimonial property rights is determined at the date of marriage, and the lex successionis at the time of death, the law applicable to each may differ. There is then a real risk that the surviving spouse will gain double recovery, if the lex successionis does not recognise community of property and provides fixed indefeasible heirship rights.307 Of course, the converse situation could also arise, where the law applicable to succession makes no provision for the surviving spouse on the basis that it considers that he already has community of property rights, but the law applicable to matrimonial property regimes rules that he does not.308 Where the spouses subsequently divorce, there may likewise be an issue of whether the community of property existing under the law of the matrimonial domicile should be taken into account by an English court in formulating an order for financial relief. Refusal to do so when invoking the discretion to grant relief under the Matrimonial Causes Act 1973 can lead to serious anomaly.309 The English court may also vary a pre- or ante-nuptial settlement pursuant to a divorce, or a decree of nullity or judicial separation under section 24(1), Matrimonial Causes Act 1973; or in support of a foreign divorce of decree under section 17, Matrimonial and Family Proceedings Act 1984.310

304 Dicey and Morris, 1076, considering Re Cooke’s Trusts (1887) 56 LT 737; Cooper v. Cooper (1888) 13 App Cas 88; Viditz v. O’Hagan [1900] 2 Ch 87. See the discussion in section 4, above. 305 Cheshire and North, 1025. 306 Viditz v. O’Hagan [1900] 2 Ch 87. 307 D Hayton, “Cross Border Estates: Part 2: the Problem of Diversity” [1994] Private Client Business 250. He goes on to explain how the problem is solved in America by the Uniform Disposition of Community Property at Death Act which deprives the spouse of succession rights where they have acquired rights of community under another law. See also Jaffey on the Conflict of Laws, 2nd edn., 467; Beaudoin v. Trudel [1937] 1 DLR 216. 308 Hayton ibid. explains how California overcomes this problem by providing in its Probate Code that where community of property would have arisen had the property been acquired in California, the right of succession of the surviving spouse is to one half of that property. 309 Jaffey on the Conflict of Laws, 2nd edn., 467–8. 310 These are fully considered in the discussion of Art. 8(2)(h), Hague Convention in Part Two of this book.

Transnational Trusts in Offshore Jurisdictions 65

21. THE CREATION OF TRANSNATIONAL TRUSTS IN OFFSHORE JURISDICTIONS 311

It will be seen that offshore legislation is, generally speaking, more pragmatic than the law of England.312 The settlor’s autonomy is very largely protected and trusts are, insofar as possible, “insulated” against claims which might arise under a foreign law. Hayton describes the approach of such states thus:313 “In the interests of attracting business to an island, say Suntopia, the island likes to boast a selection of medals: the settlor’s capacity medal (to try to avoid problems that the settlor’s spouse or heirs might create); the protection against creditors’ medal (to protect the trust fund as much as possible against creditors of the settlor); the settlor’s control medal (so that, as a matter of law—as opposed to substance—a sham trust does not arise in circumstances where a settlor can almost still ‘have his cake and eat it’); the perpetuation of the settlor’s wishes medal (to prevent beneficiaries frustrating the settlor’s wishes); the confidentiality medal (so as not to give high hopes to beneficiaries, creditors or the Revenue); the everyone in the world medal (so that discretionary trusts or powers for everyone—or everyone but a small excepted class—are valid, despite any administrative unworkability); the non-charitable purpose trust medal (to achieve the same ends as the confidentiality medal and to enable noncharitable purposes to be achieved); the everlasting trust medal (to enable the trust to endure for as long as desired); the low level of fiduciary duty medal (to attract trustees and protectors);314 and the ouster of judicial jurisdiction medal (e.g. allowing trustees to determine what is capital or income or to resolve any conceptual uncertainties or permitting special persons to consent to breaches of trust or compromises, thereby ousting the need for the judicial appointment of a guardian ad litem for incapacitated, 311 Very useful collections of trusts material relating, though not limited to, offshore jurisdictions can be found in J Glasson (ed.) International Trust Laws, vol. 2, part D and in M Lupoi, Trust Laws of the World (Rome, ETI, 1996). See also M Lupoi, Trusts: A Comparative Study, ch. 4 for a concise overview of the systems of some 25 offshore states. 312 For a recent (three-part) discussion of the position offshore and its likely development (as well as the position onshore), see A Duckworth, “Trust Law in the New Millennium: Part 1: Retrospective” (2000) 7(1) Trusts and Trustees 12; “Part 2: Prospective” (2000/1) 7(2) Trusts and Trustees 11; “Part 3: Fundamentals” (2000/1) 7(3) Trusts and Trustees 9. See also D Campbell (ed.) “Offshore Trusts” [1995] Comparative Law Yearbook of International Business: Special Issue (the Hague, Kluwer, 1996); A Kaplan (ed.) Trusts in Prime Jurisdictions (The Hague, Kluwer, 2000). See R Venables, Non-Resident Trusts, 8th edn. (London, Key Haven, 2001); P Soares, Non-Resident Trusts, 5th edn. (London, Sweet & Maxwell, 1998); W Diamond, D Diamond and V Kaplan, International Trust Laws and Analysis (Boston, Warren, Gorham and Lamont, 1995– (loose-leaf)); W Diamond, D Diamond and J Sullivan, International Trust and Laws Analysis (The Hague, Kluwer, 2001– (loose-leaf)). 313 D Hayton, “Whither Trusts in the Twenty-First Century: Part 2” [2000] Private Client Business 163. 314 See E Arosemena, “Concept and Regulation of the Protector in Offshore Jurisdictions”, [2000] Private Client Business 287; A Conder “The Office of the Protector” (1995) 4 Trusts and Trustees 12; R Ham, “Protectors” in J Glasson (ed.) International Trust Laws, ch. B3; R Lawrence, “The Role of the Protector—an Insulator for Corporate Beneficiaries?” (1993) 2 J Int Corp P 88; P Matthews, “Protectors: Two Cases; Twenty Questions” (1995) 9 Trust Law International 108; D Waters, “The Protector: New Wine in Old Bottles?” in A Oakley (ed.) Trends in Contemporary Trust Law (Oxford, Clarendon Press, 1996), ch. 4, 63.

66 Transnational Trusts in Offshore Jurisdictions unborn or unascertained beneficiaries and for judicial approval of a compromise proposed by such guardian).”

Until relatively recently, the “rocket-launching” aspects of trusts were governed in most offshore jurisdictions by common law conflict of laws principles. More recently, statutory intervention has sought to bring greater clarity to these issues.315 Whether it has been fully successful in so doing will be considered. We will focus here on the legislation of the Cayman Islands, Bermuda, Jersey and the Isle of Man.316

(A) The Cayman Islands317 The Trusts (Foreign Element) Law 1987 (hereafter “TFEL”) was the first major statute in its field and has been used as a model in other jurisdictions. Its enactment (and subsequent adaptation elsewhere) represents a rejection of the rules of the Hague Trusts Convention, which may be seen offshore as insufficiently protective of settlor autonomy, as insufficiently precise in its choice of law rules and as providing excessive derogations from the applicable law. In that respect, asset protection is not optimised by the Convention.318 The Act’s provisions (as amended) are now incorporated into the Cayman Islands Trusts Law 1967 (1998 Revision), Part VII.319 The statute states that all questions relating to a trust shall be determined by the law of the Cayman Islands if Cayman Islands law is the governing law. In other words, if Cayman Islands law governs the rocket itself, it will generally govern the “rocketlaunching” process too. Section 89 of the 1998 Revision provides thus: “(1) In determining the governing law of a trust, regard is first to be had to the terms of the trust and to any evidence therein as to the intention of the parties; and the other circumstances of the trust are to be taken into account only if the terms of the trust fail to provide such evidence. (2) A term of the trust expressly selecting the laws of the Islands to govern the trust is valid, effective and conclusive regardless of any other circumstances”. 315 See J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations”, ch. 11, 240–5. 316 Note that discussion of certain forced heirship provisions of offshore jurisdictions is contained in the analysis of Art. 15(1)(c), Hague Convention in Part Two of this book. 317 A Duckworth, “National Digest for the Cayman Islands”, in J Glasson (ed.) International Trust Laws, ch. A7, 23–7; see also A Duckworth, “Forced Heirship and the Trust”, ibid., ch. B1, 43–5; A Travers and J Appleyard, “Trusts in the Cayman Islands”, in A Kaplan (ed.) Trusts in Prime Jurisdictions, ch. 11; M Lupoi, Trusts: A Comparative Study, 209–10. 318 See further A Duckworth, “The Offshore Trust in Transition” in D Kozusko and J Schoenblum (eds.) International Estate Planning: Principles and Strategies” (Chicago, American Bar Association, 1991) 129, 138–9; J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 247; O Merren in D Campbell (ed.) “Offshore Trusts” [1995] Comparative Law Yearbook of International Business: Special Issue, 78, 79–80. 319 Hereafter “the 1998 Revision.”

The Cayman Islands 67 In other words, the parties can expressly choose Cayman Islands law to govern the trust, including questions of the settlor’s capacity to create the trust, thereby avoiding any incapacity that the settlor might have by his lex domicilii or by the lex situs. Section 90 states that: “All questions arising in regard to a trust which is for the time being governed by the laws of the Islands or in regard to any disposition of property upon the trusts thereof including questions as to— (a) the capacity of the settlor; (b) any aspect of the validity of the trust or disposition or the interpretation or effect thereof:. . . are to be determined by the laws of the Islands.

Cayman Islands trusts are insulated against claims arising under a foreign law. Section 91 provides that: “no trust governed by the laws of the Islands and no disposition of property to be held upon the trusts thereof is void, voidable, liable to be set aside or defective in any fashion, nor is the capacity of the settlor to be questioned, nor is the trustee, any beneficiary or any other person to be subjected to any liability or deprived of any right”

either because (a) the laws of a foreign jurisdiction prohibit or do not recognise the concept of a trust; or (b) the trust avoids or defeats the rights of forced heirs existing by a foreign law. It follows that Cayman Islands trusts are, at least in Cayman Islands courts, safe from the claims of forced heirs.320 The legislation applies even to foreign judgments. This is less easy to defend, involving as it does a rule of blanket non-recognition which values self-protection far more than the concerns of comity. It also involves a routine review of the substance of a foreign judgment.321 Perhaps most seriously of all, states which follow this model may do so without considering whether non-recognition of judgments is compatible with their pre-existing statutory obligations.322 320 But see Lemos v. Coutts & Co (1992–3) CILR 460; A Duckworth, “Attacking International Trusts—the Problem of ‘Blowing Hot and Cold’” [1996] Private Client Business 97; see also A Travers and J Appleyard, “Trusts in the Cayman Islands” in A Kaplan (ed.) Trusts in Prime Jurisdictions, ch. 11, 239, 244. The Bahamas has also sought to clarify the extent of protection from forced heirship claims: see H Thompson, “The Bahamas: Amendments to the Trusts (Choice of Governing Law) Act 1989” (1997) 3 Trusts and Trustees 4. On the Bahamian Trustee Act 1998, see P Maynard, “The Bahamas: the New Trustee Act” (1999) 16 Amicus Curiae 29. 321 Something not normally sanctioned in the conflict of laws: see e.g. Art. 29, Brussels Convention; Godard v. Gray (1870) LR 6 QB 288. The Brussels Convention is to be replaced by a Council Regulation (EC) No 44/2001 of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters ( “the Brussels Regulation”) with effect from 1 March 2002: OJ 2001 L 12/1. However, an identical provision is contained in Art. 36 of the Regulation. 322 M Lupoi, Trusts: A Comparative Study, 249 gives the examples of Belize and Anguilla, who are bound by (their versions of what is in the United Kingdom) the Foreign Judgments (Reciprocal Enforcement) Act 1933, which sets out the conditions under which recognition of a foreign judgment may be refused. Another problem case is the Isle of Man, which is bound by the same Act, and which nonetheless states in its Trusts Act 1995 (s. 5(b)(ii)) that a trust is protected from the fact that it “contravenes any . . . foreign judicial or administrative order. . .” which gives effect to heirship rights.

68 Transnational Trusts in Offshore Jurisdictions Section 89(3) states that if the trust contains a term to the effect that Cayman Islands law is to apply to a particular aspect of the trust, or that the Cayman Islands courts are the forum for the administration of the trust, or any like provision, there will be a rebuttable presumption that Cayman Islands law is intended to govern the entire trust, subject to any express provision to the contrary. This also suggests that a settlor could choose a different law to govern different parts of the trust. Arguably, he could even provide that the issue of capacity alone was to be governed by Cayman Islands law and this choice would be treated as valid in the Cayman Islands courts. Moreover, a settlor may change the governing law to that of the Islands, if the change is recognised by the governing law of the trust previously in effect, or change from the Island’s laws, if the new governing law recognises the validity of the trust and the beneficiaries’ interests.323 The whole effect of the Act is, prima facie at least, to give great freedom to a settlor to confer capacity on himself to dispose of property. In so doing, it might be said to value pragmatism more than consistency with conflict of laws principles on capacity and the transfer of property.324 However, there are certain restrictions on the scope of the applicable law. These are set out in section 90. Three are of particular interest for present purposes. First, the section does not validate any disposition of property not owned by the settlor or subject to a power vested in the settlor. Secondly, as regards a corporation’s capacity, it is provided that Cayman Island’s law “does not affect the recognition of the laws of its place of incorporation.” The word “recognition” is open to different interpretations. On one view, it might suggest that the capacity to create a trust is governed solely by the law of the place of a company’s incorporation; on another view, that a corporation must have capacity both by Cayman Islands law and the law of the place of incorporation; and on a third view, Cayman Islands law still determines capacity, subject to a discretion to apply the law of the place of incorporation where it is thought appropriate. Unfortunately, no further guidance is provided as to which meaning is correct. Thirdly, Cayman Islands law does not validate any trust or disposition of immovable property situated in a jurisdiction other than the Islands which is invalid according to the laws of such jurisdiction. This means that the vesting of property in the trustee will be determined by the lex situs; presumably it also refers capacity to the lex situs, since a disposition of immovable property will not be valid according to that law if the settlor lacked capacity to dispose of his property. Nor does the statute state whether a transaction which is valid by the lex situs can be invalidated by Cayman Island’s law. In principle, it seems that it should be, since, with the exception of the section 90 criteria, the scope of 323 S. 89(4). See also the discussion of Art. 10, Hague Convention in Part Two of this book (Art. 10 does not, of course, apply in the Cayman Islands, which has not adopted the Convention). 324 It has, however, been widely followed: examples include the Bahamas (s. 7, Trusts (Choice of Governing Law) Act 1989), Barbados (ss. 16 and 17, International Trusts Act 1995) and the Isle of Man (s. 4, Trusts Act 1995 (on which see below)). For further examples, see M Lupoi, Trusts: A Comparative Study, 244.

Bermuda 69 Cayman Island’s law when applicable is almost unlimited. On the other hand, it would mean that the rocket would have to be launched by both the lex situs and Cayman Island’s law in relation to land, which would tend to mitigate against the validity of trusts of land. Moreover, since the whole point of the reference to the lex situs in relation to land is that it is futile to decide a case in a manner different from the courts of the situs for immovable property, it makes little sense to refer other than exclusively to that state’s law, at least on the question of whether the settlor may alienate any property right. Furthermore, the statute makes no reference to the capacity of the trustee to hold on trust as being governed by Cayman Islands law when that is the governing law. Presumably, however, this is to be determined by the governing law, since it is regarded as part of the rocket itself even under the Hague Convention,325 and this statute provides considerably more party autonomy than does that Convention. Nor does the Cayman Island’s legislation tell us which law is to govern capacity and the vesting of property in the trustee when Cayman Islands law is not the governing law. In principle, one might expect this also to be governed by the applicable law of the trust. On the other hand, the decision not to legislate at all in relation to such trusts might suggest that the omission is material and that these issues should continue to be governed by the common law principles considered above; in which case the unbounded application of the governing law will be hard to defend. Finally, it is unclear whether renvoi is included when Cayman Islands law is chosen. It might be assumed that it is not, since section 90 states that the matters which it has previously mentioned “are to be determined according to the laws of the Islands, without reference to the laws of any other jurisdiction with which the trust or disposition may be connected.” However, to adopt renvoi is to employ the Cayman Island’s own private international law rules, which might state that a foreign law is applicable. Accordingly, the provision is not obviously conclusive. A more principled argument against renvoi is that once party autonomy is allowed, one must respect the intentions of the parties. If a settlor states that Cayman Islands Law is to govern, he almost certainly means Cayman Islands domestic law alone; in which case there is no role here for the renvoi doctrine. (B) Bermuda326 The Recognition of Trusts Act 1987 has been extended to Bermuda by the Recognition of Trusts Act 1987 (Overseas Territories) Order 1989. However, 325 At least in respect of the “specific” capacity to act as a trustee: see s. 5 above and Art. 8(2)(a), Hague Convention. Of course, the Cayman Island’s Act was not framed with the Hague Convention provisions in mind, as the Cayman Islands has not signed the Convention. 326 See J Campbell, “National Digest for Bermuda” in J Glasson (ed.) International Trust Laws, ch. B5, 22–3; A Duckworth, “Forced Heirship and the Trust”, ibid., ch. B1, 50–3; M Lupoi, Trusts: A Comparative Study, 208–9.

70 Transnational Trusts in Offshore Jurisdictions even though it is bound by the Hague Convention, the principal statute in Bermuda addressing conflict of laws questions in relation to trusts is the Trusts (Special Provisions) Act 1989, Part I. Many aspects of the Act are similar to the Cayman Islands legislation, but Part I is not limited to trusts governed by the law of Bermuda.327 The general rule is that the settlor may choose the governing law of the trust (which choice can be implied from the terms of the trust instrument, interpreted, if necessary, in the light of the circumstances of the case).328 In default, the trust is governed by the law with which it is most closely connected.329 Trusts which are governed by the law of Bermuda are protected by section 11. This provision is in slightly different terms to the exclusion in the Cayman Islands.330 It prevents Bermuda trusts from being set aside or varied by foreign rules concerning forced heirship, the personal and proprietary effects of marriage, or (and additional to legislation elsewhere) the protection of creditors in matters of insolvency. The only exception is where “the law of Bermuda has corresponding laws or public policy rules”.331 Such legislation very largely negates the impact of a provision such as Article 15, Hague Trusts Convention. However, section 11 stops short of laying down a blanket exclusion provision in respect of the recognition of foreign judgments. Section 11 arose for consideration in Garner v. Bermuda Trust Co Ltd.332 A settlor died allegedly domiciled in Mexico and the settlor’s nephew claimed heirship rights under Mexican law. The Bermudan Court of Appeal ruled that the nephew had failed to show authority for the proposition that a Bermudan court would set aside a trust for infringement of foreign succession rights “and we are not persuaded that there would have been power so to do.” A list of matters subject to the applicable law of the trust is provided in section 7, but this does not include either capacity (for which there is separate provision) or the vesting of the property in the trustee by the settlor (for which there is no provision). One might assume from this that the transfer to the trustee will 327 S. 3, Trusts (Special Provisions) Act 1989 provides that “this Part applies to trusts created voluntarily and evidenced in writing and also to any other trusts of property arising under the law of Bermuda or by virtue of a judicial decision whether in Bermuda or elsewhere.” It is suggested that this must mean that all voluntary trusts evidenced in writing, no matter what the governing law, are included, (as well as all trusts arising by judicial decision) and also trusts which do not fulfil these criteria, but which are nonetheless governed by the law of Bermuda. Since (i) the wording of s. 3 is taken from the UK Order in Council which extends to Bermuda the provisions of the Hague Trust Convention (with modifications), (ii) the Convention is clearly international in scope, and (iii) many of the provisions of Part I are taken directly from the Convention, the conclusion that the Act is not limited to trusts whose proper law is that of Bermuda appears overwhelming. See also the discussion in Part Two of this book of Art. 3, Hague Trusts Convention and the United Kingdom’s extension of the Convention in s.1(2), Recognition of Trusts Act 1987. 328 S. 5, Trusts (Special Provisions) Act 1989. 329 Ibid., s. 6. 330 See further, M Lupoi, Trusts: A Comparative Study, 246. 331 The meaning of this phrase is considered by A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, p. 51. 332 Noted (1992) 1 J Int Corp P 133; see also A Duckworth, ibid., 51–3.

Bermuda 71 be governed by the lex situs. Against this, the Act does not contain an express provision excluding “rocket-launcher” aspects of the trust and does expressly deal with capacity elsewhere; in which case one might assume that the constitution of the trust will be governed by the law applicable to the rocket aspects of the trust. Yet there is no express indication that this is the case in the Act, even where the governing law is that of Bermuda. Application of any law other than the lex situs has very little to commend it in principle, given that the transfer of property is an “ordinary” incidence of the property rules of the conflict of laws.333 As to capacity, there is express provision in section 10(1) that, subject to section 10(2), a person has capacity to create a trust as follows: “(a) where the trust property is movable— (i) in the case of an inter vivos trust, if he has capacity to create a trust of movable property by the law of Bermuda; (ii) in the case of a testamentary trust, if he has capacity to create a trust of movable property by the law of his domicile; (b) where the trust property is immovable, if he has capacity to create a trust by the lex situs of the immovable”.334

Section 10(2) further provides that all questions relating to the capacity of the settlor arising with regard to a trust governed by the law of Bermuda are to be determined according to the law of Bermuda only, subject to exceptions, including that: (i) the property must originally have been owned by the settlor; (ii) the provision takes effect subject to any express term to the contrary; (iii) in relation to the capacity of a corporation, the provision does not affect the recognition of the laws of the place of incorporation. This contains the same ambiguity as the Cayman Island legislation discussed above; (iv) the recognition of foreign laws prescribing formalities for the disposition of property remains unaffected; (v) the provision does not validate any trust or disposition of immovable property situated outside Bermuda which is invalid according to the lex situs. This is another provision whose scope is uncertain in a similar manner to the Cayman Islands provision discussed above; (vi) a testamentary trust or disposition invalid according to the law of the testator’s domicile will not be validated. Overall, this is a rather curious set of provisions. The basic capacity rule for inter vivos trusts of movables seems to be a blanket application of the law of Bermuda, whatever the governing law of the rocket itself. It appears that the law of Bermuda is simply applied as the law of the forum. The application thereof is 333

At least in the case of inter vivos transfers. This approach can also be found in Barbados (see s. 15, International Trusts Act 1995). See further M Lupoi, Trusts: A Comparative Study, 243. Contrast the much more liberal position in Guernsey, where s. 11(A)(1)(b), Trusts (Guernsey) (Amendment) Law 1990 states that the settlor needs capacity to create a trust governed by Guernsey law only under one of the following laws: Guernsey law; the law of the settlor’s domicile or nationality; or the law governing the transfer on trust. On the law of trusts in Guernsey, see R Ashton, An Analysis of the Guernsey Law of Trusts (London, Key Haven, 1998). 334

72 Transnational Trusts in Offshore Jurisdictions very hard to justify in principle. It is true that capacity rules have mandatory characteristics and this may lead a Bermuda court to feel that capacity restrictions by its law should never be avoided when litigation occurs in the jurisdiction. But the settlor may have no real connection with Bermuda.335 Indeed, by litigating in Bermuda, he may be able to avoid an incapacity which he possesses by his lex domicilii or by the lex situs. It can hardly be argued that Bermuda capacity rules merit application in all instances. The rules look particularly curious when one notes that the lex situs holds complete sway in relation to immovables. Whilst this is understandable, one would have thought that if it were felt necessary to apply Bermuda law to all capacity questions in relation to movables, it ought also to have at least a role in relation to immovables. The capacity rule for inter vivos trusts of movables is also curious, given that capacity to create a testamentary trust, be it of movables or immovables, is not subjected to the law of the forum in all cases.

(C) Jersey336 Whilst Jersey harmonised its rules with those of the Hague Convention in 1991,337 and became party thereto in 1992,338 the “rocket-launching” provisions of Jersey law are largely to be found in the Trusts (Jersey) Law 1984 (inserted by the Trusts (Amendment) (Jersey) Law 1989).339 The approach taken therein is striking. A person domiciled outside Jersey is deemed to have capacity to make a lifetime gift340 to a Jersey trust341 if he is of full age and of sound mind by the law of his domicile.342 The lex domicilii rule will apply even in the case of inter 335 Given the rules on jurisdiction to be found in s. 9, it is not necessary that he has such a connection. 336 See P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., ch. 4; K Baker and E Devenport, “National Digest for Jersey”, in J Glasson (ed.) International Trust Laws, ch. A15, 29–31; A Duckworth, “Forced Heirship and the Trust”, ibid., ch. B1 45–7; M Lupoi, Trusts: A Comparative Study, 215–17; K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 9, 25. 337 Trusts (Amendment No 2) (Jersey) Law 1991. 338 Ratification of the Hague Convention was extended by the United Kingdom to Jersey on 20 December 1991. The Convention entered into force in Jersey on 1 March 1992. 339 See also the provisions on the enforceability of foreign trusts contained in Art. 45, Trusts (Jersey) Law, 1984, which recognise the validity of such trusts, but which go on, in sub-paragraph (2), to state that the foreign trust shall be unenforceable in Jersey to the extent that it purports to do anything illegal by the law of Jersey or applies directly to immovable property situated in Jersey, or to the extent that it infringes Jersey public policy. 340 Compare s. 83(1), Trustee (Amendment) Act 1993 in the British Virgin Islands, which applies both to inter vivos and to testamentary capacity. 341 It appears that the trust must be governed by Jersey law, since Art. 8(A) of the 1989 Amendment is inserted into the 1984 Law, Part II, Art. 6 of which states that “this Part shall apply only to a Jersey trust.” “Jersey trust” is defined in Art. 1 as a trust whose proper law is the law of Jersey. 342 Art. 8A(2)(a). Although the wording might suggest that this provision applies only to dispositions to pre-existing Jersey trusts, P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 45–6 argue that it is also intended to cover the creation of a new trust. Contrast the much more liberal approach in Guernsey: s. 11(A)(1)(b), Trusts (Guernsey) (Amendment) Law 1990 (both are

Jersey 73 vivos trusts of land.343 One might pause to reflect on the desirability of this position. Suppose that a settlor domiciled in state A owns land in state B, which he purports to transfer to trustees under a trust expressly stated to be governed by Jersey law. He has capacity to do so according to the law of state A and the law of Jersey. However, the settlor lacks capacity to dispose of the property according to the law of state B. He will nevertheless be able validly to transfer legal title to the trustee, even though, according to the law of state B (the state with exclusive control over the land itself), the settlor still owns the property absolutely.344 Article 8A(2)(b) provides that, where a settlor domiciled outside Jersey has capacity by the law of his domicile, no rule relating to inheritance or succession (including forced heirship rules)345 of any system of law shall affect the transfer by that settlor of the property or the validity of the trust.346 However, this provision is limited to inter vivos trusts and to settlors domiciled outside Jersey.347 Article 8A(2) does not deal with Jersey domiciliaries, though it would be highly surprising in the light of the above, and in the light of the general preference that we have seen above in offshore jurisdictions for the application of the law of the forum, if their capacity was not governed by Jersey law. The result, however, is that overseas domiciliaries might be able to transfer on trust property located in Jersey, even though a Jersey domiciliary would not have capacity to do so. Moreover, Jersey domiciliaries will be subject to the forced heirship rules existing in Jersey law.348 The statute also does not address capacity when the proper law is not that of Jersey. Regard will have to be had to common law principles. Matthews and considered by M Lupoi, Trusts: A Comparative Study, 243). See also the Age of Majority (Jersey) Law 1999, discussed by P Hargreaves, “Problem Children: a Jersey Perspective on Assets Held by and for Minors” [2000] Private Client Business 222. 343 Compare s. 83(1)(a), Trustee (Amendment) Act 1993 in the British Virgin Islands, which applies the law of the settlor’s domicile to capacity questions, but only in respect of personal property. 344 “Thus a domiciled Frenchman, if of full age and sound mind according to French law, is deemed by Art. 8A(2) to have capacity to give Italian land to a Jersey trust, whatever French or Italian law may have to say on the matter. . .”: P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 45 (emphasis in original). 345 Note that this is not as extensive as the Cayman Islands’ exclusion of foreign law (s. 91, Cayman Islands Trusts Law 1967 (1998 Revision), considered above), which also refers to protection of trusts from foreign laws which do not know or recognise the trust concept; nor does the Jersey exclusion refer to foreign judgments conferring forced heirship rights. Compare also s. 11(A)(1), Trusts (Guernsey) (Amendment) Law 1990, which lies somewhere between the Cayman Islands and Jersey approaches. For further discussion (including a list of states which have followed the respective approaches of the Cayman Islands, Jersey and Guernsey), see M Lupoi, Trusts: A Comparative Study, 245–6. 346 Even if the transfer is not valid by the law of his domicile. Compare s. 83(1)(b), Trustee (Amendment) Act 1993 in the British Virgin Islands, which expressly provides in respect of a settlor that “no rule relating to inheritance or succession of the law of his domicile shall affect any such transfer or disposition [of personal property to the trustee of a trust] or otherwise affect the validity of such trust”. (Emphasis added). 347 See the pre- Trusts (Jersey) Law 1984 case of Rahman v. Chase Bank & Trust Company (CI) [1991] Jersey Law Reports 113, where the capacity of the settlor to defeat a forced heirship claim was raised but not decided. 348 See the Wills and Succession (Jersey) Law 1993. See also S Meikeljohn in D Hayton (ed.), European Succession Laws, ch. 4, 117.

74 Transnational Trusts in Offshore Jurisdictions Sowden argue that the general conflict of laws rule is that the settlor’s capacity to dispose of the property is governed by the lex domicilii for movables and the lex situs for immovables.349 They cite the case of AG v. Bellilos350 in support of the proposition. Yet not only was that case not really about capacity, but they largely ignore all the rest of the English authorities on capacity, which, particularly in commercial subject areas, give domicile a very limited role. They are rightly concerned about the free evasion of capacity restrictions by application of an expressly chosen proper law; but that fear does not lead logically to the conclusion that the lex domicilii, as opposed to e.g. the lex situs, is the appropriate law.351 It appears that Matthews and Sowden, are amongst the very few writers who (rightly) distinguish between capacity to transfer property and capacity to create a trust. The rule that they suggest appears to apply to the former question. It seems that they might be prepared to accept that the proper law governs capacity to create the trust structure rather than some other.352 It should be noted that Article 8A(2) only refers to transfers or dispositions to a trust; it does not appear to deal with the settlor’s capacity to declare himself a trustee. Yet that, no less than a transfer on trust, involves the disposition of the settlor’s beneficial interest and ought, accordingly, to be governed by the same principles.353 Jersey also has provisions in Article 45(2)(a), Trusts (Jersey) Law 1984 limiting the effectiveness of foreign trusts in Jersey. In particular, a foreign trust shall be unenforceable in Jersey to the extent that it requires the doing of something contrary to Jersey law, or purports to apply directly to immovable property located in Jersey. Today, these provisions sit most uncomfortably with the Hague Trusts Convention, which requires trusts governed by foreign law to be given effect to in contracting states (subject only to the tightly controlled provisions on mandatory rules and public policy)354 and with the right which it confers on the settlor to choose the law applicable to a trust, even a trust of land.355

349

P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 44–5. [1928] 1 KB 798, 826. 351 See also P Matthews, Trusts: Migration and Change of Proper Law (London, Key Haven, 1997), 56. 352 P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 45. 353 E Sykes and M Pryles, Australian Private International Law, 3rd edn., 45–6, agree. 354 Art. 45(2)(b), Trusts (Jersey) Law 1984 contains a provision rendering trusts governed by foreign law unenforceable to the extent that they are immoral or contrary to public policy. This tends to suggest that Art. 45(2)(a) of the Law is concerned with something other than simply rules of public policy, but it is not clear that there is any room for such a provision under the Hague Trusts Convention. See further the discussion of Arts. 15, 16 and 18, Hague Convention in Part Two of this book. 355 Art. 45(2)(a), Trusts (Jersey) Law 1984 is set out and considered in more detail in the discussion of “the offshore dimension” to Art. 15(1)(c), Hague Convention in Part Two of this book. 350

The Isle of Man 75

(D) The Isle of Man356 The Isle of Man’s Trust Act 1995 (hereafter “IMTA”), which came into operation on 17 January 1996, is largely inspired by the Cayman Islands legislation (TFEL). There are significant similarities, such as that: (i)

Generally, all questions relating to the trust are to be determined by Manx law if Manx law is the governing law of the trust.357 (ii) The provisions for determining if Manx law governs and for allowing changes to and from that law are similar to those of the TFEL.358 (iii) Manx trusts are similarly “insulated” from claims arising under a foreign law. No trust governed by Manx law may be impugned by reason that (a) a foreign jurisdiction does not recognise the trust;359 or (b) rights are conferred by a foreign law or foreign judicial or administrative order upon a person “by reason of a personal relationship to the settlor or by way of heirship. . .”360 This is troubling, since the Isle of Man is bound by the scheme of the Judgments (Reciprocal Enforcement) (Isle of Man) Act 1968 to recognise on a reciprocal basis money judgments from states which are party to the scheme. That scheme includes several civilian states,361 which might, of course, have forced heirship rules capable of impugning the trust. If all relevant conditions are met, a judgment delivered in one of these states should, in principle, be recognised in the Isle of Man, even if it orders the payment of a sum of money362 from an Isle of Man trust, on the basis of heirship rights conferred by local law. The Act requires the Isle of Man to register judgments which comply with its requirements, and sets out363 the (only) grounds on which registration may be set aside. The only possible argument would be that recognition of foreign judgments hinging on such matters as forced heirship infringes Manx public policy. However, this is to give an extremely broad meaning to the concept of public policy, one which goes far beyond what one might expect (and invites foreign courts asked to recognise Manx “insular” trusts judgments to play a tit-for-tat game and refuse recognition), and is very hard to reconcile with the spirit and approach of Article 15(1)(c), Hague Trusts Convention, which instructs a 356 See J Smalley, “National Digest for the Isle of Man”, in J Glasson (ed.) International Trust Laws, ch. A14, 20–2; M Lupoi, Trusts: A Comparative Study, 215. 357 S. 4, IMTA 1995. 358 Ibid., s. 3. 359 Ibid., s. 5(b)(i) 360 Ibid., s. 5(b)(ii). Emphasis added. 361 For a full list of states within the scheme, see A Briggs and P Rees, Civil Jurisdiction and Judgments, 2nd edn. (London, LLP, 1997), 364 (f/ns 380–2). 362 It could be said that the order will relate to a liability to account, rather than the payment of a fixed debt. However, a civilian state may simply order the payment of a certain sum from the trust, if there are no other assets available to satisfy the forced heirship claim. 363 See s. 4 of the (UK) Foreign Judgments (Reciprocal Enforcement) Act 1933.

76 Transnational Trusts in Offshore Jurisdictions Manx court to apply the mandatory rules of the law specified (by Manx choice of law rules) as applicable to succession.364 Whilst the desire to “insulate” Isle of Man trusts in a competitive trusts market is appreciated, this author would question the legality of the Manx provision and seriously question its appropriateness. (iv) There are many similar exceptions to the general dominance of the governing law.365 (v) The Act prevents the capacity of the settlor from being questioned by reason of the fact that the laws of a foreign jurisdiction prohibit or do not recognise the trust, or a foreign right is defeated or law contravened.366 There are, however, material differences between the two Acts. In the TFEL, the provisions for determining when Cayman Islands law governs are preceded by rules laying down the governing law for any trust, not just one governed by the law of the Islands. No such provision exists in the IMTA, presumably because the general choice of law rule for trusts is dealt with in the Hague Trusts Convention, provision for the incorporation of which was made in the Isle of Man by the Recognition of Trusts Act 1988. Nor does the IMTA contain a clause to the effect that where Manx law is stated to govern a particular aspect of the trust, it shall be presumed that Manx law was intended to apply to the whole trust. Once again, this is presumably to avoid conflict with the Hague Convention rules, which defer in the absence of choice to the law with which the trust is most closely connected.367 By sections 4(1), 4(2)(a) and 4(2)(d), a number of issues, including the capacity of the settlor and the validity of a disposition, are to be determined by Manx law. Oddly, and importantly, the section does not state that this rule only applies if Manx law is the governing law of the trust. Since the Act only really deals with the choice of Manx law, one might assume that to be the case. If not, the result would be that Manx law would always be applied to the capacity question as the law of the forum. Such a choice of law rule has already been criticised in relation to the Bermuda legislation. As regards the “rocket-launching” transfer of property to constitute the trust, it appears that the Manx law will be used if it is the governing law of the trust itself, since Manx law applies to “dispositions”,368 which are defined in section 6 to include every manner in which any legal or equitable interest in property is extinguished. Consistent with the general rule considered above for the inter vivos transfer of property at common law, it might be hoped that, where Manx 364 Even if it could be said that Art. 15(1)(c), Hague Trusts Convention applies to the law designated by Manx choice of law rules on succession, and that s. 5(b)(ii), IMTA 1995 is part of its choice of law rule for succession, the blanket approach of non-recognition runs very much counter to the spirit of Art. 15(1)(c). 365 Ss. 1(1) and 4(3), IMTA 1995. 366 Ibid., s. 5. 367 Art. 7, Hague Convention. 368 S. 4(1)(b), IMTA 1995.

Conclusion 77 law is not applicable to the trust, it will use the lex situs to determine whether the settlor has alienated a proprietary interest.369 Finally, it should be noted that section 1(1) states that the Act does not apply unless, in the case of a testamentary trust or disposition, the trust or disposition is valid under the law of the testator’s domicile at the time of death.370 The Act also does not apply to trusts or dispositions of immovables, unless the trust or disposition is valid by the lex situs.371 This means that the capacity of the settlor to transfer on trust and the vesting of the property in the trustee still require consideration of the lex situs.372 It is doubtful whether the deference to the lex situs shown in relation to trusts of land should be so readily abandoned in relation to inter vivos trusts of movable property. More importantly still, section 1 states that the Act does not apply if the law of the settlor’s domicile (for testamentary trusts) or the lex situs (for trusts of immovables) is not satisfied with respect to both the disposition on trust and the trust itself. In other words, prima facie, both “rocket-launcher” and “rocket” are subjected to these additional laws. However, the Hague Trusts Convention clearly indicates that the trust itself should be subjected only to its governing law, and not to any other law.373 Section 1(1), IMTA seems inconsistent with this requirement. Overall, one is bound to say that the Isle of Man’s legislation takes insufficient account of its pre-existing international obligations, both in respect of the Hague Trusts Convention and the recognition of foreign judgments. It may be that the legislation will require further scrutiny in the future.

22. CONCLUSION

It is suggested that hitherto neither the common law nor the offshore legislation has come up with a truly convincing answer to the complex problems raised by the “rocket-launching” aspects of transnational trusts, and especially to the problem of capacity. Certainly, the offshore legislation is more protective of ettlor autonomy and tends to “insulate” trusts governed by the local law from claims arising under a foreign law, and in particular those of forced heirs. However pragmatic the legislation might be, however, it must be recalled that the choice of law process is but one aspect of transnational litigation. As important is the question of which courts will have jurisdiction to resolve any disputes

369 The general choice of law rule for testamentary transfers of movable property is the law of the deceased’s domicile at death. 370 S. 1(1)(b)(i), IMTA 1995. 371 Ibid., s. 1(1)(b)(ii). 372 It also means that a testamentary trust of land must presumably satisfy both the law of the testator’s domicile and the lex situs. 373 Save in respect of the mandatory rule provisions identified in Arts. 15 and 16, and the public provision of Art. 18. It can hardly be argued that all rules of the lex situs or the law of the testator’s domicile at death are mandatory rules.

78 Conclusion that might arise.374 It is apparent that different “rocket-launching” choice of law rules are applied in different jurisdictions around the world. In any potential litigation, this may make the choice of jurisdiction in which to litigate especially important. Imagine a case of movable property situated in England, which a domiciliary of state A seeks to transfer inter vivos to T in England, to hold on a trust governed by the law of state A. By English law, he lacks capacity to alienate the property; by the law of state A, he has capacity so to do. Suppose that according to the English conflict of laws, the choice of law rule which determines the settlor’s capacity to alienate property inter vivos is the lex situs and that by state A’s law it is the law governing the trust. If T sought directions as to his position in an English court, it would be held that no express trust had been created due to the settlor’s incapacity to transfer legal title; but if he sought directions in the courts of state A, it would be held that a valid express trust had been created. It may well be that this judgment could be enforced in other countries, including England.375 This means that, whatever the choice of law rule in England transpires to be, one must be alert to the possibility that litigation in a transnational matter could take place first in an overseas forum, where quite different choice of law rules might be applicable, and that the effects of that litigation would be unlikely to be confined to the state in which it takes place. This also means that even the most “insulated” of offshore trusts may be impugned by litigation in an overseas forum.376 Indeed, the more an offshore state “insulates” its trusts, the greater the risk is of a conflict of laws arising between that state and a foreign state and of the trustees receiving mutually incompatible directions from the courts of different states.377

374 For detailed discussion of the law of jurisdiction and the enforcement of foreign judgments in the trusts context (including an analysis of the rules in selected offshore states), see J Harris, “Transnational Trusts Litigation—Jurisdiction and the Enforcement of Foreign Judgments”, in J Glasson (ed.) International Trust Laws, ch. C1. 375 An English court will not refuse to recognise a judgment simply because it would have applied a different law had the case been heard in England. See generally, Dicey and Morris, ch. 14; Cheshire and North, ch. 15. See also the discussion of Art. 11, Hague Convention in Part Two of this book. 376 And, in “rocket” matters, strengthens the case for international recognition of the trust and for ratification of the Hague Trusts Convention. See D Hayton, “Whither Trusts in the TwentyFirst Century: Part 2” [2000] Private Client Business 163 for a discussion of how common law courts might approach offshore legislation. See also A Duckworth, “An Offshore View of Forced HeirshipGlobal Conflict and its Planning Implications: Part 2” [1995] Private Client Business 334 for planning advice on avoiding forced heirship claims. 377 For this reason, offshore trusts should normally contain an exclusive jurisdiction clause in favour of the offshore state, in order to try (insofar as possible) to ensure that (at least in disputes concerning relations between settlor, trustee and beneficiary) not only will it have jurisdiction over the trust, but as importantly, that an onshore state will not.

I

Evolution and Application 1. BACKGROUND

The trust, widely used within the domestic law of common law based systems (and elsewhere), does not uphold national boundaries.1 Frequently, trusts created in England concern assets, trustees or beneficiaries located overseas, especially in offshore trust states, and may be subjected to a law other than that of England. This creates a conflict of laws which can lead to problematic consequences, not least due to the sometimes rather individualised and “insular” conflicts rules of certain offshore states.2 However, such problems are considerably less extensive than those which arise where elements of an international trust are connected with a country in which the trust concept is wholly unknown.3 Dyer and van Loon summarise some of the difficulties for non-trust states: “Can the trust, this corpus alienum, be recognized at all? And if so, under what category or categories of the law should the trust or the different legal relationships it creates be arranged: under the law of succession, the law of contract, the law of property, the law of foundations or associations, or any other branch of the law? And if this problem of characterization or classification has been solved, and the applicable law has been determined, further questions of adaptation or conversion into civil law institutions may have to be faced. . . The result is that, when it comes to litigation, the outcome may be highly unpredictable, as the parties and the courts will frequently be in the dark as to the way the trusts should be handled.”4

Sometimes, such problems might involve the creation of the trust itself e.g. when a settlor resident in a civil law state purports to transfers assets to a trustee to be held on trust; other times, problems may arise where a beneficiary purports to invoke his rights pursuant to a validly created trust e.g. if a beneficiary of a trust 1 The reasons for the increasing internationalisation of the trust are considered by A Dyer and H van Loon, Report on Trusts and Analogous Institutions, Preliminary Document No 1, May 1982, Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable Law and Recognition, 10, paras. 113–4, pp. 65–7. 2 See the discussion in section 21 of Part One of this book of “the creation of transnational trusts in offshore jurisdictions.” 3 See M Koppenol-Laforce, “The Trust, the Hague Trusts Convention and Civil Law Countries: a Mission Impossible?” (1998) 3 Notarius International 27. See also V Bolgár, “Why no Trusts in the Civil Law?” (1953) 2 Am J Comp Law 208. 4 Dyer and van Loon Report, para. 115, p. 67.

82 Evolution and Application validly created in England according to English law seeks to terminate the trust and to ensure the transfer to him of trust assets now located in a civil law state. Moreover, even if the trust assets are not located in a civil law state, it might be necessary to seek remedies from the courts of such a state against a trustee for breach of trust, if the trustee is domiciled and resident therein, especially if the trustee is not amenable to the jurisdiction of the courts of a common law state.5 Furthermore, even if it is possible to obtain a judgment in a common law state in relation to such a transnational trust, the possibility that it will need enforcement in a civil law state is very real.6 Should such a state not know or accept the trust concept, then two real dangers exist: either, the state will refuse to enforce the foreign judgment, on the basis (if all others fail), that it infringes its public policy; or it will attempt to “translate” the trust judgment into the nearest civil law analogue.7 This risks distorting the rights, powers and obligations inherent 5 There are provisions contained in the Civil Procedure Rules (CPR), Part 6.20(11)–(15) concerning service out of the jurisdiction with the permission of the English courts in trust matters. Subrule (11) is the main provision concerning express trusts and is limited in scope. It applies where “a claim is made for any remedy which might be obtained in proceedings to execute the trusts of a written instrument where: (a) the trusts ought to be executed according to English law; and (b) the person on whom the claim form is to be served is a trustee of the trusts. Moreover, the claimant would also need to persuade the English court to exercise its discretion to serve out of the jurisdiction, which would require him to demonstrate, inter alia, that the natural forum for the litigation is England: Spiliada Maritime Corp v. Cansulex [1987] AC 460; and see Seaconsar Far East Ltd v. Bank Markazi Jomhouri Islami Iran [1994] 1 AC 438. Where the defendant may be served as of right in the jurisdiction, the defendant may be able to persuade the English court that the natural forum for the litigation nonetheless lies overseas (Spiliada). However, an English court might be expected to be reluctant to entertain such an argument where it is alleged that the natural forum in which to bring trusts litigation (or, at least, trusts litigation concerning the internal relationship between settlor, trustee and beneficiary) is in a non-trust state. Where, in civil and commercial matters, the defendant is domiciled in a contracting state to the Brussels Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters (enacted into English law by the Civil Jurisdiction and Judgments Act 1982, as amended), he may be sued in the courts of his domicile, even though the trust itself is otherwise wholly connected to another contracting state. Art. 5(6) of the Convention may also be applicable. This allows a defendant to be sued “in his capacity as a settlor, trustee or beneficiary of a trust created by the operation of statute, or by a written instrument, or created orally and evidenced in writing, in the courts of the Contracting State in which the trust is domiciled.” Such jurisdiction will not be available, however, if one of the exceptions to the defendant domicile rule is available (on which see A Briggs and P Rees, Civil Jurisdiction and Judgments, 2nd edn. (London, LLP, 1997), 16–79. The Brussels Convention is itself to be replaced by Council Regulation (EC) No 44/2001 of 22 December 2000 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (hereafter “the Brussels Regulation”) with effect from 1 March 2002: OJ 2001 L 12/1, but the trust provisions are substantively unchanged. 6 Not least if the trust concerns immovable property located in a non-trust state. 7 See further e.g. A Bureau, Le Contrat de Fiducie: Études de Droit Comparé Allemagne, France, Luxembourg http://juripole.univ-nancy2.fr/memoires/compare/Antoine_Bureau/; C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs. Fiducie in a Business Context (Brussels, Bruylant, 1999); C de Wulf, The Trust and Corresponding Institutions in the Civil Law (Brussels, Bruylant, 1965); D Hayton and S Kortmann (ed.) Vertrouwd met de Trust; Trust and Trust-Like Arrangements (Deventer, Tjeenk Willink, 1996); D Hayton (ed.) Principles of European Trust Law (The Hague, Kluwer, 1999); D Hayton (ed.) Modern International Developments in Trust Law (The Hague, Kluwer, 1999); D Hayton, “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159; M Lupoi “Trusts and Civilian Categories” in D Hayton, S Kortmann and H Verhagen (eds.), Itinera Fiduciae, Trust & Treuhand in Historical Perspective (Berlin, Duncker & Humblot, 1998); M Lupoi (and others), Fiducia, Trust, Mandato ed

The Desirability of an International Trusts Convention 83 in the trust relationship and may create a real disincentive to create such a trust in the first place. Foremost amongst the problems is the conceptual difficulty created by the treatment of the trust assets as separate from the trustee’s personal wealth and, (in the case of common law trusts), by the division between legal and equitable ownership. Such split ownership is largely unknown in most civil law states, even if the concept of a separate fund immune from the claims of one’s creditors is not.8 Concepts of English law, such as e.g. joint tenancy and the right of survivorship, would be largely meaningless. Furthermore, especially in relation to testamentary trusts, the danger of falling foul of mandatory rules of a civil law of succession is substantial. Some non-trust states might treat the trust assets as part of the trustee’s estate upon his death. Moreover, even if the assets are treated as belonging to the beneficiary’s estate, those assets may come under attack from rules of forced heirship prevalent in many legal systems, which may compel a testator to leave a proportion of his assets to his nearest and dearest.9

2. THE DESIRABILITY OF AN INTERNATIONAL TRUSTS CONVENTION

The frequency of contact between the trust and the non-trust world provides an obvious impetus for clear, harmonised approaches to the trust so that, insofar as is possible, the incidences of the trust will be the same from state to state. Agency (Milan, Giuffrè, 1991); M Lupoi, Introduzione ai Trusts (Milan, Giuffrè Editore, 1994); M Lupoi, “Trusts in the Civil Law—An Introduction” (1996) 2 Trusts and Trustees 21; M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law (Aldershot, Ashgate, 1997), ch. 12, p. 222; M Lupoi, “Effects of the Hague Convention in a Civil Law Country” (1998) 4 Trusts and Trustees 15; M Lupoi, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–1999 (The Hague, Kluwer, 2000), ch. 4, p. 35; M Lupoi, “The Civil Law Trust” (1999) 32 Vanderbilt Journal of Transnational Law 967; M Lupoi, Trusts: A Comparative Study (Milan, Giuffrè, 1997), translated by S Dix, (Cambridge, Cambridge University Press, 2001); B McCutcheon and P Soares (eds.), EuroTrusts: the new European Dimension for Trusts (Southampton, Legal Studies Publishing Ltd, 1993); F Sonneveldt and H van Mens, The Trust: Bridge or Abyss Between Common and Civil Law Jurisdictions? (Deventer, Kluwer, 1992); H Verhagen, “Trusts in the Civil Law: Making Use of the Experience of ‘Mixed’ Jurisdictions”, (2000) 8 European Review of Private Law 477; W Wilson (ed.), Trusts and Trust-Like Devices (United Kingdom Comparative Law Series, Vol. 5) (London, Chameleon Press, 1981). 8 But see M Lupoi, Trusts: A Comparative Study, who argues (at 2–3) that “the splitting of ownership rights between trustee and beneficiary, and the existence of two ownership rights in the same asset (‘legal ownership’ and ‘equitable ownership’ represent misunderstandings. . . . [T]he trustee’s ownership position is distinguished by a lack of equitable fullness . . . what is lacking in order to obtain that fullness does not necessarily belong to another person . . . [E]quitable entitlement which is lacking in the trustee is not necessarily in the possession of others.” 9 The problem may also arise where a settlor purports to create an inter vivos trust in order to evade the application of forced heirship rules on death. Such a trust risks falling prey to claw-back succession provisions of a civil law state, which treat the trust assets as a portion of the deceased’s estate for forced heirship purposes. See further the discussion of Art. 15(1)(c) below and section 17 of Part One of this book.

84 Evolution and Application Substantive harmonisation of the law would have been almost impossible in relation to an institution to which attitudes and approaches differ so very widely. However, an international Convention, drawn up under the auspices of the Hague Conference and seeking to provide private international law harmonisation, and thereby facilitate recognition of the trust overseas, was a much more realistic target. That is not to say that all trust and non-trust states were equally eager to conclude a private international law trusts convention. Jauffret-Spinosi contends that “les pays de common law sont les grands bénéficiaires de la convention”.10 Likewise, Lupoi argues that the Convention differs from all other private international law instruments in being “one-sided”.11 However, a private international law convention establishing common provisions on the law applicable to trusts and their recognition has great potential benefits for trust and non-trust countries alike. It enables civil law countries more easily to appreciate the concept of a trust and to deal with transnational trusts in a manner more likely to fulfil the expectations of the settlor and of all parties to the trust. Problems with trusts are ones which such countries “encountered with a certain amount of frequency, particularly in connection with the estates of deceased persons”.12 For a common lawyer,13 it reduces the unattractiveness of creating a trust over property situated in such a state by providing for the recognition of that trust in contracting states. But it also codifies the private international law rules applicable in common law states, which had themselves not clearly emerged in anything other than an ad hoc manner.14 This latter consideration may not have been the 10 C Jauffret-Spinosi “La Convention de la Haye Relative à la Loi Applicable au Trust et à sa Reconnaissance (1er Juillet 1985)” (1987) 114 Journal du Droit International 23, 63. She even suggests that the willingness of the civil law states to negotiate the Convention was an exercise in altruism. 11 M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–1999, ch. 4, 35. 12 Explanatory Report on the Convention on the Law Applicable to Trusts and on their Recognition by Professor M Alfred Von Overbeck (hereafter “the Von Overbeck Report”), Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition 370, para. 14, p. 373. The Von Overbeck Report is also reproduced in (1986) XXV International Legal Materials 593. References in this book will be to the page numbers in the Hague Proceedings. 13 And a lawyer in any other state having the trust in its domestic law. 14 M Lupoi, Trusts: A Comparative Study, 331 states that “common-law writing and case law were both fumbling in the dark, and there had been a chorus of complaints about the deplorable state of the law.” Discussion of the common law principles includes: D Cavers “Trusts Inter Vivos and the Conflict of Laws”, (1930) 44 Harv LR 161; W Swabenland, “The Conflict of Laws in Administration of Express Trusts of Personal Property”, (1936) 45 Yale LJ 438; W Land, Trusts in the Conflict of Laws; Validity, Construction, Administration and Taxation of Trusts: What Law Governs (New York, Baker, Voorhis & Co, 1940); P Croucher, “Trusts of Moveables in Private International Law” (1940) 4 MLR 111; G Keeton, “Trusts in the Conflict of Laws” (1951) 4 CLP 107; V Latham, “The Creation and Administration of a Trust in the Conflict of Laws (1953) 6 CLP 176”; R Lafer and A Siegel, “Trusts of Movables in the Conflict of Laws” (1961) NYULR 713; H Goldstein, Trusts of Movables in the Conflict of Laws: A Comparative Study of the Laws of the United States and Germany with Reference to the Laws of Belgium, France, Italy and the Netherlands (Köln, dissertation, 1966); A Lowenfeld, “Tempora Mutantur—Wills and Trusts in the Conflicts Restatement”, (1972) 72 Col LR 382; A Wallace, “Choice of Law for Trusts in Australia

The Desirability of an International Trusts Convention 85 main motivation for the Convention,15 but as it becomes ever easier to move assets around the globe, and given that there was so little clear-cut common law authority on choice of law for trusts, it is a considerable advantage. Furthermore, harmonisation is something which might be expected to yield benefits even between common law countries, as the deluge of European and international initiatives in other areas of private international law illustrates. The prospect of a trust’s validity, the trustee’s powers and duties and the beneficiary’s rights being different in the eyes of the various legal systems with which the trust is connected could place all parties concerned in an impossible position. Indeed, harmonisation of choice of law rules has a positive impact on the law of jurisdiction, greatly reducing as it does the incentives to forum shop.16 Contracting states to the Brussels and Lugano Conventions are bound to recognise and enforce trust judgments in civil and commercial matters.17

and the United Kingdom” (1987) 36 ICLQ 454. See also American Restatement (2d) on the Conflict of Laws (St Paul, Minn, American Law Institute, 1971) ch. 10. In contrast, major studies of the private international law of trusts have been conducted in civil law states in recent years; see e.g.: P Czermak Der Express Trust im internationalen Privatrecht (Frankfurt, Lang, 1986); G Wittuhn, Das Internationale Privatrecht des Trusts (Frankfurt, Lang, 1987); F Sonneveldt, De AngloAmerikaanse Trust in het Nederlandse Civiele en Fiscale Recht (Deventer, FED, 1990); M Lupoi (and others), Fiducia, Trust, Mandato ed Agency (Milan, Giuffrè, 1991); J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français (Paris, LGDJ, 1992), esp. ch. 2; M Lupoi Introduzione ai Trusts: Diritto Inglese, Convenzione dell’Aja, Diritto Italiano (Milan, Giuffrè, 1994); C van Boeschoten, Het Haagse Trustverdrag in Nederlands Perspectief Preadviezen uitgebracht voor de Vereniging voor Burgerlijk Recht (Lelystad, Koninklijke Vermande, 1994); E Calo, Dal Probate al Family Trust (Milan, Giuffrè, 1996); V Salvatore, Il Trust: Profili di Diritto Internazionale e Comparato (Padova, CEDAM, 1996); B Beat Aussonderung von Treugut im Schweizerischen Partikularkonkurs (Zurich, Schulthess, 1997); C Gonzales Beilfuss El Trust. Law Institución Angloamericana y el Derecho Internacional Privado Espagñol (Barcelona, Bosch, 1997); M Koppenol-Laforce, Het Haagse Trustverdrag, (Deventer, Kluwer, 1997); J Herbots, and D Philippe, Le Trust et la Fiducie—Implications (Brussels, Bruylant, 1997); T Mayer, Die Organisierte Vermogenseinheit Gemass Art. 150 des Bundesgesetzes über das Internationale Privatrecht: unter besonderer Berucksichtigung des Trust (Basel, Helbing & Lichtenhahn, 1998); T Arrigo (ed.), I Trusts nell’ordinamento Italiano (Milan, Giuffrè, 1999); J de Arespacochaga El Trust: la Fiducia y Figuras Afines (Madrid, Marcial Pons, 2000); E Corso, Trustee e Gestione dei Beni in Trust (Milan, Giuffrè, 2000). 15 P North and J Fawcett, Cheshire and North’s Private International Law, 13th edn. (London, Butterworths, 1999), 1031; but contrast M Lupoi, Trusts: A Comparative Study, 151, who argues that the development of satisfactory choice of law rules for common law states was the key factor behind the Convention and criticises the von Overbeck Report (para. 14, 373) for “taking note of this almost unwillingly.” 16 E Gaillard, “Les Enseignements de la Convention de la Haye du 1er Juillet 1985 relative à la Loi Applicable au Trust et à sa Reconnaissance”, [1990] Revue Juridique et Politique Indépendance et Cooperation 304, 306. 17 The original Brussels Convention was revised upon the accession of the United Kingdom, Ireland and Denmark to the European Community. This revised Convention was concluded in 1978. However, it did not actually enter into force in the United Kingdom until 1 January 1987, after the negotiation (although before the entry into force) of the Hague Trusts Convention. The Lugano Convention (entered into between European Community and European Free Trade Association states) was concluded after negotiation of the Hague Trusts Convention and entered into force in the United Kingdom on 1 May 1992 (shortly after the entry into force of the Hague Trusts Convention).

86 Evolution and Application Moreover, those Conventions contain bases of jurisdiction specific to the trust.18 Yet it is not wholly clear what the word “trust” means for these purposes19 and how the obligation to recognise a trust might be fulfilled in practice.20

3. A CONVENTION EVOLVES

A detailed history of the conception and evolution of the Hague Trusts Convention is not needed.21 The Convention’s genesis was a decision taken at the Fourteenth Session of the Hague Conference in 1980 to formulate an international trusts Convention.22 The Preliminary Report by Dyer and van Loon23 was followed by a series of Special Commission meetings from 1982 to 1984. An ad hoc Drafting Committee was set up in 1982 and the Special Commission adopted a preliminary draft Convention on 28 October 1983. At the Fifteenth Session of the Conference in 1984, the preparation of the Convention was entrusted to the Second Commission, a Drafting Committee and a SubCommittee on general and final clauses. The Convention was then considered at the Fifteenth Session of the Conference on 8–20 October 1984.24 The draft

18 Most notably the special jurisdiction in Art. 5(6) conferred on the courts of the place where a trust created by statute, written instrument or created orally and evidenced in writing is domiciled (This provision is reproduced as Art. 5(6), Brussels Regulation). Art. 17, Brussels Convention contains provisions regulating choice of court clauses in trust agreements (This provision is reproduced in Art. 23, Brussels Regulation). 19 It appears that Art. 5(6), Brussels Convention was firmly modelled on the English trust and not the broader notion of a trust enshrined in Art. 2, Hague Trusts Convention. Difficulties in determining the ambit of the Brussels Convention provision could still arise, however: see S Berti, “Trusts and the Lugano Convention—does it matter? and what about the Hague Trust Convention?” in N Vogt (ed.) Disputes Involving Trusts, (Saffron Walden, Trusts & Trustees; Helbing & Lichtenhahn, 1999) 9, 10–11; see also J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, ch. 2, 217. 20 Still less what a non-trust state would do if it found itself bound to take jurisdiction under the Convention as the court first seised of a trust dispute. 21 More information can be found in the von Overbeck Report, paras. 1–10, pp. 370–2. 22 Final Act of the Fourteenth Session, Part E, No. 3; Acts and Documents of the Fourteenth Session, 1980, Book 1, Miscellaneous Matters. 23 A Dyer and H van Loon, Report on Trusts and Analogous Institutions, Preliminary Document No 1, May 1982, Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition, 10. The Report is a useful introductory source of reference on the treatment of the trust in domestic legal systems (10–65) and on the private international law of trusts in common law and civil law states (65–102). However, it has been heavily criticised by M Lupoi : see e.g. Trusts: A Comparative Study, 340 “The Preliminary Report presented to the delegates was, unfortunately, full of omissions and . . . drawn from second- and third-hand information.” Readers are referred to Lupoi’s book for an up-to-date, first-rate comparative treatment of the law of trusts. See also K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, (Dordrecht, Mohr (Paul Siebeck), Tübingen and Martinus Nijhoff, 1994), ch. 23. 24 The Conference was held under the Presidency of J Schultsz. The Drafting Committee was chaired by Professor von Overbeck (see von Overbeck Report, paras. 6 and 7, p. 371).

A Convention Evolves 87 Convention was adopted unanimously on 19 October 1984 and the Final Act signed on the following day.25

25 See Proceedings of the Fifteenth Session (1984), Book II. The Official Report by von Overbeck appears at 370. Commentaries on the Convention include: A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn. (Edinburgh, W Green, 1990); J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, ch. 2, 215; G Broggini “Folgen der Ratifikation des Trusts—Übereinkommens in Italien”, 9e Journée des Professeurs Suisses de Droit International Privé, Lausanne, 7 March 1997, available at http://www.isdc.ch/e/ Broggini97.asp; J-G Castel, Canadian Conflict of Laws, 4th edn. (Toronto and Vancouver, Butterworths, 1997) ch. 30, 544–552; L Collins (ed.), Dicey and Morris, The Conflict of Laws, 13th edn. (London, Sweet and Maxwell, 2000), ch. 29; G Droz “La Conférence de la Haye et le Droit International Privé Notarial: Deux Récentes Conventions en Matière de Trusts et de Successions” [1989] Journal des Notaires et des Avocats 507; A Dyer, “International Recognition of the Trust Concept” (1996) 2 Trusts and Trustees 5; A Dyer, “International Recognition and Adaptation of Trusts: the Influence of the Hague Convention” (1999) 32 Vanderbilt Journal of Transnational Law 989; E Gaillard, “Les Enseignements de la Convention de la Haye du 1er Juillet 1985 relative à la Loi Applicable au Trust et à sa Reconnaissance”, [1990] Revue Juridique et Politique Indépendance et Cooperation 304; E Gaillard and D Trautman, “Trusts in Non-Trust Countries: Conflict of Laws and the Hague Convention on Trusts” (1987) 35 Am J Comp Law 307; D Hayton, “The Hague Convention on the Law Applicable to Trusts and on their Recognition”, (1987) 36 ICLQ 260; D Hayton, “The Hague Convention on Trusts: a Little is Better than Nothing but Why so Little?” (1994) 3 J Int Corp P 23; D Hayton, “The Significance of the Hague Conventions on Trusts and Succession: a Common Law Perspective” in A Borras, A Bucher, T Struycken and M Verwilghen (eds.), E Pluribus Unum: Liber Amicorum Georges Droz: On the Progressive Unification of Private International (The Hague, Martinus Nijhoff, 1996) 121; D Hayton, “International Recognition of Trusts” in J Glasson (ed.) International Trust Laws (Bristol, Jordans, loose-leaf), ch. C3; D Hayton, Underhill and Hayton, The Law Relating to Trusts and Trustees, 15th edn. (London, Butterworths, 1995) ch. 23; D Hayton, “The Developing European Dimension of Trust Law (2000) 11 KCLJ 48, 52; C Jauffret-Spinosi “La Convention de la Haye Relative à la loi applicable au trust et à sa reconnaissance (1er juillet 1985)” (1987) 114 Journal du Droit International 23; F Klein “A Propos de la Convention de La Haye du 1er Juillet 1985 Relative à la Loi Applicable au Trust et à sa Reconnaissance” in F Sturm (ed.) Mélanges Paul Piotet (Berne, 1990) 467; M Koppenol-Laforce, Het Haagse Trustverdrag (Deventer, Kluwer, 1997) (ch. 7 contains a summary of the book in English); M Koppenol-Laforce, “The Trust, The Hague Trusts Convention and Civil Law Countries: a Mission Impossible?” (1998) 3 Notarius International 27; H Kötz “Die 15 Haager Konferenz und das Kollisionsrecht der Trust” (1986) 50 Rabels Zeitschrift 562; H Kötz, “The Hague Convention on the Law Applicable to Trusts and their Recognition”, in D Hayton (ed.) Modern International Developments in Trust Law (The Hague, Kluwer, 1999), ch. 3, 44 (also published in A Kaplan (ed.) Trusts in Prime Jurisdictions (The Hague, Kluwer, 2000), ch. 1); K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23 (esp. pp. 27–31); M Lupoi Introduzione ai Trusts: Diritto Inglese, Convenzione dell’Aja, Diritto Italiano (Milan, Giuffrè, 1994); M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15; M Lupoi, Trusts: A Comparative Study (Milan, Giuffrè, 1997), translated by S Dix (Cambridge, Cambridge University Press, 2001), ch. 6; L Maerten, “Le Régime International du Trust après la Convention de La Haye du 1er Juillet 1985” [1988] La Semaine Juridique 3319; J Mowbray, L Tucker, N Le Poidevin and E Simpson, Lewin on Trusts, 17th edn. (London, Sweet and Maxwell, 2000), ch. 11; P North and J Fawcett, Cheshire and North’s Private International Law 13th edn., ch. 35; P Nygh and M Pryles “The Hague Convention on the Law Applicable to Trusts and on their Recognition” in Thirteenth International Trade Law Conference 1–3 November 1986 (Canberra, Australian Government Publishing Service, 1987) 199; A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn. (London, Sweet and Maxwell, 1998), ch. 25, (esp. 752–9); P Piccoli “La Convenzione dell’ Aja sulla Legge Applicabile ai Trust” in Fiducia, Trust, Mandato ed Agency (Milan, Giuffrè, 1991); C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé 7; C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, 9e Journée des Professeurs

88 Evolution and Application 4. CURRENT STATUS OF THE CONVENTION

The Convention has now been ratified by,26 and has entered into force in, Australia,27 Canada,28 Hong Kong,29 Italy,30 Malta,31 the Netherlands32 and Suisses de Droit International Privé, available at http://www-isdc.ch/e/Reymond97.asp; V Salvatore, Il Trust: Profili di Diritto Internazionale e Comparato (Padova, CEDAM, 1996), part 2, chs. 4–6; J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5; C van Boeschoten, Het Haagse Trustverdrag in Nederlands Perspectif (Lelystad, Koninklijke Vermande, 1994); A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6, 7; A von Overbeck, “Trusts: the Hague Convention and the Law of the Netherlands”, 9e Journée des Professeurs Suisses de Droit International Privé, available at http://www.isdc.ch/e/vO97.asp; A von Overbeck, “La Ratification de la Convention de la Haye sur le Trust par les Pays Bas: un Exemple pour la Suisse?” in Collisio Legum, Mélanges offerts à G Broggini (Milan, Giuffrè, 1997); D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation—Kollision und Vereinheitlichung—Conflicts and Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème Anniversaire (Fribourg, Éditions Universitaires, 1990) 591; K Williams, “Trusts and the Hague Convention” (1986) 7 Company Lawyer 95. A very helpful bibliography to the Hague Trusts Convention is provided by the Hague Conference and is updated regularly. It can be found at http://www.hcch.net/e/conventions/bibl30e.html.26 For a list of states which have signed and/or ratified the Convention and the date of entry into force of the Convention into those countries, see http://www.hcch.net/e/stat30e.html. 27 See the Trusts (Hague Convention) Act 1991. It entered into force on 1 January 1992. 28 It entered into force on 1 January 1993 in Alberta, British Columbia, New Brunswick, Newfoundland and Prince Edward Island. The Convention currently applies to the states of Alberta, British Columbia, Manitoba (entry into force in Manitoba: 1 July 1994), New Brunswick, Newfoundland and Prince Edward Island and Saskatchewan (entry into force in Saskatchewan: 1 September 1994). For details on the enacting legislation of the various states, see J-G Castel, Canadian Conflict of Laws, 4th edn., 544, esp. f/n 40. For discussion of the (tepid) reaction to the Convention in Ontario, see M O’Sullivan, “The Hague Convention on Trusts—Further Considerations” (1993) 2 J Int Corp P 65. 29 The United Kingdom had extended the application of the Convention to Hong Kong with effect from 1 January 1992. The Government of the People’s Republic of China determined that the Convention would continue to apply to the Hong Kong Special Administrative Region with effect from 1 July 1997. Details of the correspondence can be found on the Hague Conference web site at: http://www.hcch.net/e/status/stat30e.html. It is interesting to note that China has itself recently introduced legislation to promote the trust in China: see A Goodwin, “Lings to China: the new PRC Trust Law” (2001) 29 Trusts and Estates Law Journal 19. 30 It entered into force on 1 January 1992. The Italian ratification instrument can be found in M Lupoi, Trust Laws of the World (Rome, ETI, 1996), 26. There has been considerable interest in the trust in Italy. Books published include: A Gambaro, Il Trust nell Prassi Bancaria e Finanziariai (Milan, CEDAM, 1998); M Lupoi, Appunti sulla Real Property e sui Trust nel Diritto Inglese (Milan, Giuffrè, 1971); M Lupoi (and others), Fiducia, Trust, Mandato ed Agency (Milan, Giuffrè, 1991), M Lupoi Introduzione ai Trusts: Diritto Inglese, Convenzione dell’Aja, Diritto Italiano (Milan, Giuffrè, 1994); I Beneventi (ed.) I Trusts in Italia Oggi (Milan, Giuffrè, 1996); V Salvatore, Il Trust: Profili di Diritto Internazionale e Comparato (Padova, CEDAM, 1996). A fuller list can be found in F Albisinni, National Digest for Italy, ch. A51 in J Glasson (ed.) International Trust Laws, f/ns 6 and 7. See also selected papers from Il Trust in Italia: Congresso Nazionale, Rome, 19–20 November 1999, available at: http://www.il-trust-in-italia.it/webtutti/@Congress%201999/ Congresso%201999%20tabella.htm 31 It entered into force on 1 March 1996. Technically, it would be correct to say that Malta acceded to the Convention, rather than ratified it. It only became a member state of the Hague Conference on 30 January 1995. 32 It entered into force on 1 February 1996.

The Status of the von Overbeck Report 89 the United Kingdom.33 It has also been signed, though not ratified by Cyprus,34 France,35 Luxembourg 36 and the United States.37 In the United Kingdom, the Recognition of Trusts Act 1987 gives effect to the Convention.38 It appears as a Schedule to section 1 of the 1987 Act.39 Section 1(1) of the Act states that “the provisions of the Convention. . . shall have the force of law in the United Kingdom.” In the initial ratifying states of the United Kingdom, Australia and Italy, the Convention came into force on 1 January 1992.

5. THE STATUS OF THE VON OVERBECK REPORT

The Explanatory Report on the Convention by von Overbeck40 details the evolution of the Convention and provides detailed commentary on each Article of the Convention. The courts of contracting states may, of course, refer to this and may find it persuasive in resolving ambiguities. However, it has no binding status and contracting states’ courts are free to depart from the views expressed in the Report in determining the meaning or scope of provisions contained in the Convention. In the United Kingdom, the Recognition of Trusts Act 1987 contains no reference to the Report. 41

33 This includes Bermuda, British Antarctic Territories, the British Virgin Islands, the Falkland Islands, Gibraltar, Guernsey (the UK extended the operation of the Convention to Guernsey on 28 April 1993), the Isle of Man, Jersey (the UK Recognition of Trusts Act 1987 was not extended to Jersey, but Jersey became a party to the Convention on 1 March 1992, having brought its Trusts (Jersey) Law 1984 into line with the Convention by the Trusts (Amendment No 2) (Jersey) Law 1991), Montserrat, St Helena and Dependencies, South Georgia, the South Sandwich Islands and the Turks and Caicos Islands (entry into force in Turks and Caicos: 1 July 1993). See also s. 2, Recognition of Trusts Act 1987. The Convention entered into force in the United Kingdom and in the territories to which it was initially extended on 1 January 1992. 34 Signed on 11 March 1998. 35 Signed on 26 November 1991. 36 Signed on 1 July 1985. However, a Bill was introduced to implement the Hague Convention on 14 December 2000 and approved by the Luxembourg Chamber of Commerce on 14 March 2001 (see http://www.cc.lu/avis2001/2414wje.htm). See also D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. 37 Signed on 13 June 1988. 38 The United Kingdom signed the Convention on 10 January 1986 and ratified it on 17 November 1989. In the event, the Convention entered into force on 1 January 1992. 39 As Cheshire and North note (at 1031), not all provisions of the Convention are scheduled to the Act. Cheshire and North suggest that a fuller version of the Convention should have been scheduled. However, it is not clear how much substantive difference this would have made, save in the case of the omission of Art. 13 (on which, see below), to which Cheshire and North (at 1041) raise no objection. 40 Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition, 370. 41 Contrast the Official Reports on successive versions of the Brussels Convention, which s. 3(3), Civil Jurisdiction and Judgments Act 1982 states “may be considered in ascertaining the meaning or effect of any provision of the Brussels Convention and shall be given such weight as is appropriate in the circumstances.”

90 Evolution and Application

6. THE IMPLICATIONS OF RATIFICATION FOR NON - TRUST STATES AND THEIR DOMESTIC LEGAL SYSTEMS

Whilst the ratification status of the Convention thus far represents a not inconsiderable achievement, it constitutes a relatively small percentage of the countries which participated in the Fifteenth Session of the Hague Conference, at which the Convention was concluded.42 Perhaps this is not surprising. The Convention’s greatest strength is also its weakness. It is not concerned in any sense with harmonising the domestic trust law of common law and other trust countries. Still more fundamentally, the Convention does not introduce the trust into the domestic law of states which do not have the trust concept. It does not ask them to digest a concept which might appear too radical to the operation of their domestic law. In that sense, private international law is used as a middle way, introducing a concept into a country’s conflicts law, without asking it to face the full impact of the trust. There can be little doctrinal objection to the introduction of a concept for private international law purposes only. It is in the nature of private international law, and especially of the choice of law process, that states routinely have to allow actions to be brought in their courts, or foreign judgments to be recognised, in respect of causes of action which are unknown in their domestic legal systems. Of course, such claims could be dismissed, on the basis that the litigation of a cause of action unknown to the courts of the state in question is contrary to public policy. Such an approach would, however, risk undermining the entire basis of the conflict of laws. But in another sense, introduction of a private international law of trusts into a state with no domestic law of trusts and no great familiarity with the basic trust concepts might be thought to resemble building a house without foundations. The application of such a specialised scheme of trust protection in a state with no tradition of dealing with the concept will create an atmosphere of uncertainty as to how, in practical terms, that state will deal with the consequences of recognition of a trust 43—a prospect whose attractiveness may not be obvious to either the civil law state itself, nor to residents of common law 42 The represented states were: Argentina, Australia, Austria, Belgium, Canada, Cyprus, Czechoslovakia, Denmark, Egypt, Finland, France, Federal Republic of Germany, Greece, Ireland, Israel, Italy, Japan, Luxembourg, Norway, Netherlands, Poland, Portugal, Spain, Sweden, Switzerland, Surinam, Turkey, United Kingdom, United States, Uruguay, Venezuela, Yugoslavia. There was also an observer from Panama. 43 And to potential dissatisfaction in civil law states as to the unequal treatment of the trust and related, but quite distinct domestic constructs of those systems. That, in turn, might make or break the Convention: it could lead to calls for the introduction of the trust into a civil law state’s domestic legal system by providing “un point d’ancrage solide permettant de faire ultérieurement avancer l’ideé, autrement audacieuse, d’une telle réforme des droits internes de tradition continentale”: E Gaillard, “Les Enseignements de la Convention de la Haye du 1er Juillet 1985 relative à la Loi Applicable au Trust et à sa Reconnaissance”, [1990] Revue Juridique et Politique Indépendance et Cooperation 304, 314; and see C Reymond, “Réflexions de droit comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé, 7, 20–1.

The Implications of the Ratification for Non-Trust States 91 states minded to create trusts connected with a non-trust state. The von Overbeck Report notes the unique nature of this Convention. Previous Hague Conventions had to reconcile private international law differences between states as to connecting factors applicable in areas such as adoption, divorce and maintenance.44 This Convention “is more particularly intended to build bridges between countries of common law and countries of civil law”;45 bridges, which, in this case, appear to span a chasm. Furthermore, ratification of the Convention in a non-trust state might lead, in practice, to criticism of the unequal treatment in that state of the trust and the nearest comparable domestic institutions, thereby acting as a strong catalyst for introduction of the trust into domestic law. Koppenol-Laforce and Kottenhagen sum up the predicament in the Netherlands: “On the one hand [the Dutch legislator] . . . explicitly forbids the introduction of the trust in Dutch law. On the other, there is the ratification of the Hague Trusts Convention . . . into Dutch law. In a commercial setting this development will lead to uncertainty and inequality as regards the law since foreign trusts have their own status in Dutch law. In our view the legislator must realise that the consequences of the ratification of the Convention are such that new legislation is needed in which the possibility of creating a ‘Dutch trust’ is realised.”46

Others go further still. Lupoi contends that the Convention is not purely confined to matters of private international law.47 He postulates an example of an English and an Italian resident, both of whom transfer shares in an Italian company on trust to an Italian solicitor in the face of claims by a creditor. The transfer by the English resident will be valid, even though it involves extraterritorial application of the trust. Moreover, if the transfer by the Italian resident is not permitted, the difference in result looks unsatisfactory, since both are quintessentially Italian transfers.48 He suggests that this dichotomy may even be unconstitutional in Italy.49 Important as these points are, however, they do not demonstrate that the Convention necessarily substantively affects the domestic

44

E.g. preferences for nationality, domicile or habitual residence. Von Overbeck Report, para. 12, p. 372. 46 M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” in Netherlands Reports to the XVth International Congress of Comparative Law, Bristol 1998 (Antwerpen/ Groningen Intersentia Rechtswetenschappen), section II A 3, 137, 152 (also published in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs. Fiducie in a Business Context, ch. 5. Koppenol-Laforce points out (in Het Haagse Trustverdrag, 271) that “the beneficiary of a trust enjoys a better protection than a beneficiary of a Dutch fiducia cum amico.” See also H Verhagen, “Trusts in the Civil Law: Making Use of the Experience of ‘Mixed’ Jurisdictions”, (2000) 8 European Review of Private Law 477. 47 M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law (Aldershot, Ashgate, 1997) 222, 227. 48 Another example concerning acquisition of land in Tuscany on trust by English and Italian nationals respectively is given by M Lupoi in Trusts: a Comparative Study, 360 (f/n 171). A similar point is made by M Koppenol-Laforce, Het Haagse Trustverdrag, 271. 49 M Lupoi, ibid, 360. 45

92 Evolution and Application law of any state.50 It is endemic in private international law that very similar factual scenarios may lead to different results in a court, depending on which law is applicable. That, in turn, may well be influenced by the personal connecting factors of the persons involved. There will always be borderline cases which may be characterised differently, or subjected to different laws, but whose outcome may look awkward when the two cases are compared. Moreover, the Convention neither sanctions the application of the law of a non-trust state nor requires51 the recognition of trusts in purely internal contexts where the law of closest connection is that of a state not having the trust in its domestic law.52 Where an English resident is involved, the context cannot be described as purely domestic. That said, the Convention will necessitate the recognition of trusts which, although not wholly internal, nonetheless have very substantial connections with a non-trust state. It may even validate a trust whose sole “international” aspect is the choice of a trusts state’s law to govern. That is likely to render the Convention a strong catalyst for domestic evolution of the trust concept.53 Goodhart memorably cites the advent of the Hague Trusts Convention as one reason why, “It is at least possible that the trust will in the twenty-first century join those other English inventions, such as football and the steam engine, which have swept the world”.54

50 Cf. G Broggini “Folgen der Ratifikation des Trusts—Übereinkommens in Italien”, at http://www.isdc.ch/e/Broggini97.asp; who remarks that “die Ratifikationen des Trust-Übereinkommens hat in Italien die Rechtsfigur des Trusts nicht eingeführt.” 51 This depends upon whether a contracting state has enacted Art. 13 and upon how it exercises its discretion thereunder: see the discussion of Art. 13 below. 52 However, the extremely broad understanding of the trust under Art. 2 may mean that very few states are non-trust states within the meaning of the Convention, so that they will have little opportunity to refuse recognition to “internal” trusts under Art. 13: see “the characteristics considered: the ‘shapeless’ trust” in the discussion of Art. 2, below. 53 And has done in Italy, where objectively “internal” Italian trusts have been validated by choice of the law of a trust state and where plans are afoot to introduce an Italian domestic law of trusts: see further M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–1999, ch. 4, 35, 46–7. But cf. D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation—Kollision und Vereinheitlichung—Conflicts and Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème Anniversaire (Fribourg, Éditions Universitaires, 1990) 591, 609: “The economically independent civil law jurisdiction has neither need for the trust nor the doctrinal structure wherewith readily to accommodate it.” Waters goes on to state, however, that, “If unification of law in this area is ever to come, it will come in the writer’s opinion through an initial harmony at the conflict of laws level” (ibid., 609). 54 Sir W Goodhart, “Trust Law for the Twenty-First Century” in A Oakley (ed.) Trends in Contemporary Trust Law (Oxford, Clarendon Press, 1996), ch. 11, 259. The statement may mislead somewhat, since the trust has developed in many other jurisdictions outside the courts of equity and without the division between legal and beneficial ownership: see G Gretton, “Trusts without Equity” (2000) 49 ICLQ 599.

An Open-Ended Convention 93

7. AN OPEN - ENDED CONVENTION

The Convention is not reciprocal in nature. For matters within its scope, there is no territorial limit and it matters not that the law applicable to the trust is that of a non-contracting state.55 Since the Convention is largely about making sense of the trust in non-trust countries, this is hardly surprising. If the “trust” has certain key characteristics, then presumably a contracting state which wishes to recognise trusts within its jurisdiction ought not to discriminate between trusts on the basis of their applicable law. Indeed, reciprocal treatment between common law and civil law contracting states is largely meaningless in this area. A civil law state may have institutions analogous to the common law trust; but equally, it may have no such institutions, so that recognition will be a “onesided” process.56 It would, of course, have been possible to provide that civil law states need only recognise trusts governed by the law of contracting states. It was thought that this might encourage common law states to ratify the Convention.57 However, since it appeared that the common law states were willing to press ahead with a Convention even between themselves, this last argument was not convincing. Ultimately, the Convention appears to contain sufficient safeguards against tyrannical trust laws of non-contracting states that the arguments against extension to non-contracting state trusts do not appear strong. Moreover, it would have been most unfortunate if a change of applicable law from the law of contracting trust state to a non-contracting trust state during the trust’s lifetime could have affected its ability to be recognised in a non-trust state.58 And since the entire Convention is ex hypothesi not reciprocal,59 the case for exclusion of such trusts is not obvious. The leap of faith made by non-trust states would be in recognising trusts at all, not in the recognition of the trusts of a particular state whose law would be applied. Nonetheless, the reservation permitted in Article 2160 against the application of the Convention to trusts governed by the law of a non-contracting state represents a fair compromise, in that it does remove 55 But see Art. 21, which provides that “any Contracting State may reserve the right to apply the provisions of Chapter III [on recognition of trusts] only to trusts the validity of which is governed by the law of a Contracting State.” Such a reservation was not made by the United Kingdom, and it is submitted rightly so. 56 Although this may not be true, if the very broad description of the trust in Art. 2 is strictly followed, since most states will have an institution which is homologous to, or analogous to, the “shapeless” trust described therein: see further M Lupoi, “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees 15. 57 So that their trusts would be recognised in civil law contracting states: see von Overbeck, para. 33, p. 377. 58 It would not have prevented the trust’s recognition, since Art. 14 allows contracting states to adopt more liberal recognition rules if they so choose. However, the prospects of non-trust states choosing to do so were, of course, remote. 59 In the sense that it is largely concerned with common law trusts being recognised in non-trust states, rather than with the recognition of civil law homologues and analogues in trust states. 60 On which, see below.

94 Evolution and Application a disincentive that might exist in the eyes of some states from ratifying the Convention. As regards common law contracting states, the motivation for application of the Convention is as much about establishing choice of law principles for one’s own state as it is for seeking the recognition of local trusts abroad. An English court, applying the Hague Trusts Convention, might conclude that a trust was governed by the law of a non-contracting trust state. However, if that is the law designated by application of English choice of law rules, it is ex hypothesi the appropriate one to apply to the trust and it is very hard to see why the trust should not be recognised in England. Of course, that lends no guarantee as to the contents of the domestic trust law of the system identified; but the choice of law process never does give any such guarantee, and any objections as to the content of a foreign legal system should be dealt with by a public policy let out.61

8. APPLICATION OF THE CONVENTION IN THE UNITED KINGDOM : THE TIME FACTOR

Article 22 provides that “the Convention applies to trusts regardless of the date on which they were created” but permits contracting states to reserve the right not to apply the Convention to trusts created before the date on which the Convention enters into force in that contracting state.62 By section 1(5), Recognition of Trusts Act 1987, the United Kingdom purported to exercise this power in relation to “anything done or omitted before the coming into force of this Act.” Unfortunately, this wording is not as clear as it might have been. Since section 1(5) is enacted pursuant to Article 22 of the Convention, this would tend to suggest that, if the trust was formed prior to the entry into force of the Act in the United Kingdom, the Convention rules are inapplicable even to breaches committed after entry into force of the Act;63 a view which attracts the support of Lewin on Trusts.64 However, the language of section 1(5) is arguably more consistent with the view that pre-Act acts or omissions are not subject to the Convention rules; whereas post-Act breaches of pre-Act trusts would be. If this is correct, then it is unclear how far the Act extends. Section 1(5) makes it apparent that, where an obligation was required to be performed prior to the commencement of the Act, in respect of a trust created before that date, common 61 At least where the matter concerns the application of foreign law in an English court. Where recognition of a foreign state’s judgments is concerned, the substance of the decision should not be reviewed: Godard v. Gray (1870) LR 6 QB 139; Art. 29, Brussels Convention (to be reproduced as Art. 36, Brussels Regulation). 62 See further the discussion of Art. 22 below. See also the decision of the Supreme Court of New South Wales in Saliba v. Falzon [1998] NSWSC 302. 63 Put differently, the phrase “anything done . . . before the coming into force of this Act” in s. 1(5) means the creation of a trust prior to that date. 64 Lewin on Trusts, 17th edn., 287

Application in the UK: The Time Factor 95 law principles apply. But other examples are more problematic. For example, imagine an express trust created in writing in 1986. In 1999, it is alleged that the trustees invested in unauthorised share investments in breach of the express terms of the trust instrument that the trust fund should only be invested in government bonds65 and a claim is brought for breach of trust. Does the Recognition of Trusts Act apply? In one sense, the breach stems solely from the creation of a trust pre-Act. That trust defines the obligations of the trustee and what may or may not be done by him. Those obligations in turn were set against a framework of private international law principles of the common law. The demands of legal certainty mean that it is not obvious that such obligations should be affected by a change in the private international law rules of the forum. On the other hand, it is clear that the “thing done”, i.e. the act of unauthorised investment, occurred post-Act and, on a literal construction, it does appear that the Act’s rules would apply to the dispute. Still more difficult examples can be imagined. What of a trust created pre-Act, which imposes a continuous obligation on the trustee? For example, suppose that a trust is created in 1986. In 2001, it is alleged that the trustee failed in his obligation to invest the fund prudently and to diversify investments throughout the lifetime of the trust and that the value of the fund is accordingly less than it ought reasonably to be. The obligation to invest, and continually to review the portfolio of investments, is one which is ongoing.66 It will be very hard to pinpoint a time at which it was breached. And even if one could, can one really say that failure to realise a proper value for the trust fund pre-Act should be governed by the common law, whilst post-Act failure should be the subject of the Convention regime? On balance, however, it is suggested that obligations which the trustee continues to be under after the date when the Act entered into force, and which are common to trustees of any settlement67 (i.e. are not the subject of specific regulation in the trust instrument), should be regulated by the Convention rules, unless it can be clearly shown that the breach in question was committed in its entirety before the Act entered into force. Testamentary trusts also pose problems. What if a testator makes a will in 1986 containing trusts and dies in 2001? It might be said that the relevant “thing done” was the creation of the will pre-Act and that the trusts therein should 65 In English domestic law, the Trustee Act 2000 has swept way the technicalities concerning authorised and unauthorised investments in favour of a general power of investment (contained in s. 3 and schedule 1(1)(a)). This power is subject to the trustee exercising the statutory duty of care in the exercise of his powers and duties (s. 1) and to the need to apply standard investment criteria (s. 4) and to obtain advice about how to exercise his investment power (s. 5). These provisions apply to trusts existing at the time of the Trustee Act’s inception (s. 7(1)) but the general investment power is made subject to any restrictions or exclusions imposed by the trust instrument (s. 6(1)(b)). Hence the investment in the example given would continue to be regarded as unauthorised as a matter of English domestic law. 66 See the “standard investment criteria” now applicable in English domestic law (ss. 4 and 5, Trustee Act 2000). 67 In that they apply in the absence of specific provision to the contrary, or a valid exclusion or limitation of liability clause.

96 Evolution and Application accordingly be subjected to the common law. On the other hand, the will is ambulatory until the testator’s death. It might be said that what the settlor had “omitted to be done” was to change the will after the entry into force of the Act and before his death and that the Convention might accordingly apply. Needless to say, such a view could have unacceptable consequences for the testator, whose will has been drawn up against the backdrop of the common law rules and who might scarcely be expected to know that the entry into force of the Act necessitated its alteration. Accordingly, it is submitted that this example should be governed by the common law. Most difficult of all, perhaps, would be cases where the trust is not express in nature.68 For example, if a trustee of a 1986 settlement were, in 2001, to place himself in a conflict of interest situation and make an unauthorised profit from his fiduciary position, would the Act govern a claim made under English law to make him disgorge the profit? This time, the case looks stronger for saying “yes.” For, although the trustee may be viewed as acting in breach of the (implied) terms of the (express) 1986 settlement trust, he also stands in a fiduciary relationship, in respect of which a trust may be imposed over unauthorised profits. True, the constructive trust may not have arisen had the defendant not already been in a pre-existing trustee-beneficiary relationship. Furthermore, it will argued below that the law which governs the fiduciary position of a trustee should normally be that which governs the express trust.69 However, the present issue is one of timing; and the paradigm case for applying the common law rules rather than the Convention is where there is an expectation of the parties to protect that the common law would govern. The constructive trust in our example did not stem from the expressly stipulated terms of the original trust instrument itself which regulated the parties’ powers and duties. It would not appear that legal certainty would be undermined by application of the Convention rules to the trustee’s fiduciary obligations. More generally, it might be argued that a sensible way to interpret section 1(5) is to ask whether the obligation in respect of which a claim is brought was specifically regulated70 in a trust instrument created prior to the date of entry into force of the Act. Insofar as it was and is now breached, the common law rules should apply, since those rights and duties were set against a particular legal framework of domestic and private international law. True, the parties may have had absolutely no expectation as to the private international law rules which would regulate the trust; but the rights and duties were nonetheless 68 It is not certain whether, and to what extent, the constructive trust falls within the scope of the Convention. However, the Convention’s scope is extended in the United Kingdom. The Recognition of Trusts Act 1987 also applies to any trusts arising under the law of any part of the United Kingdom (and to trusts arising by judicial decision anywhere) by s. 1(2). See further the discussion of Art. 3 below. 69 See the discussion of Art. 8(2)(g) below. 70 Normally, this will be express; though there is no reason why obligations which the settlor impliedly intended to form part of the trust should not also be included, provided that the inference of intention is not drawn too lightly.

Application in the UK: The Time Factor 97 formulated expressly, and only made sense in so far as they were construed against a legal framework subsisting at that time. Where, however, the obligation arises either from more general legal principles (e.g. the duty not to profit from a fiduciary position) or ongoing responsibilities (e.g. to meet the requisite standard of care in the trustee’s conduct of his activities), arguably the Convention rules should apply to breaches committed after the commencement of the Act, even if an express trust was created prior to the commencement of the Act. It might be objected that “general” obligations and standards of care still exist only by reference to a particular domestic legal system, and that applying the Convention may lead to a different legal system being applied. However, if the trust instrument contains a choice of law clause, then it is unlikely that the Convention will result in a change to the applicable law. True, the provisions relating to e.g. mandatory rules introduced by the Convention may make a difference;71 but if a rule is designed to be superimposed onto the law applicable to the trust, then the case for its application looks very strong, whether at common law or under the Convention. If the trust instrument contains no choice of law clause, then it will be far from obvious that the parties will have any expectation to protect as to the law regulating their rights and obligations.72 In any event, it must, of course, be remembered that the trust may require recognition in an overseas state, which may not have enacted a provision pursuant to Article 22. Such a state will then apply the Convention rules to determine whether to recognise a trust created in 1986 in England against the backdrop of the common law choice of law rules. Hence e.g. a testamentary trust created in 1986 by a testator who died in 2001 might have its validity determined by different regimes in different states.

71 72

See Arts. 15 and 16, below. At least if there is no implied choice of law either.

II

The Convention’s Provisions PREAMBLE The States signatory to the present Convention, Considering that the trust, as developed in courts of equity in common law jurisdictions and adopted with some modifications in other jurisdictions, is a unique legal institution,73 Desiring to establish common provisions on the law applicable to trusts and to deal with the most important issues concerning the recognition of trusts, Have resolved to conclude a Convention to this effect, and have agreed upon the following provisions-

73 This clause should be contrasted with the much broader description of the trust which ultimately emerged under Art. 2 and which is certainly not confined to the common law trust. “The preamble is misleading where it mentions trusts that are developed in courts of equity. In the course of the negotiation the makers decided to also encompass legal concepts that are structurally analogous to the trust”: M Koppenol-Laforce, Het Haagse Trustverdrag, 265. See further the discussion of Art. 2, below. The Preamble does not appear in the schedule to the Recognition of Trusts Act 1987.

ARTICLE 1—FUNCTIONS OF THE CONVENTION CHAPTER I—SCOPE This Convention specifies the law applicable to trusts and governs their recognition. The Convention deals with two intimately connected areas: (i) the choice of law rules for determining the law applicable to trusts; and (ii) the recognition of trusts.

1. CHOICE OF LAW

As to the former, the choice of law rules codify and clarify the means for ascertaining the applicable law in “trust” states, which are much the more likely fora in which trust litigation will occur and in which the rules will fall to be applied. In many ways, it is the harmonisation of choice of law rules between common law states which is pivotal. Such states may be accustomed to the application of choice of law rules, but, in the absence of unifying choice of law rules, the prospect of a trustee being placed in an impossible position by being exposed to different legal positions in different states is a real one. That said, it is possible that a non-trust state may find itself jurisdictionally competent over trust litigation (and have no discretion to decline jurisdiction or to stay its proceedings).74 Such states have often shown considerable ingenuity when dealing with the trust; but they will almost certainly lack a detailed system of choice of law rules for it. Moreover, those states will now apply the choice of law rules directly to the trust, rather than translating or assimilating it into the nearest civil law analogue and applying the choice of law rules for that analogue.75 That might still be a complex process, since a state unfamiliar with the trust in its domestic law must satisfy itself that it is in the presence of an actual or putative trust, before going on to apply the Convention’s choice of law rules. 74 E.g. if the claimant were to sue on the basis of the defendant domicile rule under Art. 2, Brussels Convention Regulation. 75 A recent example of assimilation can be found in Switzerland, where the trust was regarded as a form of company and the choice of law rules applicable to companies applied: OD-Bank in Liquidation v. Bankrupt’s Estate WKR case No FB920075, 98 Blätter für Zürcherische Rechtssprechung (1999) No 52, 225ss; discussed in the context of Art. 11, below. However, it may be that Switzerland will go further than this and implement the Hague Trusts Convention. Recently, the Swiss government commissioned a report, which advocated the implementation of the Convention: see L Thévenoz, Trusts in Switzerland. Ratification of the Hague Convention on Trusts and Codification of Fiduciary Transfers (Zurich, Schulthess, 2001): see also D Hayton, “Some Major developments in Trust Law” [2001] Private Client Business 361, 369–70.

The Recognition of Trusts 101 Article 2 will help in this process; but it will be seen that this provides a statement of certain characteristics of a trust, rather than a definition. More significantly, however successfully private international law weaves its magic in laying down choice of law rules and however well non-trust states adapt to applying the Convention, the application by a non-trust state of these rules is a process which might be perceived by a settlor as unpredictable (especially in relation to the various limits to the role of the applicable law contained in the Convention), and might not inspire sufficient confidence in him to convince him to establish a trust straddling that civil law jurisdiction in the first place.

2. THE RECOGNITION OF TRUSTS

As to the recognition of trusts (which is the subject-matter of Article 11), the term “recognition” is itself a recipe for confusion. If choice of law principles are established, by which contracting states are bound, then it follows that the legal provisions of the law identified by the Convention as applicable must be given effect. This is, after, all, what the choice of law process is about. One can hardly imagine the courts of Civilopia,76 having identified a trust as being governed by English law77 under the Convention’s choice of law rules, then refusing to give the trust any effect, on the basis that the said institution is anathema to Civilopian domestic law and that its recognition is a quite separate matter. The word “recognition” is usually associated with foreign judgments which are sought to be recognised in another state. However, the word “recognition” cannot be associated with trust judgments in this context. The Hague Convention does not affect the rules on recognition of foreign judgments. So, the fact that a Hague Convention trust is the subject of a judgment overseas does not mean that the judgment will be recognised in another contracting state, unless the “general” rules on recognition of foreign judgments of that state are satisfied. Conversely, it is an integral feature of the recognition of foreign judgment rules in most systems that the law used to decide the dispute in the foreign court should not be investigated, so that a foreign judgment otherwise qualifying for recognition should not be denied effect because the judgment was not delivered in accordance with the Hague Convention provisions.78 Of course, a non-trust state which is bound to recognise an English trust judgment might encounter considerable practical difficulties; it might even refuse to recognise the judgment on public policy grounds. In that respect, the Hague Convention’s 76 The present author has borrowed the mythical state frequently employed by Hayton in his work. 77 Assuming that it has a substantial connection with England, so that the exception in Art. 13 cannot be invoked (on which, see below). 78 Godard v. Gray (1870) LR 6 QB 139; Art. 29 Brussels Convention (The same provision is contained in Art. 45(2), Brussels Regulation). On the Brussels Convention and Regulation, see, f/n 5 above.

102 Functions of the Convention recognition rules are to be understood as simply furnishing guidance as to what effects a Convention trust should be given in contracting states. Why, then, does the Convention use the ambiguous and confusing term “recognition”? The Conference decided that the word “recognition” was appropriate, since it was not limited in current usage to judicial decisions. But the ambiguity is most unfortunate. Apparently, simply speaking of the “effects” of a trust valid by the identified law would not do, since the poor judge in Civilopia “might stand perplexed before the pure and simple affirmation that a trust, from the mere fact that it is in conformity with the law designated by the Convention, should deploy effects in their countries”.79

It is difficult to accept, or even to understand such a point. A trust does not stand in isolation; just like a contract, its effect can only be determined by reference to the applicable law. The point is rather that certain effects of upholding a trust under the Convention could usefully be spelt out, for no reason other than assisting a judge to understand the consequences of applying that law. That is hardly an objection to the use of the word “effect” itself.

79

Von Overbeck, para. 31, p. 377.

ARTICLE 2—CHARACTERISTICS OF THE TRUST For the purposes of this Convention, the term “trust” refers to the legal relationship created—inter vivos or on death—by a person, the settlor, when assets have been placed under the control of a trustee for the benefit of a beneficiary or for a specified purpose. A trust has the following characteristics(a) the assets constitute a separate fund and are not a part of the trustee’s own estate;80 (b) title to the trust assets stands in the name of the trustee or in the name of another person on behalf of the trustee; (c) the trustee has the power and the duty, in respect of which he is accountable, to manage, employ or dispose of the assets in accordance with the terms of the trust and the special duties imposed upon his by law. The reservation by the settlor of certain rights and powers, and the fact that the trustee may himself have rights as a beneficiary, are not necessarily inconsistent with the existence of a trust.

1. INTRODUCTION

The application of choice of law rules relating to trusts may seem a daunting task for a state which does not know the institution in its domestic law. As Anton says: “Systems of private international law normally resolve legal problems by allocating them to familiar branches or categories of law, such as those of contract, delict, succession etc. But what is the judge to do when he is faced with a branch or category which is entirely unknown to his own system?”81

Even non-trust states which, prior to the Convention, have skilfully “translated” the trust into existing concepts of their domestic law82 now face the difficulty of having to recognise the trust qua trust. The Convention seeks to ameliorate the situation by elucidating the key characteristics of a trust. This is not intended as an exclusive statement of its 80 Compare the French version “Les biens du trust constituent une masse distincte et ne font pas partie du patrimoine du trustee.” As Lupoi notes, this “tends to ‘entify’ the trust”, by treating it as a distinct estate, whereas the English version suggests that the assets are within the trustee’s estate but do not belong to his “own” estate: M Lupoi, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 44. 81 A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 628. 82 See the discussion of “‘translating’ the trust into civil law systems” in relation to Art. 11, below.

104 Characteristics of the Trust characteristics. Rather, it is designed as a signpost to allow a non-trust state to identify a trust when confronted with it. Whether the characteristics are sufficiently precise to distinguish the trust from other institutions is a different matter, which we shall discuss below.83 Article 2 is not intended to amount to a definition of a trust.84 Partly, this is because of the rigidity which such a definition might introduce; but it is also explained by the impossibility, even between common lawyers, of formulating such a definition.85 That said, the detail contained within the Article comes close to a definition and it is reasonable to suppose that it will be so regarded in certain civil law systems. Reymond comments that “la convention contient ce que d’aucuns ont appelé une description du trust, mais qui me paraît personnellement être aussi proche que possible d’une définition”.86 Clearly, the paradigm English express trust falls within Article 2 and is the prime concern of the Convention, as the Preamble indicates. However, the characteristics described may be exhibited by many other institutions. This is deliberate,87 since although the trust of common law jurisdictions “is a unique legal institution”88 it is intended that the trusts of countries which do not separate legal and equitable ownership should also be covered.89 It is also intended to cover civil law institutions which exhibit certain core characteristics of the trust.90 Von Overbeck tells a brief tale of calls for analogous institutions of Egypt, Japan, Luxembourg90a and Poland to be included within the 83

See the discussion of “the characteristics considered” below. Nor is the word “trustee” defined: M Lupoi, Trusts: A Comparative Study 224. He points out that those states which have adopted what Lupoi calls the “international model” of the trust (on which see ch. 4, ibid.) e.g. Jersey, Malta, Guernsey and Mauritius have tended to define the trustee along lines closer to the traditional English model and have emphasised that “a trustee shall in relation to the trust property have all the same powers as a natural person acting as the beneficial owner of such property” (Art. 20(1),Trusts (Jersey) Law 1984). 85 Von Overbeck, paras. 36–38, pp. 378–9. And, of course, the trust’s existence is not confined to common law states, magnifying the difficulties. 86 C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé, 7, 13. 87 But see M Lupoi, “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees 15; see also M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222. 88 Preamble to the Convention. 89 D Hayton gives the examples of India, South Africa and Scotland in J Glasson (ed.) International Trust Laws, ch. C3, 6. If the distinction between legal and equitable ownership had been included in the Art. 2 description, problems could have arisen even in England. Technically, an English charitable purpose trust transfers ownership merely subject to obligations to apply the fund in a specified manner. It is, however, clear that such a trust falls within the Art. 2 description. In this respect, Nygh is incorrect to state that recognition of the trust pursuant to Art. 11 “imposes an obligation, particularly on States parties which do not know the institution of the trust such as Italy, to give effect to the separation between beneficial and legal ownership”: P Nygh, Conflict of Laws in Australia, 6th edn., 521. 90 This had not been the intention of the Special Commission drafting the Convention, but the Fifteenth Session took a different view. 90a The Luxembourg Chamber of Commerce, in giving its approval on 14 March 2001 to a Bill (introduced on 14 December 2000) to implement the Hague Convention, considered that its fiducie should be recognised overseas under the Convention (see further http://www.cc.lu/avis2001/ 2414wje.htm). 84

The General Characteristics 105 Convention.91 However, “it is not enough if the institution in question is ‘functionally analogous’ [to a common law trust] . . . it has to be ‘structurally’ similar. . .”92

2. THE GENERAL CHARACTERISTICS

The first paragraph of Article 2 introduces the essential protagonists in a trust relationship: settlor, trustee and beneficiary. However, it also makes clear that purpose trusts shall be included within the Convention’s scheme. Whilst some provision for purpose trusts is necessary, the unrestricted scope of this provision immediately takes the understanding of a trust beyond the scope of the traditional English model, where the restricted category of charitable purpose trusts exists, but where most non-charitable purpose trusts are void.93 The term “trust” refers to the “legal relationship” between the parties. The term “fiduciary relationship” is not used. “This formulation was inspired by the fiduciary duties that English and American law impose on a trustee. Certain experts from civil law countries however had objections to the use of the term ‘fiduciaire’, which had no meaning in French law. They preferred to speak of ‘legal relationships’ ”.94

It is somewhat unfortunate that no further reference to the fiduciary nature of trusteeship appears in Article 2, nor in Article 11, and that it is unclear to what extent the choice of law and recognition rules of the Convention extend to fiduciary duties of the trustee.95 Of course, if a trust does not necessarily import a fiduciary relationship, the Convention understanding of the word “trust” extends much further than the English trust. Article 2(1) “moves away from the English trust model. . . . In this way, the definition of the shapeless trust begins to emerge. . . .”96 Moreover, as Lupoi, points out, Article 2 does not explain the relationship between the trustee and other parties, or clarify the extent of his ability to pass title to them.97

91 Von Overbeck Report, para. 26, pp. 375–6. D Hayton, (1987) 36 ICLQ 260, 262, also mentions trust-like institutions of, inter alia, Israel, Sri Lanka, Argentina, Quebec and Louisiana as potentially falling within the Convention. Compare the views of M Lupoi and D Waters, discussed in “the characteristics considered”, below. See also M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, 231, who points out that many such institutions are not “analogous” to the common law trust but “are homologous to the English-model trust properly defined.” For other analogous concepts in civil law systems, see D Hayton (ed.) Principles of European Trust Law (The Hague, Kluwer, 1999) and D Hayton (ed.) Modern International Developments in Trust Law, especially chs. 1–3. 92 H Kötz, in D Hayton (ed.) Modern International Developments in Trust Law, ibid., ch. 3, 40. 93 See further M Lupoi, Trusts: A Comparative Study, 334. 94 Von Overbeck Report, para. 40, p. 379. See also M Lupoi, ibid., 336. 95 See the discussion of fiduciary duties in Art. 8(2)(g) below. 96 M Lupoi, Trusts: A Comparative Study, 335. 97 M Lupoi, ibid., 336.

106 Characteristics of the Trust Reference is made to the settlor as “a person.” The temptation to think that this might exclude corporate settlements is obvious. However, von Overbeck makes clear that this is not the case and that settlor, trustee and beneficiary may all be companies. In other words, “person” means “natural or legal person.” That being so, it is perhaps unfortunate that this was not made rather clearer in the text of the Convention itself.98 The assets must be placed under the “control” of the trustee. It is very curious that “ownership” need not be transferred and the wording used risks including within the Convention forms of agency and mandate. “The notion of ‘control’ is of very little legal importance. The explanation that it was not possible to be more precise for fear of alarming civil-law countries seems to me to be ridiculous (although it is likely true)”.99

In fact, Hayton suggests that it is unlikely that agency and mandate would fall within the Convention, since Article 2(2)(a) requires that the assets constitute a separate fund.100 But this is far from certain; the fund could be separate from the trustee’s own estate precisely because it is still owned by the settlor.101 Of course, nominee situations must be accommodated in any description of the trust, so that one cannot speak exclusively of “ownership” passing to the trustee: but this could be achieved by using the phrase “place under the ownership of a trustee (or of a nominee on behalf of the trust).” The declaration by a settlor of himself as trustee does not sit well with the idea of the settlor “placing” assets under the control of a trustee.102 Literally, declarations of trust would appear to be excluded from the Convention.103 However, the matter is not free from doubt. In principle, there seems to be no reason why a declaration of oneself as trustee should be excluded; provided that the trust is validly created,104 its applicable law and its entitlement to recognition should be identical. Put differently, the Convention does not determine how the trust is launched; and once it is validly launched, the Convention should have no interest in the manner in which it was launched. This author would strongly favour the inclusion of declarations of oneself as trustee within the Convention.105 98

Von Overbeck Report, para. 41, p. 379. M Lupoi, Trusts: A Comparative Study, 334. 100 D Hayton, “The Developing European Dimension of Trust Law (2000) 11 KCLJ 48, 52 101 Art. 2(2)(b) would support this view, since title to the assets may be “in the name of another person on behalf of the trustee.” 102 Who is assumed to be a different person from the settlor: M Lupoi, Trusts: A Comparative Study, 335. However, Lupoi points out that the reference to a settlor in Art. 2 was requested by civilian delegates and that the common lawyers did not consider it necessary, so it may be argued that little store should be attached to this fact. It should not necessarily be taken to mean that the settlor and the trustee cannot be the same person. 103 See the United Kingdom delegate’s remarks: Actes et Documents 237, and M Lupoi ibid., 335. See also M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, 228. 104 Which is a matter outside the scope of the Convention by Art. 4. 105 M Lupoi also argues that “On balance, the arguments in favour of extending the Convention to include self-declared trusts seem to me to be convincing”: Trusts: A Comparative Study, 335. 99

The Specific Characteristics 107

3. THE SPECIFIC CHARACTERISTICS

Turning to the second paragraph of Article 2, sub-paragraph (a) emphasises that trust assets constitute a separate fund which is not part of the trustee’s personal estate. Exactly the same point is repeated, presumably for the avoidance of doubt, in Article 11. However, since the recognition provisions only extend to trusts ex hypothesi within the Convention, the repetition is strictly unnecessary. Whilst it is stated that the trust assets must constitute a separate fund, nowhere is it required that legal and equitable ownership should exist separately, or that the concept of equitable ownership be known at all. “It will be seen that it is the trustee’s ownership of a segregated trust fund . . . that is at the core of the trust . . . There is no necessity for beneficiaries to have in rem proprietary rights, characterised as ‘equitable’ rights in Anglo-American jurisdictions where the trustees have legal rights”.106

The trust of e.g. Scotland would hence clearly be included within the Convention.107 Remedies for the beneficiary which exist for breach of trust need not include the assertion of proprietary rights to trust assets in the trustee’s hands.108 Indeed, the concept of a fund separate from one’s own wealth is a feature of many civil law systems which, at least outside the context of Article 2, Hague Convention, would not be considered trust states. Hayton gives a number of examples, such as the lawyer’s client account, treated as a distinct asset not susceptible to claims by personal creditors of the lawyer in Denmark, Germany and the Netherlands. In Germany, the assets purchased by an investment company with a client’s funds are equally immune from personal creditors.109 Kötz gives further examples, including the fideicomiso of Latin American systems and the Quebec fiducie110 as structures which would fall

106 D Hayton in D Hayton (ed.) Modern International Developments in Trust Law, ch. 2, 23. See also D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 943. M Lupoi argues (ibid., 328) that the term “Anglo-American trust” should be rejected, as no such unified law of trusts exists. Nonetheless, the term is used in the Conference proceedings. 107 For a brief discussion of the impact of the Hague Trusts Convention in Scotland, see S Rae, “Trusts: the Scottish Experience” in N Vogt (ed.) Disputes Involving Trusts, 303, 336–8. See also K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 5–6. See further the discussion of “the characteristics considered” below. 108 But see Art. 11(3)(d), below. 109 D Hayton in D Hayton (ed.) Modern International Developments in Trust Law, ch. 2, 23. 110 For discussion of the Quebec fiducie, see D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation— Kollision und Vereinheitlichung—Conflicts and Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème Anniversaire, 591 (esp. 594–605) (especially on the question of who owns the assets in question); see also the Canadian Supreme Court decision in Royal Trust Company v. Tucker [1982] 1 SCR 250. See further K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 6.

108 Characteristics of the Trust within the Convention.111 German law knows the Treuhandverwaltung and Japan, by virtue of the Japanese Trust Act 1922, a not dissimilar concept of a fund not forming part of the trustee’s private estate. Sub-paragraph (b) of Article 2(2) is designed to make clear that title to the trust assets need not necessarily be held by the trustee, providing that it is held by a person nominated by the trustee to hold on his behalf. Hayton points out the danger that the sub-paragraph is broad enough to cover some forms of agency or mandate e.g. the transfer by an owner to an agent of control (but not ownership) of assets for the benefit of the owner and others.112 This construction appears unlikely, given the existence of sub-paragraph (a)113 and the requirement that there should be a separate fund. Nevertheless, “to avoid problems of construction”114 Article 1 of the Principles of European Trust Laws describes the trust in different terms.115 (1) In a trust, a person called the “trustee” owns assets segregated from his private patrimony and must deal with those assets (the “trust fund”) for the benefit of another person called the beneficiary or for the furtherance of a purpose. (2) There can be more than one trustee and more than one beneficiary;116 a trustee may himself be one of the beneficiaries. (3) The separate existence of the trust fund entails its immunity from claims by the trustee’s spouse, heirs and personal creditors. (4) In respect of the separate trust fund a beneficiary has personal rights and may also have proprietary rights against the trustee and against third parties to whom any part of the fund has been wrongfully transferred. 111 The introduction into French law of the concept of the fiducie is under consideration: see F Mascré in J Glasson (ed.) International Trust Laws, ch. A10; J Dubard, “ ‘La Fiducie’: Trusts in France” (1992) 3 International Company and Commercial Law Review 324; H Dyson, “The Proposed New Law of Trusts in France” [1992] Conv 407; M Grimaldi and F Barrière, “La Fiducie en Droit Français” in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs. Fiducie in a Business Context, ch. 3; I Koele, “France: a Law Proposal Introducing ‘la Fiducie’” in F Sonneveldt and H van Mens (eds.) the Trust—Bridge or Abyss between Common and Civil Law Jurisdictions, ch. 6. However, the institution will be substantially different from the trust: “. . . nous ne tomberons pas dans l’erreur d’assimiler trust et fiducie. Lex deux institutions demeurent profondément différentes dans leur structure, dans leur fonctionnement et, plus encore peut-être, dans l’esprit qui les anime”: C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé, 7, 24; see C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, at http:// www.isdc.ch/e/Reymond97.asp. 112 D Hayton, in J Glasson (ed.) International Trust Laws, ch. C3, p. 6; see also D Hayton “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159, 167. 113 And of Art. 11. 114 D Hayton in D Hayton (ed.) Modern International Developments in Trust Law, ch. 2, 22; see also D Hayton, “Whither Trusts in the Twenty-First Century? Part 1” [2000] Private Client Business 94. 115 See D Hayton (ed.) Principles of European Trust Laws; a succinct account of the Principles by the same author is contained in D Hayton (ed.) Modern International Developments in Trust Law, ch. 2 and see (2000) 8 J Int Corp P 159, 168–172; see also D Hayton, [2000] Private Client Business 94. 116 Art. 2, Hague Convention refers to the singular: “a trustee” and “a beneficiary” (though it is surely inconceivable that the existence of more than one trustee or beneficiary would lead any court to conclude that it was not faced with a “trust” within the meaning of Art. 2, Hague Convention).

The Specific Characteristics 109 Such a description is intended, inter alia, to emphasise that (legal) ownership of the assets must vest in the trustee. Does this deal adequately with the bare trust, where the property is held by a nominee? Hayton argues that the trustee still has “ownership” of the claim to assets in the hands of the nominee and that such claim is itself a trust asset, so that it is unnecessary to speak of ownership by a person other than the trustee.117 Technically, this might be true, but it may be unlikely to be understood too readily by the courts of certain legal systems unfamiliar with the trust. It is true that the task of the Principles of European Trust Law is to develop trust principles. The Hague Convention is broader in its scope and designed to scoop up other so-called “analogous” civil law constructs.118 It may accordingly be that one should expect a different description of the trust in the Convention to that in the Principles. However, for both the Convention and the Principles, it may have been better to have made express provision as to the need for ownership to be in the trustee or in the nominee on behalf of the trust. This is clearer: in the case of the Hague Convention, it minimises the dangers of the overextension of the “trust” into straightforward cases of agency; in the case of the Principles, it prevents a “trust” not being regarded as such merely because there is a nominee and “ownership” is considered not to vest in the trustee. Sub-paragraph (c) of Article 2(2), Hague Convention subjects the trustee to obligations specified in the trust instrument, but also to duties imposed on him by law. His accountability in this respect is intended to emphasise that the beneficiaries may have recourse to the courts to ensure that these powers and duties are properly exercised. The phrasing is weak, however, since it is not expressly stated to whom the trustee is accountable. It does not appear that the trust is confined to cases where the trustee’s accountability is solely to the beneficiaries, as in the typical common law trust.119 On one view, the trustee could even be made accountable to the settlor.120 The non-charitable purpose trust of jurisdictions 117

D Hayton in D Hayton (ed.) Modern International Developments in Trust Law, ch. 2, 21–2. But see M Lupoi, Trusts: A Comparative Study, 333, who is fiercely critical of the Conference: “Although there are occasional mentions of functional and structural analogies, they represent nothing more than a vague exercise. The English model was not compared with a series of civil-law structures which it was claimed could be compared with it, nor was a category developed which was feasible from a comparative viewpoint and which included both English-model trusts and other institutions. On the contrary, the huge problems with a task of this nature, and the opportunity to avoid even having to undertake it, were referred to explicitly.” 119 D Waters, “The Concept Called ‘the Trust’” (1999) 53 Bulletin for International Fiscal Documentation 118, 124 regards an essential element of the trust as that “From the moment of the creation of the trust there must be an ability of the beneficiary to secure an accounting or, to put it another way, a power in the beneficiary to enforce the discharge of his duties.” 120 Although as M Lupoi rightly asks “How could an ‘institution’ be ‘analogous’ to the AngloSaxon trust if the trustee were to be accountable to the ‘settlor’?” “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees 15, 16. But D Waters, ibid., 120–1 comments that “. . . the Convention appears as reluctant as the present writer to say that a reserved power of control or disposition of the trust property, automatically makes the trust a ‘sham.’” He points out that some offshore centres give the settlor extensive powers to intervene in the running of the trust and that “This development not only embarrasses the principal mainland jurisdictions . . . but it has encouraged civilians in their conceptual conclusion that the ability or disability of the trustor to intervene is irrelevant” (ibid., 125). See 118

110 Characteristics of the Trust which allow an “enforcer” to uphold the terms of the trust would also meet the requirement of accountability.121 Moreover, to a common lawyer, the word “accountable” emphasises the personal obligations of a trustee in respect of the trust assets. This once again underlines the fact that a Hague Convention trust can exist without any proprietary claims or remedies being available to the beneficiaries against the trustee.122 The final paragraph of Article 2 indicates that a trustee may himself be a beneficiary, although not, of course, the sole beneficiary. Equally, the settlor may be a beneficiary of the trust. However, the “reservation by the settlor of certain rights and powers”123 is also permitted. This is of limited help, as it is not stated which rights or powers may be reserved, so that “we have no way of evaluating whether in any given circumstances where ‘prérogatives’ are retained by the settlor a legal relationship is a trust or not.”124 Moreover, we saw above that a trust may exist within the meaning of Article 2 where assets are placed only under the “control” of the trustee and that title to the assets may stand “in the name of another person on behalf of the trustee.” A “settlor” may accordingly retain ownership of the fund, retain other rights and powers and yet there may still be a trust for the purposes of Article 2. This may have the unfortunate effect of “steer[ing] a course towards the civilian mandate and deposit and the very many types of contracts which are based on them”.125 It is true that, when deciding whether the reservation of certain rights and powers by the settlor is inconsistent with the trust, some common law states might adopt a construction of the term “trust” which excludes what would be regarded in the common law as sham “trusts”.126 Equally, however, the lack of clarification on the matter is likely to result in contracting states adopting different views on what is a genuine “trust”, with the result that the scope of the Convention’s application will vary from one contracting state to another. further s. 86(1), (British Virgin Islands’) Trustee (Amendment) Act 1993; s. 13(C), (Cook Islands’) International Trusts Act 1984 (inserted by s. 6, s. 11, International Trusts Amendment Act 1989, as amended by s. 12, International Trusts Amendment Act 1995–6). 121 See D Hayton, Underhill and Hayton, The Law Relating to Trusts and Trustees, 15th edn., ch. 23, 932. 122 Although the trust property itself may, in principle, be recovered from the trustee (where he has mixed it with his own assets in breach of trust) and from a third party: Art. 11(3)(d). 123 Contrast the reference in French to “certaines prérogatives” (Actes et Documents, 167); M Lupoi, Trusts: A Comparative Study, 338. 124 Ibid., 338. 125 M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, 230. 126 See Snook v. London & West Riding Investments Ltd [1967] 1 All ER 518; the Jersey decision in Clothilde Rahman v. Chase Bank (CI) Trust Co Ltd and Others [1991] JLR 103; and the litigation in the Bahamas, (and subsequent litigation in England and Jersey) in Private Trust Corp. v. Grupo Torras [1997/98] 1 OFLR 443. See also W Goodman, “Vulnerability of Trusts to Attack— Retention of Powers and Interests by Settlor” in J Glasson (ed.), International Trust Laws, ch. B8; A Saker and J McKeown, “Reservation of Powers to Settlors” [2001] Private Client Business 154. On the position of sham trusts in a non-contracting state (the Cayman Islands), see N Lawton, “The Cayman Islands: Sham Trusts” (1999) 17 Amicus Curiae 31.

The “Shapeless” Trust 111

4. THE CHARACTERISTICS CONSIDERED : THE

“ SHAPELESS ”

TRUST

It is undoubtedly true that the characteristics of a trust described in Article 2 lack a certain ambition. They are rather general. Certainly, they do not incorporate anything like all the features of the English trust.127 In particular, no stress is placed on the transfer of ownership to the trustee;128 nor on the proprietary entitlement and other rights of the beneficiary; and reference to the fiduciary nature of the trust, so pivotal to its essence in the common law world, is excluded.129 Of course, the advantage of the looseness of the characteristics is their ability to include a number of civil law constructs. That may, however, be something of a mixed blessing. Lupoi argues that the Article 2 has “turned squares into circles and whites into blacks”.130 He suggests that the looseness of Article 2 was not warranted and points out that many states outside the AngloAmerican legal systems which were assumed to be “non-trust” states in fact have highly developed trust notions in their domestic law.131 These constructs are not merely analogous to the common law trust; they share many of its core characteristics and are described by Lupoi as homologous thereto. They are laws which define the trust in terms of a transfer of title to the trustee on the terms of the trust132 and protect the trust assets from the claims of creditors to 127 As M Lupoi shows (Trusts: A Comparative Study, 331), the initial plans were for the Convention to cover the “English model” trust (Actes et Documents, 141–4), but the Special Commission’s draft came to include the so-called “analogous institutions.” Many states called for the inclusion of civilian analogues. It is somewhat curious, then, that M Koppenol-Laforce, Het Haagse Trustverdrag, 265 states that “Art. 2 gives a definition of a trust which is completely based on the Anglo-American trust.” Art. 2 is neither a definition nor can it be seen to be wholly based on such a trust (leaving aside for present purposes Lupoi’s contention that the “Anglo-American” trust does not exist). 128 Only “control” is mentioned. 129 “How ‘analogous’ could an ‘institution’ be should it lack a fiduciary foundation?”: M Lupoi, “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees, 15, 16. 130 M Lupoi, ibid., 15; M Lupoi, “Effects of the Hague Convention in a Civil Law Country”, (1998) 4 Trusts and Trustees 15; see also M Lupoi, Trusts: A Comparative Study. 131 He gives the examples of Law no 62, 21 April 1922 of Japan; Art. 897 of the Personen- und Gesellschaftsrecht of Liechtenstein and Art. 314 of the Ley general de Bancos of Peru. Elsewhere, he lists also Argentina, Colombia, Ecuador, Israel, Italy, Louisiana, Mauritius, Panama, Russia, Scotland, the Seychelles, South Africa and Venezuela: see M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, f/n 32. Lupoi points out that he does not include Ethiopia or Quebec in the list, but that the latter is regarded as homologous to the common law trust by Waters. Lupoi would, however, include the “analogous institutions” of Ethiopia, Quebec, Scotland, South Africa, the Netherlands, Ghana and Nigeria within the understanding of a “shapeless” trust for the purposes of Art. 2 of the Convention: see Trusts: A Comparative Study, 320, f/n 66 (and ch. 5). See also M Lupoi, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, esp. 38–40. See further, K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, pp. 5–10. 132 Lupoi informs us that Peru is the sole exception amongst the states which he discusses. See also D Waters, “The Institution of the Trust in Civil and Common Law” (1995) 252 Recueil des Cours 117, 427–35; D Waters, “The Concept Called ‘the Trust’” (1999) 53 Bulletin for International Fiscal Documentation 118.

112 Characteristics of the Trust the trustee’s personal assets. Many contain references to the fiduciary role of the trustee. Lupoi argues that the sophistication of the trusts of such civil law systems, and the extent of their similarities to the common law trust, were ignored by the drafters of the Convention. Indeed, he points out that the Convention began life largely as a means of facilitating the recognition of common law trusts overseas, a fact borne out by the Preamble’s reference to the trust “as developed in courts of equity in common law jurisdictions” as being “a unique legal institution.” He criticises the Preparatory Report133 presented to the Hague members at the Fifteenth Session as failing to study civil law trusts in their context, grouping them as analogous institutions to the trust, without making any sustained attempt to compare or contrast them. He argues that, had the Report done so, it would have appreciated that a diluted statement of characteristics such as that contained in Article 2 was not necessary.134 Instead, faced with an uncertain, but potentially overwhelming number of “analogous institutions”, Article 2 was diluted. Lupoi sees the characteristics of the trust identified in Article 2 as a sop offered by common law states to civil law states—it attempts to be uncontroversial and in intelligible language135 and hence an easier pill for them to swallow. It departs very substantially from the Preamble’s reference to the “unique” institution of the common law trust as developed in the courts of equity.136 “The drafters of the Hague Convention set happily to work in order to shift the object of the Convention from the Anglo-Saxon trust . . . into the undefined realm of analogy”.137

He points to the relative lack of ratification of the Convention by civilian states138 as evidence that this plan failed. And, in seeking to be all things to all people, the common law trust lost its soul.139 No longer is there even a requirement that the 133 Report on trusts and analogous institutions, Preliminary Document No 1, May 1982 (the Dyer and van Loon Report), contained in Acts and Documents of the 15th Session of Hague Conference on Trusts, vol II, 10. Lupoi argues that the analogous institutions were bundled together without individual analysis of their similarities and differences from the common law trust and from each other. 134 A more benevolent view is taken by A Oakley, Parker and Mellows, The Modern Law of Trusts 7th edn., 12–14 and by G Gretton, “Trusts without Equity” (2000) 49 ICLQ 599, 614–5. 135 One example which M Lupoi gives is the proposal of the US delegate (Actes et Documents, 238) that the idea of a “transfer” to the trustee would be more readily understood if the Convention spoke of assets which were “placed under the control” of him, a view which has stripped the trust of one of its core characteristics: see Trusts: A Comparative Study, 327. 136 See also M Lupoi, ibid., 337. However, it is not wholly accurate to say that “the Hague Convention . . . caring nothing for equity leaves it out of its ‘description of a trust’.”: J Goldsworth, “A Better Definition of Trusts” (2000) 6 Trusts and Trustees 34, 35 (Emphasis added). As Lupoi demonstrates, the reasons for the extension of the Convention beyond the common law trust described in the Preamble are more complex and subtle than this. 137 M Lupoi, (1995) 1(3) Trusts and Trustees, 15, 16. 138 Save Italy and the Netherlands. 139 “Attempts to be intelligible to those unfamiliar with the myriad practical application of the [trust] institution too easily result in an adoption of obvious and generalized positions: . . . This was the path chosen by the authors of the Hague Convention. . .” M Lupoi, Trusts: A Comparative Study, 5. Only the Preamble now contains reference to the common law trust.

The “Shapeless” Trust 113 legal title to the property passes to the trustee;140 and a trust shorn of fiduciary duties is barely analogous to a common law trust at all. The result is what Lupoi calls a “shapeless trust”,141 devoid of sufficient backbone to be recognisable specifically as the institution of any state, whilst allowing any number of institutions to fit within its parameters.142 Indeed, so broad is its scope that he argues that “non-trust” states within the meaning of the Convention are extremely rare, or even non-existent143 and that the provisions of Articles 5, 6 and 13, which refer to the law of such states are rendered either meaningless or pointless.144 Moreover, the looseness of Article 2 means that it “would be well-nigh impossible to describe the many civilian (and non-civilian) contractual shapes which fit snugly within the Convention definition”.145 Furthermore, it should be noted that some offshore states have used Article 2 as the basis for the definition of a trust in their domestic laws and, in doing so, “have either revolutionized their trust law or carried out a purely cosmetic operation”.146 140 M Lupoi, ibid., explains that this was so as not to “scare” civil law countries, an explanation proffered by C Jauffret-Spinosi in “La Convention de la Haye Relative à la Loi Applicable au Trust et à sa Reconnaissance”, (1987) 114 Journal de Droit International, 23, 27. 141 “The trust which I call ‘shapeless’ . . . represents a conscious choice by the Conference”: M Lupoi, ibid., 332; see also A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 2 Trusts and Trustees 6. 142 M Lupoi maintains (in Trusts: A Comparative Study, 341) that commentators have erroneously interpreted his views on the shapeless trust as a criticism. It is difficult to read his article “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15 and not to reach this conclusion. However, he asserts (in Trusts: A Comparative Study, 340–1) that the real failing which he criticises is the lack of a sophisticated comparative approach to the trust demonstrated in the Preliminary Report or during the Conference, which relied on “second- and third-hand information. No serious research was conducted into non-common law systems. The failure to conduct a comparative analysis has led to the definition of a legal structure whose relevance was not understood, and which today has problems of acceptance. This is not, however, a criticism; it is an interpretation of the rules of the Convention.” In “The Shapeless Trust” (ibid., 16) he comments that “The idea was conveyed that a crowd of ‘analogous institutions’ existed but no attempt was made to ascertain what those institutions had in common nor if degrees of analogy could usefully be drawn.” 143 He describes Italy as a trust state under the terms of Art. 2 and argues that the shapeless trust can be found “from Ecuador to Russia to the Philippines to the waqf of Muslim law.” But C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, at http:// www.isdc.ch/e/Reymond97.asp, who argues that the Swiss fiducie would not fall within the Convention’s ambit, because not creating a separate fund. 144 This is not quite true, since Art. 13 might still be used to refuse recognition to categories of “trusts” not known in a legal system, (although not in the United Kingdom, which has not enacted Art. 13): 145 M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, 231. A somewhat more restricted view of Art. 2 appears to be taken by D Waters in (1995) 252 Recueil des Cours, 117, 435–443. See also D Hayton “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159, 167. 146 M Lupoi, Trusts; A Comparative Study, 238. He notes that Bermuda has adopted Art. 2 verbatim. Those offshore states which have adopted their own definition and not followed Art. 2, Hague Convention have generally been more successful. Lupoi points out the advantages of the Guernsey approach in s. 1, Trusts (Guernsey) Law 1989 (also used in e.g. Jersey, Malta and Belize) which states that “A trust exists where a person (a ‘trustee’) holds or has vested in him, or is deemed to hold or have vested in him, property which does not form, or which has ceased to form, part of his own estate: (a) for the benefit of any person (‘a beneficiary’) whether or not yet ascertainable or in existence: or (b) for any valid charitable or non-charitable purpose which is not for the benefit of

114 Characteristics of the Trust Lupoi submits that: “an appropriate definition of the trust in comparative law terms would include the following elements: (1) the transfer of property to the trustee, or a unilateral declaration of trust (2) the lack of commingling between said property and other elements of the trustee’s estate (segregation); (3) the loss of any power of the settlor over said property; (4) the existence of beneficiaries or a purpose, and the resulting functionalism of the exercise of the right transferred to the trustee; and (5) the imposition of a fiduciary component upon the exercise of the trustee’s rights, with principal reference to conflicts of interest”.147

These criteria represent the result of a highly detailed and sophisticated comparative study by Lupoi and it is inappropriate to seek to evaluate them here in any detail. However, it may be noted that Lupoi’s criteria undoubtedly have certain clear advantages over the Article 2 approach. They emphasise: the effectiveness of declarations of trust; the loss of power by the settlor (although, of course, the settlor may be a beneficiary of a trust, provided that it is not a sham); and the fiduciary component so noticeably missing from Article 2. He argues that Waters’ attempted description of the trust,148 which also mentions the importance of tracing by beneficiaries, takes an overly “proprietary” view of the position of beneficiaries149 and that civil law systems may provide other forms of protection similar to that which may be permitted pursuant to the tracing process.150 Waters would also include the need for transfer of full ownership to the trustee, a view rejected by Lupoi on the basis that the trustee may hold a the trustee.” Lupoi points out (at 238–9) the strengths of this definition: it does not make reference to the settlor (an omission which is required if constructive and resulting trusts are to be included in the definition); it speaks of the “vesting in the trustee” (since a transfer of assets will not occur in many cases); it rejects use of the unsatisfactory notion of passing “control” to the trustee; and it states that the assets are not part of the trustee’s own estate (as does the English version of Art. 2(2)(a), Hague Convention), thereby indicating that the assets are part of his estate but segregated from his personal wealth. 147 M Lupoi, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 36–7. Contrast J Goldsworth, “A Better Definition of Trusts” (2000) 6 Trusts and Trustees 34, 35 who would favour a definition placing greater emphasis on the beneficiaries’ rights: “a transfer of assets to a beneficiary or beneficiaries where certain rights of ownership, specified by the settlor on making the transfer, in connection with the enjoyment of the income and capital of the assets transferred are under the control of another as a settlor determines for such time in such a way that the transfer of the ownership of the assets by the settlor is made without, at that time, transferring the whole of the incident of ownership to the beneficiaries, although the beneficiaries have substantial control subject to the settlors [sic] express intentions over the fulfilment by the trustees of the settlors [sic] wishes.” This definition should not be supported. Amongst its flaws, it does not begin to explain the ownership of the trustee (the word “control” is used), the nature of his powers and duties, the fiduciary nature of his office, or his powers and duties in respect of third parties. See also the definition given in K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 3. 148 For which see D Waters, “The Institution of the Trust in Civil and Common Law” (1995) 252 Recueil des Cours, 117, 229. See also D Waters, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118. 149 See further M Lupoi, “The Civil Law Trust”, 37, 42–3. 150 Although the tracing process does not confer any proprietary entitlement per se (and remedies of a claimant who can trace wealth into the hands of the defendant may be personal or proprietary).

The “Shapeless” Trust 115 right short of ownership, such as possession, on trust.151 Whilst Lupoi’s point is a valid one, he speaks only of “the transfer of property to the trustee”, not even mentioning that the trustee takes full title to whatever right is transferred to him. Whilst this may be implicit from the loss of any power by the settlor over the property, this writer would personally prefer the Waters’ approach on this point.152 Of course, the “shapeless trust” of Article 2 may provide a degree of reciprocal benefit to civil law states in allowing their institutions to be recognised in common law trust states. And if the trust within the description of Article 2 is indeed as prevalent as Lupoi suggests, the scope and importance of the Convention is greater than it might appear. But the “shapeless trust” is also a recipe for uncertainty, inconsistent application of the Convention in different contracting states and for fear and mistrust of the unknown. This concern should also be present in common law states. Article 2 may let so much through the gateway of the Convention that it risks fuelling anxiety, given the extreme difficulty in predicting what “analogous institutions” of Civilopia lurk beyond, which a contracting state is bound to recognise. All that said, the radical implications of Lupoi’s cogent argumentation may be thought unlikely to be followed to their logical extreme by the courts of contracting states.153 The intention principally to regulate the common law trust is clear from the Preamble to the Convention, even if it was subsequently decided to make provision for the “analogous” institutions existing elsewhere. As Hayton notes, the possible inclusion of e.g. agency within the Convention would be difficult to reconcile with the existence of a separate Hague Convention on that subject and would have led to a very different direction to the discussions at the drafting stages. Purposive interpretation of the Convention by common law states would suggest that the most irreducible core of the trust would be that ownership of the assets154 passes to the trustee,155 but that the assets do not form part of his personal patrimony.156 But this view cannot be stated with confidence; 151 D Waters, (1995) 252 Recueil des Cours, 117, 221; M Lupoi, “The Civil Law Trust”, 37. However, there is much shared ground between Lupoi and Waters: “We agree that there is a fiduciary relationship between the trustee and the beneficiary (or trust purpose), that the trustee has title to the trust property, and that the trust property is not subject to claims other than by creditors of the trustee acting (or omitting to act) in his capacity as a trustee . . . [We] also envisage that the trust instrument, if any, will essentially leave control of the trust property, its required management and eventual disposition, to the trustee. It also seems likely that there is agreement between us that, if definition is insisted upon, the trust constitutes the dedication of segregated property to the attainment of a stated object”: D Waters, (1999) 53 Bulletin for International Fiscal Documentation 118, 121. 152 The other main difference in the approach of Waters is that he emphasises the independence of the trustee from the settlor and beneficiary (save (arguably) where the trust instrument reserves a power to the settlor; or the beneficiary seeks to enforce the terms of the trust instrument): (1995) 252 Recueil des Cours, 117, 226. But as Lupoi notes (ibid., 37–8) this may tend to understate the importance of the trust for purposes and direct the civil lawyer to thoughts of the contract for third parties. See also J Langbein, “The Contractarian Basis of the Law of Trusts” (1995) 105 Yale LJ 625. 153 Although see the “application in Italian courts” of Art. 13, below. 154 Or at least, that legal title to whatever interest is subjected to a trust is vested in the trustee. 155 Or to a nominee on behalf of the trust. 156 D Hayton, “The Developing European Dimension of Trust Law” (2000) 11 KCLJ 48, 53.

116 Characteristics of the Trust still less can it confidently be asserted that the scope of Article 2 will be uniformly interpreted from one contracting state to another.

5. CHARACTERISATION

One of the most feared spectres in the conflict of laws is that of characterisation, i.e. the exercise of attributing a cause of action, issue or legal rule to a particular category for private international law purposes, in order to determine which choice of law rules to apply thereto.157 The classification of the matter may be especially complex where the cause of action, rule or institution does not exist in the domestic law of the forum, or where it exists but is given a different classification to that attributed to it by a foreign law. This problem is, of course, particularly pertinent in the case of the trust (whose nature may differ from state to state) and the so-called civil law “analogues.” However, more recently, many international instruments, which have brought a degree of harmonisation in private international law have, if not simplified the characterisation process, then at least moved the goal posts. It is very common to speak, in the context of European instruments, of a “European autonomous” meaning to particular terms. In the jurisdictional context, for example, the scope of the Brussels Convention, in its application to civil and commercial matters,158 must be uniform and not vary from state to state. Likewise, terms such as “matters relating to contract” are also accorded an autonomous meaning under the Brussels Convention.159 Lupoi argues that when deciding whether to recognise a trust governed by foreign law “a state must first evaluate whether the relationship is a trust. This cannot be achieved by applying the foreign law to the relationship . . . because for the purposes of recognition it is the Convention which determines whether or not one is in the presence of a trust”.160 This author would disagree. Article 2 explains certain key characteristics of a trust; it is not a definition. It is not able to provide an exhaustive, autonomous definition of the trust and a state called upon to apply the Convention should determine whether it is in the presence of a trust.161 Article 2 will be their guide, but it will not be their bible. It might be thought that the Hague Trusts Convention eliminates characterisation problems, precisely because it does not attempt to define the trust and 157

For more detail, see Dicey and Morris, ch. 2; Cheshire and North’s Private International Law,

ch. 3. 158

Art. 1, Brussels Convention; case 814/79 Netherlands State v. Rüffer [1980] ECR 3807. Art. 5(1), Brussels Convention; case 34/82 Peters Bauunternehmung GmbH v. Zuid Nederlandse Aannemers Vereiniging [1983] ECR 987. However, it is not clear whether the word “trust” bears a European autonomous meaning under the Brussels Convention: see S Berti, “Trusts and the Lugano Convention—does it matter? and what about the Hague Trust Convention” in N Vogt (ed.) Disputes Involving Trusts, 9, 10–11. 160 M Lupoi, Trusts: A Comparative Study, 338. 161 Or at least, should satisfy itself that an arrangement which shows the characteristics described in Art. 2 is a trust. 159

Characterisation 117 merely describes its key characteristics. However, conceptually, the fact that no definition of the term “trust” is provided cannot mean that national courts can avoid a process of ascertaining to which legal relationships it should apply the Hague Convention. A legal relationship may have the characteristics specified in Article 2; but it does not appear to follow that, if a relationship does have these characteristics, it is inevitably to be classified as a trust. Nor is it expressly stated in the Convention that, for relationships having the characteristics detailed in Article 2, the Convention “shall” apply.162 That said, whilst a wholly uniform practice of interpretation is almost impossible when dealing with a subject on which there is limited shared international understanding, what does not seem at all desirable is that the same institution of a given legal system may or may not be regulated by the Convention, depending on the courts of which contracting state happen to be considering, and characterising it. Insofar as possible, an approach to characterisation that tends towards uniformity should be adopted. How then should an English court conclude the process of characterisation? A number of possibilities come to mind: (a) a court should determine whether, according to the domestic law of the forum, the relationship with which it is dealing is a trust This is an option to which no credibility can be attached. Since the point of the Convention is to incorporate trusts from a range of states world-wide, and also to extend to analogous institutions of the civil law, an English court clearly cannot refuse to apply the Convention to a legal relationship satisfying the characteristics of Article 2, simply on the basis that it does not meet the English (domestic) requirements of a trust, any more than it could e.g. refuse to accept a relationship as a contract solely on the basis that it does not import good consideration.163 (b) a court should determine, according to the private international law rules of the forum, whether the relationship with which it is dealing is a trust It might be possible for an English court, using its own private international law understanding of a trust, to apply the Convention rules to a legal relationship falling within such category, assuming that it also shows the characteristics described in Article 2. This would result in the application of the Convention to trusts of the Anglo-American legal systems, and would probably be sufficient to apply also to the trusts of certain other states, such as India, Scotland or South Africa. But even if it were to go beyond this, and to extend the private international law category to any state which has (what that state calls) a “trust” law, there is a grave danger of an English court excluding from the category a number of civil law 162 Contrast Art. 1, Rome Convention. But see s. 1(1), Recognition of Trusts Act 1987: “The provisions of the Convention set out in the Schedule shall have the force of law in the United Kingdom” (emphasis added). 163 Re Bonacina [1912] 2 Ch 394.

118 Characteristics of the Trust homologues and analogues exhibiting the characteristics listed in Article 2 and, as such, undermining the Convention’s intended scope.164 (c) a court should determine, according to the applicable law, whether the relationship is considered to be a trust This characterisation by the governing law (lex causae) encounters the familiar objection that, until the process of characterisation has been completed, the applicable law cannot be ascertained. That necessitates, at the very least, a preliminary classification by the law of the forum (lex fori) of the legal relationship as one to which the Hague Convention might apply and a consideration of its putatively applicable law. On the other hand, it is equally true that, until one has ascertained the law applicable to a legal relationship, one cannot determine what the full implications of the relationship are and, in turn, conclusively determine whether it shows the characteristics specified in Article 2. Inevitably, some characterisation by the applicable law cannot be avoided. If it is alleged that a civilian construct is sufficiently analogous to a trust that, for private international purposes, it should be classified as a trust, then it must be examined in its context, and by reference to the legal system which regulates it. However, we are concerned with a slightly different suggestion here: namely that the applicable law, once identified, should itself conclusively determine whether it is a “trust” for Hague Convention purposes. Such a suggestion must be rejected. In relation to civil law constructs which are not labelled a “trust” for domestic purposes in the state in question, it may be impossible to determine whether that state regards the institution as a “trust” for Convention purposes, there being no conclusive authority on the matter. Furthermore, to ask how, as a matter of private international law, the applicable law would characterise the legal relationship (as opposed to an English court simply analysing that construct in its context in order to determine for itself whether it thinks that construct is a Hague Convention “trust”) might be thought to be conceding too much control of the characterisation process to the governing law. (d) a court should determine, according to the applicable law, the characteristics of the legal relationship and consider whether it shares sufficient characteristics with the common law trust structure This is a more sensible middle ground and one which, in practice, an English court is subconsciously likely to pursue. It involves three distinct stages. First, a preliminary decision needs to be reached whether the legal construct of Civilopia shares enough characteristics with the common law trust for its applicable law to be identified in accordance with the Convention’s 164 Particularly if other states treat the “shapeless” trust of the Convention as including any number of analogous and homologous institutions to the common law trust. See further the discussion of “the characteristics considered” above; see also M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15.

Characterisation 119 provisions. Of course, the ascertaining of that applicable law may involve looking at the terms of the legal relationship, which may themselves be in issue. Rather than simply assuming the validity of all such terms at this stage, an English court should ask itself, as a matter of preliminary assessment, whether it is seriously arguable that such terms are a part of the legal relationship. Moreover, it must be satisfied that the relationship in question appears to be valid according to the law by which it will be governed. Having done this, it is then necessary to determine the legal rights and obligations created by the legal construct of Civilopia, by reference to the applicable law. Finally, it then becomes possible to ask whether that construct shares sufficient similarities with a common law trust to fall within the Hague Convention’s scope. The chief advantage of this approach is also its main weakness; namely, that it leaves an element of discretion to the English judge. It means that a legal construct of Civilopia which shows the characteristics of Article 2 of the Convention may, but need not, fall within the Convention. Such discretion may be warranted, for the characteristics in Article 2 are not detailed165 and might scoop up legal relationships under which a separate fund is placed under the control of a party, but where the rights and duties in question are otherwise quite unrecognisable from the normal incidences of a common law trust. In other words, this interpretation would not allow the “shapeless” trust of Article 2 to prevail.166 It would recognise that the Convention is intended to extend beyond the common law trusts model; but it would also acknowledge that, as is clear from the Convention’s Preamble, the common law trust is the principal concern of the Convention; and it would recognise the fact that, as Lupoi demonstrates, any number of institutions of the civil law are “homologous” to the trust and will still fall within the Convention’s scope.167 But such an approach also makes inconsistent application of the Convention in different states likely, if not inevitable. To which, one might retort that this is inevitable in any event, since the likelihood of civil law and common law states uniformly interpreting the Convention is low, especially as regards the exceptions contained therein. But the fact that perfect uniformity is unattainable is no argument for not achieving as much uniformity as possible. Furthermore, it is desirable that, as between onshore and offshore states, something very close to uniformity of application of the Convention is achieved. (e) a court should determine the characteristics of the legal relationship in question and decide that, if it meets all the characteristics specified in Article 2 of 165

Indeed, they are significantly less detailed than the subsequent Principles of European Trust

Law. 166

M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15. M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222. 167

120 Characteristics of the Trust the Hague Convention, it will automatically apply the Hague Convention’s rules to that institution This option does not include the element of discretion present in option (d). It involves saying that, if a legal relationship exhibits the characteristics specified in Article 2 of the Convention, the Convention rules must be applied to it. Whilst the Convention does not require this option, it is certainly the most naturally complementary approach. It means that if the characteristics are demonstrated, the legal arrangement is, for private international law purposes, a trust; if not, it is not. Even here, the characterisation process is more subtle than it would at first sight appear. If the parties dispute whether the Convention applies to the legal construct in question, there is much for the English judge to do. For, when immediately confronted with a civilian construct allegedly governed by the law of Civilopia, an English court has to form an initial view that this is something to which the Hague Convention might apply. It must then look initially at the apparent terms of the arrangement, in order to form a preliminary view as to the applicable law. Having ascertained that applicable law, it then becomes a matter of determining, according to that law, what the constituent ingredients of the relationship actually are. Only once this has been resolved can it be decided whether the arrangement has the characteristics specified in Article 2. If it does, then it is a trust for Convention purposes. This non-discretionary approach is effectively an “international autonomous” characterisation of a trust. The “characteristics” of a trust are turned, if not into a definition of a trust, then into a firm basis for classification. This is a basis which will assist the quest for uniformity and reduce the time spent disputing the application of the Convention. But it is also a basis which allows the full implications of the “shapeless” trust to be felt. It may fuel uncertainty and apprehension as to what application of the Convention entails, both locally, and in any onlooking would-be contracting state considering whether to ratify the Convention. It is not easy to imagine an English court treating e.g. agency as a Convention “trust”, on the basis that it construes Article 2 as being satisfied. On balance, this author would reject option (e) and conclude that option (d) is the most viable one for an English court to adopt. 6. TRUSTS AND THE ROME CONVENTION

A further pitfall of the characterisation process lies in the danger of overlap between the definition of a trust for Hague Convention purposes and of “contractual obligations”, within the meaning of the Rome Convention on the Law Applicable to Contractual Obligations.168 When one considers the remarkably 168

Enacted into English law by the Contracts (Applicable Law) Act 1990.

Trusts and the Rome Convention 121 broad description of “matters relating to contract” adopted for the purposes of Article 5(1), Brussels Convention in Jakob Handte GmbH v. Traitements Mecano-Chimiques Des Surfaces169 as “an undertaking freely entered into by one party in relation to another”, then it is apparent that such a description could, at face value, include vast areas of the law of express trusts. However, the Rome Convention states that it does not apply to the constitution of trusts, nor to the relationship between settlor, trustee and beneficiary.170 Moreover, Article 25 of that Convention states that it “shall not prejudice the application of international conventions to which a contracting state is, or becomes, a party.” It might thus be thought that any overlap between the Rome and Hague Conventions is excluded and that the Hague Convention takes precedence. But matters are not entirely straightforward, at least where a trust arises in pursuance of a contractual obligation. This might occur e.g. in the case of a contractual agreement to create a trust. In the event that such an obligation is breached, would an action brought by one contracting party, the promisee, against the promisor, be governed by the Rome Convention? On the one hand, the claim does involve the relationship between the putative settlor, trustee and beneficiary and the need to constitute the trust. On the other hand, the claim is not about constitution; rather, it is essentially one for breach of contract. It is, effectively, a preliminary question prior to the operation of the trust, and hence appears to be outside the Hague Convention.171 Again, if the remedy sought for breach of contract is specific performance, the claim is effectively one to enforce a trust, through reliance on a contractual right. If one asks whether, according to the applicable law of the contract, Civilopia, a party can be compelled to create a trust to be governed by the law of Commonia, the answer may well be in the negative. But, on balance, the claim does look contractual in nature; and since it is indirectly (and not directly) concerned with the constitution of the trust, it may still be a matter for the Rome Convention. True, the decision of the court will have consequences for any relationship between the putative settlor, trustees and beneficiaries, and true, the substance of the claim, not the form in which it is dressed up, should fall to be characterised; but the fact remains that, without the contractual obligation, no claim would exist at all; for whichever law governs the trust, equity is unlikely to assist the volunteer. An equally complex problem arises where the conclusion of the contract is alleged itself to create a trust obligation This will normally be in the form of a constructive trust, in the case of a specifically enforceable contract. What choice of law rules should be used to decide whether such a constructive trust arises: the applicable law of the contract, or of the putative trust? Although Article 10(1)(c), Rome Convention subjects the availability of remedies to the law 169 Case C–26/91 Jakob Handte G.m.b.H v. Traitements Mecano-Chimiques Des Surfaces [1992] ECR I–3967, para. 15. 170 Art. 1(2)(g), Rome Convention. 171 See Art. 4, Hague Convention, discussed below and Part One of this book.

122 Characteristics of the Trust applicable to a contract, the Rome Convention applies (only) to “contractual obligations”.172 It can apply to personal claims based on breach of contract;173 but it does not determine the point at which title passes. Insofar as the trust can be said to have been “created voluntarily and evidenced in writing” within the meaning of Article 3, Hague Convention, or to arise “under the law of any part of the United Kingdom” under section 1(2), Recognition of Trusts Act 1987, it must be the provisions of the Hague Convention which determine whether, and if so when, the constructive trust arises.174

172

Art. 1(1), Rome Convention. Though not all such claims. Art. 10(1)(e), Rome Convention, which subjects claims based on the nullity of a contract to the putatively applicable law of the contract, has been excluded by the United Kingdom in s. 2(2), Contracts (Applicable Law) Act 1990. 174 However, it is debatable whether the constructive trust imposed in the case of a specifically enforceable contract is within the Hague Convention at all, although this author argues that it is: see the discussion of constructive trusts under Art. 3, below. 173

ARTICLE 3—TYPES OF TRUST GOVERNED BY THE CONVENTION The Convention applies only to trusts created voluntarily and evidenced in writing. The Convention restricts itself to trusts which reflect the will of the settlor, and of which there is written evidence. Unfortunately, the meaning of Article 3 is not straightforward. We shall look in turn at the questions: (i) which formalities will suffice for the trust to fall within the Convention; and (ii) which types of trust fall within the Convention? 1. FORMALITIES

As to the formalities necessary for a Convention trust, one can readily see the reasons for the “evidenced in writing” requirement. It seeks to minimise doubts as to the intention of the settlor to create a trust. This will, in particular, facilitate recognition in states unfamiliar with the trust in their domestic law. It prevents arguments in the courts of Civilopia concerning the alleged intentions of the parties in relation to a largely alien creation. But, as with all formality requirements, it runs the risk of limiting the application of the Convention rules to trusts for which there is ample evidence and of frustrating the recognition of genuinely voluntary settlements. Since the trust normally involves a unilateral act by the settlor, it does not appear necessary that the agreement of the trustee to act in that capacity need be evidenced in writing: the trust will not fail for want of a trustee. It is much less clear whether a written trust instrument need also be signed by the settlor. Of course, this will not be an issue where the trust is created by will and must comply with testamentary formalities. Where the trust is created inter vivos, however, one might expect that a signature would be required, in order to safeguard against fraud. It would be legitimate to argue that “evidenced in writing” means sufficiently evidenced to represent the will of the settlor and that a signature may be required for that purpose. That said, a trust created orally may fall within the Convention, if it is subsequently evidenced in writing. In this context, Lewin on Trusts argues that “written evidence that comes into existence after the trust is declared but before it comes to be enforced will bring it within the Convention, and that it need not be signed by or on behalf of the settlor but may emanate from the trustee”.175 An obvious concern, however, is whether this can be proven to represent the genuine intention of the settlor. 175 Lewin on Trusts, 17th edn., 287. This appears to accord with the view of von Overbeck, para. 52, p. 381.

124 Types of Trust Governed by the Convention It is submitted that the “answer” must be that no specific formal requirement is imposed on the trust by Article 3, other than that there be written evidence of the settlor’s intention. Hence it cannot be said that the written evidence of the trust must be signed by any particular party. However, a state addressed should take the view that there must be sufficient evidence that the settlor intended to create the trust, which may require more than the mere existence of a written document stating that fact. That evidence might most readily be provided by a signature; but it need not be if the court is otherwise satisfied that the trust represents the will of the settlor. That said, it may be open to the court addressed to apply its own rules of evidence to the matter and to decree e.g. that it will not admit extrinsic evidence of the settlor’s intentions. For that reason, the “safer” option is undoubtedly for the trust instrument to be signed by the settlor. It should be noted that Article 3 requires only that the trust itself be evidenced in writing. It is not stated that the terms thereof must be so evidenced. Article 6 provides that a choice of law clause must “be express or be implied in the terms of the instrument creating or the writing evidencing the trust, interpreted, if necessary, in the light of the circumstances of the case.” No formalities are specified as to other terms of the trust. Presumably, only the minimum constituent terms of a valid express trust need be evidenced in writing: there must be certainty of intention, subject-matter and object. However, evidence of additional terms affecting the powers and duties of the trustee need not apparently be contained in the written document.176 Nor does it seem practical to require written evidence of the act of constitution (even though it is, of course, necessary to bring the trust into effect). Indeed, the transfer of legal title to a trustee is better seen as a preliminary, “rocketlaunching” matter and not regulated by the Convention at all.177 Finally, it may be remarked that the lack of detailed formality provision in Article 3 makes it hard to see it as laying down an exhaustive rule of uniform law. Accordingly, it may also be open to a court to consider additional formalities imposed upon the trust by national law; and in particular, those of the law applicable to the trust. This choice of law question is considered in the context of Article 8 below.178

2. WHICH TYPES OF TRUST ARE WITHIN THE CONVENTION ?

(A) Voluntary Trusts Two obvious questions immediately present themselves: which trusts are included within Article 3; and why should this delimitation have been introduced into the Convention? 176 Though it may be preferable for all trust terms to be evidenced in writing, since it cannot easily be predicted what extrinsic evidence the courts of any given contracting state might admit of additional terms of the trust. 177 See further Part One of this book; and the discussion of Art. 4 below. 178 In the section “formal validity of the trust.”

Which Types of Trust are Within the Convention? 125 As to the first question, the category of trusts created “voluntarily” does not exist as such in English law and would appear to be a wider one than the more familiar express trust.179 It might seem to be unnecessary to determine the precise meaning of this category, given the extension of the scope of the Convention in England beyond Article 3’s boundaries.180 But very often, the prime consideration will be which sorts of English trusts will be recognised in other contracting states (which have not extended the Convention’s scope and apply it only to trusts falling within Article 3) and in that respect the boundaries of Article 3 are crucial. It appears that the trust must at the very least reflect the will of the settlor (even if he did not expressly declare and constitute that trust), so excluding constructive trusts imposed by the court irrespective of the settlor’s will and trusts created by operation of law.181 This is despite the fact that trusts arising by operation of law, such as those on intestacy, may be perfectly easy to enforce overseas.182 Trusts voluntarily created as a means of discharging an obligation imposed by law, e.g. to discharge a maintenance obligation, are not, however, excluded.183 Moreover the fact that aspects of an expressly created trust require execution by the court does not lead to the trust’s exclusion, e.g. where the court has to intervene to nominate a trustee.184 Article 3 will also cover express trusts varied by court order, in pursuance, for example, of a marriage settlement or powers under the Variation of Trusts Act 1958 or use of the cy-près doctrine in the case of charitable purpose trusts.

(B) Resulting Trusts As to resulting trusts, it appears that those of the automatic type fall within the Convention.185 The trust in question need not itself be expressly constituted, it seems, providing that a trust arises from the express declaration of trust. One cannot help but observe that, in the light of the exclusion of trusts arising by operation of law, the inclusion of resulting trusts appears rather curious.186 179 See also M Lupoi, Trusts: A Comparative Study, 14, 341: cf. the understanding of a contract for the purposes of Art. 5(1), Brussels Convention as “an undertaking freely entered into by one party in relation to another” in case 26/91 Jakob Handte G.m.b.H v. Traitements MecanoChimiques des Surfaces [1992] ECR I–3967, para. 15. 180 S. 1(2), Recognition of Trusts Act 1987 (on which, see below). 181 Although the desirability of these exclusions is not unquestionable: “I am unable to comprehend why trusts established by law should have been categorically excluded from the Convention” (M Lupoi, Trusts: A Comparative Study, 341). Similar views were expressed by the United Kingdom delegate (Actes et Documents 146). 182 M Lupoi, ibid., 341–2. 183 Von Overbeck Report, para. 49, p. 380. 184 Ibid., para. 50, p. 380. 185 Ibid., para. 51, pp. 380–1. 186 This author would agree with M Lupoi, Trusts: A Comparative Study, 342, who observes that “the (erroneous) conviction was developed which saw resulting trusts falling within the subject-matter of the Convention, because they were established voluntarily, whereas constructive trusts remain

126 Types of Trust Governed by the Convention Although von Overbeck discusses the inclusion of “resulting trusts” (without further elaboration), he appears to be referring only to automatic resulting trusts and not to the category of presumed resulting trusts.187 But the justification for the inclusion of the former category can hardly be that the automatic resulting trust gives effect to the settlor’s intentions, even if one accepts for present purposes that this is the theoretical basis of such trusts. If it were, it would seem curious not also to give effect to other forms of implied trust apparently based on the settlor’s assumed intentions, including the presumed resulting trust and many constructive trusts. And although it will be apparent to a judge in Civilopia that an express trust was to be created, and that, upon its failure, some arrangement has to be accommodated with regard to the trust property, thereby making the trust more “visible” than a trust created by judicial decision might be, the fact remains that an automatic resulting trust is not an express trust and that the rights and duties of the parties in relation to it will probably be quite different.188 Transparency alone cannot justify the inclusion of automatic resulting trusts, since many other trusts arising by operation of law (e.g. that arising in English law in the event of intestacy) are scarcely less transparent. A better justification for the inclusion of automatic resulting trusts is provided by Stevens. He argues that the law which determines that an express trust fails should also be used to determine the aftermath of that failure: “If a restitutionary analysis is adopted it can be strongly argued that the system of law which determines that the enrichment is unjustified should also determine its consequences”.189 Of course, such dichotomies are not unknown in the conflict of laws: the law applicable to a contract determines whether a contract is void, but it does not necessarily determine the restitutionary aftermath thereof;190 nor, in a contract of sale, does it determine the point at which title to goods passes. But where they can be avoided and a single law coherently applied, so much the better. The automatic resulting trust is a case where this might legitimately be excluded because they are ‘created’ by a court . . . [C]ontrary to the predetermined purpose of the drafters, certain types of constructive trust might be included, while certain types of resulting trust might not.” 187 Von Overbeck Report, para. 51, pp. 380–1. But cf M Lupoi, ibid., 343: “if one makes will the controlling argument, the application of the Convention to resulting trusts might be prejudiced.” For a different view again, see K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 28: “Voluntary trusts clearly include resulting trusts but exclude constructive and statutory trusts.” 188 See also M Koppenol-Laforce, Het Haagse Trustverdrag, 265 who is equally unconvinced that the automatic resulting trust should fall within Art. 3. 189 R Stevens, “Resulting Trusts in the Conflict of Laws” in P Birks and F Rose (eds.) Restitution and Equity, vol. 1: Resulting Trusts and Equitable Compensation (London, Mansfield Press, 2000), 147, 157. 190 Art. 10(1)(e), Rome Convention, which would have referred the consequences of nullity to the law (putatively) applicable to the contract, was not enacted in the United Kingdom (see s. 2(2) the Contracts (Applicable Law) Act 1990). In Scotland, application of the putatively applicable law of the contract has been rejected by the Outer House of the Court of Session in favour of a more flexible (but less certain) test based upon the objective proper law of the (putative) restitutionary obligation: Baring Bros & Co Ltd v. Cunninghame District Council [1997] CLC 108.

Which Types of Trust are Within the Convention? 127 done. And if the same law is to be applied, it is better that the recognition of both should fall within the Hague Convention. As to resulting trusts of the presumed variety,191 such as that which might arise where a jointly purchased property is registered in the name of one party only, matters are less clear. Hayton postulates a scenario where S and T purchase a property in England, which is conveyed into T’s sole name, which T then sells. T reinvests the proceeds of sale in property situated in Civilopia (a contracting state to the Hague Convention). An English court holds that a presumption of a resulting trust arises. If a Civilopian court was faced with a claim by S to be recognised as a beneficiary under this presumed resulting trust, should Civilopia recognise the trust under the Convention? Even if the trust can be said to have been created “voluntarily”, the resulting trust does not appear itself to be “evidenced in writing.” Nevertheless, Hayton argues that the phrase “evidenced in writing” imposes a requirement of sufficiency of evidence of the trust, rather than imposing a strict formality requirement. A Civilopian court will not investigate oral evidence of an alleged resulting trust, if it is itself asked to determine the existence thereof; but if an English court has already declared that T held the property and proceeds of sale on presumed trust, Hayton contends that the English’s court declaration is itself the necessary written element evidencing a voluntarily created trust, so that its recognition in Civilopia should fall within the Convention.192 With respect, it is not easy to accept this argument. Article 3 makes it reasonably clear that the trust itself must be voluntarily created and evidenced in writing. In a sense, one can argue that the presumption of a resulting trust is premised on an assumed agreement between S and T that, as co-contributors to the purchase price, both should be beneficiaries of a trust. However, that is but a legal fiction which strains the idea of voluntary conduct to breaking point.193 The parties may have to accept the legal consequences of joint contribution to a property; but that does not mean that the “settlor” voluntarily agrees to create a trust. This is a trust created by a presumption of law (albeit a presumption which is unrebutted on the facts), reaffirmed by judicial decision. It is not easy to describe it as “voluntary”; it is harder still to say that the existence of the trust was manifested by written evidence of the settlor. It is accordingly suggested that presumed resulting trusts lie outside the scope of Article 3. 191 The questions whether the Recognition of Trusts Act 1987 extends to presumed resulting trusts was left open by Peter Gibson LJ in Lightning and Anr v. Lightning Electrical Contractors Ltd, judgment of 23 April 1998. C Jauffret-Spinosi “La Convention de la Haye Relative à la Loi Applicable au Trust et à sa Reconnaissance (1er Juillet 1985)” (1987) 114 Journal du droit international 23, 35 argues for the inclusion of all resulting trusts: “ . . . si l’on estime que seule l’intention du constituant n’a pas été expressément manifestée, mais qu’elle existait dans son esprit.” 192 D Hayton, (1987) 36 ICLQ 260, 264; D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 944. 193 “The resulting trust which arises in favour of S is not a trust ‘voluntarily created’ at any stage as it arises by operation of law”: R Stevens, “Resulting Trusts in the Conflict of Laws” in P Birks and F Rose (eds.) Restitution and Equity: Resulting Trusts and Equitable Compensation 147, 157.

128 Types of Trust Governed by the Convention (C) Constructive Trusts Matters are considerably more complex with regard to the constructive trust, given its many guises, spheres of operation and the intensity of the debate as to its nature. However, it is possible that some forms of constructive trust will fall within Article 3. One cogent example postulated by Lupoi is where duly executed written mutual wills are made.194 True, the parties may not have expressly declared the particular trust which binds them to the agreement. Nevertheless, there will be written documentation which reflects the voluntary intention of the parties to be bound by the mutual wills and, if necessary, to have them enforced by a constructive trust, which should suffice to bring the trust within the Convention. A related scenario is where a written contract is made which is specifically enforceable. This may give rise to a constructive trust from the time of conclusion of the contract.195 It could be argued that the parties voluntarily contract, that they know that a constructive trust will arise and that there is written evidence of the agreement which generates that trust. It should follow that the constructive trust arising in these circumstances falls with the Convention. Against this, it could be said that there has been no declaration of trust and that whilst the parties may choose to enter into a written contract, the constructive trust is imposed pursuant to that contract. On balance, this author would favour the inclusion of such trusts within the Convention. A contract creates a voluntary obligation and the parties willingly submit themselves to a legal framework; if that framework decrees that a constructive trust arises, then so be it. The parties have agreed to the framework, and it is evident from looking at the subject matter of the written contract that the trust arises. Lupoi suggests that where there is a written declaration of obligations assumed by a party, short of a declaration of trust, (such as obligations undertaken by a seller of property which are not contained in the final sale instrument), any constructive trust imposed on that party to effect those obligations should fall within the Convention.196 The constructive trustee may not want the trust imposed on him, but he can hardly complain that he did not freely undertake obligations which would, in default, bring that constructive trust into existence, or that there is not written evidence to such effect. Equally clearly, the personal remedy imposed upon a stranger to the (express) trust making him liable to account as a “constructive trustee” for losses or gains made as if he were an express trustee cannot be within Article 3, since “The Convention . . . is [concerned] with trusts of specific property”.197 194

M Lupoi, Trusts: A Comparative Study, 342–3. In English domestic law, “From the very moment that a vendor enters a specifically enforceable contract to sell property, he holds it on Constructive trust for the purchaser”: R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn. (London, Butterworths, 1998), 262. 196 M Lupoi, Trusts: A Comparative Study, 342. 197 D Hayton, (1987) 36 ICLQ 260, 264. 195

Which Types of Trust are Within the Convention? 129 It must be the case that, where the assets of an expressly created trust change during the lifetime of the trust, any substituted assets in the trustee’s hands now representing the trust fund fall within the Convention’s scope. It would be unworkable if recognition of an expressly created trust overseas depended on the trust assets remaining the same as when the trust initially came into effect.198 It would also appear that claims to recover trust assets of an express trust (evidenced in writing) wrongfully dissipated to a third party who is alleged to hold the assets on constructive trust should be included within the Convention. Hayton points out that the ability to recover trust assets from a third party is pivotal to the common law trust199 and argues that this constructive trust must remain within the Convention.200 He notes that following and tracing provisions are contained in the Convention, Article 11(3)(d) of which states that a consequence of trust recognition is that “the trust assets may be recovered when the trustee, in breach of trust, has mingled trust assets with his own property or has alienated trust assets.” However, one might question the soundness of Hayton’s view: such constructive trusts are very difficult to reconcile with the need for the trust in issue to be the result of voluntary creation. Article 11(3)(d) in fact goes on to state that “the rights and obligations of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum.” In other words, the Convention rules do not necessarily apply unless a state chooses to extend them to claims against third parties. Against that, it could be argued that the fact that third party claims are mentioned at all in the Convention shows that such constructive trusts are within the Convention’s scope. Article 11(3)(d) envisages that, in principle, express trust assets may be recovered from third parties; it merely goes on to make clear that this rule is not absolute.201 In effect, the constructive trust in this scenario is treated for present purposes as an incidence of the express trust initially created under the Convention.202 On balance, then, it is suggested that such constructive trusts should be regarded as within the Convention. Beyond these examples, the scope of Article 3 is still less clear. Constructive trusts which impose an obligation to hold specific property may be affirmed or created by a court and may or may not reflect the actual or imputed intentions of the parties. Whilst von Overbeck informs us that the Convention excludes “constructive trusts imposed by the courts”,203 Hayton argues that “the Convention should apply where persons voluntarily intend to create a trust 198 A similar view is expressed by D Hayton, “International Recognition of Trusts” in J Glasson (ed.) International Trust Laws, ch. C3, 7. 199 Although Art. 2 extends beyond the common law trust: see above. 200 D Hayton, (1987) 36 ICLQ, 260, 266. 201 See further the discussion of Art. 11(3)(d) below. 202 But not all agree. M Koppenol-Laforce, Het Haagse Trustverdrag 266 argues that this constructive trust is only included where a state has enacted provisions pursuant to Art. 20 extending the scope of the Convention. This view is highly tenable and may even be preferable in theory; but as a matter of construction of the Convention itself, this author inclines to think that this constructive trust is within the Convention’s scope. 203 Von Overbeck Report, para. 49, p. 380.

130 Types of Trust Governed by the Convention which cannot take effect as an express trust because of a failure to comply with the requisite statutory formalities, but the court intervenes to vindicate the express trust by imposing a constructive trust so as to prevent the property holder from fraudulently retaining the property for himself”.204 Equity, in its steadfast refusal to allow a statute to be used as a cloak for fraud, would thus be capable of transcending the national boundaries of Civilopia. One might also think that this view is consistent with the inclusion of automatic resulting trusts in the Convention, since both types of trust arise pursuant to a failed express trust. With respect, however, it is difficult to agree with Hayton’s view. The automatic resulting trust arises because there have to be legal effects given to a trust which is validly declared, but which fails in whole or in part, in order to prevent beneficial rights existing in a vacuum. In contrast, the constructive trust of the type described by Hayton upholds the sprit of a trust purported to be created voluntarily; but it is not a trust itself created voluntarily. Rather, it is, of course, a trust which one party does not wish to see imposed, and one which does not need to be imposed at all. Furthermore, Reymond points out that the constructive trust is used to perform a number of functions which civil law systems deal with using separate provisions, such as those relating to unjust enrichment, and that it should normally be excluded from the Convention: “Car le recours au trust dans ces hypothèses s’appuie sur une certain division du droit qui n’est pas la nôtre”.205 Moreover, Hayton’s suggestion that a court decree would itself be sufficient written evidence of the trust is slightly unsettling. If the affirmation of a constructive trust by judicial decision provides the requisite formalities as to the existence of that trust, then it is not easy to see why “judicial trusts” needed to be excluded from the Convention in the first place. And Hayton’s suggestion might make it possible for the courts of Civilopia to recognise a constructive trust reaffirmed by an English court; but it will scarcely help the courts of Civilopia, faced with a choice of law problem as to whether to impose a trust itself, to decide whether to do so.206 It is also notable that Nygh unequivocally states that “. . . trusts created by law such as: implied, resulting and constructive trusts, remain outside the Convention”.207 Béraudo also takes the view that all 204 D Hayton, (1987) 36 ICLQ 260, 264; see also P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7. 205 C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé 7, 19. 206 Curiously, C Reymond, ibid., 19–20, states that a trustee could not use the Convention to obtain directions from a civil law contracting state’s courts as to the operation of a trust, since the trust is not known or administered in such states: “C’est dire que le trustee obtient le statut d’un visiteur privilégé, mais pas celui d’un résident.” However, if a court has in personam jurisdiction over the defendant and is asked to give effect to a trust governed by the law of England, it is submitted that the Convention does necessitate that court to uphold the trust in accordance with English law. Otherwise, the Convention would be a pure exercise in passive recognition in civil law states, where no assistance administering the trust could be obtained. 207 P Nygh, Conflict of Laws in Australia, 6th edn. (Sydney, Butterworths, 1995); no further words of explanation of this interpretation are proffered. The present authors view is that some constructive trusts (considered above) do fall within the Convention, but that most do not.

Which Types of Trust are Within the Convention? 131 constructive trusts are outside the scope of the Convention.208 Implicitly, both must reject Hayton’s views. Further specific cases are considered by Hayton. One such case is the fully secret trust created where a beneficiary under a will agrees during his lifetime to hold his inheritance on trust for another.209 Once again, Hayton suggests that, if the promisor then refuses to hold the property on trust and an English court declares him bound so to do, the trust’s recognition falls within the Convention, because the judgment is itself sufficient written evidence of the trust. Once again, this writer is bound to disagree. One might engage in lengthy debate about whether such a trust is express or constructive210 but, either way, it is fair to say that it is created “voluntarily.”211 But if the “evidence” that it is was created is that the court thought so, the limits of Article 3 are remarkably wide. More to the point, what should the courts of Civilopia do if they find themselves taking jurisdiction and faced with a claim on the merits that a secret trust arises by the applicable law of that alleged trust, English law? The difficulties with which it will be confronted appear to be precisely of the type that the exclusion in Article 3 was designed to remove. The half-secret trust may be an even more complex case. If the existence of the trust appears in the will, but not its terms, does that make the trust voluntarily created and evidenced in writing? Put differently, how much of the trust need be in writing? Once again, it does not appear to be helpful to answer this question by asking the same theoretical questions as one might in English domestic law: is the trust testamentary or does it operate dehors the will; is it express or constructive etc? The trust is voluntarily created and, this time, there is written evidence of its intended creation rather than a court decision to that effect. However, the trust’s validity depends upon the terms being communicated at an appropriate time, in an appropriate manner, and the trust being sufficiently certain. It is undoubtedly unfortunate that a court in Civilopia may have to wrestle with a trust whose terms are not set out in the will, and indeed, have to consider the validity of the trust212 which is one of the problems that Article 3 sought to avoid. However, the trust seems to fall within the spirit of Article 3. A suitable compromise might be to include the trust within the Convention provided that: (i) the terms are communicated according to the requirements of the putative applicable law; (ii) the terms necessary to establish the trust’s validity are themselves ultimately communicated in, or evidenced in writing; and (iii) those terms are sufficiently certain and the trust is duly constituted. 208 J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, ch. 2, 219. He points (at 220) out that the French courts have, exceptionally, recognised the existence of a constructive trust: see Clunet 1935, 651. However, in contrast to Nygh, Béraudo states that all resulting trusts are within the Convention’s scope. 209 D Hayton, (1987) 36 ICLQ 260, 265. 210 See e.g. R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn., ch. 7 for a consideration of the cases and of the voluminous academic literature. 211 See also W Binchy, Irish Conflict of Laws (Ireland, Butterworths, 1988), 508. 212 Including the time at which the terms were communicated to the (half-secret) trustee.

132 Types of Trust Governed by the Convention Hayton also raises the question of the “family home” constructive trust, where a common intention to share a property is backed by detrimental reliance. Once again, he regards a court order establishing the trust as sufficient written evidence of the trust. Once again, it is difficult to agree. The trust itself is not created voluntarily. True, it apparently stems from an agreement between the parties short of an express trust. But given that, in many cases, a shared desire between the parties to share the property beneficially appears to be lacking,213 talk of a voluntarily created trust appears strained. And once more, can the court order really provide sufficient written evidence? What if the question for the courts of Civilopia is not whether to give effect to a trust imposed by an English court, but whether, as a choice of law question, itself to impose such a trust? Such consideration of the existence of the family home constructive trust is precisely the sort of difficulty that the confining of the Convention to “trusts created voluntarily” might be expected to avoid. Hayton concludes that it is only constructive trusts of the type imposed regardless of the intention of the parties which should be excluded from Article 3. An example might be the case of an unauthorised profit made by a fiduciary. However, where that fiduciary is the trustee of a written, express trust, it could be argued that it is better to regard the constructive trust imposed in respect of unauthorised profits as an incidence of his trusteeship under the express Convention trust. Although the trust imposed is part of the “general” law on fiduciaries, it arises on the facts because of the profit holder’s status as an express trustee. If the trustee voluntarily accepted the position, he can be said voluntarily to have accepted the fiduciary obligations that go with the office. But against this, the position can be distinguished from e.g. the constructive trust created by a specifically enforceable written contract. In the latter scenario, the constructive trust arises immediately and is evident from the face of the contract; in the present case, it will not be evident from the trust instrument whether a constructive trust exists.214 That will depend upon whether: (i) profits have been made; and (ii) whether they are unauthorised; and if the constructive trust arises, it does so not because the trustee is a trustee, but because the trustee is a fiduciary. On balance, it is suggested that the constructive trust imposed upon a fiduciary, (whether that fiduciary is the trustee of an express trust or not), lies outside the scope of Article 3.215 Ironically, however, application of the Convention to institutional constructive trusts imposed irrespective of the parties’ intentions may be easier for a court of Civilopia than application of it to the “intention” based constructive trusts. In relation to unauthorised profits made by a fiduciary, difficult questions 213 Especially where one party makes an excuse to the other as to why their name should not appear on the legal title: see e.g. Eves v. Eves [1975] 1 WLR 1338; Grant v. Edwards [1986] 2 All ER 426. 214 Only that it might exist, if the trustee subsequently makes an unauthorised profit. 215 See further the discussion of Art. 8(2)(g) below, where it is argued that, in any event, the law applicable to the express trust should determine the fiduciary obligations of the trustee.

Which Types of Trust are Within the Convention? 133 as to whether the trust should be imposed or affirmed are, on the whole, less likely to arise, given the blanket nature of the trust in issue216 and the fact that the parties’ intentions do not need to be discerned.217

(D) Statutory Trusts Hayton then turns to the question of “statutory trusts”, such as that which arises upon a person’s intestacy,218 which are automatically triggered by certain scenarios (in contrast to trusts which may be declared by the court in the exercise of its statutory jurisdiction, such as pursuant to section 24(1)(b) Matrimonial Causes Act 1973). Hayton argues that “by choosing to die intestate the deceased has voluntarily authorised his administrators to vest his property in trustees for his widow and children as evidenced in the written assent”.219 With due respect, this is a remarkably broad view to take of voluntary conduct. If every automatic legal consequence which follows from a failure to act in a particular manner can be said to be voluntarily incurred, then no wonder equity will not normally assist the volunteer. At best, what one sees in Hayton’s example is a failure to prevent an (avoidable) trust from coming into existence. But that trust is imposed without the consent of the deceased; and the deceased was probably completely unconscious that a trust would come into effect on his intestacy, if, indeed, he intended to die intestate at all.220 Inadvertent, negligent, and perhaps even reckless failure to follow a course of action do not amount to a voluntary desire to accept the consequences of that failure. In short, it is suggested that such a statutory trust lacks the necessary element of consensus221 and must be regarded as outside the Convention.

(E) Hayton’s Views Considered We have seen above a number of points of agreement and disagreement between Hayton and the present author as to the scope of Article 3. More generally, one feels that Hayton’s discussion of the scope of Article 3 is subject to a deeper problem. Quite rightly, he sees the Convention’s scope as an issue of importance primarily in allowing a judge in a civilian court to understand what to do with 216 At least in states which do not have a remedial constructive trust and an element of discretion as to whether to impose a constructive trust or to confine the claimant to a personal claim. 217 Of course, difficult questions about whether the fiduciary duty has been breached may still arise. 218 S. 33, Administration of Estates Act 1925. 219 D Hayton, (1987) 36 ICLQ 260, 267. See also D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 945–6. 220 As opposed to having intended to make a will but having not got around to doing so before his death, or having made a will which is invalid because it is not properly signed or witnessed. 221 In this author’s view, it is not even close.

134 Types of Trust Governed by the Convention a trust valid under the common law, and whether to recognise that trust in Civilopia. For those purposes, it might well be sufficient that, although there is no express trust instrument setting out the terms agreed by the settlor, there is a court order declaring the terms of the trust. But the Convention fulfils another function, as a provider of a set of choice of law rules. It may be that the courts of Civilopia themselves are the ones asked to rule on the alleged existence and validity of a trust which one party alleges exists and is governed by English law. In that scenario, a court of Civilopia might well be able to cope with an express trust evidenced in writing. But if it is asked to place itself in the shoes of an English judge and rule on the existence e.g. of a constructive trust, even one alleged to arise from the intention of one or more parties, it may very quickly feel that it is out of its depth. And if there is a such a risk, then there is good reason for such states not to extend the Convention to such trusts. In other words, the scope of the Convention must be the same whether the issue facing a civilian court is recognition of a common law trust already declared by an English court to exist or a choice of law question where the Civilopian court is itself asked to rule as to the existence of a trust. The “front door” provision for entering the Convention is the same. True, it is likely to be rare that the courts of Civilopia will actually be asked to resolve a choice of law question of the second kind; but if it were asked to do so, the problems that it might face with an alleged trust of which there is no written evidence are very considerable and might even put that state off ratifying the Hague Convention at all. It is for this reason above all, it is submitted, that a far narrower view of the scope of Article 3 should be taken than that suggested by Hayton.

3. WHY EXCLUDE JUDICIALLY CREATED TRUSTS ?

(A) General The exclusion of trusts created or affirmed by judicial decision is not selfevidently desirable.222 The argument which won the day at the Hague Conference was that express trusts respected the intentions of the parties; whereas trusts created otherwise offered no such guarantee. Moreover, the range of trusts declared or affirmed by judicial decision risked cutting across categories used in civil law states which seek e.g. the reversal of unjust enrichment, by other means. Against that, however, the intervention of a court to impose or to affirm a trust might even lead to more inherently reasonable, clear obligations ultimately being imposed than might arise under an express trust: “. . . tels trusts sont des 222 The English and French delegations lobbied for their inclusion: L Maerten, “Le Régime International du Trust après la Convention de La Haye du 1er Juillet 1985” [1988] La Semaine Juridique 3319.

Why Exclude Judicially Created Trusts? 135 trusts sains, et non frauduleux, qu’étant soumis à la lex magistratus, il n’y a pas de crainte que le constituant ait fraduleusement choisi une loi étrangère”.223 Moreover, the Brussels Convention imposes an obligation on contracting states to recognise and enforce civil and commercial judgments delivered in England. The trust which is subject to litigation, and which is declared or affirmed by court judgment, is ironically the one most likely to be recognised under that scheme; an express trust in respect of which no court decision has been given will not give rise to a recognition obligation under the Brussels Convention.

(B) The “Problem” of Recognition of Foreign Judgments in Relation to Trusts Nevertheless, it is clear that the perceived problems of recognition of foreign judgments to a large extent lay behind the restriction of the Convention to trusts created voluntarily. Hayton argues that trusts declared by judicial decision were excluded so as not to open “the ‘Pandora’s box’ of the recognition of foreign judgments and the bases for allowing or disputing such recognition”.224 In apparent contrast, the von Overbeck Report, in discussing the desirability of the inclusion of judicially created trusts, comments simply that “further arguments were based on the Brussels Convention on jurisdiction and the recognition of judgments”.225 Von Overbeck appears to have in mind the fact that EC states would be bound in any event to recognise trust226 judgments of other contracting states under the free recognition principle of the Brussels Convention. Those states might have little cause for complaint about recognising judicial trusts created in other contracting states to the Brussels Convention.227 However, on closer inspection, neither the Hayton nor the von Overbeck observation wholly convinces. Both are a product of the ambiguous use of the word “recognition” in the Hague Convention. The fact that a contracting state to the Brussels Convention might have to recognise an English judgment creating a trust is inevitable, given that, by Article 29 of that Convention, “under no circumstances may a foreign judgment be reviewed as to its substance”.228 In that sense, the consequences of an English trust judgment in relation to assets located in another contacting state to the Brussels Convention must already be confronted by non-trust states party to the Brussels Convention. Nor, under the 223 C Jauffret-Spinosi “La Convention de la Haye Relative à la Loi Applicable au Trust et à sa Reconnaissance (1er Juillet 1985)” (1987) 114 Journal du Droit International 23, 33. 224 D Hayton, (1987) 36 ICLQ 260, 267. 225 Von Overbeck, para. 167, p. 408. 226 Note that, “The fact is, however, that the Brussels Convention trust . . . is the English model trust” (M Lupoi, Trusts: A Comparative Study, 344) and has a quite different meaning to the “trust” covered by the Hague Convention. 227 Even if an English court declares or imposes a trust governed by the law of a state outside Europe, a contracting state to the Brussels Convention is bound to recognise that judgment under the free recognition rules of the Brussels Convention. 228 Reproduced as Art. 36, Brussels Regulation.

136 Types of Trust Governed by the Convention Brussels Convention, can it be a basis to refuse recognition of a trust judgment that the trust itself is unknown in the state of recognition; indeed, if unknown causes of action of the state of origin allowed another state to refuse recognition, the Brussels Convention scheme would be torn apart. But the Hague Convention does not create rules relating to the recognition of foreign judgments; rather, it spells out to a court the typical consequences of giving effect to a trust falling within the Convention’s scope. Inclusion within the Convention of trusts created by judicial decision in England would not have entailed that the decision itself need be recognised in Civilopia. In contrast, the “recognition” that is referred to in the context of the Hague Convention is quite different and potentially much further reaching. Imagine that judicially created trusts were within the scope of Article 3, Hague Convention. A Civilopian court would be bound to recognise a foreign trust judgment if, but only if, Civilopia’s rules on recognition of foreign judgments were satisfied. The Brussels Convention’s obligation to recognise English trusts judgments itself seems to impose no requirement at all that a choice of law Convention should also cover judicially created trusts. However, the Hague Convention could theoretically also require civilian courts with jurisdiction themselves to determine whether a trust should be judicially created, and not merely to give effect to a trust already declared to exist by the judgment of the courts of a trust state. If a claimant alleges in the courts of Civilopia that a defendant should be required to hold on trust for him because English law (the law putatively applicable to that trust) would so decree, the civilian court might itself have to rule as to whether the trust existed and, if so, to recognise the trust as having certain effects under Article 11, Hague Convention. This means that if an English court would declare or impose a trust on the facts, applying English law, then the courts of Civilopia should do likewise, as they are instructed to apply English law to all questions of substance within the Convention.229 That is a matter squarely within the Hague Convention; yet it is a far more direct application of “English” judicial trusts than the recognition of foreign judgments entails, because it involves the courts of Civilopia themselves exercising their minds as to how an English judge, faced with these facts, would rule on whether a trust should be imposed. That might be the type of decision which judges in civilian states might be unwilling to make, involving as it does a Civilopian judge placing himself in the shoes of an English judge. Of course, if the courts of a non-trust contracting state to the Brussels Convention have jurisdiction, they could potentially be asked to impose a trust with or without the Hague Convention. However, without the Hague Convention, a non-trust state may refuse to do so, on the basis that the trust is unknown to it in domestic law and for choice of law purposes. The likelihood of a claimant seeking to litigate in a non-trust state is substantially increased if 229 This, of course, assumes in the case of the constructive trust that it is neither procedural nor remedial in nature. On this, see Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd [1981] Ch 105.

Why Exclude Judicially Created Trusts? 137 that state has choice of law rules derived from an international Convention to deal with trusts. If so, this is a rather more satisfactory reason for excluding judicially declared trusts from the Hague Convention regime than any reason based on the rules for recognising foreign judgments.230

(C) “Remedial” Judicial Trusts Equally puzzling is von Overbeck’s statement that: “in the common law countries judicial trusts were most often a method of providing a legal remedy . . . Now, it was unthinkable to require recognition in trust form of something which would not be recognised in the form of an ordinary judgment for the payment of a sum of money”.231

In the first instance, it may be observed that the fact that a judicially created or affirmed trust might have remedial consequences does not make the trust itself remedial, at least if the phrase “judicial trust” is understood to include trusts whose existence is affirmed by court order, rather than created at the court’s discretion by said order.232 It is true that the remedial constructive trust would be remedial in a more meaningful sense, in that it is dependent on the exercise of the judge’s discretion, rather than being automatically triggered by a set of conditions;233 but outside this example, one could hardly say that “most” judicial trusts were a legal remedy, since no element of court discretion is permitted. Unless, that is, the category of “judicial trusts” is taken to cover only remedial constructive trusts and trusts imposed at the court’s discretion and to exclude the conventional “institutional” scenarios in which a court might find a constructive trust in England e.g. in the case of a specifically enforceable contract, or an unauthorised profit made by a fiduciary. If that is correct, then the subsequent extension of the Convention in England to judicial trusts,234 if similarly construed, would be far narrower than one first might imagine and leave the bulk of the law of “institutional” constructive trusts untouched,235 and therefore still governed by common law rules.

230 In contrast, when the courts of Civilopia are called upon to recognise a Brussels Convention judgment delivered in England, the law applied by the English court is a stage removed, and it is the judgment itself which is the focus of attention. 231 Von Overbeck Report, para. 167, p. 408. 232 R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 287–93. 233 See P Birks, “Proprietary Rights as Remedies” in P Birks (ed.) The Frontiers of Liability (Oxford, University Press, 1994), vol. 2, ch. 16, 214. 234 On which, see the section below. 235 Or, more accurately, the bulk of institutional constructive trusts not governed by the law of a part of the United Kingdom (as s. 1(2), Recognition of Trusts Act 1987 also applies to such trusts).

138 Types of Trust Governed by the Convention

(D) Trusts Imposing a Continuing Obligation Von Overbeck explains that the Commission also considered a proposal to include “trusts of a continuing administrative nature which are created by a judicial instrument”.236 This would have covered “lasting” trusts such as an order that a testamentary executor be appointed trustee of property left to the deceased’s children. It would not cover constructive trusts “which are simply a transfer of assets”.237 No doubt, the striking over-simplification of this description, when one considers the constructive trust in all its many guises, would not impress most common lawyers. Moreover, the idea that the constructive trust involves transferring assets (which are ex hypothesi at some stage owned beneficially by the defendant), rather than recognising their existence, suggests a restitutionary classification of the constructive trust, a point which will often be highly debatable. But leaving these objections aside, the suggested proposal would have a certain appeal. A trust which the courts of Civilopia could see had an enduring status (and in respect of which the trustee’s rights and duties may have come to share certain characteristics with their powers and duties in respect of the day-to-day operation of an express trust) might well be one to which it would be minded to give legal effect. On the other hand, the very fact that the trust is of a “continuing administrative nature” is a reason for the courts of Civilopia to steer clear, lest, rather than being asked merely to recognise decisions made in a common law state, they were themselves asked to exercise administrative powers. In fact, the reason given by von Overbeck for the ultimate rejection of the proposal is that “it would involve the recognition of foreign judgments”.238 Once again, this is an objection which does not stand up to scrutiny. Insofar as a trust of this nature arises and is supervised by court order, there will indeed be a judgment of the English court which may need to be enforced in Civilopia. But whether the trust is recognised there is simply a question of the rules relating to the recognition of any foreign judgment. Inclusion of such trusts within the Hague Convention would surely not necessitate the recognition of trusts from a court not otherwise jurisdictionally competent in the eyes of Civilopia. Otherwise, the intervention of an English court in a matter in which it had no jurisdiction in the eyes of Civilopian law would require its decision to be recognised abroad. Instead, a better point would have been that trusts of a “continuing administrative nature” require court intervention. It would make little sense for the courts of Civilopia to recognise the trust itself, but then potentially be unable to recognise decisions of an English court in relation to that trust, because they do not satisfy the rules on recognition of foreign judgments. 236 237 238

Von Overbeck Report, para. 168, pp. 408–9. Ibid., para. 168, p. 409. Ibid., para. 168, p. 409.

Statutory Extension of Convention Rules in the UK 139 Indeed, if that scenario were to arise, it might be necessary to seek an order in relation to the trust in Civilopia itself; a prospect which such a state is hardly likely to relish.

4. THE STATUTORY EXTENSION OF THE CONVENTION RULES IN THE UNITED KINGDOM

(A) The Relevant Provisions The mandatory exclusion from the Convention of presumed resulting trusts and of most constructive and statutory trusts, as well as those created by judicial decision would have been particularly unfortunate in trust states, which are comfortable with such trusts and which would be quite willing to apply the laws of another trust state in deciding whether to impose such a trust. Article 20(1),239 was the compromise. It provides that: “Any Contracting State may, at any time, declare that the provisions of the Convention will be extended to trusts declared240 by judicial decisions.”

In a sense, such a statement hardly seems necessary. For matters which are outside a Convention’s scope, it is always possible, of course, for a state to choose to apply the same choice of law rules as exist within the Convention. Nevertheless, for the sake of clarity, it is enormously beneficial that any provisions extending the scope of a Convention be made as clear as possible.241 The United Kingdom242 was quick to take advantage of the opportunity afforded by Article 20. Section 1(2), Recognition of Trusts Act 1987 states that: “Those provisions [of the Convention] shall, so far as applicable, have effect not only in relation to the trusts described in Articles 2 and 3 of the Convention but also in relation to any other trusts of property arising under the law of any part of the United 239

See the discussion of this Art. below. Contrast the French “créés.” The English version suggests that the Convention may be extended to trust whose existence is affirmed by the court; the French version tends to suggest that the trust must be imposed by the court. 241 Witness e.g. the debate as to which law should be applied to capacity to contract, a matter (largely) excluded from the Rome Convention; see Dicey and Morris, The Conflict of Laws, 13th edn., Rule 179, 1271–6; and see section 4 of Part One of this book. 242 As were the states of Canada where, “All the provinces that have implemented the Convention have extended it to trusts created by virtue of judicial decision, including constructive trusts and resulting trusts”: J-G Castel, Canadian Conflict of Laws, 4th edn., 545. See the legislation in Alberta (s. 1(6), International Convention Implementation Act, SA 1990); British Columbia (s. 3(1), International Trusts Act, RSBC 1996); Manitoba (s. 4(1), International Trusts Act SM 1993); New Brunswick (s. 6(1), International Trusts Act SNB 1988); Newfoundland (s. 5(1), International Trusts Ac RSN 199); Prince Edward Island (s. 3(1), International Trusts Act RSPEI 1988); Saskatchewan (s. 5(1), Trusts Convention Implementation Act SS 1994). The following subsection of each of these states’ legislation provides that trusts declared by judicial decision need nonetheless not be recognised if their is a cogent reason for refusing to give effect thereto (No such provision appears in the United Kingdom in the Recognition of Trusts Act 1987). 240

140 Types of Trust Governed by the Convention Kingdom or by virtue of a judicial decision whether in the United Kingdom or elsewhere.”

These provisions have been said to: “have the . . . advantage of ensuring that the UK courts have one set of conflicts of law rules to apply to all trusts arising under the law of any part of the UK even if they are oral trusts and irrespective of any uncertainties that might arise over the application of the Convention to some unusual trust”.243

Certainly, much of the above debate about which trusts should fall within Article 3, Hague Convention concerned the perceived difficulties which nontrust states might face with certain categories of trust. Those problems are much less pertinent when the question relates purely to the choice of law rules which an English court should apply to an international trust, or to the types of trust from other trust jurisdictions which should be recognised in England and it is accordingly perfectly sensible for the scope of the Convention to be extended in this way in England.

(B) A Choice of Law Rule for Resulting and Constructive Trusts? Insofar as the Convention has been extended to cover resulting and constructive trusts, “It means that the applicable law . . . is the law with which it is most closely connected”.244 This must be correct; and it might be thought somewhat remarkable that that the Hague Convention’s choice of law rules receive almost no discussion in the few English cases where the existence of a constructive trust in a transnational context has been at issue.245 However, where a party alleges that a third party recipient of trust property holds on constructive trust, Castel 243

D Hayton, (1987) 36 ICLQ 260, 268. J-G Castel, Canadian Conflict of Laws, 4th edn., 551; save, of course, in the rare case where an express or implied choice of law can be found (an example might be a purported express trust which contained a choice of law clause for English law, where the trust failed for want of formalities, but where equity stepped in to impose a trust to prevent a statute being used as a cloak for fraud). 245 E.g. El Ajou v. Dollar Land Holdings plc [1993] 3 All ER 717 (reversed on different grounds: [1994] 2 All ER 685); Arab Monetary Fund v. Hashim (No 9), The Times, 11 October 1994; Trustor AB v. Smallbone, judgment of 9 May 2000 (CA); Kuwait Oil Tanker v. Al Bader, [2000] 2 All ER (Comm) 271 (and see G Virgo, “Interest, Constructive Trusts and the Conflict of Laws” (2000) 8 RLR 122); Grupo Torras v. Al-Sabah [2001] Lloyd’s Rep Bank 36 (noted by J Garton, (2001) 15 Trust Law International 93). In Arab Monetary Fund v. Hashim (No 9), Chadwick J remarked that “[I]n cases involving a foreign element in which an English court is asked to treat a defendant as a constructive trustee of assets which he has acquired through misuses of his powers, the relevant questions are: (i) what is the proper law which governs the relationship between the defendant and the person for whose benefit those powers have been conferred; (ii) what, under that law, are the duties to which the defendant is subject in relation to those powers; (iii) is the nature of those duties such that they would be regarded by an English court as fiduciary duties; and (iv) if so, is it unconscionable for the defendant to retain those assets.” This curious examination of the lex causae in its context, so that the lex fori can decide whether to treat the relationship as fiduciary and whether to impose a constructive trust, makes no reference to the Hague Convention. 244

Statutory Extension of Convention Rules in the UK 141 points out246 that by Article 11(3)(d)247 the “rights and obligation of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum”; in particular, the question whether the third party is a bona fide purchaser for value without notice would be subjected to the law of the place where the property is situated (lex situs).248

(C) Trusts Arising Under the Law of any Part of the United Kingdom Unfortunately, the closer one inspects the wording of section 1(2) of the Act, the less clear one becomes as to what is included within it. Take, in the first place, the extension to “any other trusts of property arising under the law of any part of the United Kingdom.” This appears to be a genuinely all encompassing provision, applicable to all trusts, whatever their guise.249 Some types of trust would fall within its catchment on any view. This would include express trusts created in, and governed by, English law which are not evidenced in writing. It would also include trusts created in English law pursuant to a domestic statute, or to a statutory discretion.250 Equally clearly, an express trust not evidenced in writing and created by a law outside the United Kingdom would remain excluded from the Act.251 But outside these clear cut examples, when does a trust “arise under the law of any part of the United Kingdom”? Three possibilities immediately present themselves. (a) The Convention is extended to apply only to trusts which arise as a matter of the domestic law of a Part of the United Kingdom. It may be that a trust must not merely be recognised as valid, but that it must actually be governed by the domestic law of a part of the United Kingdom, in order to fall within the statutory extension of the Convention. This is certainly the most natural construction of the Act. Indeed, it is only the complexity that

246

J-G Castel, Canadian Conflict of Laws, 4th edn., 551. On which, see below (especially the discussion of “third parties and the residual role of the conflicts rules designated by the law of the forum”). 248 Macmillan Inc v. Bishopsgate Investment Trust plc (No 3) [1996] 1 WLR 387 (on which, see A Briggs, “From Complexity to Anticlimax: Restitution and Choice of Law” (1996) 4 RLR 88; J Stevens, “Restitution or Property? Priority and Title to Shares in the Conflict of Laws” (1996) 59 MLR 741; R Stevens, “The Law Applicable to Priority in Shares” (1996) 112 LQR 198; J Bird, “Choice of Law Rule of Priority Disputes in Relation to Shares” [1996] LMCLQ 57; W Swadling, “A Claim in Restitution?” [1996] LMCLQ 63; C Forsyth, “Characterisation Revisited: an Essay in the Theory and Practice of the English Conflict of Laws” (1998) 114 LQR 141). 249 Indeed, many trusts arising by judicial decision in the United Kingdom will be caught by both the first and second limbs of s. 1(2), Recognition of Trusts Act 1987. 250 Such as the trust in the case of co-ownership (on which see ss. 34 and 36, Law of Property Act 1925, as amended by the Trusts of Land and Appointment of Trustees Act 1996, schedule 2, paras. 3–4), or the trust arising on intestacy pursuant to s. 33, Administration of Estates Act 1925, as amended by the Trusts of Land and Appointment of Trustees Act 1996, schedule 2, para. 5). 251 See also Dicey and Morris, 1088. 247

142 Types of Trust Governed by the Convention may arise in determining the answer to that question252 which leads one even to consider other interpretations. Theoretically, trusts governed by English or Scottish law, though showing very marked differences, should hold no fears for the courts of either country and there is no reason not to extend the Act to them. In contrast, a trust governed e.g. by the law of Ontario would not fall within this provision, in part because any domestic rules of Ontarian law concerning the imposition, affirmation or operation of a trust are not ones which an English court would necessarily wish to seek to apply.253 There then follows a very difficult problem. For what if a trust is alleged to exist, for which not only the validity, but also the applicable law, is in issue? If the dispute is between the applicability of e.g. English or Scottish law, then there seems no problem in saying that the trust arises under the law of the United Kingdom. But what if the dispute is as between e.g. English law and that of Ontario?254 The essential problem is similar to one which is familiar in the private international law of contract: namely the chicken and egg problem of ascertaining the applicable law of an institution whose validity is also in question. In domestic law, one would ask the following questions: (1) is there a trust; and (2) if so, what are its terms? But here, it appears that it is necessary to know the terms of the trust initially, in order to know what the applicable law of that trust is. Only then can the application of the Recognition of Trusts Act 1987 to that trust be determined. This is indeed a messy problem, but it is neither unique nor insurmountable. Choice of law problems do arise as preliminary matters.255 It is not reasonable to expect a judge confronted with such an issue to form a conclusive view on the applicable law, in order to answer the question “which rules of private international law should be used to determine the applicable law?” It is, however, quite possible for him to ask, “on a balance of probabilities, is there a trust which is (a) valid and (b) governed by the law of a part of the United Kingdom?” That might also require the preliminary resolution of certain sub-issues, if the terms of the trust are themselves in issue. This approach might, in particular, require the judge to ask whether a choice of law clause allegedly agreed between the parties is valid256 and to decide by 252

As we are about to see. But, of course, trusts arising by judicial decision in Ontario would fall within the second limb of the extension in s. 1(2), Recognition of Trusts Act 1987. 254 R Stevens argues that s. 1(2) includes cases where “the only choice is between the systems of law of the constituent parts of the United Kingdom”: “Resulting Trusts in the Conflict of Laws”, in P Birks and F Rose (eds.) Restitution and Equity, vol. 1: Resulting Trusts and Equitable Compensation 147, 158. For the reasons given below, this author does not adopt this position. 255 Often at a much earlier stage when determining a court’s jurisdiction e.g. in deciding whether to permit service out of the jurisdiction, the applicable law may become relevant; as it may in deciding whether a serious issue arises on the merits, according to the applicable law. See e.g. Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc [1990] 1 QB 391. 256 This, of course, will be rare, since choice of law clauses will normally be evidenced in writing and voluntarily created express trusts evidenced in writing fall within the “main” Convention provisions under Art. 3. It might be relevant, however, e.g. in the case of a purported express trust which fails to satisfy the requisite formalities, where the agreement is given effect instead by the constructive trust. 253

Statutory Extension of Convention Rules in the UK 143 which law he should resolve that question;257 it might, in the absence of such a valid clause, require him to form a preliminary assessment of the terms of the trust, in order to ascertain its law of closest connection. What happens if an English court forms a preliminary view that a trust is governed by Scottish law and proceeds to apply the Act, only then to discover that as a matter of Scots law, such terms of the trust as are valid point rather to an overseas law as the law of closest connection? Should the court then rule that, after all, the Act does not apply? This is a most difficult question. On the one hand, if one thinks of the scenario where an English court takes jurisdiction on the basis of a choice of court clause in a contract whose validity is in question, it cannot be correct to say that the court’s jurisdiction was unwarranted, if it ultimately transpires that the contract containing the clause is void. That particular example might be distinguished by some, on the basis that the choice of court clause may stand separate to the contract and hence logically survive the contract’s invalidity; whereas here the question is of determining the law of closest connection from clauses whose existence depends solely on the trust. Nonetheless, the example indicates a point: choice of law is not an exact science and the fact that a preliminary decision is made on a matter in the course of deciding a larger question does not render that larger question imperfectly decided, if the finding on a preliminary issue turns out ultimately to be incorrect. On the other hand, it remains the case that a constructive trust governed by the law of an overseas state does not appear to be governed by this limb of section 1(2) of the Act.258 The finding that the Act applies can itself only be preliminary until the applicable law is known; but the finding as to the applicable law can only be preliminary until the scope of the Act is known. They are two sides of the same coin and it is not obvious that the answer to either should depend on which side the coin lands. How then might the scope of the Act be determined? In theory, the following series of questions is suggested:259 a. According to the law of the forum, does there appear, on a balance of probabilities, to be a valid trust?260 b. If so, does it appear, on a balance of probabilities, to be governed by the law of a part of the United Kingdom?261 In answering that question, it may be 257 On which see the discussion of “consent to the choice of law clause” in connection with Art. 6, below. 258 And, insofar as the trust does not arise by judicial decision of a foreign court, it will not fall within the second limb of s. 1(2), Recognition of Trusts Act 1987 either. 259 Of course, in practice, a more impressionistic approach may be more likely to be taken by a court. 260 This is a relatively high initial threshold; but where one party alleges the existence of a trust and the other denies it, it is dictated by the concern of acting even-handedly as between them. 261 Of course, a question arises as to whether this preliminary assessment of the applicable law should be made applying an English court’s common law rules or the Act. Since it is not yet established that the Act is putatively applicable, it is suggested that the common law rules should be applied. This might seem rather impractical. However, it must be reiterated this is a preliminary exercise and only an assessment of the prima facie applicable law is needed.

144 Types of Trust Governed by the Convention necessary to ask: “according to the law of the forum what, on a balance of probabilities, appear to be the terms of that trust?” If the trust does not appear to be governed by the law of a part of the United Kingdom, the Act does not apply; if it does, the Act is putatively applicable and, in order conclusively to determine the matter, it should be asked; c. Applying the choice of law rules contained in the Act, does it still appear that the trust is governed by the law of a part of the United Kingdom? If so, then the Act clearly applies; if not, the Act is inapplicable and the common law rules on the validity of the trust should be applied. (b) The Convention is extended to apply to any trust which arises as a matter of private international law of a Part of the United Kingdom. Under this approach the word “law” in section 1(2), Recognition of Trusts Act 1987 would be interpreted to include any trust whose existence and validity would be upheld by an English court, applying its choice of law rules. Once again, carts and horses would be displaced: it would be tantamount to saying that a trust which is valid by English conflicts rules is governed by the Act. It would also lead to such a broad interpretation of the scope of the 1987 Act that almost anything would fall within the parameters of the phrase “arising under the law of any part of the United Kingdom”, thereby making it unclear what further purpose is served by the phrase “or by virtue of a judicial decision whether in the United Kingdom or elsewhere.” (c) The Convention is extended to apply to any trust which is imposed or affirmed in the United Kingdom. This interpretation would include all trusts arising in the United Kingdom, irrespective of their applicable law. It would, however, exclude trusts created abroad, irrespective of their applicable law. This is an interpretation which should be rejected. The natural wording of section 1(2) does not relate to the place of creation or affirmation; and the difficulties involved in attributing a place to where e.g. a constructive trust arises are potentially enormous.262 Moreover, the point of the Convention is to lay down choice of law rules; the place of creation is not, and should not be treated as, a relevant matter. On balance, it seems that interpretation (a) must be correct and that only trusts governed by the domestic law of a part of the United Kingdom fall within the Convention. That only creates a fresh set of problems as to how to tell if this is the case.

262 Compare, in the context of the Brussels Convention, Casio Computer Co Ltd v. Sayo, judgment of 20 December 2000, The Times, 6 February 2001; Casio Computer Co Ltd v. Kaiser, judgment of 11 April 2001; [2001] EWCA Civ 661; Dexter Ltd v. Harley, judgment of 8 March 2001, The Times, 2 April 2001 (Lloyd J). These cases are discussed by T Yeo, “Constructive Trustees and the Brussels Convention” (2001) 117 LQR 560.

Statutory Extension of Convention Rules in the UK 145

(D) Trusts “Arising” by Judicial Decision (i) A source of complexity When one then turns to trusts arising by judicial decision “whether in the United Kingdom or elsewhere”, the questions that arise are equally complex. It is important to notice that, this time, no reference is made to the law under which such a trust arises. Furthermore, the place where the trust arises is expressly stated to be immaterial.263 (ii) Constructive trusts The authors of Lewin on Trusts take the view that this extension catches almost all constructive trusts. They argue that where a constructive trust is imposed by a foreign law, “the judgment will serve as written evidence of the trust within Article 3 of the Convention and . . . [that] the trust arises by virtue of the decision within section 1(2) of the Act”.264 With respect, this writer would beg to disagree. The trust will not necessarily satisfy Article 3, since (i) it may not be a trust “voluntarily” created by a settlor; and (ii) it is very dubious whether the judgment itself constitutes sufficient written evidence of the trust. In any event, it is unnecessary for a trust caught by section 1(2), Recognition of Trusts Act 1987 to satisfy the formal requirements laid down in Article 3.265 But it is not clear that all constructive trusts fit within section 1(2) either. If a court judgment affirms (retrospectively) the existence of a constructive trust, it does not obviously make the trust “arise” within the meaning of section 1(2). Accordingly, the matter merits closer examination. How then does one tell if a trust arises by judicial decision? Presumably, the “statutory jurisdiction trusts”266 would be covered by this provision. Hence, if an English (or foreign) court were to utilise its jurisdiction to impose a trust as part of a settlement pursuant to a divorce order, such a trust would fall within the Act. Indeed, it seems that if the court has the discretion to impose a trust in a given factual scenario, rather than the duty to do so, the trust can be said to arise by judicial decision. When one passes beyond that point to trusts declared to exist by a court, matters turn decidedly more grey. Article 20 speaks of extending the Convention to trusts “declared” by judicial decision and it is far from clear that the word “arising” means the same thing as “declared”: arguably “one . . . seems to refer to dispositive judgments, while the other refers to declaratory judgments”.267 263

As is the applicable law of the trust. Lewin on Trusts, 17th edn., 288. 265 See the discussion of “constructive trusts” above. 266 To adopt D Hayton’s terminology. These are “trusts created pursuant to a court order made under a statutory jurisdiction and dealing with unique factual circumstances . . .”: (1987) 36 ICLQ 260, 266. 267 M Lupoi, Trusts: A Comparative Study, 343. The remark is in fact made in contrasting the French word “créés” with the English “declared” in the two versions of Art. 20. 264

146 Types of Trust Governed by the Convention Does a trust whose existence is in dispute “arise” when the court rules that it does exist? If X alleges that he contributed substantially to the upkeep of Blueacre on the basis of an understanding with Y that he would obtain a beneficial share in the property (which was conveyed into Y’s name solely), and the parties dispute whether X has an interest by virtue of a constructive trust, is the court’s finding that the trust does exist within the Act? In England, it appears that the (institutional) constructive trust ordinarily takes effect from the time when the circumstances justifying its existence arise and not when the court rules upon the matter. So much is said by BrowneWilkinson J in Re Sharpe (a Bankrupt)268 and is “generally accepted . . . in the absence of a court order to the contrary”.269 Matthews observes that: “The institutional constructive trust is that which is imposed if the stated truth conditions are satisfied. It arises at that moment, rather than when later declared by the court to exist”.270

However, it is logically possible that the constructive trust may take effect from a date prior to a judgment and yet still requires a court order before it can take effect. Its operation would simply be retrospective. As Parker and Mellows point out, neither Goulding J in Chase Manhattan Back v. Israel (British) Bank271 nor the Supreme Court of Canada in Rawluk v. Rawluk272 thought a court order necessary to bring a constructive trust into effect. However, they point out that the former case was largely concerned with the position under New York law and that the latter case did not concern a traditional institutional trust. They argue that a constructive trust should require a court order, not least because otherwise “the beneficiary may find himself liable to income tax and capital gains tax on an interest arising under a constructive trust which he has never claimed or enjoyed”.273 On the other hand, it would be unreasonable to insist upon a court order where both parties agree that an interest by constructive trust has arisen. Whether, for private international law purposes, institutional constructive trusts can be said to “arise” by judicial decision depends upon how purposive an interpretation of the Act is taken. If they do not, very few trusts would be caught by this provision. Statutory jurisdiction trusts and the remedial constructive trust would be ripe for inclusion, on the basis that they clearly require a court decision to impose the trust. But why should it be desired to enact a section which extends the scope of the Convention beyond the realm of express trusts evidenced in writing to the other extreme of trusts arising at the court’s discretion, whilst missing 268 [1980]1 WLR 219, 225; see also A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., 269. But see also Bannister v. Bannister [1948] 2 All ER 133. 269 Parker and Mellows, ibid., 269. 270 P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7, 8. Emphasis added. 271 [1981] Ch 105 272 (1990) 65 DLR (4th) 161. 273 Parker and Mellows, 270.

Statutory Extension of Convention Rules in the UK 147 out constructive and resulting trusts arising by operation of law?274 After all, if the limitation to the scope of the Convention was based on the desire not to expose courts to the uncertainties that lurk beyond clearly evidenced express trusts, what could be more uncertain for an English court than to have to ask itself whether, e.g. a Canadian judge, applying his local law might, in his discretion, impose a trust on the facts? Nevertheless, the position appears to be sufficiently spelt out in the Act that trusts whose existence is merely reaffirmed are excluded. Dicey and Morris, in the case of constructive trusts, seem to regard this as so self evident that they proceed immediately to discuss which choice of law rules of the common law might apply to them, given that “there seems to be no English or Commonwealth authority” on the matter and conclude that it “seems best”275 to apply the choice of law rules applicable to claims for restitution and to use the proper law of the obligation.276 Technically, that seems incorrect in relation to the remedial constructive trust (and also in relation to the few institutional constructive trusts which fall within Article 3, Hague Convention).277 In contrast, Barnard takes the view that constructive trusts are included within the Act. Unfortunately, she offers no real reason for saying so. Moreover, she states that “s. 1(2) provides that the Convention applies to both express and constructive trust”.278 With respect, section 1(2) says no such thing. And whilst it might seem overly technical to ask whether a particular trust arises by judicial decision or is merely reaffirmed by such decision, the fact remains that it would not have been at all difficult for the wording of section 1(2) to have been extended to trusts “declared”279 or “arising or affirmed” by judicial decision. The institutional constructive trust, which takes effect prior to a court order, and in respect of which it cannot be said unequivocally in domestic law that a court order is required, cannot comfortably be said to “arise” by judicial decision. On balance, it is suggested that such trusts should be regarded as excluded from the Act.280

274 This is not quite true, since trusts arising under the law of a part of the United Kingdom will fall within the Act regardless of their nature (covered as they are by the first limb of s. 1(2), Recognition of Trusts Act 1987); but constructive trusts governed by an overseas law would be excluded. 275 Dicey and Morris, 1096. See also R Stevens in F Rose (ed.) Restitution and the Conflict of Laws (Oxford, Mansfield Press, 1995), ch. 5, 216–8; P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7. 276 Ibid., as detailed in Rule 200, 1485. 277 See the discussion of “constructive trusts” above. 278 L. Barnard, “Choice of Law in Equitable Wrongs: a Comparative Analysis” [1992] CLJ 474, 479. 279 The word used in Art. 20. 280 It is curious that P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7, having earlier (rightly) pointed out (at 8) that the institutional constructive trust arises at the time when the necessary conditions are met and not when declared to exist by a court later concludes (at 12) without explanation that the statutory extension in the United Kingdom extends to cover such trusts. It is suggested that this latter view does not accord with the wording of s. 1(2), Recognition of Trusts Act 1987.

148 Types of Trust Governed by the Convention (iii) Presumed resulting trusts It will follow also from the above argument that presumed resulting trusts will not be covered by the extension. The presumption takes effect from the time that the transferor transfers property in circumstances where he does not dispose of his equitable interest in the property transferred.281 True, the presumption can be rebutted and a court’s intervention may be needed to declare that this is the case. Nevertheless, the fact that evidence of the trust’s existence may later need to be considered by a court does not alter the fact that the trust itself arises from the transferor’s intentions282 at the time that the property was transferred. The position is likewise where a presumption of advancement operates283 and the transferor successfully persuades the court that the presumption was rebutted. Accordingly, it is suggested that presumed resulting trusts are not covered by the statutory extension in the United Kingdom.284

(E) Judgments Entitled to Recognition On one view, the extension to judicial trusts arising anywhere is extraordinary. It might be taken to mean that the trust must be recognised even if, according to English rules on recognition of foreign judgments, the judgment does not qualify for recognition. A trust arising by judicial decision in a state outside Europe, in which the defendant is neither present nor resident, and to which he has not submitted, might require recognition in England. Nowhere does the Act contradict such a view; indeed it seems on its wording to encourage it. But the implication would be remarkable: namely that as long as a judgment of a state falls within section 1(2), Recognition of Trusts Act 1987, it would fall to be recognised, even though an English court might otherwise regard the foreign court as jurisdictionally incompetent, and even though there may be no guarantee of procedural propriety in the courts of that foreign state. Of course, it might be possible to state that such a judgment infringed English public policy; but the inherently negative process that is entailed therein is both unattractive and inappropriate to be used routinely. Stevens argues that the Convention does not affect recognition of trust judgments in England, since it “. . . is only concerned with the recognition of the institution of the trust by those systems of law which do not possess it. . .”285 With respect, this is not correct. The “shapeless trust” of the Convention may well lead 281 See R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn., ch. 8 (esp. 227–49). 282 Or lack of intention to give absolutely. 283 R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 238–48. 284 Unless, of course, they arise under the law of a part of the United Kingdom. 285 R Stevens, “Resulting Trusts in the Conflict of Laws”, in P Birks and F Rose (eds.) Restitution and Equity, vol. 1: Resulting Trusts and Equitable Compensation 147, 158.

Statutory Extension of Convention Rules in the UK 149 to calls upon the English courts to recognise civil law “analogues”.286 However, Stevens goes on to argue that “[the Convention] does not affect the English courts’ recognition of foreign judgments”.287 This latter statement is, it is suggested, correct.288 The Convention is about choice of law rules and the impact of trusts conforming to those rules; it is not about the jurisdiction of any courts to rule on trusts matters and says nothing about when an English court should regard a foreign court as jurisdictionally competent in trust matters. Section 1(2), Recognition of Trusts Act 1987 extends the scope of application of the Hague Convention, but it does not otherwise alter this position. Accordingly, it is suggested that any judgments creating trusts by judicial decision cannot be recognised in England if they do not satisfy the English rules on recognition of foreign judgments. Of course, if they do satisfy such rules, they would require recognition in any event, even without the extension in section 1(2) of the Act, so that the provision adds very little to the pre-existing position on recognition. It may, however, be relevant where an English court is itself asked e.g. to impose a (remedial) constructive trust using the choice of law rules contained in the Convention. (F) Conclusion In conclusion, it is suggested that Article 3 of the Convention should be construed as applying to express trusts evidenced in writing and (because the intention of the drafters to do so is apparent) to automatic resulting trusts. It should also apply to a few, but not to most, institutional constructive trusts. Section 1(2), Recognition of Trusts Act 1987 should be regarded as extending the Convention’s provisions within the United Kingdom to all trusts which appear, on a balance of probabilities, to have been validly created under the domestic law of a part of the United Kingdom, whatever their nature and also to trusts which an English court is asked to impose by virtue of a judicial decision anywhere, provided that the court’s judgment also satisfies the English rules on recognition of foreign judgments. But section 1(2) does not cover express trusts not evidenced in writing which arise by the law of an overseas state. Nor does it cover the institutional constructive trust arising under a foreign law, whether arising by common intention of the parties289 or otherwise, or the presumed resulting trust.290 Trusts arising by operation of statute overseas would also be 286

M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15. R Stevens, “Resulting Trusts in the Conflict of Laws” 147, 158. J-G Castel, Canadian Conflict of Laws, 4th edn., 551 states that “the recognition and the enforcement of . . . trusts of property arising by judicial decision have nothing to do with conflict of laws rules covered by the Convention”; see also M Koppenol-Laforce, Het Haagse Trustverdrag 266, who argues that the normal rules on recognition of foreign judgments must still be satisfied. 289 M Lupoi would approach the matter slightly differently here (although he is discussing Art. 20 and not s. 1(2) Recognition of Trusts Act 1987): see Trusts: A Comparative Study, 344–5. 290 In Lightning and Anr v. Lightning Electrical Contractors Ltd, judgment of 23 April 1998, Peter Gibson LJ took the view that the same result would be reached whether the presumed resulting trust in issue fell within the Recognition of Trusts Act 1987 or was governed by the common law. 287 288

150 Types of Trust Governed by the Convention excluded, although trusts created by the exercise of a court’s jurisdiction to create a trust (whether in the United Kingdom or overseas) should be included, as should the remedial constructive trust. Although this constitutes a relatively narrow construction of the Convention291 and of the Act, it is apparent that, when one moves beyond the context of the express trust satisfying certain requisite formalities, the reasons for the imposition of a presumed resulting trust or a constructive trust are often to be found in justifications other than protecting party autonomy and often rooted in policies of a particular state less universally accepted than the autonomy principle. That being so, it is less than obvious that a harmonised scheme of rules should be applied to them, or that the conditions for the recognition of such trusts should be regulated by the Convention rules.

291 This author agrees with M Lupoi that “the Convention should be interpreted according to the errors which it contains” (Trusts: A Comparative Study, 344).

ARTICLE 4—EXCLUSION OF PRELIMINARY MATTERS; THE ROCKET-LAUNCHER AND THE ROCKET The Convention does not apply to preliminary issues relating to the validity of wills or of other acts by virtue of which assets are transferred to the trustee.

1. PRELIMINARY MATTERS EXCLUDED FROM THE CONVENTION

The Convention is applicable only to matters which concern the validity of the trust itself. But, of course, for a trust to be brought into existence, it must be validly declared and constituted. When contained in a written instrument, that written instrument must itself be valid, both as to form and substance. When involving the transfer of property into the hands of the trustee, the property must have been validly transferred. When the type of trust in question can be created only by certain types of person, then the settlor must have the capacity to create the trust. These issues are all preliminary ones. Without them, a valid trust may not come into existence, but they do not relate to the trust itself. Article 4 deems them to be outside the parameters of the Hague Convention. “The image employed was that of the launcher and the rocket; it will always be necessary to have a ‘launcher’, for example a will, a gift or another act with legal effects, which then launches the ‘rocket’ the trust. The preliminary act with legal effects, the ‘launcher’, does not fall under the Convention’s coverage”.292

The “rocket launching” aspects of a trust are of critical importance and, in some cases, themselves give rise to very complex and unsettled choice of law problems. Detailed consideration is devoted to them in Part One of this book and only some basic comments about the inter-relationship of the “rocket” and “rocket-launcher” are provided here.293 Von Overbeck explains that questions of a formal or substantive nature as to the transfer of assets to the trustee are excluded and that this includes the act by which a settlor declares himself, and turns himself into, a trustee of the property.294 Where the transfer of property is ineffective by its governing law, the 292

Von Overbeck Report, para. 53, p. 381. See also J Harris, “Launching the Rocket: Capacity and the Creation of Inter Vivos Transnational Trusts”, in J Glasson (ed.) International Trust Laws, ch. C2. 294 Von Overbeck, paras 54–57, pp. 381–2. He explains that a proposal to make express the fact that the phrase “assets are transferred to the trustee” was to include declarations by a settlor of himself as trustee was rejected, perhaps largely due to the cumbersome words proposed to convey this idea. However, M Lupoi argues (and this author would support his view) that the declaration by a settlor of himself as trustee does fall within the Convention (Trusts: A Comparative Study, 335). The better view is that the trust must still be launched by the settlor successfully converting himself into a trustee; but that once launched it then falls within the Convention. See the discussion of “the general characteristics” of Art. 2 above; and the discussion of the declaration of trust in Part One of this book. 293

152 Exclusion of Preliminary Matters fact that it was valid by the law putatively applicable to the trust is irrelevant.295 On the matter of capacity, von Overbeck explains a division that appeared between civil lawyers who regarded capacity as a preliminary matter ripe for exclusion and “common law countries which tend to subject capacity to the law governing the validity of the act which is in question”.296 One can only remark at this stage that the reluctance of an English court to apply the putatively applicable law of a contract297 to a party’s capacity to contract makes one strongly doubt whether it would have been appropriate to have included the settlor’s capacity within the Convention scheme and to have allowed the law chosen by the settlor to govern the trust also to determine his capacity. In any event, the compromise reached was that the capacity of the trustee would be included in the Convention,298 whilst other issues of capacity would be excluded.299 The rocket-launcher/ rocket dichotomy is an ingenious way of protecting the “purity” of the constituent elements of trusts (in recognising that they consist both of an alienation of property and the creation of a trust), whilst at the same time not scaring non-trust jurisdictions with the prospect that the trust might turn from an ingenious property management device into a device for evasion and even fraud. “Thus, the problem presented was to enable the uniform choice of law rules for trusts to coexist with the conflicts rules of the forum”.300 Despite this, Schoenblum criticises the exclusions in Article 4. He argues that a transfer of assets on trust situated in a civil law state may fail, on the basis that the trust institution is not known therein.301 But it is suggested that this view is mistaken. The “rocket-launching” issue for the lex situs is whether the settlor may dispose of his property at all, not the more specific question whether he may create the trust structure, which should be regarded as a matter for the applicable law of the trust.302 Schoenblum also argues that Article 4 means that a settlor in a civilian state may fail to create a trust on the basis that the law applicable to succession (lex successionis), which will frequently be the law of his domicile, may not recognise the validity of the transfer. This will frustrate attempts by civilians to create trusts. But the Convention does not set out positively and directly to encourage the use of trusts by domiciliaries in non-trust states, a step which may have 295 But see Municipality of Canterbury v. Wyburn [1895] AC 89, 96; see also the example given by P Nygh, Conflict of Laws in Australia, 6th edn., 513–4. Despite the intimation of Nygh, the law putatively applicable to a trust which has not ex hypothesi been launched should not necessarily be used to determine the effects of the trust not having being validly created. 296 Von Overbeck Report, para. 59, p. 382. 297 On which, see further Part One of this book. 298 Art. 8(2)(a). 299 On the capacity of a beneficiary, see Lewin on Trusts, 17th edn., 305–6. 300 E Gaillard and D Trautman, “Trusts in Non-Trust Countries: Conflict of Laws and the Hague Convention on Trusts” (1987) 35 Am J Comp Law 307, 329. 301 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 6; J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations (Oxford, Clarendon Press, 1992) ch. 11, p. 245. 302 See Part One of this book.

Distinguishing between the “Rocket” and “Rocket-Launcher” 153 proved so radical as to have scuppered prospects of ratification in those states. Again, the view seems mistaken, since the lex successionis need only recognise that the settlor may dispose of his assets, not that he may create the trust structure with them. It is extremely hard to see why a settlor who is not free by his lex successionis to alienate property outside his immediate family (by any means, be it by gift, transfer on trust or other method) should be permitted to do so by choosing a law applicable to the trust under which he may do so. If Schoenblum’s concern is, at it appears to be, that offshore jurisdictions are much more prosettlor autonomy and that the Convention will make trust states which adopt it uncompetitive in the trusts marketplace relative to those offshore states, it must be seriously questioned whether an international Convention designed to facilitate transnational co-operation can be seen to adopt rules which are almost a recipe for conflict with those of states having forced heirship rules.303 Without the “rocket-launcher” exclusions, one might suspect that the Convention’s prospects of ratification outside trust states would be substantially diminished, effectively turning it into a choice of law package for existing trust states.

2. DISTINGUISHING BETWEEN THE “ ROCKET ” AND THE “ ROCKET - LAUNCHER ”

Schoenblum also contends that “the simplistic analogy to the launcher and the rocket . . . explodes upon closer scrutiny”.304 Certainly, it must be conceded that the distinction between “rocket-launcher” and “rocket” will prove very difficult to make in practice. Rules restricting the disposition of property may be regarded as affecting the “preliminary” question of the ability to constitute a trust by transferring the property to trustees at all, or as relating to the ability to create the trust structure and hence “rocket” issues to be determined by the Convention. Moreover, a trust which is ostensibly valid by its applicable law may nonetheless fall foul of other “preliminary” matters in respect of which the forum would apply different choice of law rules. For matters appertaining to the trust itself, the forum is free to apply its mandatory rules pursuant to Article 16;305 but in so far as a matter is treated as preliminary and outside the scope of 303 Moreover, many offshore states are bound by the Hague Convention, yet still have highly pragmatic “rocket-launching” rules; see, in particular, the discussion of offshore states in Part One of this book. 304 J Schoenblum, (1994) 3 J Int Corp P 5, 12. But the example that he gives does not illustrate his point. He considers the removal of a testamentary trustee and the duty to account, which are treated by Art. 8 as trusts matters, even though they would be considered to relate to probate in the United States. Of course, different states will have differing views as to what is part of the preliminary process of trust creation; but the Convention makes it abundantly clear that these two matters are henceforth to be considered by contracting states as “rocket” matters. Much more complex are cases (not considered by Schoenblum) where the Convention is silent and the characterisation might differ from state to state (e.g. forced heirship). 305 And the domestic mandatory rules of the law designated by the forum’s choice of law rules in related areas of law pursuant to Art. 15.

154 Exclusion of Preliminary Matters the Convention, the choice of law rules designated to that issue will apply in toto, mandatory or not, and may prevent the trust from receiving its fuel. Hayton gives the example of the rules against perpetuities,306 which is especially difficult to classify and may or may not be regarded as a “rocketlaunching” matter.307 He imagines a testator domiciled in England making a will “in English form” creating a trust of movable property in Manitoba for the present and future issue of his brother, who lives in Manitoba, pursuant to a trust governed by Manitoban law. With respect to the perpetuity issues raised by the trust, he favours a “rocket” classification, since “English law has no strong policy interest in denying the application of Manitoban law [to the perpetuities rule]”.308 With due respect, such an “interest analysis” technique for classifying an issue as being, or not being, a preliminary matter has little pedigree in the English conflict of laws. Moreover, by choosing a rather extreme example of a trust where the assets and beneficiaries are all located overseas and the applicable law is the law of that overseas state, he is able to reach a conclusion which seems relatively uncontroversial. But a perpetuity rule must, initially, be classified either as a preliminary matter or as a matter relating to the trust itself. To a considerable extent, that depends on the particular rule of a given legal system which it is claimed applies on the facts. Is it a rule designed to limit the rights of a settlor to alienate property at all in a certain manner for an indefinite period; or is it a rule designed specifically to limit the duration of a trust structure? That is a question of construction of the legislation of the state in question, in order to ascertain its nature. The former type of rule is a “rocketlauncher”; the latter a “rocket” matter. It is accordingly simply not possible to say, in the abstract, that a rule against perpetuities is or is not part of the “rocket” itself. Such a rule may in state A be of a quite different nature to a perpetuities rule in state B. Whether that legislation is then designed to apply to a particular factual scenario is a further stage of construction of the legislation and its intended scope, once it is decided that that law is, in principle, relevant. In England, section 164(1), Law of Property Act 1925 provides that “No person may by any instrument or otherwise settle or dispose of any property in such manner that the income thereof shall” exceed the specified accumulation period.309 This is a provision which appears to affect property dispositions in general, and not trust dispositions in particular and ought arguably to be subjected to the law which governs such transfers.310 Nevertheless, it is suggested that both the 306

See section 8 of Part One of this book. J-G Castel, Canadian Conflict of Laws, 4th edn., 543 points out that, at common law, the law applicable to the trust determined whether it infringed the rule against perpetuities, citing Freke v. Lord Carbery (1873) LR 16 Eq 461. 308 D Hayton, (1987) 36 ICLQ 260, 269. P Matthews, Trusts: Migration and Change of Proper Law (London, Key Haven, 1997), 55 likewise favours a “rocket” classification. Elsewhere, Hayton calls for legislative clarification of whether perpetuities appertain to the “rocket-launcher” or to the “rocket”: “Developing the Law of Trusts for the Twenty-First Century” (1990) 106 LQR 87, 94, f/n 14. 309 Emphasis added. See also the Perpetuities and Accumulations Act 1964. 310 Transfers of title to land will be governed by the lex situs, whether inter vivos or testamentary: see Part One of this book. 307

Distinguishing between the “Rocket” and “Rocket-Launcher” 155 English rules against perpetuities and accumulations should be treated as “rocket” matters and applicable when (but only when) English law is the law governing the trust.311 Although perpetuity rules are not historically restricted to the trusts context, as Cheshire and Burn note, “. . . the problem of future interests, apart from questions of construction and of tax liability, essentially resolve themselves into the question of the application of the perpetuity rule to equitable interests under trusts”.312 They are, in practice, predominantly trust focused.313 Gaillard and Trautman take the issue of forced heirship rules as another difficult case. It is necessary “to determine whether the forced share interest of the spouse of a settlor limits the conveyance attempting to set up the trust or limits the trust itself”.314 If the objection is to the disposition of one’s property outside the family, then the rule looks like a question of the ability to transfer assets and thus a “rocket-launcher” question. But again, everything depends on the legislation in question and the limits which it imposes.315 The uncertainty is evident in the discussion in Lewin on Trusts, which initially appears to favour a “rocket-launcher” classification. In their discussion of preliminary acts to vest immovable property on trust, the authors state that “The law applied under this process decides, for instance, whether the testator can leave the asset out of the family or whether some members have forced heirship rights . . .”316 However, in their later discussion of Article 15(1)(c), which (for matters falling within the scope of the Convention) preserves the application of the mandatory rules of the lex successionis, they comment that “Forced heirship rules are in this category . . . because a settlor or donor cannot infringe them”.317 Yet if forced heirship rules fall outside the scope of the Convention as affecting the validity of the transfer itself, how can the mandatory rule provisions inside the Convention be applied to the trust? 311 This view is supported by the decision of the Supreme Court of New South Wales in Saliba v. Falzon [1998] NSWSC 302, where a testamentary disposition infringed the rule against accumulations of English law, but not New South Wales law and where the court felt bound to consider the law applicable to the trust to determine whether the accumulation period was infringed. The case is discussed more fully in the consideration of “common law authorities on implied choice of law” in respect of Art. 6, below. 312 E Burn, Cheshire and Burn’s Real Property, 16th edn. (London, Butterworths, 2000), 309. See also Re Fitzgerald [1904] 1 Ch 573. 313 See also Law Commission Report No 251 on the Rules against Perpetuities and Excessive Accumulations (London, HMSO, 1998). The Report proposes that the rule against perpetuities should be restricted to interests arising under wills and trusts. It proposes a single fixed perpetuity period of 125 years. 314 E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 329. P Matthews, Trusts: Migration and Change of Proper Law, 55 treats forced heirship as a “rocket-launching” matter. See also D Hayton, “Trusts and Forced Heirship Problems”, (1993) 2 J Int Corp P 3 (also published in (1993–4) 4 KCLJ 12). 315 See section 17 of Part One of this book. 316 Lewin on Trusts, 17th edn., 278–9. 317 Ibid., 296. R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7(5) Trusts and Trustees 24, 27 appears to treat forced heirship as a “rocket-launching” matter, but then goes on to say (at 28) that one reason for changing the governing law of a trust might be to avoid forced heirship claims (which would make no difference if forced heirship is, ex hypothesi, not a “rocket” matter).

156 Exclusion of Preliminary Matters At this point, a problem of circularity potentially arises, due to the need to characterise the legislation before the court. For it cannot be known which legislation to have regard to until one knows the law applicable to a certain issue. In order to ascertain that law, it is necessary to characterise the issue at hand, be it as an issue of trusts or of another category. Yet, until one knows which legislation is applicable and is able to construe it, it may be impossible to characterise that legislation as regulating a trusts matter (the “rocket”) or as restricting the right to alienate property (the “rocket-launcher”).318 It follows that the court needs to form a preliminary view as to whether e.g. forced heirship rules or rules against perpetuities are trusts matters or not. In order to circumvent such problems, the court needs to approach the various issues separately. If it asks first “according to the law identified by our choice of law rules for testamentary dispositions, are there any rules which restrict a party’s right to alienate his property as he/ she chooses?”, then reference is immediately made to a single law for its answer to that question. The fact that that law might also have e.g. perpetuity rules relating to the trust itself is not relevant, for that law is not necessarily the law applicable to trust issues properly so called. Then the court may proceed to ask “if the transfer of property is prima facie valid, are there any restrictions on the ability of a party to create the trust structure with his property, according to the law applicable to the trust”? In other words, the question is not “are forced heirship issues to be characterised as trust issues” but “what restrictions, according to the applicable law of the trust, are there on a party’s ability to create a trust structure?” Once a court has identified the law applicable to a particular issue, it then becomes a matter of construction of that legislation to determine how much, if any, of its law to apply. For example, if it is determined that the applicable law of a trust is Utopian, the Utopian legislation must be construed, in order to determine if it is, properly interpreted, a restriction on the creation of the trust structure. In principle, that question, which concerns the process of characterisation, is one to be answered by the law of the forum. But it is almost certain to lead to “limping” trusts and distortion of that foreign law if no attempt is made to discover whether a Utopian judge would consider the Utopian law to be a trusts rule, as opposed to a general rule of property law or succession. In other words, once a preliminary classification of the issue at hand has been made by the law of the forum, the law identified ought to be considered in its context. Otherwise, it is likely to be applied when it was not intended to be, or vice versa. It is thus quite unhelpful to say that e.g. forced heirship rules are or are not preliminary issues; everything depends on what the forced heirship rule of the law identified actually provides. In conclusion, it is possible to say that certain issues are clearly part of the “rocket-launching” process and outside the Convention. These include the 318 It might be that two legal rules of different systems have strong claims to apply on the facts and that, until the issue at hand is characterised, it cannot be determined to which rule one should have regard: see e.g. Re Cohn [1945] Ch 5.

Distinguishing between the “Rocket” and “Rocket-Launcher” 157 transfer of property to trustees, the validity of a will as to form and substance and the capacity of the settlor to create the trust. But it is not possible to say that other, more specific legal rules, such as rules on forced heirship or rules against perpetuities, are or are not outside the Convention. Insofar as they relate to the right to transfer property at all, they are outside; in so far as they relate to the creation of the trust structure, they are inside. That is a question of interpretation of the legislation, to be carried out against the backdrop of the legal system in which it is to be found. Only then can one finally ask whether that legislation is intended to apply to a certain factual situation.

ARTICLE 5—NON-TRUST JURISDICTIONS The Convention does not apply to the extent that the law specified by Chapter II does not provide for trusts or the category of trusts involved. Article 6(2) states that where a law has been chosen to govern a trust and that law: “does not provide for trusts or the category of trusts involved, the choice shall not be effective and the law . . . [applicable in the absence of choice] . . . shall apply.”

1. APPLICATION

The combined effect of these provisions is that if the settlor chooses a law which does not have the trust concept in its domestic legal system, that choice shall be disregarded and the law applicable in the absence of choice shall be ascertained under Article 7. If application of Article 7 points to the application of the law of a non-trust state, the Convention’s rules 319 shall not, after all, be applicable. In other words, one needs to ascertain the governing law under the Convention, in order then to ascertain whether the Convention actually applies. Moreover, if the country whose local law is applicable knows the trust concept, but not the particular category of trusts in issue, the Convention will also be inapplicable. Von Overbeck rightly points out that problems with this Article are likely to be rare.320 Even a settlor determined to evade the clutches of a law of closest connection to a trust is unlikely to enter into the uncertain terrain of choosing the law of a non-trust state. That must be true. But one might wonder how often it will be that a law will be found not to “provide for trusts.” Much hinges on whether this means that the law designated does not have the trust in its domestic law, or whether it does not have the trust at all, even for the purposes of private international law. If the law of Civilopia, a non-contracting state to the Convention is chosen, then, in all probability, it will have neither a domestic law of trusts nor a set of choice of law rules to deal with the trust. But if Civilopia is a contacting state, then it may have no domestic law of trusts, but it will have a set of choice of law rules for the trust. In that sense, the law does “provide for trusts”, because it has a private international law of trusts. However, it might be retorted that it would be absurd and unworkable to say that, provided a trust is recognised in the private international law of a state, it is outside Article 5. The Convention identifies the domestic law of a particular state as applicable to a trust. It then becomes necessary to apply that system’s domestic law and it is, of course, impossible to do that if the system identified has no trust laws. 319 320

Including, paradoxically, Art. 7. Von Overbeck Report, para. 61, p. 383.

Application 159 So the trust must exist in the domestic system identified. But that begs a secondary question, which is: “in the context of Article 5, what does ‘trust’ mean”? Presumably, it means what “trust” means throughout the Convention. If it was intended to mean anything different, one would have expected a statement to such effect. In which case, presumably, a trust is a legal relationship showing the characteristics identified in Article 2. Because that is a “shapeless” trust 321 and because it covers a broad range of “structurally analogous” institutions, then, if Article 2 is taken at face value,322 it is possible to imagine that the number of “non-trust” states, in the sense of countries which do not have the common law trust or a homologue or structural analogue thereto, will be small. Of course, it will be far from ideal for the courts of Civilopia to find that their law has been selected to apply to what is, in all other respects, a common law trust. It simply will not have domestic rules of its legal system to determine the rights and duties of the parties and the interests created. True, it has to confront the importing of the common law trust into its jurisdiction when required to recognise trusts pursuant to Article 11. Indeed, the courts of many civil law jurisdictions demonstrate great ingenuity in fitting the trust into their domestic law concepts, when called upon to recognise an Anglo-American trust.323 But that is quite different to finding that such a trust is actually governed by Civilopian domestic law. Moreover, translation324 of the common law trust into the nearest analogue would be complicated by the fact that many aspects of the trust are spelt out in the Convention, in particular in Article 11, and must be demonstrated even if the governing law of that institution is Civilopian law. Nevertheless, the fact remains that a state cannot be regarded as not knowing the trust simply because it does not know the classic English trust. A settlor who chooses the law of Civilopia to govern a settlement may expect it to have the effects specified in Article 11, and may expect that the issues set out in Article 8 will be determined by the applicable law. Beyond that, he must accept that the “trust” which he creates, when transplanted into the legal system which he has chosen, may look little like a common law trust and may have unexpected consequences.325 What if a choice is made of the law of a state which genuinely does not know the trust or a structural analogue? At that point, the choice is deemed to be invalid. One might expect that, if a trust is invalid by its applicable law, it is 321 M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15; see also the discussion of “the characteristics considered” in relation to Art. 2, above. 322 Whether it should be is considered in the discussion of “the characteristics considered: the ‘shapeless’ trust” in respect of Art. 2, above and the discussion of Art. 13, below. 323 See e.g. D Hayton, “International Recognition of Trusts”, in J Glasson (ed.) International Trust Laws, ch. C3, 17–23. 324 The process of translation is supposedly ended by the Hague Convention, because a trust must be recognised qua trust. But if a state lacking the common law trust has to apply its choice of law rules to it, it is difficult to see what else it can do. 325 Whether he might be able to use dépeçage to alleviate some of those consequences will be considered in the discussion of Art. 9, below.

160 Non-Trust Juisdictions invalid full stop and no trust exists.326 Indeed, those who dabble with danger by choosing the law of Civilopia might not expect a great deal of sympathy if that were the legal consequence. However, the Convention takes a more benevolent line and states that, if the chosen law is that of a non-trust state, the applicable law should then be determined in accordance with the rules of Article 7, as if in the absence of a choice of law. And perhaps rightly so; choice of the law of a non-trust state is unlikely often to be motivated by evasionary tactics; and if it is, the parties can hardly complain if the law of closest connection is imposed on the trust in lieu. On the other hand, a set of rights and obligations understood by the parties to be governed by one law to which the settlor and trustees have agreed ought not obviously to be referred to a different law, if it transpires that the choice was not valid. It is likely that, if the chosen law (Civilopian) finds that the choice of its law is invalid, it will also hold that, by Civilopian law, the entire trust is invalid. Of course, Civilopian law may not provide for trusts in its domestic legal system, but it might still permit the trust to be effective for private international purposes, if it is valid by its law of closest connection (assuming that this points to the law of a trust state). The reference to the objectively applicable law would, at its purest, involve a reference not to what English law considers to be the law of closest connection, but to the law which Civilopian law would itself apply in the absence of choice, assuming that it would proceed to the law of closest connection.327 Its main advantage is that it would result in the same law being applied in an English or a Civilopian court, consequently reducing the attractiveness of forum shopping. But since such a reference would effectively be an application of renvoi, and since in any event when the settlor chose a law, he would almost certainly have had in mind the domestic law of Civilopia, so that an application of Civilopian private international law could not be justified on the basis of giving effect to party expectations, it is not obvious that this path should be taken. Of course, it would also encounter enormous difficulties of proving what law a Civilopian judge would apply in the absence of choice. In the event then, Article 6 makes clear that the law of closest connection should be determined by an English court by reference to the Article 7 factors. But when von Overbeck comes to discuss the law of closest connection in this context, he remarks: “one would think that, given the broad power to assess the situation that article 7 confers on a court, the judge will have a tendency to conclude that a trust is most closely connected with a State which has this institution”.328

326 Compare Art. 8(1), Rome Convention, which subjects the validity of a contract to the putatively applicable law. 327 Assuming that Civilopia is a non-contracting state to the Hague Convention and has its own choice of law rules. 328 Von Overbeck Report, para. 61, p. 383. See also Report of the Special Commission, Preliminary Document No. 9 of 9 March 1984, No. 50.

Dépeçage 161 In policy terms, such an approach may be commendable. It also attracts the support of Nygh.329 However, the approach is also curious. Article 7 is concerned with ascertaining the law of closest connection. It is very difficult to see how such a question can be affected by the fact that, if that law of closest connection is Civilopian law, the trust will fail, whereas if the law of closest connection is English law, it will be valid. Von Overbeck, whilst stressing the flexibility of Article 7, is suggesting something close to a presumption of validity. Respect for settlor autonomy, and a policy which views the upholding of a trust in some respect as better than its failure, have considerable appeal. But they seem to belong to a choice of law rule concerned with a search for the implied intention of the parties, a matter with which Article 7 is not concerned. If Civilopian law is the law of closest connection, it is hard to see how it ceases to be so simply because it renders the trust invalid.330

2. DÉPEÇAGE

Von Overbeck then turns to consider the, at first sight, improbable situation where dépeçage applies and a severable part of the trust is governed by the law of a non-trust country.331 He has in mind particularly the law governing the administration of a trust, which, depending as it does in particular on the place of administration, the habitual residence of the trustees and location of the assets, may point to a non-trust state. Even then, it is indeed going to be rare, when a trust instrument contains no choice of law clause, for the law applicable to administration to be different from the law applicable to the validity of the trust. Nevertheless, an Australian proposal was put forward specifically to deal with the matter,332 to the effect that if a law of a non-trust state was chosen to govern an issue other than validity, the choice should be ineffective. The proposal was accepted, but rendered less important by the decision to include a general provision rendering any choice of the law of a non-trust state ineffective under the Convention. The result is that a choice of the law of a non-trust state, whether to govern the trust in whole or in part, will be inoperative. That only begs a second question, namely: “which law should apply in the event that dépeçage points to the application of the law of a non-trust State?” If, as regards validity, a trust state’s law is chosen, by which law the trust is valid, then the question arises what to do if the law of a non-trust state is chosen as regards e.g. matters of administration. Two obvious possibilities present themselves: to ascertain the applicable law in the absence of choice in accordance 329 P Nygh, Conflict of Laws in Australia, 6th edn., 516, placing reliance on Augustus v. Permanent Trustee Company (Canberra) Ltd (1971) 124 CLR 245. 330 See further the discussion of “validity and the law of closest connection” in relation to Art. 7, below. 331 Von Overbeck Report, para. 62, p. 383. 332 Working Document No. 6. See Von Overbeck Report, paras. 95–96, p. 391.

162 Non-Trust Juisdictions with Article 7, specifically in relation to the question of administration;333 or alternatively, to extend the law applicable to questions of validity to questions of administration, thereby eliminating the dépeçage complication. The point is not decided by the Convention, but von Overbeck argues that the Convention goes “in the direction of making an objective connection under article 7 for the severable aspect”.334 This seems correct, since Article 6(2) merely states that where a choice of law is made for a non-trust state, Article 7 should apply. It does not indicate that that failed choice has to be in respect of the entire trust and it seems that Article 6(2) applies to a choice which fails in whole or in part. In principle too, there is much to be said for this. Whilst more complex to apply than simply using the law applicable to validity, it is arguably more likely to give some effect to the settlor’s intentions, or at least the settlor’s negative intentions. Von Overbeck points out that the contrary viewpoint would be that: “the choice of a law which does not provide for trusts could be considered as an absurdity which should be completely ignored, not only as concerns the choice itself but also as concerned the principle of dépeçage”.335

In effect, the settlor’s intentions, having chosen such a law, can no longer be taken seriously at all. But this does not seem a strong enough reason not to apply dépeçage. It must surely be accepted that some attempt to give effect to the settlor’s intentions is better than none. Presumably, a settlor who specifies that the law governing validity should not also govern administration has in mind that there are good reasons why administration should not be governed by the “dominant” law, and to go ahead and apply that law anyway may be exactly what he would not have wanted. Furthermore, the principle of dépeçage is supported if the court looks under Article 7 not for the law of closest connection to the trust as a whole, but for the law of closest connection to the question of its administration. If the severable part of the trust’s applicable law must be determined by Article 7, it follows that if application of that Article points to a non-trust country, for those issues, the Convention will be inapplicable. However, since this may have dire consequences in relation to the recognition of a trust, whose validity may be recognised in Civilopia under the Convention, but not its administrative requirements, it is not unreasonable to imagine a court attempting to avoid such an outcome at all costs.336 Before leaving the subject of dépeçage, it is worth questioning what would happen if matters were reversed. What if the question of validity was stated to be governed by a non-trust state’s law; whereas all other matters were to be governed by the law of England? On matters of validity, would it be appropriate to 333 I.e. the law with which the trust’s administration has its closest connection, not the law with which the trust as a whole has its closest connection. 334 Von Overbeck Report, para. 96, p. 391. 335 Ibid., para. 96, p. 391. 336 Even if, strictly speaking, the law of closest connection should be determined entirely objectively.

Categories of Trust 163 apply Article 7 and simply look for the law of closest connection (which might be the law of a non-trust state, so that the trust is declared invalid); or would it be appropriate simply to apply English law to that question? This time, it would not appear unreasonable to apply a law which upholds the trust, in so far as a presumption of validity exists under the Convention. However, if a settlor really does go out of his way to select an applicable law under which the trust is not valid, one wonders whether it would be appropriate for a court to fight tooth and nail to find a way of nonetheless upholding the trust. On balance, it is submitted that a search for a law of closest connection, whether in relation to the whole or a part of the trust, must be just that and no presumption of validity, or against dépeçage, should be applied.

3. CATEGORIES OF TRUSTS

It is important to note that the provisions of Article 5 are triggered not merely when the application of a non-trust state’s laws is called for, but also when the state identified does not provide for the particular “category” of trusts specified. Self-evidently, an enormous amount of importance attaches to the meaning of the word “category” in this context. At the most general level, one might categorise trusts in a given state as “express” or “implied”.337 Again, one might categorise as “trusts arising by intention” or “trusts arising irrespective of intention.” When one gets more specific, and starts to regard as categories “resulting” and “constructive” trusts, the likelihood of any given state not knowing that category is magnified. More specifically still, could it be said that although the putative applicable law knows e.g. the resulting trust, it does not know the “category” of automatic resulting trusts? Or suppose that a settlor purports to create a charitable purpose trust to be governed by the law of Utopia. Utopian law recognises the category of express trusts for a purpose, providing that the trust has a named “enforcer.” However, it does not allow a trust to be created for charitable purposes. Could it be said that a purported charitable purpose trust governed by Utopian law falls outside the Convention? The importance of this issue can hardly be exaggerated, since if a very narrow, specific view is taken of what constitutes a “category”, the Convention’s application would be enormously restricted.338 Of course, in many contracting states, the Convention will be concerned only with trusts created “voluntarily”;339 but even then a state may have to determine e.g. whether an automatic resulting trust is a “category” of trust;340 or whether the “category” of valid 337 The Hague Convention seems rather to categorise trusts as voluntarily and involuntarily created: Art. 3. But the reference to categories of trust in Art. 5 seems to show that those trusts falling within the Convention constitute more than one category. 338 Ironically making it arguably “safer” and, accordingly, in one sense more attractive for a civilian state to ratify the Convention. 339 Art. 3. 340 Unless, of course, the word “category” has a very narrow meaning.

164 Non-Trust Juisdictions non-charitable purpose trusts exists.341 Problems of classification of this sort may be still more complex in states which have taken advantage of the right to extend the Convention’s scope to a broader range of trusts. Against this, categories which are too broad may be a recipe for disaster. Imagine that a charitable purpose trust was stated to be governed by Utopian law. Imagine that an English court were to find that, although such trusts were not permitted by Utopian law, the broad “category” of purpose trusts did exist there and upheld the choice of law. If the choice of law was permitted, it would then fall to that law to determine whether the trust itself were valid, in accordance with Article 8. In this case, the question would be: “according to Utopian law, is the creation of a charitable purpose trust governed by Utopian law valid?” The answer would be “no” and the trust would fail. In contrast, if it were found that the “category” for Convention purposes was the charitable purpose trust, then an English court would find that such a category of trusts did not exist in Utopian law and that, accordingly, it should reject the choice of Utopian law, but should proceed to identify the applicable law in the absence of choice. That may well enable it to identify a law of closest connection to which the charitable purpose trust is known and to uphold the instant trust. But there is a further complication. For however broad or narrow a view of the word “category” is taken, the classifications that will be used from state to state will, in all probability, vary enormously. That is a recipe for inconsistency, and also again makes the likelihood of the trust falling within a recognised category in a particular system highly unpredictable. True, the Convention is intended to apply to both the trust strictly speaking and to “structural analogues” to the trust. Presumably, if there is a structural analogue to the particular type of trust sought to be imposed, a state entering into the spirit of the Convention should be minded to find that the category does exist in its legal system. But to say that this process is no exact science is a great understatement. A further question then arises as to which state’s categories of trust an English court should use: those of English law or those of the (putative) applicable law. If characterisation is, in the ordinary course of events, a matter for the law of the forum, then it might be thought that English categories should be applied in an English court. But against this, ignoring the putative applicable law’s categorisation of trusts may be a recipe for inconsistent application of the Convention from state to state. It could lead to the upholding in England of the choice of Utopian law to govern a trust, which a Utopian judge would regard as invalid. The effect of categorising by English law might be that an English court would treat as valid a choice of Utopian law to govern a trust, only then to find that the trust itself is declared invalid by Utopian law pursuant to Article 8, precisely because it does not allow that category of trust to be created in its domestic law. That argues for the use of the putative applicable law’s categorisation of trusts. 341 A difficult question where, as in England, the validity of such trusts is the exception, rather than the rule.

Categories of Trust 165 That way, if the category does not exist by Utopian law, it is possible to proceed to look for a law of closest connection under which such a trust may be valid. This categorising by the putatively applicable law also accords with the wording of Article 5. But if it is hard enough for an English judge to determine what a “category” of trusts is in English law, it may be many times harder for him to determine what is a “category” of trusts in Utopian law. Arguably, the word “category” is ripe for something which the drafters of the Convention, for reasons of perceived impossibility, impracticality and undesirability sought to avoid: namely autonomous categorisation. Were the Convention (or the Official Report) to give some indication of what a “category” is, a clear frame of reference, at the very least, would exist, which could be applied in the courts of all contracting states.342 This could be done by providing e.g. that: “categories of trusts might be those that are created expressly for beneficiaries,343 those that are created expressly for charitable purposes, those that are created expressly for non-charitable purposes, those that arise by assumed common intention of the parties, those that arise by operation of law or statute irrespective of the parties’ intentions and those that arise by judicial discretion.”

Against this, autonomous classification might well lead to the upholding of a choice of the law of a state deemed to provide for a certain category of trust, only then to find that, in the domestic legal system of that state, such a trust itself is invalid. If upholding the validity of trusts is to be regarded as a major purpose of the Convention, then the best approach is arguably for the question of what constitutes a “category” of trust to be determined not by the law of the forum or by autonomous classification, but in accordance with the applicable law. If the category does exist by the putatively applicable law, the likelihood of the trust itself being essentially valid by that law is greatly increased. If the price to pay for that validity is inconsistency in application of the Convention and additional work for the judge of a state trying to wrestle with the categories of a foreign legal system, then it might be thought a price worth paying. In default of evidence as to the categories existing in the putatively applicable legal system, an English court would presumably assume them to be the same as those of the forum and apply its own categories of trust.

342 Although, of course, the United Kingdom would still need to resolve which “categories” are covered by s. 1(2), Recognition of Trusts Act 1987, which extends the Convention’s scope of application. 343 The Convention speaks in Art. 3 of trusts created voluntarily, not of express trusts. But everything that falls within Art. 3 cannot constitute a single category; otherwise, there would be little point in Art. 5 speaking of states which do “not provide for trusts or the category of trusts involved.”

ARTICLE 6—EXPRESS OR IMPLIED CHOICE OF LAW CHAPTER II—APPLICABLE LAW A trust shall be governed by the law chosen by the settlor. The choice must be express or be implied in the terms of the instrument creating or the writing evidencing the trust, interpreted, if necessary, in the light of the circumstances of the case. Where the law chosen under the previous paragraph does not provide for trusts or the category of trusts involved, the choice shall not be effective and the law specified in Article 7 shall apply.344

1. THE TRIUMPH OF SETTLOR AUTONOMY

In Article 6(1) lies the statement of the autonomy principle. It is a principle which is pragmatic and attractive.345 Castel comments that: “In the field of trusts, the most important objective or policy is the fulfilment of the justified expectations of the creator of the trust in the disposal of the trust property. Thus, conflict of laws rules should attempt to carry out this objective or policy by giving pre-eminence to the intention of the creator of the trust. . . .”346

Trusts created voluntarily347 are the province of the Convention and if a person chooses to dispose of his property by creating a trust, it is not obvious that he should be precluded from choosing the law by which to do so. After all, the right of the settlor to alienate the property and the validity of the transfer itself will already have been sanctioned by the “rocket-launching” process.348 It is not apparent that the question of what structure to create with that property disposition is a matter over which further control is necessary, save insofar as any relevant mandatory rules or public policy requirements are infringed. Of course, that is not to say that there may not be real difficulties if e.g. the lex situs does not accept the trust, or even recognise the trust as creating a separate fund not forming part of the trustee’s personal wealth. But there is a reasonably strong case for saying that a settlor who chooses to create a trust over property situated in a non-trust state, governed by the law of a trust state, runs the risk of encountering problems. 344

Art. 6(2) is discussed in connection with Art. 5, above. On the relationship between policy and pragmatism, and the argument that pragmatism is a policy, see M Whincop, M Keyes, R Posner, Policy and Pragmatism in the Conflict of Laws (Aldershot, Dartmouth, 2001) 346 J-G Castel, Canadian Conflict of Laws, 4th edn., 537. 347 At least where they are evidenced in writing. 348 On which see Part One of this book. 345

The Triumph of Settlor Autonomy 167 The argument for party freedom to choose the governing law is one which was long fought349 in contract law.350 Arguments in that context that allowing party autonomy to prevail was effectively to allow the parties to place themselves above, and outside the control of, the English legal system, never had any great conviction. The sanctioning of autonomy is a matter of English private international law. Indeed, any choice of law rule applied in England ultimately draws its life blood from English law; it is the forum which selects the connecting factor. It would be better to focus attention on the question whether the freedom to choose the governing law is appropriate to that particular area of law. In contract, no one can seriously doubt today that it is. It is the law which the parties manifestly expect to apply and by which their rights and duties are framed. It is by far the most practical law to apply in international commerce, and the “interest” of laws other than the chosen law is unlikely to be especially strong in contractual matters. Contract rules do not especially concern the status of persons domiciled in the jurisdiction;351 they do not raise fundamental questions appertaining to civil liberties such as may arise in the law of tort; and they do not concern the control of specific assets over which the situs may have a strong influence. But it may be different with the law of closest connection. This will contain a set of rights and duties which would have been applied in the absence of choice and the choice of a foreign law may be motivated by little more than evasionary tactics. Whatever the degree of “interest” of the law chosen by the parties, some degree of restriction of party autonomy appears necessary to deal with choices which are essentially made purely for negative, evasionary reasons. In some ways, the dominance of party autonomy in contract law makes its adoption in the trust context relatively unsurprising. But it is worth considering how much the two institutions have in common from a choice of law perspective. They share certain characteristics, at least in the case of the express trust evidenced in writing, in that they create personal rights and obligations which, since they arise voluntarily, ought arguably to be regulated by whichever law the settlor wishes, and expects, to apply.352 Without the decision to create a trust, one would ordinarily not be imposed by a court.353 Indeed, the voluntarily created trust might be thought on the whole less susceptible than the contract to the inclusion of a choice of law clause “oppressive” to a particular party. Some settlement is better than none, from the point of view of the beneficiary; and the 349

And fought long ago. See P Nygh, Autonomy in International Contracts (Oxford, Clarendon Press, 1999), ch. 1. 351 Some rules, such as those on capacity, may do so (and different laws may apply in this case), but most contract rules do not raise issues of personal status. 352 See D Hayton, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96. See also A Duckworth, “Trust Law in the New Millennium: Part 1: Retrospective” (2000) 7(1) Trusts and Trustees 12; “Part 2: Prospective” (2000/1) 7(2) Trusts and Trustees 11; “Part 3: Fundamentals” (2000/1) 7(3) Trusts and Trustees 9. 353 Save in specific cases, such as e.g. an express trust created in anticipation of a divorce settlement. 350

168 Express or Implied Choice of Law trustee typically does not need to accept his office. True, there will be difficult cases, such as where the trust is created in contemplated fulfilment of a legal requirement e.g. pending a financial settlement; but the “obligation” to set up a trust is typically weaker than the obligation on e.g. a consumer to contract to buy goods for domestic use. But against this, the express trust partakes of contractual and proprietary characteristics. Hayton favours a definition of the trust for beneficiaries in English law as: “an equitable obligation binding a person (‘the trustee’) to deal with property owned by him as a trust fund segregated from his private patrimony for the benefit of persons (‘the beneficiaries’) of whom he may himself be one, any one of whom has the right to enforce the obligation . . .”354

Obligations of the trustee arise in respect of property. And where property rights are concerned, it is anything but obvious that autonomy should be allowed free reign. True, the “rocket-launching” parts of the trust will not be determined by the chosen law—the property must be vested in the trustee, which normally requires application of the lex situs.355 But the ability to create the trust structure in relation to that property will be subjected to the chosen law, as will the operation of the trust once “launched”; and in so far as that trust creates rights and duties between trustee and beneficiary with respect to that property, it is far from obvious that the lex situs should have nothing to say. Differences in e.g. the scope of the right to retain trust property between the chosen law and the lex situs, or the beneficiaries’ right to terminate the trust and claim the property absolutely, are not obviously desirable and may be difficult to enforce in the situs.356 Nor, where the matter relates to a question which might be classed as one of administration, is it obvious that the law of the residence of the trustees should be ignored. After all, an order compelling trustees situated in an offshore state e.g. to divest themselves of investments improperly retained may prove futile if: either (i) the trustees are not amenable to the jurisdiction of the English courts; or (ii) they are so amenable, but the order would be unenforceable in the courts of the trustees’ habitual residence. Nevertheless, it is clear that the dominant role given to the chosen law reflects the common law approach to express trusts.357 In the Australian case of 354

D Hayton, (2001) 117 LQR 96, 107. Or, in the case of testamentary transfers of movables, application of the law of the deceased’s last domicile. 356 Or more accurately, difficult to enforce where the situs would apply its own law to the matter, rather than the law applicable to the trust. 357 See e.g. Re Hernando (1884) 27 Ch D 284; Re Fitzgerald [1904] 1 Ch 573; Iveagh v. IRC [1954] Ch 365; Trustees Executors and Agency Co. Ltd v. Margottini [1960] VR 417. M Lupoi, Trusts: A Comparative Study, 349 points out that even the Cour d’appel of Paris, (10 January 1970, [1971] Revue Critique de Droit International Privé 518) accepted the freedom to choose the applicable law of a trust created by a Frenchman over securities situated in the United States. On the reception of trusts in France, see further K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 18–21. 355

Trusts of Land 169 Augustus v. Permanent Trustee Company (Canberra) Ltd,358 Walsh J clearly expressed the view that the general choice of law principles applicable to contracts at common law359 should extend to trusts. That meant that the settlor should be free to select the law applicable to the trust. The importance of giving effect to a settlor’s intentions, without which the trust would not exist at all, necessitated such a rule.360 The American Restatement (2d) adopts a similar approach with regard to movables, once the “rocket” has been launched.361

2. TRUSTS OF LAND

(A) A Need for the Lex Situs? Not the least merit of applying the law chosen by the settlor is that it allows a single law to govern the trust. A lex situs approach might well lead to a number of different laws being applied, where all the trust assets are not neatly located in a single state. But where a trust concerns land, use of the situs approach looks considerably more attractive. The American Restatement (2d) provides in relation both to inter vivos and testamentary trusts simply that: “the validity of a trust of an interest in land is determined by the law that would be applied by the courts of the situs”.362 Undoubtedly, the lure of the lex situs is strong. If the validity of a will creating a trust and the vesting of property in the trustee are to be determined by the lex situs because, in the case of land, there is very little point in deciding a dispute in a manner incompatible with the lex situs, then it is far from clear that the same arguments do not extend to the trust itself.363 If the lex situs does not recognise the validity of the trust, then it may regard an effective transfer of property to the “trustee” as an absolute disposition and as giving the “trustee” absolute ownership of the property. For the “trustee”, that allows him to deal with the property as he may choose;364 but it might also 358 (1971) 124 CLR 245; P Nygh, Conflict of Laws in Australia, 6th edn., 515; E Sykes and M Pryles, Australian Private International Law, 3rd edn. (Sydney, Law Book Co, 1991), 713–4. 359 See, in particular, Vita Food Products Inc v. Unus Shipping Co Ltd [1939] AC 277; Bonython v. Commonwealth of Australia [1951] AC 201. 360 Nor does it seem that a narrow construction of the choice of law clause would be taken at common law: see A Wallace, “Choice of Law for Trusts in Australia and the United Kingdom” (1987) 36 ICLQ 454, 460–1. 361 Restatement (2d) on Conflict of Laws, §269–70. See also Hutchinson v. Ross 262 NY 381; Shannon v. Irving Trust Co 275 NY 95. 362 Ibid., §278. This view is supported by E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn. (St Paul, West Group, 2000), 1030. See First National Bank in Mitchell v. Daggett 242 Neb 734, 497 NW 2d 358 (1993). 363 Although, in the case of inter vivos trusts, application of the proper law of the trust appeared to be favoured even at common law for questions of validity: see Dicey and Morris, 11th edn., 1987, 1077–8; E Sykes and M Pryles, Australian Private International Law, 3rd edn., 718–19. In the case of testamentary trusts of land, however, the lex situs held sway on questions of validity: Nelson v. Bridport (1846) 8 Beav 547; Sykes and Pryles, 720–1. 364 Of course, a mala fides “trustee” may simply claim the property for himself in its situs; but he will, of course be regarded in Hague Convention states as having acted in breach of trust.

170 Express or Implied Choice of Law imperil the assets, in that they might be subject to attack by the “trustee’s” personal creditors in the situs. Alternatively, the lex situs may take the view that the “rocket-launching” transfer of the assets to the trustee fails altogether, because those assets were only intended to be conveyed pursuant to a valid trust, so that the entire property remains with, or results back to, the settlor. Either way, the fact will remain that it will be impossible for the intended beneficiary to enforce rights against the land itself; and an in personam order compelling the trustees to deal with the land in a manner inconsistent with the title recognised by the situs may prove futile in practice. It might be argued that freedom of choice should be allowed free reign, subject to the application of the international mandatory rules of the lex situs and the forum’s public policy.365 That might seem a reasonable compromise between party autonomy and the legitimate interests of the situs. After all, why should the lex situs be applied, when the courts of the situs themselves would be happy for another law to govern a particular issue? The response to this would be twofold: first, there is the familiar difficulty in ascertaining, in statutes which are silent on the matter, their intended application to trusts governed by a foreign law;366 and secondly, insofar as such intention can be inferred, it is reasonable to assume that most statutes of the situs which affect the nature and scope of an interest in land are intended to apply to land situated within the jurisdiction.367 There comes a point where application of the mandatory rule exception becomes the rule itself.368 In that case, it is better simply to subject the entire trust to the lex situs as a choice of law rule.369 True, it is possible to imagine some matters to which the situs’ law need not be applied because they are not intrinsically proprietary. Examples would be the appointment or removal of trustees, or delegation of duties.370 Moreover, the construction and interpretation of the trust instrument may be capable of being resolved by the law chosen by the settlor.371 Furthermore, the personal rights of trustee and beneficiary under a trust of land could potentially be subjected to a law other than that of the place where the property is situated. It might also be said that the beneficiary of an English trust could sue in England under English law to restrain the trustee from dealing with land in Civilopia, notwithstanding 365 See also A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., 753–4 for a consideration of possible use of public policy in relation to foreign trusts of English land. 366 A fortiori if the situs does not even know the trust concept. 367 Such as the Trusts of Land and Appointment of Trustees Act 1996. 368 See e.g. Art. 45(2)(iii) of the Trusts (Jersey) Law, 1984, which states that a trust governed by a foreign law shall not be enforced in Jersey insofar as it purports “to apply directly to immoveable property situated in Jersey.” Sub-paragraph (3) goes on to state that “any person in whom the title to such immoveable property is vested shall not be, and shall not be deemed to be, a trustee of such immoveable property.” 369 Not least because applying the chosen law and the lex situs requires two laws to be satisfied by a claimant. 370 See further the discussion of “the Trusts of Land and Appointment of Trustees Act 1996” below; see also J Harris “Ordering the Sale of Land Situated Overseas” [2001] LMCLQ 205, 213–14. 371 Cf. Restatement (2d) on Conflict of Laws, §277.

Trusts of Land 171 that the trust is not known in that latter state.372 But against this, the trustee could equally obtain a declaration in Civilopia that he was absolutely entitled to the land, as a matter of Civilopian law. The potential for a conflict of laws is all too apparent; and it is a struggle in which the state with exclusive control over the assets ought naturally to prevail.

(B) The Trusts of Land and Appointment of Trustees Act 1996 In order to consider further the proper role of the lex situs with respect to trusts of land, let us take the example in England of the Trusts of Land and Appointment of Trustees Act 1996.373 It contains no statement of its intended territorial scope. It contains rules which are “administrative” in nature relating to such matters as consultation with beneficiaries374 and the appointment and retirement of trustees.375 No harm is obviously done by subjecting such matters to a law other than the lex situs. The same might also be said of the provisions on obtaining of consent,376 or consultation with beneficiaries.377 They may be enforced by in personam order and affect the land itself only indirectly. But would an English court really allow a transfer of land situated in England, which is subject to a trust governed by Utopian law, to be effected without consultation with the beneficiaries, on the basis that such consultation is not required by Utopian law? These personal obligations on the trustee are provisions to ensure the appropriate alienation of the property itself; and in so far as the lex situs takes a different view on them to the chosen law, the risk of their ineffectiveness in the courts of the situs is real. In Pollard v. Ashurst,378 a property situated in Portugal was jointly owned by Mr and Mrs Pollard. Upon Mr Pollard’s bankruptcy, the trustee in bankruptcy brought an action seeking the sale of the property with vacant possession. Parker LJ ruled that the claim was not caught by the exclusive jurisdiction provisions of Article 16(1), Brussels Convention,379 since it had as its object380 not rights in rem in immovable property, but “personal issues as between the Trustee on the one hand and Mr and Mrs Pollard on the other.”381 He quoted the words of Jacobs J at first instance: 372 R White, “Equitable Obligations in Private International Law: the Choice of Law” (1986) 11 Sydney LR 92, 111–2. That, of course, requires a basis of in personam jurisdiction over the trustee and, where service out of the jurisdiction is needed, for the claimant to persuade the court to exercise its discretion. 373 See further J Harris, [2001] LMCLQ 205, 213–4. 374 S. 11, Trusts of Land and Appointment of Trustees Act 1996. 375 Ibid., s. 19. 376 Ibid., s. 10. 377 Ibid., s 11. 378 [2001] 2 WLR 722; J Harris, [2001] LMCLQ 205. 379 To be replaced by Art. 22(1), Brussels Regulation. 380 Which his Lordship interpreted ([2001] 2 WLR 722, 730–4) to mean “the principal subject matter” of the claim, relying on the Advocate General’s Opinion in Case C–294/92 Webb v. Webb [1994] ECR I–1717, paras. 11, 27, 28 and 46. 381 Pollard, ibid., 733.

172 Express or Implied Choice of Law “What the [1996] Act does not say is that the court cannot act in relation to trust property held abroad or that similar orders as can be made under the Act cannot be made by virtue of the court’s jurisdiction over property held under an English trust”.382

Jacobs J was specifically addressing section 14, Trusts of Land Act 1996 (which deals with applications for a court order), although his comments appear applicable to the whole Act. Yet even section 14 includes a provision383 conferring a right in interested parties to seek a declaration of “the nature or extent of a person’s interest in property subject to the trust.” Such a declaration will, needless to say, run the serious risk of being ineffective in the situs if the lex situs begs to differ. Other provisions of the 1996 Act impact still more directly on the land itself. For example, section 12 provides a statutory right in beneficiaries entitled to an interest in possession to occupy the land, if the purposes of the trust include making the land available for such occupation. Section 13 sets out the conditions in which one or more beneficiaries may be excluded from occupation. Such occupation rights are themselves a form of proprietary right which affects the alienation of the property, and which it is difficult to imagine an English court not applying where land is situated in England, irrespective of the law stated by the settlor to govern the trust. Equally, if an English trust of land situated in Utopia were created, there would be little point in an English court ruling that, pursuant to section 12, the beneficiary had a right to occupy the land, if the land is situated in Utopia.

(C) Trusts for Sale A particularly problematic example arises in relation to co-ownership and interests in the proceeds of sale of land. As to the former, prior to the 1996 Act, the beneficial interest of a co-owner of land was treated as an interest in the proceeds of sale of the land and hence as an interest in personalty.384 In Re Berchtold,385 the devolution of a beneficial interest in English land subject to a trust for sale arose in determining the inheritance rules applicable to the assets of an intestate who died domiciled in Hungary. Russell J accepted that the interest would, as a matter of English law, be treated as an interest in personalty, but nonetheless proceeded to apply the choice of law rules applicable to succession to immovable property. After all, the land was currently unsold and what was 382

[2001] 2 WLR, 726, citing from the first instance ruling: [2000] 2 All ER 772, 778. S. 14(2)(b). 384 “The whole purpose of the trust for sale is to make sure, by shifting the equitable interests away from the land and into the proceeds of sale, that a purchaser of the land takes free from the equitable interests. To hold these to be equitable interests in the land itself would be to frustrate this purpose”: per Cross LJ in Irani Finance Ltd v. Singh [1971] Ch 59, 80. See further E Burn, Cheshire and Burn’s Real Property, 16th edn., 218–19; 230–1. 385 [1923] 1 Ch 192. 383

Trusts of Land 173 in issue was the intestate’s interest in unsold land. Indeed, had the interest been treated as a movable and subjected to Hungarian law, that law itself may have had quite different views of the rights which the interest created. And since those rights would need to be vindicated against the property itself, any Hungarian decision which infringed the principles of English land law and affected the alienation of the property ran the risk of being refused enforcement in any event.386 Indeed, failure to apply the lex situs to trusts of land can lead to curious and unsatisfactory results, as in Re Piercy.387 There, a testator died domiciled in England, leaving land in Sardinia to trustees on trust for sale, to be held subject to life interests for his children, remainder for their issue. The proceeds of sale were likewise to be held. The disposition was valid by English law. However, Italian law regarded such trusts as invalid and would treat the children as absolutely entitled to the property. North J acknowledged that unsold land must be treated as immovable property prior to sale and subject to Italian law. But this did not stop him from directing the sale of the land and that the proceeds be held according to the precise terms of the will. He reasoned that Italian law was applicable prior to sale, but that once the property was sold, the interest of the beneficiaries would be solely in the proceeds of sale, which were movables and hence subject to English law. However, such an argument cannot withstand serious scrutiny.388 The judge made an order, with respect to the disposition of Italian land as yet unsold, which was unauthorised by that law. True, that order could be enforced in England by in personam proceedings; but in so far as an order is illegal in the situs, its enforcement in that latter state is most unlikely.389 In English domestic law at least, the doctrine of conversion has now been abolished and the beneficiary’s interest under a trust for sale is treated prior to sale as one in the land itself.390 That only strengthens the in rem nature of the trust; and where interests in rem relating to land are concerned, attempts to avoid the lex situs rule are doomed to failure. That said, where a trust for sale is created under which land in Utopia is to be sold and the proceeds to be remitted to Ruritania, the application of the lex situs may appear less crucial. Indeed, the American Restatement (2d) provides that: 386

See also the discussion in section 9 of Part One of this book. [1895] 1 Ch 83; considered by K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 17–18. See also Phipson-Stow v. IRC [1961] AC 727. In America, see Moore v, Livingstone 148 Ind App 275, 265 NE 2d 251 (1970). 388 See Cheshire and North, 934–5. Contrast P Nygh, Conflict of Laws in Australia, 6th edn., 514 who states that “The reasoning in this case is dubious, but the result eminent common sense.” See also Lewin on Trusts, 17th edn., 299. 389 Indeed, the entire question of whether there is a conversion of a beneficiary’s interest should be a matter for the situs: Restatement (2d) on Conflict of Laws, §278, 216. See also E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1029 where the authors argue that that the lex situs should be applied where a trust is created to hold a particular piece of land, but not necessarily where the land is treated as an investment. 390 S. 3, Trusts of Land and Appointment of Trustees Act 1996. 387

174 Express or Implied Choice of Law “When a trust of land is created, either by will or inter vivos, and the trustee is directed to sell the land and remit the proceeds to another state to be held there in trust, the trust is valid if valid at the place of administration even though it would be invalid under the local law of the situs, provided that it does not violate a strong policy of the state of the situs”.391

However, the fact remains that the direction to sell must be valid by the lex situs. If there is, for example, a statutory right of occupancy for the beneficiaries under another law, the risks of unenforceability in the situs are evident. It would be preferable for the question of whether the land may be sold and the status of the equitable interest therein prior to sale to be remitted to the lex situs. Once the interest is, by the situs, duly transferred into the proceeds of sale, then there can be little objection to that trust of the proceeds being subjected to another law. In England, that law would be the law chosen by the settlor to govern the trust.

(D) Other Interests in Land Further difficult cases are imaginable in the case of trusts of land, where the beneficial interest in question does not relate to the freehold. The leasehold interest is one such case.392 Although regarded in domestic law as personal property, it is treated in England for conflict of laws purposes as an interest in immovables.393 That must be correct. The principal justification for the use of the lex situs is the requirement of enforceability. An interest pursuant to a lease may create an exclusive occupation right in the land itself. What benefit can be served in a finding that, by the law chosen to apply to a trust, B can reside on the property, if the lex situs takes a different view? Whether the interest is technically one in real or personal property is not the point. The fact is that a leasehold directly affects real property; it should be classified accordingly as immovable and subjected to the lex situs. It is not obvious that the lex situs should be discarded simply because the leasehold interest happens to be the subject of a trust. Likewise with mortgages, which have been classified in England for conflict of laws purposes as immovable.394 The “characteristic” part of the mortgage may be the debt itself; and arguably it is the law applicable thereto which should determine the rights of mortgagor and mortgagee. However, the mortgage creates a secured right over the land; and if that right is to mean anything, then it must be enforceable in the situs. Once again, it is not obvious that the position changes when a trust enters the picture.

391

Restatement (2d) on Conflict of Laws, §278, 215. Which is treated in some parts of the United States as subject to the law on movable property: see ibid., §278, 218. 393 Freke v. Carbery (1873) LR 16 Eq 461; Duncan v. Lawson (1889) 41 Ch D 394; Pepin v. Bruyere [1900] 2 Ch 504. 394 Re Hoyles [1911] 1 Ch 179. 392

Trusts of Land 175

(E) A Look Offshore Even the rules of certain offshore trust jurisdictions, which are often particularly protective of trusts governed by the local law, recognise the importance of the lex situs to trusts of land more explicitly than does the Hague Convention. Two examples may illustrate this. In the Bahamas, the Trusts (Choice of Governing Law) Act 1989 provides in section 4 that the parties may expressly choose the law of the Bahamas to govern a trust.395 Where the law of the Bahamas is chosen, it is stated to apply to most aspects of the trust.396 However section 7(2)(a)(ii) states that Bahamas law shall not validate: “any trust or disposition of immovable property situate in a jurisdiction other than the Bahamas in which such trust or disposition is invalid according to the laws of such jurisdiction.”

Effectively, this subjects trusts of land to both the chosen law and the lex situs for questions of validity and attempted dispositions of the land. Indeed, it is such a general provision that it almost introduces a double actionability397 into trusts of land. There is no suggestion that the “laws” of the situs need be mandatory in a domestic or international sense.398 In the Cayman Islands, the Trusts (Foreign Element) Law 1987 contains a more general provision validating choice of law.399 There follow specific provisions relating to the scope of application of Cayman Islands law to trusts stated to be governed by that law.400 But, as in the Bahamas, Cayman Islands law may “not validate any trust or disposition of real property 401 situate in a jurisdiction other than the Islands which is invalid according to the laws of such jurisdiction”.402

(F) Where no Law is Chosen to Govern the Trust Where no law is expressly chosen to govern the trust, the case for applying the lex situs is yet stronger. It may be possible to infer that a trust of immovables 395

No such provisions are made in respect of a choice of a foreign law. S. 7, Trusts (Choice of Governing Law) Act 1989. 397 I.e. the claimant asserting the validity of such a trust of land must satisfy two laws (Bahamas law and that of the situs) in order to succeed. 398 It is also not stated whether “laws” means domestic or private international laws of the situs. If the justification for use of the lex situs is to reproduce the conditions applicable in that state, then it ought to be the latter. 399 S. 4(1), Trusts (Foreign Element) Law 1987. The legislation is now incorporated into Part VII of the Trusts Law (1998 Revision), s. 89. 400 Ibid., s. 5 (s. 90, 1998 Revision). 401 The curious reference to “real property” in a conflict of laws statute may limit the scope of the provision and exclude its application to trusts of e.g. leasehold or mortgage interests. 402 S. 5(e), Trusts (Foreign Element) Law 1987 (s. 90, 1998 Revision). 396

176 Express or Implied Choice of Law was intended by the settlor to be governed by that law. If there is no evidence even of an implied intention, then there is unlikely to be an expectation on the part of settlor, trustee or beneficiary for the law to protect and it is difficult to imagine that any other law, however close its connection, could have a stronger interest in the matter. That said a state’s “interest” is not the benchmark for ascertaining the law of closest connection. Where the applicable law is ascertained in the absence of any express or implied choice, the law of closest connection could transpire to be other than the lex situs. It might be thought, however, that this will rarely be the case, since even if the trust has no other significant connections to that state, the situs should weigh especially heavily in ascertaining the law of closest connection to any trust of land.403 Nevertheless, whilst the lex situs proved of some relevance at common law, the objective proper law test has been determined by reference to a range of factors, and the situs not accorded overwhelming significance.404 Take but one example: in Re Fitzgerald,405 a marriage settlement had been entered into between an English and a Scottish domiciliary, consisting principally of immovable property situated in Scotland. Scottish law was held to be the proper law, partly because this was the lex situs, but also because of the wife’s Scottish prenuptial domicile and the Scottish instrument creating the trust. Indeed, the existence of the Scottish document might today be taken as an indicator that the settlor impliedly intended Scottish law to govern. It is not obvious what would have happened had the Scottish settlement related to immovable property located overseas in e.g. Ontario, but one might surmise that it would have made no difference and that Scottish law might still have governed. Insofar as a court seeks to ascertain an implied intention of the settlor, the role of an “objective” factor such as the situs may be very small (especially as the settlor may precisely have intended to avoid restrictions which exist in the situs). It will be only where no implied choice can be ascertained that the situs may become relevant but, as we shall see, the situs is not accorded anything like conclusive weight when ascertaining the law of closest connection under Article 7, Hague Convention.406 Further support for the marginal importance of the lex situs to trusts of land is provided in a more recent decision, Lightning and Anr v. Lightning Electrical Contractors Ltd, (LEC),407 which arose in the context of a presumed resulting trust. LEC purchased a property in Scotland, which it registered in its own 403 A Wallace, (1987) 36 ICLQ 454, 471. See also (in the context of unjust enrichment claims in respect of land) Batthyanay v. Walford (1887) 36 Ch D 269; Dicey and Morris, Rule 200(2)(b), 1485, 1497; cf. G Panagopoulos, Restitution in Private International Law (Oxford, Hart Publishing, 2000), 151–3; 221–5. 404 See also the discussion of the common law cases below. 405 [1904] 1 Ch 573. 406 Note that Art. 7, Hague Convention puts the situs only second in an apparently hierarchical list of factors. See further the discussion of “the four factors considered” in relation to Art. 7 below. 407 Unreported, judgment of 23 April 1998, CA. See also R Stevens, “Resulting Trusts in the Conflict of Laws” in P Birks and F Rose (eds.) Restitution and Equity, vol. 1: Resulting Trusts and Equitable Compensation, 147, 156.

Trusts of Land 177 name. Mr Lightning, the managing director of LEC, claimed that he had provided the entire purchase price for the property and that it was accordingly held on resulting trust for him by LEC. Both Lightning and LEC were resident in England. A question arose as to whether the trust would be governed by English or Scottish law. Their Lordships did not decide whether the Recognition of Trusts Act 1987 applied to such a trust, taking the view that the result would not differ, whether applying the Act or the common law. They ruled that English law was applicable. Millett LJ commented that: “If A provides money to B, both being resident in England, to purchase landed property in his own name but for and on A’s behalf, and B does so, the consequences of that transaction are governed by English law. It would be absurd if they were governed by the law of the place where the property in question happened to be located. Such a rule would lead to bizarre results if, for example, A’s instructions were to buy properties in more than one jurisdiction, for the consequences of the same arrangement might then be different in relation to the different properties acquired.”

Such an unequivocal statement cannot be wholly endorsed. It is very far from obvious why it would be “absurd” for the (absolute) beneficial entitlement of A to land in Utopia to be governed by English law, that being the law (a) which third parties might assume to be applicable to determine any rights existing in the Utopian property; and (b) which will need to recognise the existence of the right. Is this manifestly more absurd than application of the lex situs to a purported absolute transfer of Utopian land inter vivos by one English resident to another; or to the validity of a testamentary transfer by an English testator of Utopian land to an English legatee? Moreover, Millett LJ’s point concerning land in several jurisdictions proves too much, since a will devising land in several countries requires the application of the lex situs of each state separately, without causing untold difficulty. In the same case, Gibson LJ considered the decision in Webb v. Webb.408 In that case, an English farmer bought land in France in the name of his son. Some twenty years later they fell out and the father sought to sue the son in England, claiming equitable ownership on the basis of a resulting trust. The ECJ ruled that the English courts were jurisdictionally competent and that exclusive jurisdiction did not vest in the French courts. Whilst Webb was a decision unequivocally on jurisdiction, Gibson LJ in Lightning v. LEC commented on the English court’s reference to the ECJ thus: “it seems to me implicit that the English court not unnaturally regarded English law as applicable to the relationship between the parties before it in the absence of any event governed by the lex situs destructive of the equitable interest being asserted.” What is meant by the latter part of this sentence is very far from clear and was not elucidated by his Lordship. The more important point for present purposes is that his Lordship, admittedly in the context of a presumed resulting trust, applied a choice of law rule of closest connection almost identical to that which one might apply to any express trust 408

[1991] 1 WLR 1410; Case C–294/92 [1994] ECR I–1717.

178 Express or Implied Choice of Law in the absence of choice, and showed not the slightest inclination to treat the lex situs as a significant, let alone dominant or determinative, factor.

(G) Trusts of Movable and Immovable Property Would the case for the lex situs be weaker if the trust related both to movable property and immovable property?409 After all, “The prospect of a range of different laws being applied to different parts of the property would scarcely appeal to the settlor”.410 Where a court is looking for an implied choice of law, it might be seen as unlikely that a settlor would have wished for movable and immovable property under a single settlement to be subjected to different laws. Where the bulk of the trust property is movable and all situated in a single state (Utopia) and the immovable property is located in Ruritania, the case for applying Utopian law to the entire trust is apparent. Of course, such convenience is overlooked in the field of succession, where different choice of law rules apply to movables and immovables. However, in succession law, no choice of law is permitted at all; insofar as express or implied intention is permitted in the law of trusts, it is quite legitimate to assume an intention to avoid dépeçage. Where no choice can be inferred in relation to a trust of movables and immovables, should the search be for a single law of closest connection? Certainly, convenience still supports such an approach. However, it is necessary to ascertain the closest connection for each issue arising under the trust, there being no expectation to protect.411 Furthermore (as the choice of law rules for inter vivos transfers of property and for succession reflect), the attraction of the lex situs to immovable property is exceptionally strong (so much so that different laws are applied to succession to movables and to immovables). Accordingly, it is suggested that no such presumption of uniformity should operate. Indeed, a helpful principled starting point may be that, in the absence of choice, the lex situs of immovables should be presumed to be the law objectively applicable to trusts of immovables, whether or not movable property is also subject to the trust. In practice, however, neither the common law nor Article 7, Hague Convention attaches overwhelming weight to the situs of immovable property in ascertaining the law of closest connection; and, unfortunately, it can only be imagined that the lure of the situs will be perceived as weaker still where the trust consists of movable and immovable property.

409

A Wallace (1987) 36 ICLQ 454, 471. W Binchy, Irish Conflict of Laws, 513. 411 Save, perhaps, that of third parties, who might expect that any interests in land will be determined by the lex situs. 410

A Need for an Objective Connection to The Chosen Law? 179

3. A NEED FOR AN OBJECTIVE CONNECTION TO THE CHOSEN LAW ?

In earlier negotiations of the Convention, the idea had been floated that the chosen law must be one with which the trust had an objective connection.412 A Greek proposal that the choice be disregarded where no such connection existed was considered by the Fifteenth Session.413 Its ultimate rejection was in part due to the lack of criteria for determining whether the requisite contact existed. One might wonder whether it would have proved insurmountable to have furnished such criteria. Presumably, the factors which determine the applicable law in the absence of choice pursuant to Article 7 would be those to which one would have had regard. In the event, the Convention contains a form of compromise arrangement. A choice of law will not be struck down for having no objective connection with the trust. However, Article 13414 excuses a state from recognising a trust which has (save for its chosen law, the place of administration and the habitual residence of the trustee) its greatest objective connection to a non-trust state. Accordingly, the settlor is quite free to choose the law of trust state A, even though the trust is otherwise wholly connected to trust state B. Whilst Article 13 is a serious restriction on settlor autonomy, examples of its application may in practice be fairly rare415 and will not, for example, avail a settlor who chooses the law of one trust state (say, the British Virgin Islands) to govern a trust wholly connected with another trust state (say, England). The general principle, that the chosen law need otherwise have no connection with the trust, is more liberal even than the rules on party autonomy set out in the Rome Convention in respect of international contracts. There, Article 3(3) provides that: “The fact that the parties have chosen a foreign law, whether or not accompanied by the choice of a foreign tribunal, shall not, where all the other elements relevant to the situation at the time of the choice are connected with one country only, prejudice the application of rules of the law of that country which cannot be derogated from by contract, hereinafter called ‘mandatory rules.’ ”

This approach again involves upholding a choice of law clause, irrespective of whether the law identified has a connection with the contract. However, it recognises that some choices of law may be motivated by evasionary tactics; and indeed, that, whatever the parties’ motivation, a law of very strong objective connection to the contract has a legitimate interest in the matter. Hence, it superimposes onto the applicable law those domestic mandatory rules which 412 The idea was considered and rejected by the Special Commission: see Report of the Special Commission No 54. 413 See Von Overbeck Report, para. 65, p. 384. 414 This Art. is not scheduled to the Recognition of Trusts Act 1987. It is discussed below. 415 Witness the reluctance to use Art. 13 in Italy (see the discussion of “application in Italian courts” in relation to Art. 13, below).

180 Express or Implied Choice of Law could not have been derogated from had that law of closest connection also been the applicable law. Certainly, the scope of Article 3(3) of the Rome Convention is narrowly drawn.416 The choice must be of a “foreign” law 417 and all other objective factors must point to another law. It may also prove problematic, given the difficulty in determining which factors should be treated as “relevant to the situation”.418 But if choice of a wholly unrelated law is not permissible in contract law, it is not obvious why it should be permissible in trusts. Moreover, the common law position, both in contract419 and in trusts420 probably did require at least that the choice of law was made bona fide. Arguably, a more restrictive approach is called for in trusts law: “Indeed, there may be some argument that the special nature of a trust and the traditional emphasis the courts place upon protecting the beneficiaries’ interest warrant a stricter approach to trusts than to contracts”.421

Of course, beneficiary protection is not always the most important consideration. A voluntary settlement is normally better than no settlement as far as a (volunteer) beneficiary will be concerned. But where the settlement is in fulfilment of prevailing legal obligations, such as a financial settlement pending divorce, or in satisfaction of the forced heirship rules of a given system, subjecting the trust to a law protective of the beneficiary’s rights should arguably be a significant consideration. So too should be the protection of the trustee, if he will be subjected to burdensome duties and standards of care, or if the trust is subject to an extremely long perpetuity period.422 Of course, the trustee does not need to accept office, but one would not wish to make his conditions so unfavourable that he is subsequently moved to resign, or that additional or replacement trustees could not be found. But above all, the proprietary nature of a trust renders it essentially different in nature to a contract. Whilst a contract may, of course, have proprietary consequences, there is no suggestion of such consequences being subjected to the chosen law. In trusts, the nature and scope of the interest in question is a matter in which the situs has an obvious interest; a fortiori if other objective factors point to that same law.

416 It is concerned with parties to a purely domestic contract trying to evade the application of their state’s law by inserting a foreign choice of law clause. 417 See Cheshire and North, 557–8. 418 See the discussion in Golden Acres Ltd v. Queensland Estates Pty Ltd [1969] Qd R 378. 419 The requirement comes from Vita Foods Products Inc v. Unus Shipping Co Ltd [1939] AC 277, 290. 420 Dicey and Morris had suggested at common law that the public policy of the objective proper law could not be evaded by choice of another law: see 10th edn., p. 680; A Wallace (1987) 36 ICLQ 455, 463. 421 A Wallace, ibid., 462. 422 Insofar as the perpetuity period affects the trust itself. If it applies to “general” alienations of property, it is argued in Part One of this book and in the discussion of Art. 4 above that perpetuities should rather be seen as a preliminary matter excluded from the scope of the Convention.

A Need for an Objective Connection to The Chosen Law? 181 The American Restatement (2d), on Conflict of Laws is more restrictive of party autonomy. It requires, in relation to inter vivos trusts of movables, that the chosen law has “a substantial relation to the trust.” Even if this is the case, the chosen law should be disapplied to the extent that it infringes the strong public policy of the state with which the trust has its most significant relationship.423 “A state has a substantial relation to a trust when it is the state, if any, which the settlor designated as that in which the trust is to be administered, or that of the place of business or domicil of the trustee at the time of the creation of the trust, or that of the location of the trust assets at that time, or that of the domicil of the settlor, at that time, or that of the domicil of the beneficiaries”.424

It may not be difficult to satisfy this “substantial relation” test but it does provide a connecting factor with some form of legitimate interest in the trust. It cannot simply be argued that, because a settlor is allowed to designate the applicable law of the trust, he should be able to designate any law, regardless of its connection, if there is another law whose interest is sufficiently strong and significant.425 That is especially so where all objective factors point to a different law and makes a provision such as Article 3(3), Rome Convention attractive, extending autonomy as it does to broad, but circumscribed limits. Indeed, one might think that the failure to include a provision relating to domestic mandatory rules of the law which has overwhelmingly the closest connection with the trust (but for the choice of law clause)426 is a recipe for difficulty when it comes to the recognition of the trust. This applies a fortiori to trusts of land.427 Of course, the counter-argument would be that non-trust states to whom this degree of settlor autonomy is unacceptable can refuse to recognise trusts objectively connected to a non-trust state under Article 13. States to whom recognition of trusts governed by a law with which they have no objective connection is acceptable should be allowed to recognise them if they see fit. This is not conclusive however, since Article 13 limits the circumstances in which 423

Restatement (2d) on Conflict of Laws, §270, 215. Ibid., comment to §270, 164–5. The list is a non-exhaustive one. 425 A Wallace, (1987) 36 ICLQ 454, 463. 426 Art. 16(2) of the Convention allows a court to apply the international mandatory rules of a state having “a sufficiently close connection” to the trust “in exceptional circumstances.” However, this provision was not enacted in the United Kingdom. Moreover, if a trust is objectively connected wholly to one state, one might expect its domestic mandatory rules to be applied as well, on the basis that, had that law been the applicable law, those provisions could not have been derogated from. There is no scope for applying domestic mandatory trust rules of a third country in the Convention (Art. 15 applies to domestic mandatory rules, but only to those of the governing law in related areas of law). In this respect, the views of E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 323 that Arts. 15 and 16 provide enough restriction on settlor autonomy are open to doubt. 427 See the discussion of “trusts of land” above. Even J Schoenblum, who strongly criticises the Convention for the number of derogations which it contains from the chosen law, fears that the absence of a substantial connection requirement will lead to use of public policy in relation to trusts of land to mitigate against apparently evasionary choices of law: “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 8. 424

182 Express or Implied Choice of Law recognition may be denied (discounting the place of administration and the habitual residence of the trustee). Moreover, states which wish to adopt more favourable rules for the recognition of trusts are in any event permitted by Article 14 to do so. Finally, the recognition rules will not permit a state to deny recognition (whether or not it has enacted Article 13) where a settlor chooses the law of one trust state to govern a trust entirely objectively connected to another trust state, and where that choice is purely motivated by evasionary tactics. It might accordingly have been preferable to have included in the Hague Convention at least some restriction on the chosen law based on objective connection. In its absence, choice of law is likely to be controlled more negatively by the use of public policy and the superimposition of mandatory rules of the forum.

4. “ INTERNATIONAL ” TRUSTS

The drafters of the Convention also considered whether to restrict settlor autonomy to truly “international” trusts. In the event, no such requirement was specified. Of course, recognition of a trust “presupposes the insertion into the legal system of a phenomenon which would otherwise only operate in a different, and therefore foreign, legal system”.428 But that interaction of legal systems can be provided even where the parties, the trust assets and the place of administration are all initially located in the same state, since one or more elements may subsequently become connected to another state, in which recognition may be sought. Furthermore, the rejection of the need for a substantial connection between the trust and its applicable law means that “there is nothing to prevent all the elements of the trust from being concentrated in Italy or in Holland, and yet have the trust governed by, for example, English law”.429 In other words, what appears to be a wholly domestic relationship can be rendered sufficiently “international” for the Convention to apply merely by the choice of the law of another trust state to govern the trust. It might be argued that a settlor should not be able to create a conflict of laws simply by choosing a foreign law to govern what is essentially a domestic trust.430 In France, for example, prior to the enactment of the Rome Convention, choice of law clauses were only permitted in “international” contracts;431 and the Hague Convention on the Law Applicable to Contracts for the International Sale 428

M Lupoi, Trusts: A Comparative Study, 353. M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 44. 430 Von Overbeck Report, para. 66, p. 384. See also J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 245; M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in P Jackson and D Wilde (eds.) The Reform of Property Law 222, 224. 431 E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 319; H Batiffol & P Lagarde, “Droit International Privé” (1983) LGDJ, vol. 2, no. 575. For an analysis of the word “international” in the contractual context, see P Nygh, Autonomy in International Contracts, 47–55. 429

“International” Trusts 183 of Goods 1955 states that a choice of law agreement between the parties is not in itself sufficient to render a contract “international”, which is necessary for that Convention to be applicable.432 However, practical difficulties might be posed by such a requirement in the trusts context, given that even a purely domestic trust might subsequently need recognition overseas if e.g. the trustee wishes to invest trust assets abroad, or if the trustees themselves change. Nonetheless, one might wonder whether a “domestic” English trust should ever be subject to anything other than English law, no matter that the parties have designated another law to govern. If it is possible to agree to create a conflict of laws, it is possible to agree to be subject to a particular choice of law rule. Arguably, that places the settlor above the (private international) law. On the other hand, insofar as the rules of a domestic legal system are not mandatory, it is not clear that there is a real objection to their circumvention by choice of another law. A better approach would be to respect the freedom of choice even in such a “domestic” trust, but subject it to the domestic433 mandatory rules of the law of closest connection. That way, all rules which could not have been derogated from had there been no choice of law clause will continue to be applied. The effect of the lack of an “international” requirement has been keenly felt in Italy. Lupoi points out that trusts which are objectively entirely connected to Italy are being created by the choice of the law of a trust state. He lists a number of developments, including: the formation of a Consultative Group on Trusts (“Consulta nazionale sui trusts”) under which the opportunities presented by the Convention have been considered; a Ministry of Finance report on taxation of trusts; rules to permit governmental officials to place assets on trusts governed by foreign law were adopted by the lower Chamber of Parliament; a Bill to provide for trusts (governed by the law of a trust state) for handicapped children; and, most significantly of all, another Bill on trust management which would not require the trust to be governed by a foreign law.434 At present, 432 Art. 1(4), Hague Sales Convention 1955 (which also provides that a choice of court or arbitration clause would not be sufficient in itself to render the Convention applicable); see also the Hague Convention on the Law Applicable to Contracts for the International Sale of Goods 1986 (not yet in force), Art. 1(b) of which provides that a choice of law clause is not in itself sufficient to render the Convention applicable, even if accompanied by a choice of court or arbitration clause. 433 This broader category of mandatory rules is called for, because it reflects all those laws which could not have been derogated from had there been no choice of law. 434 See further M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 46–7. There is also a new journal entitled Trusts e Attività Fiduciarie. See also D Hayton, “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159, 164; P Matthews, “Italian Trust for Debenture Holders” (1997) 11 Trust Law International 20. But see M Graziadei, “Trusts in Italian Law”, in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs. Fiducie in a Business Context, ch. E4, 286–7, who comments that Italian courts might yet use Art. 13 in respect of trusts objectively connected with Italy but governed by a foreign law. See further P Matthews, “Un Trust per l’Italiano” (1998) 12 Trust Law International 104; P Matthews, “New Draft Italian Trust Law” (2000) 14 Trust Law International 33.

184 Express or Implied Choice of Law “Italians are content to have foreign laws govern their trusts.”435 However, they have also developed practices and approaches unique to Italian law which enable the trust to “settle” more appropriately into a civilian system.436 The Convention, in promoting the autonomy of the settlor to choose an applicable law with no connection to the trust, has directly led to the development of a flourishing trust in Italy which Lupoi calls the “trust interno.”437 At the same time, would-be contracting states who might regard such a development as a cause for apprehension, rather than excitement, might always refuse to recognise such trusts under Article 13.438

5. CHOICE OF THE LAW OF A STATE COMPRISING SEVERAL TERRITORIAL UNITS

The choice must be that of an individual legal system. Article 23439 provides that: “where a State compromises several territorial units each of which has its own rules of law in respect of trusts, any reference to the law of that State is to be construed as referring to the law in force in the territorial unit.”

This establishes the validity of a choice of e.g. the law of Ontario to govern a trust.440 It does not, however, explain what should happen if the settlor chooses the law of the country, rather than a territorial unit. Presumably, a choice of e.g. Canadian law to govern a trust would be treated as meaningless and ignored.441 That having been said, it may be possible to rescue such a choice, at least where factors relating to the trust are connected with only one state e.g. if trust assets are situated in Ontario, as is the settlor, but the trustees located in New York, the choice of “Canadian law” might conceivably be sufficient to allow the court to infer that the settlor intended the law of Ontario to govern the trust. Where, however, Canadian law is chosen and the trust is connected with both Ontario 435

M Lupoi, ibid., 47. Ibid., 47–9. 437 See further, M Lupoi, Trusts: A Comparative Study, ch. 7. It is, of course, not the ultimate domestic trust, which would be one where the applicable law was also Italian. See also Il Trust in Italia: Congresso Nazionale, Rome 19–20 November 1999. Many of the papers presented from this conference are available at http://www.il-trust-in-italia.it/webtutti/@Congresso% 201999/Congresso%201999%20tabella.htm. Important contributions include: D Hayton, “The ‘trusts interni’ seen from England”: http://www.il-trust-in-italia.it/webtutti/@Congress% 201999/Hayton.htm; M Lupoi, “I trusts interni”: http://www.il-trust-in-italia.it/web tutti/@Congress%201999/Lupoi.htm; G Palermo, “It trust di Diritto Interno” http://www.il-trustin-italia.it/webtutti/@Congress%201999/Palermo.htm. 438 At least insofar as a state such as Italy may continue to be described as a “non-trust state” (see Art. 2, above). 439 On which, see below. 440 And that, in the absence of choice, it is the territorial unit of closest connection for which one must seek under Art. 7. 441 This is the view of D Hayton, “The Hague Convention on the Law Applicable to Trusts and on their Recognition” (1987) 36 ICLQ 260, 269. 436

Choice of Law of a Recognised Legal System 185 and British Columbia, it may be impossible to construe the choice of law clause at all, and it will be necessary to determine the law of closest connection in accordance with Article 7.

6. THE NEED TO CHOOSE THE LAW OF A RECOGNISED LEGAL SYSTEM

The Convention does not make clear whether the settlor must choose the law of a recognised legal system. In the contract sphere, the question arises whether the settlor may instead select the lex mercatoria to govern the contract. In trusts, one might envisage a settlor stating e.g. that the trust should be interpreted “in accordance with the maxims of equity”.442 The “contrat sans loi” involves “the absence of . . . a rule or set of rules emanating from a sovereign authority and enforced by public authority”.443 It has little support.444 Lord Diplock stated in Amin Rasheed Shipping Corp v. Kuwait Insurance Co. that “contracts are incapable of existing in a legal vacuum. They . . . are devoid of all legal effects unless they were made by reference to some system of private law . . .”445 The Rome Convention is also thought to reject the lex mercatoria.446 However, anational law is permitted only if the forum so decrees. Hence, whether one regards it as incorporated by reference into national law or not, it nonetheless derives from sovereign authority. So, for example, as a matter of English law, section 46(1)(b), Arbitration Act 1996 states that an arbitration tribunal may decide a dispute “if the parties so agree, in accordance with such other considerations as are agreed by them or determined by the tribunal.” It is the forum which remains master of its private international law destiny.447 The same argument may be made in relation to trusts. Furthermore, if the settlor can, through the process of dépeçage, split the law applicable to different parts of the trust (as Article 9, Hague Convention envisages) so as effectively to apply the sum total of a number of national laws in such a way that they do not form the recognisable whole of any legal system, it is difficult to see why he must select anational law to govern the entire trust.

442 Whilst it appears undeniable that such common maxims do exist, it is equally clear that they would not, in and of themselves, be anything like sufficient to regulate the rights and obligations arising under an express trust. See R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 17–24; J McGhee, Snell’s Equity, 30th edn. (London, Sweet and Maxwell, 2000), ch. 3. 443 P Nygh, Autonomy in International Contracts, 173. 444 Ibid., 60–3 and ch. 8. 445 [1983] AC 50, 65. 446 P Lagarde “Le Nouveau Droit International Privé des Contrats” [1991] Revue Critique de Droit International Privé 288, 300–1; Dicey and Morris, 1223–1224; Cheshire and North, 559–560. Contrast Arts. 9(2) and 10, Inter-American Convention of the Law Applicable to International Contracts 1994 (“the Mexico Convention”), considered by P Nygh, Autonomy in International Contracts, 188. 447 See also J Harris, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247, 264–6.

186 Express or Implied Choice of Law In practice, however, it is very likely that such a choice would not be permitted. There is simply no body of internationally accepted rules of equity and trusts sufficient to resolve a dispute. Moreover, Article 17, (whose prime role is to exclude the doctrine of renvoi), appears to envisage that the rules of law of a “state” must be applied. On balance, it is suggested that the Convention is likely to be interpreted to require choice of a legal system. Any settlor who chooses the lex mercatoria risks finding his choice struck out altogether and the law of closest connection being applied.448

7. MULTIPLE SETTLEMENTS

A difficult question might arise if a settlor creates a trust by vesting assets in the hands of the trustee and then subsequently another person transfers assets to the trustee to be held under the same settlement.449 What if the original settlor chooses the law of Utopia to govern the trust but the subsequent transferor of assets to the trust chooses the law of Ruritania? Although Article 6 ostensibly allows the settlor to choose the governing law, Hayton convincingly argues that this must be taken to refer to the original settlor. Where a person transfers assets to be held on the terms of a pre-existing trust, he has no choice but to submit to the law already governing that trust.450 Of course, the subsequent disposition must still be “launched” by the rules which determine the transfer of assets to a trustee.451

8. CONTRACTS AND TRUSTS 452

A difficult question potentially arises where the trust’s origin is contractual. One way that this might arise is pursuant to a contractual agreement to settle property on trust. Suppose that S contractually agrees with B to settle property on B pursuant to a contract governed by Utopian law. The contract provides that the trust thereby established will be governed by Ruritanian law. Which law determines the relationship between S and B? The answer must be that if a claim is 448 It may be possible to incorporate by reference e.g. “the maxims of equity” into a trust otherwise governed by the trust law of a state whose trusts do not have their origin in the courts of Equity. However, that is a matter for the applicable law to determine, and it may take the view that some or all of those maxims cannot be accommodated within the applicable law’s legal framework. 449 D Hayton, (1987) 36 ICLQ 260, 270 gives the examples of a unit trust and of the establishment of a pension fund, friendly society or unincorporated association. 450 Ibid., 270. One might query whether the position might differ if the initial settlement consisted of a nominal sum located in e.g. England (and the court determined that the law of closest connection under Art. 7 was English law) and the same person then transferred a vast sum of money located in Civilopia into the trust (which, had it been invested initially, would have rendered Civilopia the law of closest connection). 451 On which, see Part One of this book. 452 See also the discussion of “trusts and the Rome Convention” in connection with Art. 2, above.

Contracts and Trusts 187 brought for failure to fulfil the contractual duty to create a trust, such claim is governed by the law applicable to the contract. But what if the contract itself contains terms and conditions relating to the trust to be established? Suppose that S contractually agrees to declare himself trustee. The contract specifies e.g. the investments which may be made and it is alleged that an unauthorised investment had been made. Which law should determine S’s liability? On the one hand, the terms are contractually agreed.453 On the other hand, they form part of a valid trust whose applicable law is not ordinarily determined by its genesis.454 In principle, since the alleged breach might at once be both a breach of contract and a breach of trust, it is arguable that the claimant should have the choice which action to bring and that he may accordingly choose to which law the claim will be subjected. Any other solution artificially distorts the characterisation process; and to prefer either the “trust” or the “contract” classification would appear somewhat arbitrary. A more difficult case still might arise if SB transfers assets to T, a professional trustee, to manage for SB’s benefit. Once again, imagine that the contract is governed by Utopian law; the trust is governed by Ruritanian law. Assuming that the trust is valid by its applicable law and not regarded as a sham, which law should regulate the relationship between SB and T? The relationship is again simultaneously one in contract and trusts. In a different context, where concurrent claims exist in principle in contract and tort (as may arise e.g. where an employee is injured at the workplace), it is sometimes argued that the choice of law rules regulating the “specific” relationship of the parties (that is, the contract) should apply to the exclusion of the “general” obligations imposed by the law of tort. It “proceeds on the principle that where a specific relationship exists between the parties regulating their rights and obligations towards each other (such as one based on contract or marriage), and in the course of carrying out the obligations of that relationship an obligation based on the general law (such as one arising out of the law of delict) is infringed, the claims arising out of the breadth of the general duty should be subject to the same law as that which governs their existing relationship”.455

But the problem is more complex here, since the contract and the express trust are both “specific” relationships and there is no obvious reason for favouring one over the other.456 That being so, if the claimant has concurrent claims in 453 The argument that the applicable law of a trust arising by contract (in the context of an alleged presumed resulting trust) should be determined by application of the Rome Convention’s choice of law in contract rules was raised, though not relied upon, by counsel in Lightning and Anr v. Lightning Electrical Contractors Ltd, judgment of 23 April 1998. Their Lordships did not rule on the point. 454 The trust must be “launched” by choice of law rules falling outside the scope of the Convention; but once it has validly been launched, the rocket in orbit is freed from the shackles of its launcher. 455 P Nygh, Autonomy in International Contracts, 240, basing his definition on that of M Keller and K Siehr, Allgemeine Lehren des Internationalen Privatrechts (Zurich, Schulthess, 1986), 285. 456 The “specific relationship” test is not one which this author favours even in the context of concurrent claims in contract and in tort: J Harris, “Contractual Freedom in the Conflict of Laws” (2000) OJLS 247, 266–8.

188 Express or Implied Choice of Law contract and trusts, he must be allowed to pursue whichever claim he so chooses. An action framed as a breach of contract claim would be subject to Utopian law; a breach of trust action would be subject to Ruritanian law. True, this creates the unappealing prospect that an identical action may at once be deemed a breach of contract but not a breach of trust, or vice-versa. It also loads the dice in favour of the claimant. But insofar as two different laws govern the contract and the trust, such awkward results cannot easily be avoided.

9. CONSENT TO THE CHOICE OF LAW CLAUSE

The Convention does not state what is to happen where there is a dispute as to whether a particular law is agreed upon. Suppose that a settlor vests legal title in property in a trustee, who has agreed to hold that property on trust. All objective factors relating to the trust point to Ruritanian law. Subsequently, the settlor communicates to the trustee in writing that he wishes the trust to be governed by Utopian law. The trustee does not respond to this statement. A dispute arises as to whether the trust is governed by Utopian or Ruritanian law, which in turn depends on the validity of the choice of law clause. Which law should determine whether the clause is valid? It might be argued that determination of whether a choice exists is a connecting factor (because it goes to the question of how much evidence of an alleged choice of law the forum requires before it will contemplate the application of the law allegedly chosen) and is, as such, to be determined by the law of the forum. Against this, the choice of law clause can be seen as a term of the trust itself;457 as such, its validity should be determined by the (putatively) applicable law of the trust, namely Utopian law. Of course, this raises the familiar objection that ascertaining the applicable law is the very issue which confronts the court and that any assumption as to what law might govern places the cart before the horse. But, pragmatically at least, the use of the putative applicable law is well established. Article 3(4), Rome Convention, for example, provides that “The existence and validity of the consent of the parties as to the choice of the applicable law shall be determined in accordance with Articles 8, 9 and 11”.458 In turn, Article 8(1) requires the validity of a given term to be determined by “the law which would govern it under the Convention if the . . . term were valid.” A further problem arises where the allegation is not merely that the choice of law clause is void, but that the entire trust (of which the choice of law clause is a part) is void. Where the parties dispute the validity of the trust, but accept that, should it be held valid, so too is the choice of law clause, little harm can be done by use of the putative applicable law. But where they dispute both the validity of the trust and the choice of law clause within the trust, the theoretical objections 457 458

Or, at the very least, parasitic on the trust itself: ibid., 252–5. See also R Morse, (1982) 2 YEL 107, 119.

Consent to the Choice of Law Clause 189 to use of the law which would govern if the trust were valid are especially potent. Where a party asserts that Utopian law was chosen to govern a trust, it is very likely that he will do so knowing that Utopian law regards such a choice as having validly been made. To test the validity of the choice of law clause by Utopian law might be said to be insufficiently neutral between the parties to the litigation. This is because it involves starting from the premise that the party asserting the validity of the clause is correct. Such an approach is especially unattractive where the other party alleges that the entire trust is invalid.459 How then should the validity of a choice of law clause be ascertained? In the first place, the law of the forum should determine whether the clause in question is, properly construed, a choice of law. The law of the forum should also determine, according to its civil standard of proof,460 whether there appears to be a validly agreed trust containing that choice of law clause. Of course, this requires a rather detailed investigation by the law of the forum into the validity of the choice of law clause and the trust itself. But without it, even-handedness between the parties cannot be achieved. Why should a party alleging that a Utopian choice of law governs a trust be permitted to have the question of agreement to that clause subjected to Utopian law, without offering substantial evidence that, by the law of the forum, a valid choice has indeed been made? Until he has done this, he does not obviously have the right to have the question subjected to Utopian law at all. Once a party has shown that, by the civil standard of proof of the law of the forum, a choice of law clause has been agreed according to which the law of Utopia is applicable, then and only then should the matter of the validity of that choice of law clause be determined, finally, by Utopian law. If the forum takes the view that there appears to be a valid trust governed by Utopian law, it is no longer unfair to the party contesting the validity of the choice of Utopian law to subject the clause’s validity to Utopian law. Nor is it objectionable if it turns out that the trust (putatively containing the Utopian law clause) was invalid by Utopian law.461 Utopian law was applied because the law of the forum determined that it was the law which had the best claim finally to determine the validity of the trust. So much for the theory. In practice, however, what is a court likely to do when faced with a question of the validity of a choice of law clause, in the absence of an answer in the Hague Convention itself? Pragmatically 462 it is likely that the validity of the clause will be immediately subjected to the putative applicable law, without an initial stage of requiring substantial evidence of its validity by the law of the forum. Article 8, Hague Convention adopts the bootstraps approach of using the putatively applicable law to determine the 459

See also J Harris, (2000) 20 OJLS 247, 252–5. It is suggested that the balance of probabilities should be the requisite standard, so that the court is seen to act fairly as between the parties. 461 So that it did not, after all, contain a valid Utopian choice of law clause. 462 Although some might say, intellectually depressingly. 460

190 Express or Implied Choice of Law validity of the entire trust.463 If that larger question can be resolved by applying a law which, if the trust turns out to be invalid, ex hypothesi was not agreed upon (and which the law of the forum did not provisionally declare valid), then one might expect that the validity of the terms of the trust will also be subjected to the putatively applicable law. That means that, if a dispute arises as to whether a trust contains a Utopian choice of law, a court would apply Utopian law in order to decide whether the choice itself was valid.

10. CHOICE OF A LEGAL SYSTEM MINUS ITS MANDATORY RULES

Can the mandatory rules of the law applicable to the validity of the trust be avoided by the settlor splitting the law applicable to different parts of the trust or expressly stating that “this trust shall be subject to the laws of Utopia, except its mandatory rules” ? 464 In contract, Nygh suggests that such freedom is an ultimate extension of the autonomy principle and asks “why the parties . . . should be bound by the mandatory rules in the international sense of any State other than those of the forum, if in their contract they have made it clear that they wish to make other arrangements”.465 Does the applicable foreign law have a legitimate interest in the application of its mandatory rules in the courts of another country? The exclusion of renvoi in trusts 466 (as in contract)467 suggests that there is no attempt to apply the governing laws as a judge in that foreign state would do. Moreover, the choice will still be subject to the mandatory rule and public policy provisions of the Convention.468 If the settlor can split the trust in such a way that the sum total represents a law such as would be applied domestically in no court in the world, why should he not also be able to evade the applicable law’s mandatory rules more directly by their express exclusion? One answer might be that failure to apply the mandatory rules of the governing law can open the door to forum shoppers. A more contextual answer is that, if it is true to say that the Hague Convention requires the parties to choose one or more national legal systems to govern the trust, rather than anational law,469 and if it does not allow that law to be frozen in time,470 it is very unlikely that it would allow the applicable law to 463 Art. 8(1), Hague Convention states that “the law specified by Article 6 or 7 shall govern the validity of the trust. . . .” See further the discussion of “formation and essential validity of the trust” in relation to Art. 8, below. 464 See Kahler v. Midland Bank [1950] AC 24. 465 P Nygh, Autonomy in International Contracts, 214–5; see also J Harris, (2000) 20 OJLS 247, 263–4. 466 Art. 17, Hague Convention. 467 Art. 15, Rome Convention. 468 Arts. 15, 16 and 18, Hague Convention. 469 On which, see the discussion of “the need to choose the law of a recognised legal system”, above. 470 On which, see the discussion of “time for ascertaining a choice of law”, below.

Implied Choice of Law 191 be distorted by a clause excluding its mandatory rules. Accordingly, it is suggested that such a clause would be treated as ineffective.

11. IMPLIED CHOICE OF LAW

(A) Nature of Implied Choice Where no express choice of law clause is contained in the trust instrument, the applicable law will be that which the settlor impliedly chose, if any. “The implied choice must derive from the terms of the instrument creating, or the writing evidencing, the trust, interpreted as necessary in the light of the circumstances of the case”.471

It is quite distinct from the objective test of closest connection (determined by Article 7), which comes into operation only if no express or implied choice can be ascertained. Accordingly (and in contrast to the law of closest connection) the factors to which regard should be had in ascertaining whether the settlor impliedly made a choice of law are not those factual matters which determine the centre of gravity of the trust. Objective factors may tell us little, or nothing, about what was in the settlor’s mind. It is interesting to note the difference in wording between Article 6 and the reference to implied choice of law under the Rome Convention. Article 3(1), Rome Convention requires a choice of law in contract to be “demonstrated with reasonable certainty by the terms of the contract or the circumstances of the case”.472 As it stands, the absence of a “reasonable certainty” or similar requirement in Article 6, Hague Trusts Convention allows a court to apply a weaker standard in inferring a choice than is used in contract. But, retorts von Overbeck, “it was pointed out that it was not absolutely necessary to conform, where trusts were involved, to the Rome Convention’s formulations and that if harmonisation was being sought it was better to draw closer to the texts of other Hague Conventions”.473

Indeed, writing elsewhere von Overbeck has stated that “The Convention recognises more liberally an implied choice than the European Union Convention of 19th June 1980 on the Law Applicable to Contractual 471

Von Overbeck Report, para. 64, p. 384. The Rome Convention is, in turn, substantially more liberal than the Hague Convention on the Law Applicable to Contracts for the Sale of Goods 1955, Art. 2 of which requires an inferred choice to be determined “beyond doubt” from the contract and its terms. Art. 7(1), Hague Convention on the Law Applicable to the Sale of Goods 1986 requires a choice to be “clearly demonstrated by the terms of the contract and the conduct of the parties viewed in their entirety” (emphasis added). See further P Nygh, Autonomy in International Contracts, 109–11. 473 Von Overbeck, para. 68, p. 385. The particular analogy which von Overbeck has in mind is to the [then draft] Hague Convention on Sale of Goods 1986 (on which see the previous footnote). 472

192 Express or Implied Choice of Law Obligations . . .”474 However, it might legitimately be asked why it should be easier to infer a choice of law applicable to a trust than to a contract. Nor is it made plain in the von Overbeck Report whether the search must be limited to a genuine choice made by the settlor,475 as opposed to the imputing to the settlor of a choice which he might have made, had he considered the matter, but which he did not in fact consciously make. Indeed, Hayton argues that the difference between “subjective implied choice of law under Article 6 and the objective applicable law under Article 7 is obviously a question of degree”.476 If that were the case, then it may not in reality matter much whether the choice is deemed to be inferred by Article 6, or the law of closest connection determined by Article 7. However, there are reasons to think that the distinction between implied choice and objective applicable law is important under the Hague Convention. For one thing, a hierarchical mechanism exists for ascertaining the applicable law under Article 7 in the absence of choice, firmly based on considerations of geography and convenience. Distinguishing between an implied choice and the law of closest connection in the absence of choice may prove crucial. Indeed, whether the applicable law in the absence of an express choice is determined in accordance with Article 6 or 7 may, in extreme cases, determine the validity of the entire trust. Imagine a trust instrument made by a settlor domiciled in Civilopia, creating a trust for the benefit of Civilopian domiciliaries. The trustees are also Civilopian domiciliaries and the trust property located in Civilopia. However, the trust is in English form and contains a clause providing for the exclusive jurisdiction of the English courts. If factors relating to e.g. the parties, or the situs of the property are irrelevant, or of relatively little importance, to the implied choice under Article 6, it may be possible to infer that such a choice has been made and to hold that the trust is governed by English law; if no implied choice is found, it is possible that an English court might find that the law of closest connection of the trust is Civilopian law, under Article 7. Since this is the law of a “non-trust” country, the entire Convention may be found to be inapplicable.477 Furthermore, in the Rome Convention context, there is a conceptual distinction made between a choice which the parties actually made (albeit that they did not include a choice of law clause) and identifying the law of closest connection in the absence of choice. The relevant factors in inferring a choice in the contractual setting are much more limited than those to which regard may be had 474 A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6, 7. 475 M Giuliano and P Lagarde, “Council Report on the Convention on the Law Applicable to Contractual Obligations” OJ 1980 C282/4, pp. 15–7 (“the Giuliano and Lagarde Report”); Hellenic Steel Co v. Svolamar Shipping Co Ltd: the Komninos S [1991] 1 Lloyd’s Rep 370. 476 D Hayton, (1987) 36 ICLQ 260, 271. 477 Pursuant to Art. 5 (on which, see above). Even where a choice of English law is inferred, a contracting state could still refuse to recognise the trust if the law of closest connection is that of a nontrust state, pursuant to Art. 13. Art. 13 has not, however, been enacted in the United Kingdom.

Implied Choice of Law 193 in ascertaining the objective applicable law.478 In particular, factors which go merely to identifying the law of closest connection, such as the place of contracting, the residence of the parties or place of performance would not be relevant to inferring a choice. Giuliano and Lagarde give examples of the sorts of factors from which an implied intention can be found.479 The inclusion of a jurisdiction or arbitration clause might prove an especially strong indicator of the applicable law: after all, the court identified will be manifestly better placed to apply its own domestic law than a foreign law. Use of terminology resonant of a particular legal system would be another relevant factor, as would a standard form contract typically governed by a certain law.480 Inferences might also be drawn from the circumstances of the case, e.g. where there is a related transaction between the parties subject to a choice of law clause, or a previous course of dealing. But perhaps more revealing are the factors to which no reference is made as indicators from which a choice may be inferred. These include such factors as the place of conclusion of the contract, currency, place of delivery and the domicile or residence of the parties. If the search is for a genuine choice made by the parties, then it might be said that such factors cannot be used to demonstrate that a particular law was intended to govern. On the other hand, such objective factors might be relevant more negatively to rebut an inference that a particular law was chosen by the settlor. When a number of “subjective” factors point to a choice of a particular law (e.g. Utopian law) but the “objective” factors strongly point to another law (e.g. Ruritanian law), there may be cause to doubt whether the choice of Utopian law was really made at all.481 In Egon Oldendorff v. Liberia Corp,482 Mance J, when considering whether to permit service out of the jurisdiction, was confronted with a clause providing for arbitration in London in a case where a number of objective factors pointed to Japan. Whilst he found that there was a good arguable case that English law governed, and that the factors pointing to Japan were insufficient to indicate otherwise, he did appear to think such factors relevant in principle. The position under the Rome Convention accordingly seems to be that the search for an implied (conscious) choice is limited to a relatively small number of subjective factors. Objective factors are not capable of demonstrating, to the requisite standard of reasonable certainty, that a choice of law was actually made by the parties. However, objective factors may exceptionally be relevant to negative the existence of an implied choice, if the strength of connection to another law is so strong that it casts doubt on whether another law was really agreed upon. 478

Giuliano and Lagarde Report, OJ 1980 C282/4, p. 17. Ibid., 17. 480 Compare the common law position in Amin Rasheed Shipping Corp v. Kuwait Insurance Co [1984] AC 50; Egon Oldendorff v. Liberia Corp [1995] 2 Lloyd’s Rep 6; [1996] 1 Lloyd’s Rep 380. 481 Though one might say that the reverse is true and that the “subjective” factors should be taken to indicate that the one law which the parties did not want to govern the contract was that with which it is objectively most closely connected. 482 [1995] 2 Lloyd’s Rep 64, 69. 479

194 Express or Implied Choice of Law It is arguable that a similar principle should be applied to trusts. If the search is for an implied, conscious choice of the settlor, such intent cannot be ascertained by looking at objective factors of close connection. The fact that the “reasonable certainty” test is not included in the Convention should rather be treated as appertaining only to the standard of proof necessary to demonstrate that choice. That said, there is no obvious principled reason why the standard required to infer such a choice should be weaker than in the contractual context. Respect for party autonomy is enshrined in the Hague Convention, but it is difficult to see that it should be supported more strongly even than in contract law. The typical trust requires reference not just to the settlor’s intentions, but to the trustee’s acceptance of the intended terms of that trust;483 it has some of the elements of bilateral agreement also present in the contract. Moreover, since many trusts are testamentary, it is intrinsically more likely in trusts than in contract that the settlor will not be alive and able to be questioned as to his intent, which might lead one to be cautious about attributing intentions to him.484 On balance, it is suggested that the same high standard of proof that an implied choice of applicable law was made should be required in trusts as is required in contract by the Rome Convention. Perhaps the only exception would be where the trust does not require a[nother] trustee to agree to act on the terms of the trust. If a settlor declares himself sole trustee of his property, then the fear of holding the trustee to the terms of a trust which he did not accept disappears and the only concern is to ascertain what the settlor intended when he created the trust. The trust is, in this situation, markedly different to a contract, in that it is unilaterally brought into effect by the settlor. Some attempt to give effect to the settlor’s intention might be thought better than none.

(B) Implied Choice and the Law of Closest Connection Of course, too liberal a test creates a real danger that the finding by the court of an implied choice will be nothing but a legal fiction. Indeed, the very fact that there is a dispute as to the applicable law before a court may itself be indicative that no clear choice was made. In the contractual context, Stephenson LJ observed that, “If [the parties] . . . had been asked to agree on the application of the other’s law each would probably have refused and there would have been no contract: but there is a contract”.485 483 Although a trust may not fail for want of a trustee, somebody needs to be appointed trustee who is willing to agree to act on the terms that the settlor wishes to specify (which will include the settlor’s wishes as to what law will govern the trust). 484 Though this argument could be turned on its head. It could be said that, in the case of a deceased testator, one cannot often be “reasonably certain” what was in his mind and that a weaker standard of proof should be accepted. 485 Coast Lines Ltd v. Hudig & Veder Chartering NV [1972] 2 QB 34, 50.

Implied Choice of Law 195 The dividing line between implied intention and law of closest connection can be very thin.486 It is possible to discern three possible meanings of the phrase “implied intention of the settlor”: first, that an actual intention was formed by the settlor as to the governing law; secondly, that the settlor either had formed an intention as to the governing law or would have formed such an intention had he thought about the issue; or thirdly, that a reasonable man would have intended the trust to be governed by a certain law. The third meaning is rather close to the law of closest connection test. More fundamentally, it could be argued that a choice of law approach which distinguishes between express choice, implied choice and the law of closest connection is unnecessary and unworkable, since: “Either the parties have expressed a choice, or they have not”.487 A simple distinction could be drawn between cases where a genuine choice was made (whether by choice of law clause or otherwise) and where it was not. In that scenario, the “implied choice” category would disappear. The High Court of Australia in Akai Pty Ltd v. The People’s Insurance Co. Ltd 488 has endorsed the view that ultimately, the question is simply whether a choice was made or not. As Nygh argues, the consequence of a simple two-tiered approach would be that cases falling within the former category must represent a genuine choice by the parties; there might accordingly be less scope to rely on certain key factors, such as a jurisdiction clause, as mechanical factors from which an inference of the applicable law may be drawn.489 Nevertheless, for better or for worse the three tiered system prevails under the Hague Trusts Convention and the implied intention category survives. That means that an inference as to the intended applicable law may be drawn from certain typical factors. However, a further question arises as to exactly how many factors may be relevant to the search for an implied choice and how much importance to attribute to them. Would attention be limited, for example, to factors such as a jurisdiction clause, use of terminology peculiar to a particular legal system etc; or would it also allow consideration of e.g. the situs of the property, the place of administration etc? The difficulty is that the more factors considered and the more relevance attached to them, the more the line between an implied choice and the law of closest connection in the absence of choice is likely to be blurred.

486 See e.g. the judgments of Lords Diplock and Wilberforce in Amin Rasheed Corp v. Kuwait Insurance Co. [1984] AC 50, 85, discussed in P Nygh, Autonomy in International Contracts, 107. See also Akai Pty Ltd v. The People’s Insurance Co. Ltd (1996) 141 ALR 374. 487 P Nygh ibid., 107. See also P North, Private International Law Problems in Common Law Jurisdictions (the Hague, Martinus Nijhoff, 1993), 106. 488 (1996) 141 ALR 374. 489 P Nygh, Autonomy in International Contracts, 109.

196 Express or Implied Choice of Law

(C) Implied Choice and “The Circumstances of the Case” One possible distinction between the Convention and the common law approach considered below is that, in the former case, the choice must be “implied in the terms of the instrument creating or the writing evidencing the trust, interpreted, if necessary, in the light of the circumstances of the case”.490 The court should, if possible, confine itself to the trust instrument which, in any event, is the first point of reference in the search for an implied choice. That appears to mean that factors appertaining to the trust which are not mentioned in the trust instrument should be disregarded. So, whilst the use of terms resonant of a particular legal system, or the specification of a place of administration or choice of court clause would be relevant, the fact that e.g. the trust was created in a particular state, or that the trust deed is held in a particular state (not specified in the deed) would not appear to be factors which can provide any evidence of implied intention under the Convention. Extraneous factors may become relevant, but only insofar as they shed light on terms specified in the trust instrument itself. Even then, extraneous circumstances should only be referred to “if necessary.” This is “intended to express the idea that circumstances should only be consulted if the writings do not allow for a clear answer”.491 This appears to be a materially different approach to that taken in the Rome Convention context. There, Giuliano and Lagarde state that factors relevant to inferring a choice of law include related transactions between the parties, or a previous course of dealing.492 If for example, a settlor has previously settled property on the same trustee for the same beneficiaries, and that earlier settlement contained a choice of law clause, one might expect such a factor to be of great importance in ascertaining the applicable law in a later settlement that contains no such clause. But such previous dealings do not, strictly speaking, shed light on the terms of the trust themselves and it may be that they should be disregarded under Article 6. However, such an approach has very little to commend it in principle. If the search is for a genuine choice by the settlor of an applicable law, then exclusion of a potentially powerful indicator such as a previous settlement involving the same parties is positively obstructive. What is less obvious is whether the “circumstances of the case” can be used negatively to demonstrate that what appears from the terms of the trust to amount to an implied choice of law is not in fact sufficient when viewed in the context of extrinsic factors which do point towards a different law. An English jurisdiction clause in the trust instrument, for example, might indicate an implied choice of English law to govern a trust; but would the fact that the settlor had orally directed that the property should be invested exclusively in 490

Art. 6(1) (emphasis added). Von Overbeck Report, para. 69, p. 385; see also Report of the Special Commission, nos. 58 and 59. 492 Giuliano and Lagarde Report, OJ 1980 C282/4, 17. 491

Implied Choice of Law 197 Utopia be permitted to cast doubt on the matter? The question is not easy to answer: on the one hand, if there is genuine evidence to cast doubt on the apparent implied choice, it is difficult to turn a blind eye to it and proceed to affirm that an implied choice had been made; on the other hand, the Convention itself, partly for reasons of proof, limits itself to trusts created or evidenced in writing; and the introduction of evidence as to terms of that written trust which are themselves not evidenced in writing (or other extraneous factors) might be thought rather to undermine the functioning of the Convention. Von Overbeck comments that circumstances of the case may not be used to identify an implied intention “while completely ignoring the writings”;493 but equally, it is difficult to see how the writings can be used to infer an intention whilst completely ignoring extraneous circumstances. Ultimately, the significance of the written terms may often only be assessed accurately by considering their context; and if that context shows a number of external factors pointing to another law, those are “circumstances of the case” which should not be ignored. Accordingly, it is submitted that the search for an implied choice under Article 6 should not be limited strictly to the terms stated in writing, but considered in the light of all other extraneous factors, including the settlor’s purpose and any orally agreed terms of the trust. As Hayton says, “Courts tend to strive towards an applicable law that will give effect to, and not frustrate, the intentions of the settlor;”494 those intentions can only be ascertained by considering all the circumstances leading to the creation of the trust. In any event, the United Kingdom has extended the Hague Convention’s provisions beyond the scope of trusts evidenced in writing, to include “any other trusts of property arising under the law of any part of the United Kingdom or by virtue of a judicial decision whether in the United Kingdom or elsewhere”.495 In that context, there may be no writing evidencing the trust at all; yet, strictly speaking, the restrictions in Article 6 on the ascertaining of the applicable law still apply. The United Kingdom ought to have amended Article 6 to reflect its extension of the Convention rules beyond voluntary trusts evidenced in writing. In the absence of such amendment, a teleological construction of Article 6 is called for. Can it really be said that, in the case of, for example, an oral express trust allegedly governed by English law,496 no express or implied choice can be found, because the choice is not contained in writing? Such a result would be patently absurd, since it would lead to all such trusts being determined by the applicable law in the absence of choice. Rather, an express or implied choice should be sought from the oral declarations of the settlor, or from any relevant surrounding circumstances which demonstrate what the settlor’s intentions

493

Von Overbeck Report, para. 69, p. 385. D Hayton, (1987) 36 ICLQ 260, 271. 495 S. 1(2), Recognition of Trusts Act 1987 (see the discussion of “the statutory extension of the Convention rules in the United Kingdom” in connection with Art. 2, above). 496 Which will fall within the extended scope of the Convention in the United Kingdom. 494

198 Express or Implied Choice of Law were. Providing that these circumstances are not used to impute to the settlor an intention which he did not really have, the case for their consideration by a court is extremely strong.

(D) The Settlor’s Purpose—Implied Choice of a Law by which the Trust is Valid? Enquiry into the settlor’s purpose in creating the trust is not explicitly part of the search for his intentions. Indeed, a proposal to include a reference to such purpose was rejected.497 Nevertheless, if the function of Article 6 is to give effect to the settlor’s intentions, it is difficult to see how enquiry into the nature and purpose of the trusts which he sought to create can be excluded. Most fundamentally, of course, the settlor can be taken to have intended to desire the application of a law under which the trust is valid. Presumably, the settlor’s purpose can be considered as one of the “circumstances of the case” relevant to shedding light on the construction of the written trust document. However, such a purpose cannot itself be conclusive as to the intended applicable law, since Article 6 requires that the choice derives from the written instrument itself, as interpreted by reference to the surrounding factors.498

12. COMMON LAW AUTHORITIES ON IMPLIED CHOICE AND THE OBJECTIVE PROPER LAW

(A) Relevance In the absence of any further guidance in the Convention or von Overbeck Report as to how to ascertain an implied choice, the inclination of an English court to turn to common law case law will be strong. At common law, a number of factors have proved relevant in ascertaining the proper law in the absence of choice.499 Unfortunately, the cases do not clearly distinguish between factors relevant to inferring a choice of law and factors relevant to ascertaining an objective proper law in the absence of an implied choice. This becomes evident when one looks at the factors to which importance was attached in the search for implied intention. We shall now turn to these cases. 497

Working Document No 11; discussed in the von Overbeck Report, para. 67, p. 384. And not vice-versa. P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn. (London, Key Haven, 1993) 43, state that extrinsic evidence should only be admitted if the trust contains an ambiguity, citing the Jersey cases of Re Wardlow-Milne dec’d (1970) JJ 1539 and re Matthews dec’d (1980) JJ 139 in support. 499 Compare the position in America, discussed by E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1031–4; see e.g. Hutchison v. Ross 262 NY 381, 187 NE 65 (1933); Shannon v. Irving 275 NY 95, 9 NE 2d 792 (1937); Lewis v. Hanson 36 Del CH 235, 128 A 2d 819 (1957); Ford v. Newman 77 Ill 2d 335, 396 NE 2d 539 (1979). 498

Common Law Authorities 199

(B) Testamentary Trusts Cheshire and North state that it was assumed at common law that testamentary trusts were intended to be governed by the law of the testator’s domicile.500 Some of the cases cited by the authors in reliance are more convincing than others. In Peillon v. Brooking,501 a testator left a bequest to his daughter “free from the control of her . . . husband, without power of alienation or anticipation.” The daughter and her husband had been domiciled in France, but subsequently acquired a Sardinian domicile. Upon separation, a deed in English form was executed by which the daughter purported to settle an income on her husband and their children out of the property held under her father’s will. Evidence was brought that by Sardinian law the restriction on alienation in the will could be disregarded. Nevertheless, Romilly MR applied the law governing the will trust, which he assumed to be that of the testator’s domicile, to what he characterised as an issue relating to the interest created in the daughter by the will.502 However, the case contains little discussion of party intention or of the distinction between the will itself and the trust created therein. Similarly, in Re Afanoor’s Trusts,503 Romer J said, in connection with a dispute about beneficial entitlement under a will, that “English law adopts the law of the domicil in a case like this”.504 But again, there was no real search for implied intention, or considered separation of the choice of law applicable to the trust and to the will. In Re Lord Cable,505 a testator domiciled in India left a will in English form establishing a trust of which four of the five trustees were English residents. Most of the trust assets were initially invested in British Government securities prior to exchange control legislation entering into force in India. Slade J nonetheless remarked that: “It seems to me that there is nothing to displace the presumption that . . . [the testator] intended Indian law, the law of his domicile, to be the governing law of the will for all purposes in relation to his movable property”.506

This constitutes a much clearer reference to presumed intention than anything found in the aforementioned cases. However, no authority is cited by Slade J in support of this proposition. Moreover, the testator had expressly declared his domicile in the will. It appears that it was this step which led the judge to 500

Cheshire and North, 1035. (1858) 25 Beav 218. 502 This is not an obviously correct characterisation. True, the alleged restriction on alienation derived from a will; but the issue in the case was the validity of an inter vivos settlement. 503 (1895) 64 LJ Ch 521. 504 Ibid., 523. A specific problem arose in the case due to a change in the law between the time of death and the bringing of the case. Romer J ruled that the law of the domicile should be applied as it stood at the testator’s death. 505 [1977] 1 WLR 7. 506 Ibid., 20. 501

200 Express or Implied Choice of Law “presume” that the testator’s personal law should apply; the case is not a clear authority for the proposition that such a presumption of an intention to subject a will trust to the law of one’s domicile applies generally to will trusts which contain no choice of law. Nor does the judgment sit easily with the case of A-G v. Campbell,507 which had been cited in argument. In that case, the testator was domiciled in Portugal but made his will in English form. He directed his executors to sell property in Portugal and invest the proceeds in England in order to pay a life annuity to his sister (resident in Ireland), remainder to his children. Lord Westbury saw the importance of distinguishing between questions going to succession and administration of the estate on the one hand, and the interests created under the trust arising under the will on the other hand.508 “There is, therefore, a settlement made of the property which is brought into this country and invested here in such mode of investment as gives to the property whilst it remains here the character of English property in respect of its locality. That settlement so made, undoubtedly becomes subject to the rules of English law under which it is held, by virtue of which it is enjoyed and under which it will be ultimately administered . . . [I]f a party taking this distributive part comes to this country and invests it upon trusts, it assumes the character of a British settlement and British property”.509

This represents a clear preference for a law which is both the situs of the assets and the place of administration. But the language of implied intention is not used. If anything, the tone of the judgment is that such a trust “belongs” to English law. The court appeared influenced by the fact that the testator’s property had been sold and the proceeds diverted to England; thenceforth, the trust of the proceeds took on a quasi-inter vivos characteristic. More recently, in Saliba v. Falzon,510 the Supreme Court of New South Wales dealt with a will made in Malta by an English domiciliary. Probate had been granted in the New South Wales court. The English High Court had granted letters of administration in respect of the same will.511 The will appointed the testator’s daughter as “universal heir”, but went on to provide that she was not fully entitled to the property until she attained the age of forty. One problem which arose was the differing laws on accumulations in England and New South Wales. The period of accumulation exceeded that permitted in English law, but not New South Wales law.512 The court considered a number of common law decisions on the law applicable to testamentary trusts.513 However, Young J remarked that: 507

(1872) LR 5 HL 524. In modern terminology, the difference between the “rocket-launcher” and the “rocket”. 509 (1872) LR 5 HL 524, 530–1. 510 [1998] NSWSC 302. 511 Administration of the will was still in progress. 512 Implicitly, the court treated the accumulation issue as a “rocket” matter affecting the trust itself. Compare the discussion of the rule against perpetuities in respect of Art. 4, above and section 8 of Part One of this book. 513 Jewish National Fund Inc v. Royal Trust Co. (1965) 53 DLR (2d) 577; Fordyce v. Bridges (1848) 2 Ph 497; Re Mitchner [1922] St R Qd 39; Public Curator of Queensland v. The Union Trustee Co of Australia Ltd (1922) 32 CLR 66; Re Mitchner (No 2) [1922] St R Qd 252. 508

Common Law Authorities 201 “. . . nowadays courts should not be put off by the novelty of propositions or that such propositions might conflict with what was said in a decision of only persuasive authority 60 years or more ago. In many cases the decision made 60 years ago or so was made for social reasons which were relevant at the time, but which are no longer relevant. In the administration of trusts and private international law things have moved swiftly. . . . It seems to me that it is now sensible to adopt the rule . . . that prima facie a sub-trust in a will which is to be administered in a particular jurisdiction and which the testator has shown he chooses it to be administered according to the law of that jurisdiction, is to be administered by the law of that jurisdiction . . . It follows . . . that the court of the domicile will now recognize this and not lightly declare the trust invalid because it might so be regarded by the law of the domicile. . . .”514

Young J went on to rule that the testator had impliedly chosen New South Wales law to govern the trust and that no problem of accumulation accordingly arose. Together, the cases do not seem to amount to anything like conclusive authority to the effect that a settlor may be taken impliedly to have intended the law of his domicile to govern. Nor is it obvious why it should. Where a testator leaves specific directions that his assets should be invested in another state, where they should be administered for the beneficiaries, the inference may indeed be quite the contrary. Re Lord Cable was a case where the testator drew attention to his domicile; that in itself told the court something of his intentions. But the other cases do not support an inference from the testator’s domicile as to his intentions. It is suggested that domicile should only become relevant when one turns to consider the law of closest connection to a testamentary trust in the absence of choice.515 For Article 6 purposes, it should not routinely be regarded as a part of the search for an implied intention.

(C) Inter Vivos Trusts When one turns to inter vivos trusts, there appears to be no one factor which has been consistently relied upon by the courts in seeking to determine the settlor’s implied intention. Once again, however, consideration of the case law is complicated by the frequent omission516 to distinguish between those factors relevant to ascertaining an implied intention and those which demonstrate the objective proper law. 514

Young J endorsed the view of P Nygh, Conflict of Laws in Australia, 6th edn., 569–70. Although, as we shall see below, the law of the settlor’s domicile or residence is not included in the (non-exhaustive) hierarchical list of relevant factors for obtaining the law of closest connection under Art. 7. 516 The word “omission” should not be taken to imply criticism. Many common law cases do not explicitly distinguish, where no express choice of law is made, between implied intention and the objective proper law. It is simply that a distinction between implied choice under Art. 6 and the law of closest connection under Art. 7 needs to be drawn for Convention purposes and the factors relevant to each may differ. Ambiguity in common law case law means that they provide correspondingly less assistance in determining which matters are relevant to each method of ascertaining the applicable law. 515

202 Express or Implied Choice of Law A particularly interesting case in point is the Irish decision in Revenue Commissioners v. Pelly.517 In that case, an Irish domiciliary died in England leaving a will consisting of property principally located in England, with the remainder located in Ireland. He bequeathed his residuary estate to his executrix. Following intervention by the testator’s widow, an agreement was entered into in the High Court in England whereby the executrix would create an inter vivos trust for the widow, part of which consisted of a life interest, with remainder to the widow’s children, who were minors domiciled in Ireland. The trust deed was prepared by English solicitors and stamped in England. Two trustees, one English and the other Irish, were appointed. The assets were duly invested in England and the deed of settlement lodged with an English bank. When the widow died, an application was made by the Revenue Commissioners in the Irish Court for the trustees to provide an account for succession duty. This could be required only if the children’s succession was governed by Irish law. Maguire P had “no difficulty in drawing the inference that the intention of the parties was that the rights created by the deed . . . were to be regulated by the laws of England”.518 Interestingly, the judge attached particular significance to the fact that the settlement was made pursuant to an agreement reached in the English High Court. “Unless it were expressly otherwise provided it seems to me to follow that it was the intention of the parties that any questions . . . should be decided in the English courts and in accordance with English law”.519

Presumably, the same would apply to a trust deed which contained an English jurisdiction clause. O’Byrne J also saw the search as one for the parties’ intentions and was likewise influenced especially by the English court’s jurisdiction over the settlement.520 Maguire P also pointed out that the widow was resident in England, the bulk of the assets were in England and the deed was prepared by English solicitors.521 This last factor was also considered important in another Irish decision, Cloncurry’s Case.522 But Maguire P made clear that such factors were to be accorded less weight than the submission of the executrix and widow to the English jurisdiction. Maguire P indicated that factors pointing to Ireland, such as the domicile of the two children and the fact that one of the trustees was Irish, were relevant, but not weighty enough to displace the inference that English law applied. The interesting point is that such ostensibly “objective” factors were nonetheless regarded by the judge as relevant to the ascertaining of intention. This merging of implied intention and objective proper law is especially evident in the judgment of Hanna J. He stated that the determining of the proper law 517 518 519 520 521 522

[1940] IR 122; see also W Binchy, Irish Conflict of Laws, 509. Ibid., 127. Ibid., 127. Ibid., 129–31. Ibid., 128. [1932] IR 687; see also W Binchy, Irish Conflict of Laws, 509.

Common Law Authorities 203 “resolves itself into a question of a legal balance sheet, to determine on which side the preponderance of the evidence lies.” He went on to look at the domicile of the testator, settlor, trustees and beneficiaries, the situs of the assets, the English proceedings, the use of English solicitors, the stamping of the deed in England and its lodging with an English bank. “On these facts weighed against each other,523 I am of opinion that the intention of the parties524 was that it should be considered as an English settlement”.525 Such an approach merges the objective proper law and implied intention into one category. It suggests that an almost limitless range of factors may be considered in the quest for implied intention. Harris Investment Ltd v. Smith 526 is cited by Cheshire and North as a case where the place of making the original deed was considered relevant to ascertaining the settlor’s intentions.527 In that case, a contract in deed form was made in British Columbia and British Columbian law was held to be that which was intended to govern the trust. However, MacDonald CJBC also emphasised that the property which had been mortgaged was also in British Columbia, where most of the obligations under the trust were to be performed and where the mortgagee was resident. Although the trustees named were resident in Oregon “there is nothing in the deed to particularly indicate that Oregon law is to be resorted to by the trustees.”528 The case is a rather weak authority on the relevance of the place of the creation of the deed; it is arguably a rather stronger authority for the relatively minor importance to be attached to the residence of the trustees, at least where other factors point to a different law. In A-G v. Jewish Colonization Association529 the settlor, an Austrian domiciliary, executed a deed in English language and form under which movable assets were to be used to benefit Russian Jews. The assets were transferred to a limited company which he formed for this purpose. The company was incorporated in England, although all its business was transacted in Paris. The company memorandum contained a clause specifying that, in the event of it being wound up, the assets should be transferred to an institution fulfilling similar objects and that, in default of selection, the English High Court had jurisdiction to make such a selection. At the time of the settlor’s death, only a minority of the assets was situated in England. When the liability of the company to succession duty arose, the proper law of the trust became important. In its search for the implied intention of the settlor on this matter, the Court of Appeal attached particular importance to the trustee of the settlement. Collins LJ commented that, “I think it is clear beyond possibility of doubt that the company was not 523

The judge gives no indication of the relevant weight to be accorded to each factor in the judg-

ment. 524 525 526 527 528 529

Emphasis added. [1940] IR 122, 129. [1934] 1 DLR 748 Cheshire and North, 1035. [1934] 1 DLR 748, 750. See also J-G Castel, Canadian Conflict of Laws, 4th edn., 541. [1901] 1 KB 123.

204 Express or Implied Choice of Law only in name but in fact an English company, deliberately created as such by its founder”.530 The jurisdiction clause in the memorandum reinforced this view, as did the use of the English language and English form in the settlement deed.531 The importance attached to the English trust company stands in marked contrast to the subsequent decision in Revenue Commissioners v. Pelly,532 where the domicile of the trustees was treated as a relatively minor factor. However, there is no conflict between the cases. In the latter case, the trustees’ identity had revealed nothing about the settlor’s intentions: there were two trustees from two different jurisdictions.533 Likewise, AG v. Jewish Colonization Association534 does not stand as an authority for the overwhelming relevance of the trustees’ domicile to the search for implied intention. It was the fact that the settlor specifically established a holding company to act as trustee which revealed something about the settlor’s intentions; certainly, it is hard to imagine the appointment of personal trustees being accorded such weight. An equally interesting aspect of the Jewish Colonization case is the factors which were specifically discounted by the Court of Appeal in the search for the settlor’s intention. Stirling LJ mentioned five such factors: 1. 2. 3. 4. 5.

the domicile of the settlor; the deed was executed abroad; the assets were largely located abroad; the company business was largely transacted abroad; the intended beneficiaries were foreigners.

He commented: “As to the first, it seems to me that the effect of the domicil of the donor was exhausted when . . . the funds were placed under the legal control of the defendants. As to the second, it does not seem to be one to which any great weight can reasonably be attached . . . As to the third and fourth, it is to be borne in mind that the defendant cannot escape from the jurisdiction of the English courts by continuing or selecting investments which may be locally situate out of the jurisdiction, or by carrying on operations in a place without the jurisdiction. As to the fifth, I see no way in which the rights of the beneficiaries, whether native or foreign, can be completely enforced except through the intervention of Her Majesty’s Attorney-General, or, it may be, the Charity Commissioners.”535

530

[1901] 1 KB 135. Ibid., 144–5. 532 [1940] IR 122. 533 In the United States, the appointment of a trust company is regarded as “persuasive evidence that the settlor of either an inter vivos or a testamentary trust intended that the trust should be administered in the state where the trust company is incorporated”: M Rosenberg, P Hay and R Weintraub, Cases and Materials: Conflict of Laws, 10th edn. (New York, Westbury, 1996), 813 (relying on the Restatement (2d) on Conflict of Laws § 217–2; and on Boston Safe Deposit & Trust Co v. Alfred University 339 Mass 82). 534 [1901] 1 KB 123. 535 Ibid., 145–6. 531

Common Law Authorities 205 These five factors would not obviously be relevant to a quest for implied intention. But Stirling LJ appears to suggest that they have no relevance at all, even to the ascertainment of an objective proper law. But again, everything must depend on the settlement in question. In a suitable case, one or more of these factors may be helpful in ascertaining an implied choice of law or the objective proper law. The domicile of a beneficiary for instance, may be crucial where the trust is established with protectionism in mind.536 Drawing attention to one’s domicile, as in re Lord Cable, reveals something of the settlor’s intent. Likewise, it is difficult to believe that the situs of movables should be accorded so little weight. Stirling LJ remarked that as the property in the instant case was situated in a number of jurisdictions “it cannot have been intended that the law of each of those countries should apply”.537 Maybe not; but where a settlor orders assets to be transferred to a specific jurisdiction, that may tell us something of his intentions; and as for the objective proper law, it is difficult to see how the situs can be dismissed, at least where the bulk of the property is in one jurisdiction, given the importance attached to the situs in absolute property transfers.538 In the Australian case of Lindsay v. Miller (No 1),539 a settlor domiciled in Scotland executed a deed in Scotland in Scottish form. The deed itself was kept in Melbourne, where the trust was administered. The trust assets were shares situated principally in Victoria, with the residue held elsewhere in Australia, although the income was partly paid to beneficiaries in England and Scotland. Two of the trustees were resident in Australia, one in Scotland and one in China. In the absence of an express choice of law, the Supreme Court of Victoria found the proper law to be that of Victoria. This time, the situs of the assets was stressed, along with the importance of the place of administration and retention of the trust deed. The concentration of the assets in Victoria doubtless made the situs a more important factor than it might otherwise have been. Moreover, although administration may take place in a state other than the situs, administrative powers operate in tandem with the assets themselves; and where both administration and situs coincide, the case for that law being the proper law is very strong. In Iveagh v. IRC,540 a settlor domiciled in Ireland settled property on trust for his son for life, remainder to his children. The son and children were also domiciled in Ireland. The deed, which was drawn up by English solicitors, permitted investment in land in England. The trust was administered in England. The assets were located in a number of states, although the share certificates were held in England. Upjohn J saw his role as to establish the settlor’s intentions,541 but was prepared to look at a wide range of “objective” factors relating to the 536 The trust in the instant case was intended to protect Russian Jews, but since it was designed to help them leave Russia, obviously application of Russian law would not have been appropriate. 537 [1901] 1 KB 123, 144. 538 See the discussion of Art. 7 below; see also section 7 of Part One of this book. 539 [1949] VR 13. 540 [1954] 1 Ch 364. 541 He relied upon Duke of Marlborough v. AG (No 1) [1945] Ch 78 on this point.

206 Express or Implied Choice of Law settlor, the deed, the situs of the property, the place of administration and the investment provision. He acknowledged the significance of all such factors but did not find on the facts that they pointed in any particular direction. The assets were scattered across the world. The administration of the trust mattered little because, on the facts, Upjohn J ruled that the beneficiary could sue in England or in Ireland, as the trustees had residences in both countries, so that the trust would be enforceable whichever law governed the trust. But, he reasoned, a settlor domiciled in Ireland who created a trust of property for the benefit of Irish beneficiaries also domiciled in Ireland over an extended period might be taken to have intended Irish law to govern the trust. Accordingly, he found Irish law to be the proper law. The case represents clear support for the domicile of the parties;542 again, however, it would be wrong to make too much of this. All the parties were domiciled in the same state and there was no particular connection with any one other state. Moreover, the fact that the settlement was a family one led the judge to place more weight on “personal” connecting factors. In contrast, in Perpetual Executors & Trustees Association of Australia Ltd v. Roberts,543 the settlors of two trusts and the beneficiary of them were all domiciled in England; but the Victorian court found that Victorian law applied, because a host of other “objective” factors relating to the trustees, administration, the trust assets and the trust deed pointed to that state. More generally, one might expect that personal connecting factors may prove rather less significant in a more commercial setting. Personal connecting factors have little relevance in more commercial areas of the conflict of laws. Furthermore, a settlor may be taken as having been unlikely to want his personal law to govern a commercial trusts transaction. Perhaps the most well known case in recent years is the decision in Chellaram v. Chellaram.544 Two brothers (who were assumed to be Indian domiciliaries) created a discretionary trust for their family. Some members of the family were resident in India or elsewhere overseas. Of the three original trustees, two resided in England;545 the third, though an Indian domiciliary, held a British passport and had substantial connections with England. The settlement was in English form and in the English language, although it was drawn up in India by an Indian lawyer and signed by the two settlors in Singapore and Nigeria respectively. The trust assets consisted of shares in a Bermudan company. Administration of the trust had been effected in England. The claimants sought the removal of the trustees. The defendant trustees argued that the English court 542 Personal connecting factors relating to the settlor and beneficiaries are not even mentioned in Art. 7 in the (non-exhaustive list) of relevant factors in ascertaining the law of closest connection. 543 [1970] VR 732. 544 [1985] Ch 409; noted by D Evans, “Administering Foreign Settlements” (1986) 102 LQR 28; D Hayton, “Trusts in Private International Law” (1985) 135 NLJ 18; see also A Wallace, (1987) 36 ICLQ 454, 479–482. 545 Both of whom were subsequently replaced, so that the all the trustees were, at the time of litigation, domiciled in India. However, all the trustees (at the time of litigation) had substantial connections with England.

Common Law Authorities 207 had no jurisdiction to remove trustees of a foreign settlement; alternatively, they sought a stay of the English proceedings on forum non conveniens grounds. Both contentions were supported by the argument that the proper law of the trust was Indian.546 Although Scott J ruled that the court had jurisdiction irrespective of the proper law of the trust, he did nonetheless consider at length obiter what the proper law might be. The judge commented that “most important of all” in the search for implied intention was “the identity of the three original trustees”, which, in combination with other factors meant that: “the inference is inescapable that the parties to the settlement contemplated that administration thereof would take place in London . . . The question why, if the parties intended the settlements to be governed by Indian law they should have arranged for an English administration, is a difficult one to answer. The parties’ contemplation of an English administration seems to me to point strongly in favour of an English proper law”.547

He cited with approval from re Pollak’s Estate.548 In that case, a domiciliary of the Transvaal had left a will consisting of movable property in England, South Africa and elsewhere, leaving property to beneficiaries, most of whom were domiciled in England. He appointed an English bank as executor and trustee, which had no branch in South Africa. In the Transvaal court, Davis J had ruled that the appointment of the English bank to administer the trust showed that the testator intended English law to govern its administration. Scott J in Chellaram regarded the case, not as establishing that administration of a will trust is governed by the law of the place of administration, but simply that a settlor of any trust, whether inter vivos or testamentary, can select the proper law.549 The judge went on to reject the view that a different choice of law rule should apply to matters of administration from that governing essential validity. Instead, the place of administration would be one of the most powerful indicators of the law intended to govern the entire trust. In the event though, since Scott J found that nothing turned on whether the proper law on the facts was English or Indian and that the English court would have jurisdiction in either case, he expressed no final view on the proper law. But the central role which he gave obiter to the place of administration, which he in turn ascertained very largely from the connection of the original trustees with England, is reminiscent of A-G v. Jewish Colonization Association.550 That latter case was in many ways a clearer one for reliance on such a factor, given that the trustee there was a specially formed company. Certainly, an express specification of the place of administration will have a very 546 The stay contention was supported by the argument that the natural forum to hear a case governed by Indian law was India. 547 [1985] Ch 409, 425. 548 [1937] Transvaal Provincial Division 91. 549 Rule 121 of Dicey and Morris, 10th edn., had stated (683) that the administration of trusts was apparently governed by the law of the place of administration. 550 [1901] 1 KB 123.

208 Express or Implied Choice of Law strong influence in ascertaining the applicable law in the absence of express choice. Moreover, where such a place is not specified, a reference to administrative provisions of a particular legal system may be a powerful indicator that that law should govern the trust, even if it is only to exclude those provisions.551 But more generally, the place of administration and the domicile or residence of the trustees are unlikely always to be determinative. Evans suggests that “It is difficult to imagine a court which had to deal with a dispute over the extent of the beneficial interests giving such weight to either of these factors”.552 We have seen the relevance attached to other factors in the earlier common law cases and that, in any given context, factors relating e.g. to the settlor or beneficiary, or the situs of the assets, may reveal something about the settlor’s intentions.

(D) Implied Intention to Create a Valid Trust? One final factor bears mentioning: can the settlor be presumed to have chosen a law by which the trust is valid? In Augustus v. Permanent Trustee Co (Canberra) Ltd553 one reason for the High Court’s finding that a trust was governed by the law of New South Wales was that by the other potentially applicable law, that of the Australian Capital Territory, the trust infringed the rule against perpetuities. In relation to American law, Scoles and Hay argue that the “. . . policy of giving effect to the settlor’s intention tends to sustain the settlor’s disposition whenever possible. Exceedingly strong contrary local policy is necessary to overcome this preference towards validating the disposition in trust”.554

Whilst a party might expressly choose the law of a state by which the trust is invalid,555 it seems reasonable to conclude when trying to infer what a settlor might have intended that he would not have ordinarily intended to choose the law of a state by which the trust was invalid. However, the converse is not true; it cannot be a sufficient reason to infer that a settlor chose a particular law simply that the trust is valid by that law, since he may not have thought about the matter, or, indeed, might have equally wished to choose another law by which the trust is valid. 551 D Hayton gives the English examples of a provision qualifying or excluding ss. 23 and 30 of the Trustee Act 1925 or the Trustee Investments Act 1961 (which has in any event largely been superseded by the Trustee Act 2000): see (1987) 36 ICLQ 260, 270. 552 D Evans, “Administering Foreign Settlements”, (1986) 102 LQR 28, 31. One cannot discount the possibility that a court may do this in practice, but the point made is curious in principle, given that Scott J expressly envisaged a single law governing all substantive issues arising under a trust (as, of course, does the Hague Convention, in the absence of dépeçage). D Hayton appears to take a different view, arguing that a statement of the place of administration will be a very strong indicator that the law of that place should apply to matters of administration and that since this “presupposes a valid trust to be administered, hence to such law governing validity . . .” : ibid., 271. 553 (1971) 124 CLR 24; and see P Nygh, Conflict of Laws in Australia, 6th edn., 516. 554 E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1028. See also Vansant v. Roberts 3 Md 119 (1852) 555 Or even choose the law of a non-trust state, so that the choice is struck down by Arts. 5 and 6(2).

Common Law Cases Evaluated 209

13. THE COMMON LAW CASES EVALUATED

It is apparent that the common law cases do not draw a rigid distinction between implied intention and objective proper law. In essence, their approach is two-tiered, in that they simply distinguish between cases where an express choice has been made and cases where it has not. In most of the cases, the latter category is presented as a search for the implied intention of the settlor; but the word “intention” is stretched to breaking point and the courts’ approach is often to find that intention hidden in the factors with which the trust has a close connection.556 Although such an approach has been described as “eminently sensible”,557 it could rather be seen as an unhelpful intellectual distinction. It fails to distinguish real choices from cases where there is, in reality, clearly no choice. Such a distinction cannot, of course, be avoided under the Hague Convention.

14. CONCLUSIONS ON IMPLIED CHOICE OF LAW

To conclude at the end of this discussion of the relevant cases that no one factor can be conclusive as to the governing law in the absence of an express choice does not obviously take us too far. Nonetheless, it is manifestly true; and insofar as the attempt is to find out what the settlor genuinely intended, it has to be thus. For example, where a settlor takes a particular step, such as setting up a trust company,558 or expressly declaring his domicile,559 it draws attention to factors which might otherwise have been considered of little importance in determining the settlor’s intention. Nonetheless, some conclusions can be formed as to how the common law cases might assist the search for an implied choice under Article 6, Hague Convention.560 1. The search under Article 6 must be for a law genuinely intended by the settlor to govern the trust. In this respect, some of the common law cases which do not draw a rigid distinction between an implied choice of law and the law of closest connection in the absence of choice should be approached with caution when their relevance to the application of the Hague Convention is considered. 556

E.g. Lindsay v. Miller (No 1) [1949] VR 13. A Wallace, (1987) 36 ICLQ 454, 468. 558 As in A-G v. Jewish Colonization Association [1901] 1 KB 123. 559 As in Re Lord Cable [1977] 1 WLR 7. 560 Compare the conclusions of A Wallace, (1987) 36 ICLQ 454, 468–9. However, those conclusions are in relation to a two-tiered approach, whereby no distinction need be drawn between the implied intention and the absence of intention. See also the factors listed by W Swabenland, “The Conflict of Laws in Administration of Express Trusts of Personal Property” (1936) 45 Yale LJ 438, 442–3. 557

210 Express or Implied Choice of Law 2. In the case of testamentary trusts, it appears that the deceased’s last domicile is influential in ascertaining the settlor’s intentions. However, it is far from determinative561 and many of the common law cases where the testator’s domicile was considered relevant may be better decided today under the Hague Convention as cases where there is no express or implied choice and the law of closest connection needs to be found under Article 7. 3. In the case of inter vivos trusts, the insertion of a jurisdiction clause will be a powerful indicator of the settlor’s intention.562 Furthermore, if the trust is created by virtue of, or pursuant to, a judicial settlement,563 the inference that the law of the forum was intended to govern will be strong.564 4. Use of terminology resonant of a particular legal system, or of a form known in only one system, may be an indicator of the settlor’s intentions. However, it has not proved to be anything like determinative and a different law has been found to have been intended to apply in a number of the cases.565 It is likely that a similar approach will be taken under Article 6. 5. The place of administration will be given great prominence.566 However, since under the Hague Convention the principle focus is on the trust instrument itself, this factor should be treated as especially important only where the place of administration is expressly specified therein. Even where such specification is not deemed sufficient to infer a choice of law, the place of administration is likely to be considered an important factor in ascertaining the law of closest connection in the absence of choice pursuant to Article 7. 6. Little inference can normally be drawn from the domicile or residence of the original trustees.567 However, there are two obvious exceptions to this principle. First, where all the original trustees are domiciled or resident in one state, and their identity was stipulated in the trust instrument, this may itself lead to an irresistible inference as to where the settlor intended the trust to be administered.568 That place of administration itself becomes a key factor in the search for an implied intention. Secondly, where the trustee is a company established specifically to hold the trust assets, the inference is that the law of the place of incorporation of the trust company will govern.569 7. The situs of the assets may prove to be an important factor where the bulk of the trust property is immovable. However, where movable property is 561

See, in particular, A-G v. Campbell (1872) LR 5 HL 524; Saliba v. Falzon [1998] NSWSC 302. As in A-G v. Jewish Colonization Association [1901] 1 KB 123. 563 Since, in the United Kingdom, the Hague Convention has been extended to apply to such trusts by s. 1(2), Recognition of Trusts Act 1987. 564 As in Revenue Commissioners v. Pelly [1940] IR 122. 565 Peillon v. Brooking (1858) 25 Beav 218; Re Lord Cable [1977] 1 WLR 7; Lindsay v. Miller (No 1). [1949] VR 13; Revenue Commissioners v. Pelly, ibid.; Iveagh v. IRC [1954] 1 Ch 364. 566 As in Chellaram v. Chelleram [1985] Ch 409, Lindsay v. Miller (No 1), ibid. 567 Revenue Commissioners v. Pelly [1940] IR 122; Iveagh v. IRC [1954] 1 Ch 364. 568 Chellaram v. Chelleram [1985] Ch 409. 569 As in A-G v. Jewish Colonization Association [1901] 1 KB 123. 562

Conclusions on Implied Choice of Law 211

8.

9.

10.

11.

12.

13.

concerned, the situs appears to be a relevant, but not especially crucial factor.570 Naturally, its relevance increases if the great bulk of the assets are in one state only.571 In any event, the situs of the assets is unlikely to say much about the settlor’s implied intentions in relation to movable property, unless he specified that the assets must, for all times, be kept in a particular jurisdiction. Otherwise, the relevance of the situs is likely to be to the ascertaining of the applicable law under Article 7 in the absence of choice. Personal connecting factors relating to the settlor or beneficiary are unlikely to be given much weight in the case of inter vivos trusts. However, their relevance is likely to increase significantly where the trust is a family settlement and all parties to the trust are domiciled or resident in one state.572 The place where a deed was executed may be given some weight.573 However, it is more likely that such a factor would today be considered relevant to the ascertaining of the applicable law in the absence of choice. Although there appears to be no clear common law presumption of validity, it may be thought intrinsically unlikely that a settlor would choose a law under which the trust is invalid.574 Strictly speaking, the search should be confined to the terms of the trust itself. Article 6 indicates that extraneous circumstances should only be considered “if necessary”, and then only to shed light on the terms of the written instrument. However, it is suggested that it would be undesirable in principle for a court to disregard any such relevant circumstances, providing that extraneous evidence is only used to draw an inference in cases where it gives strong grounds to indicate an intended law, or to cast doubt upon whether a particular law was intended. The policy underlying the trust should be considered where it sheds clear light on the settlor’s intentions. However, a court should be slow to speculate as to what the purpose of the settlor may have been in anything but clear cut cases.575 Any other factor may prove relevant on the facts of a given case, especially where the settlor specifically draws attention to that matter.576

570 A-G v. Campbell, (1872) LR 5 HL 524; Lindsay v. Miller (No 1) [1949] VR 13; Iveagh v. IRC [1954] 1 Ch 364. 571 As in Lindsay v. Miller (No 1), ibid. 572 As in Iveagh v. IRC [1954] 1 Ch 364. 573 Duke of Marlborough v. AG (No 1) [1945] Ch 78; Harris Investment Ltd v. Smith [1934] 1 DLR 748; Lindsay v. Miller (No 1) [1954] 1 Ch 364; Iveagh v. IRC, ibid. 574 Augustus v. Permanent Trustee Co (Canberra) Ltd (1971) 124 CLR 24. This appears to accord with the policy of the Convention, whereby a choice of a non-trust state is disregarded altogether and the law of closest connection is used (Art. 6(2)). 575 This is suggested by the comments of von Overbeck, para. 67, p. 384. 576 E.g. the affirmation of domicile in Re Lord Cable [1977] 1 WLR 7.

212 Express or Implied Choice of Law

15. WHICH LAW DETERMINES WHETHER A CHOICE CAN BE INFERRED ?

One final point should be noted. Different laws may adopt a differing approach as to the strength of inference to be drawn from e.g. a jurisdiction clause; or they may differ as to the forms of evidence which may be admitted in order to demonstrate a choice.577 The search for an implied intention is treated in the cases as at all times one to be conducted solely by reference to the law of the forum. If an implied intention to subject a trust to New York law is found, it is not asked whether the facts constitute, as a matter of New York law, sufficient evidence to demonstrate a choice. Why should this be so?578 The obvious answer is that identification of an express or implied choice by the settlor is a connecting factor and that the interpretation of connecting factors is always determined by the lex fori: “This is elementary, axiomatic and could not be otherwise”.579 But whilst this is true, it is not conclusive. After all, where there is an express choice of law clause, the forum’s decision to allow that clause, in principle, to determine the applicable law is a selection by the forum of the chosen law as a connecting factor. Nevertheless, if it is alleged e.g. that Utopian law was chosen, we still go on to ask whether that clause was agreed upon by Utopian law.580 Put differently, the law of the forum decides if a choice of law may be made and whether there is prima facie evidence of such a choice; the putative applicable law decides whether that law was validly agreed upon. By analogy, what should happen if there is no express choice of law clause, but an English court decides that it appears that the settlor impliedly intended Utopian law to govern the trust? Imagine that, by Utopian law, the facts would not constitute sufficient evidence of an implied choice. To close one’s eyes to that fact entails upholding a trust governed by Utopian law, even though, by Utopian law, no such choice has been made.581 The choice of law shopper, who wants the trust to be subjected to Utopian law, should come shopping in an English court. Accordingly, it is suggested that, strictly speaking, the search for an implied intention should involve two stages; first, that the forum satisfies itself, to the standard of proof required by Article 6, Hague Convention, that the settlor intended the law of Utopia to govern the trust; if so, then secondly, that according to Utopian law, this law was indeed chosen. This two-staged test is admittedly more cumbersome to apply and more likely to result in a finding that no valid choice has impliedly been made, since it must clear two hurdles. But 577

And, in particular, on the admissibility of evidence of subsequent conduct. P Nygh, Autonomy in International Contracts, 113. 579 J Collier, [1989] All ER Rev, 61. 580 Compare Art. 3(4), Rome Convention. We do not, however, ask whether Utopian law allows freedom of choice as a connecting factor. 581 This is quite different from application of the doctrine of renvoi: there is no question of applying the law that a Utopian court would apply. The enquiry is limited to asking whether, by Utopian law, a choice has been made. If not, the forum should apply its choice of law rules in the absence of choice. 578

Time for Ascertaining an Implied Choice Of Law 213 without it, the finding of an implied choice of law by an English court, and the application of that law where it would not even be applied in its local courts on the facts, is a real prospect.

16. TIME FOR ASCERTAINING AN IMPLIED CHOICE OF LAW

The Convention is silent as to the time at which factors must be present from which inferences as to the settlor’s intention may be drawn. In particular, should such factors be ones persisting at the time of the creation of the trust, or should subsequent evidence which is alleged to shed light on the settlor’s intentions be considered? 582 The Hague Convention applies to trusts created voluntarily and evidenced in writing,583 with reference to surrounding circumstances to ascertain the settlor’s implied choice being permitted only “if necessary”.584 This might suggest that extraneous evidence of the terms of the trust should be kept to a minimum. Moreover, it would seem that geographical factors585 which change after the trust is created, such as the replacement of a trustee domiciled in one state by a trustee domiciled in another state, or a change in the situs of the assets, cannot be relevant. They tell us nothing about what the settlor might have intended at the time of creation of the trust and cannot alter the applicable law.586 On the other hand, subsequent events which shed light on the settlor’s intentions at the time of creation of the trust (e.g. evidence that the settlor intended the trust to be created for a particular purpose, or that his reasons for inserting a jurisdiction clause for a particular state were unrelated to the law which he expected to be applied in that state) may be thought necessary for the interpretation of the written trust document. As a matter of English common law, at least in the contractual setting, the position appeared to be that subsequent conduct could not be taken into consideration. In Whitworth Street Estates (Manchester) Ltd v. James Miller and Partners Ltd,587 the House of Lords held that, in determining whether a contract contained an implied choice, subsequent conduct of the parties could not be considered. The uncertainty and potential elongation of proceedings which may result from admission of such evidence may be at the root of such an approach. However, it is unclear if the rule survives the enactment of the Rome Convention, which is 582 Even more strikingly, perhaps, the time factor is not specified for ascertaining the applicable law in the absence of choice pursuant to Art. 7. See the discussion of “time for assessing the law of closest connection” below. 583 Art. 3. 584 Art. 6. 585 Insofar as these are relevant to the search for an implied choice of law (on which, see the discussion of the common law cases above). 586 Such a view is expressed by Upjohn J in Iveagh v. IRC [1954] 1 Ch 364, 370. A Wallace (1987) 36 ICLQ 454, 470 points out that it is possible that the settlor may have evinced an intention that, should such objective factors change, the applicable law should also change, although in practice this is likely to be a rare occurrence. 587 [1970] AC 583.

214 Express or Implied Choice of Law silent on the matter.588 Furthermore, it appears that the admission of evidence of subsequent conduct is well established in America and a number of civil law jurisdictions.589 It is contended that subsequent evidence of the settlor’s intentions at the time of the creation of the trust should be permitted in England for several reasons. First, the United Kingdom has extended the application of the Convention rules in England beyond voluntarily created trusts evidenced in writing.590 This means that an English court will have to ascertain the applicable law of trusts for which there is no written instrument. This will necessitate reference beyond the terms of a trust instrument for evidence as to the settlor’s intentions; and it would seem curious to freeze that search in time. Secondly, other contracting states which have not extended the Convention’s scope nonetheless adopt more liberal rules on admission of subsequent evidence.591 Thirdly, and most tellingly, if the search is for a choice genuinely made by the settlor, it seems almost perverse to shut one’s eyes to any substantial evidence which indicates whether such a choice was made.592

588 Although Giuliano and Lagarde suggest that, where the applicable law is to be ascertained in the absence of choice, “factors which supervened after the conclusion of the contract” may be considered: OJ 1980 C 282/4, 20. See also P Lagarde “Le Nouveau Droit International Privé des Contrats” (1991) 80 Revue Critique de Droit International Privé 287. 589 F Mann, “Subsequent Conduct as Aid to Construction”, (1973) 89 LQR 464. See also Dicey and Morris, 1215–6; P Nygh, Autonomy in International Contracts, 111–2. 590 S. 1(2), Recognition of Trusts Act 1987 (considered in relation to Art. 3 above). 591 E.g. the Italian Civil Code, Art. 1362 of which permits evidence of subsequent conduct in the contractual context (see P Nygh, Autonomy in International Contracts, 112). 592 See Dicey and Morris, 1216 and P Nygh, ibid., 112, for a similar view in relation to contracts.

ARTICLE 7—THE APPLICABLE LAW IN THE ABSENCE OF CHOICE Where no applicable law has been chosen, a trust shall be governed by the law with which it is most closely connected. In ascertaining the law with which a trust is most closely connected reference shall be made in particular to— (a) (b) (c) (d)

the place of administration of the trust designated by the settlor; the situs of the assets of the trust; the place of residence or business of the trustee; the objects of the trust and the places where they are to be fulfilled.

1. ORIGIN AND APPLICATION

Article 7 will determine the applicable law either where no express or implied choice has been made, or when the settlor has chosen the law of a non-trust state, a choice which will be disregarded under Article 6(2). The latter scenario continues to fall within the Convention;593 only if the law of closest connection turns out to be that of a non-trust state will the trust then be expelled from the Convention, losing any claim to recognition thereunder.594 The principle of the law of closest connection applying in the absence of choice accords with common law doctrine, although as we have seen, common law cases do not draw a clear distinction between the implied choice of law and the applicable law in the absence of choice. In principle,595 a broader range of objective factors should be considered relevant to ascertaining the law of closest connection than would be relevant when trying to discern a settlor’s intentions. The Convention specifies factors which are of especial importance in ascertaining the law of closest connection.596 No doubt, these factors are intended to install an element of clarity in what could otherwise be a most unpredictable process. But the Convention stops short of adopting fixed criteria for determining the law of closest connection. A Spanish proposal suggested a fixed hierarchy of 593 This “pro-Convention” solution was the product of an Israeli proposal: Working Document No 39. 594 But, of course, national states may be willing to recognise this apparently curious construct and Art. 14 allows them to adopt more liberal recognition rules, if they see fit. 595 Subject to what is said below. 596 Compare J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 223 who states that “the general ‘American’ approach to the validity of inter vivos trusts of movables [is] as follows: the principal elements for consideration are the place of administration, the domicile of the settlor, the situs of the corpus, and the place of execution of the trust instrument.”

216 Applicable Law in the Absence of Choice connecting factors which would determine the applicable law, at the top of which would have been the law of the place of administration. If that law did not provide for trusts, the Convention would have proceeded to the next connecting factor in the hierarchical chain to determine the applicable law, until such time as the law of a trust state was identified by one of the connecting factors in the hierarchy.597 Such an approach had the perceived merits of clarity and a rule-based approach which mitigated very strongly in favour of the validity of trusts. However, common lawyers were concerned by the inflexibility of a mechanical approach to the law of closest connection to trusts and the proposal was rejected. A less mechanical approach, and one which has found favour in a different private international law context, would have been to adopt a presumption as to what the law of closest connection might be.598 Although such a presumption might have eased matters for the courts of non-trust states when interpreting the Convention, common law experience would suggest that the absence of a presumption is wise. For we have seen that no one factor was consistently conclusive in the search for the applicable law in the absence of an express choice.599 A presumption that was frequently rebutted would add little in the way of certainty; one which could only be rebutted in the face of overwhelming evidence would risk applying the “wrong” law, that is one of little connection. A radical proposal was put forward by the American delegation. In the absence of an express or implied choice of law, it would have advocated dépeçage writ large. Each and every issue would have been subjected to the law with which it had its closest connection.600 The proposal was initially approved, but subsequently withdrawn amid dissatisfaction, especially from civil law states, as to such a fragmented approach to the choice of law process. Pragmatically, it would be difficult to think of a much more daunting task for a state unfamiliar with the trust than trying to ascertain a law of closest connection for each and every issue raised in proceedings, or to determine whether to recognise such a trust. The result is that dépeçage is still permitted by the Convention,601 but in principle, at least, the initial search is for a law which governs the entire trust. In the end, the four factors identified are intended to be highly flexible,602 but nonetheless to contain an element of hierarchy which was certainly not explicit 597

Working Document No 29, discussed in the von Overbeck Report, para. 76, p. 386. Compare the presumption of characteristic performance contained in Art. 4(2), Rome Convention (whereby the law of the habitual residence or place of central administration of the characteristic performer (who, in a contract of sale, would be the seller) is presumed to apply in the absence of a choice of law in contract). In the trusts context, the most sensible presumption, looking at the common law case law and the ultimate emphasis of Art. 7, Hague Convention, would probably have been one in favour of the law of the place of administration of the trust. 599 Where the search was for an implied choice of law/the law of closest connection and no clear distinction was drawn between the two. Here of course, Art. 7 only becomes relevant where there is no implied choice. 600 Working Document No 12: discussed in the von Overbeck Report, para. 94, p. 390. See further the discussion of Art. 9 below. 601 Art. 9. 602 Maybe even too flexible: see D Hayton, “The Hague Convention on Trusts: a Little is Better than Nothing but why so Little?” (1994) J Int Corp P 23, 25. 598

Origin and Application 217 in the common law approach. As to the four factors mentioned, it is clear that none can be conclusive and that the list of factors which may be considered is not closed. However, they are singled out as of especial relevance,603 even if von Overbeck describes them as simply “examples”604 of relevant factors. As to the hierarchy of the four factors mentioned, it is a case of all factors being equal, but some being more equal than others. “These criteria are all in principle on the same footing: however, the conference has given them their place by order of importance so that it might be said that there is among them a certain hierarchy”.605

Von Overbeck later comments that, “The delegations of the civil law countries also agreed that the second paragraph, by the order in which the criteria were set forth, comprised an implicit hierarchy which would satisfy the needs of practice. But the text will also allow the judges of the common law countries to take into account, as they have the habit of doing, all factors at the same time”.606

The importance of this hierarchy is shown by the discussions which took place as to which order they should be listed in and by the relegation during negotiations of factor (d) (the objects of the trust and the places where they are to be fulfilled) from its proposed position as the first mentioned criterion.607 Nevertheless, Article 7 still contains too much flexibility for some. Hayton comments that “In reality, Article 7 is of little assistance and leaves the position as flexible as it is under common law choice of law rules”.608 Schoenblum criticises the flexibility of Article 7 as an example of “the inclination to compromise away fundamental differences and arrive at solutions which are unworkable in practice”.609 He has in mind that the four factors are likely to be applied much more rigidly and hierarchically in civil law states than in the common law states. “This divergence as to how the governing law is to be determined in the most basic of situations exemplifies the lack of guidance or predictability afforded by the Convention and the lack of coherence lurking just beneath the surface”.610 603 An Irish proposal to include a number of other relevant factors was rejected: Working Document No 41, discussed by von Overbeck, para. 75, p. 386. 604 Von Overbeck, para. 72, p. 386. 605 Ibid., para. 72, p. 386. 606 Ibid., para. 77, p. 387. 607 Ibid., para. 75, p. 386. 608 D Hayton, in A Borras, A Bucher, T Struycken and M Verwilghen (eds.), E Pluribus Unum, 121, 124. 609 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 6, f/n 8. 610 Ibid., 9. Schoenblum also slightly misinterprets the von Overbeck Report (para. 80, p. 387) as saying that the relevant factors under Art. 7 should be interpreted at different times. Although von Overbeck does suggest that the factors may need to be considered at different times, he goes on to point out that discussion of the issue of timing was deferred and ultimately never decided at the Conference. See also J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 237–8.

218 Applicable Law in the Absence of Choice This is, it is submitted, something of an over-statement. Of course, in an ideal world, precise, fixed common factors would be agreed and uniformly applied in all contracting states.611 However, the price of seeking such rules might be: (i) failure to reach any kind of agreement; (ii) inflexible rules; and (iii) nonratification by disaffected states. The fact is that the Convention made a substantial advance in agreeing to only four named factors and, insofar as some differences in application of Article 7 may emerge, can some diversity not be accepted if, as Schoenblum also argues, the Convention already demonstrates “undue preoccupation with multilateral harmony . . .”?612

2. THE FOUR FACTORS CONSIDERED

So, the list of relevant factors is not closed. Nonetheless, it is notable that Article 7 states that reference shall be made to these four factors.613 Of course, a judge would be free to consider the relevance of each factor and then disregard one or more of them on the facts. However, the language of Article 7 suggests that he is bound at least to consider such factors.

(A) Place of Administration The particular relevance of the place of administration of the trust will surprise few common lawyers. Certainly, it is very much in keeping with the dominant role accorded to it in Chellaram v. Chellaram,614 albeit that there the search was ostensibly for an implied choice of law. Dyer and van Loon comment that: “To borrow an analogy from the conflict of laws of contracts, the administration of property carried out by the trustee or trustees is the ‘characteristic performance’ under the trust instrument”.615

The question then arises whether it is sufficient to ascertain where, in fact, the trust is being administered or whether importance should be attached to this place only where it is specified by the settlor in the trust instrument. On this point, the Convention is unambiguous: the relevant factor is the place of administration designated by the settlor. As von Overbeck explains, this was very much a conscious decision.616 The place of administration would not be worthy of such a lofty place 611 The enumeration of some factors is surely better than having none at all: see D Hayton, (1994) J Int Corp P 23, 24. 612 J Schoenblum, (1994) 3 J Int Corp P 5, 22. 613 Von Overbeck explains that the word “may” was replaced by “shall” to meet constitutional objections of the Austrian Delegate: para. 75, p. 386. 614 [1985] Ch 409; and in Lindsay v. Miller (No 1) [1949] VR 13. 615 A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 185, p. 95; see Art. 4(2) Rome Convention. 616 Von Overbeck Report, para. 78, p. 387.

The Four Factors Considered 219 in the hierarchy of Article 7 if it were to include cases where nothing was said on the matter by the settlor. This is at once understandable, appealing and curious. It is, of course, understandable to want to distinguish between cases where the place of administration is and is not specified: the former situation is one where the place of administration is given particular emphasis by the settlor himself; the latter situation does not manifestly merit a rank above e.g. the situs of the assets. But against this, where the place of administration is designated by the settlor, one would have thought that this would be a very strong indicator, indeed arguably the strongest available indicator, of an implied choice of law pursuant to Article 6; in which case, the cases where Article 7(2)(a) may be relied upon to establish the law of closest connection will indeed be rare. Moreover, whether a factor is objectively significant to the search for the law of closest connection should not obviously depend upon whether it was expressly designated by the settlor;617 either it is of territorial significance or it is not. The place of administration designated by the settlor is often likely to coincide with the place of residence of the trustees. The place of administration is also likely to be that where any disputes between the parties are heard. In the unlikely event that the settlor decrees that the day to day running of the trust is to take place in one state, but that any disputes should be resolved in the courts of another state, it may be inappropriate to accord much weight to the place of administration. Likewise, if the trustees are all resident in another (single) jurisdiction, it may be that an administration clause will not be enough to enable the courts of the place of administration to take jurisdiction and that any litigation would have to take place in the court of the trustees’ residence. Again, in that case, the place of administration is likely to be considerably less important.618 Where the place of administration is not specified by the settlor, Article 7(2)(a) does not apply. But what relevance, if any, should then be accorded to the place of administration? Schoenblum fears that the answer to this is “none.” “Thus, an important criterion referred to in common law jurisdictions for determining the governing law is essentially jettisoned by the Convention”.619 It certainly appears that it should be ranked as a less important factor than the remaining three factors stated in Article 7(2)(b)-(d). Moreover, it may be that the place of administration cannot be determined at the time when the trust is created, so that the place which subsequently turns out to be that where administration is carried out cannot be relevant to the determination of the law of closest connection at the time that the trust is created.620 That said, Chellaram v. 617

This risks confusing subjectivity and objectivity. In Lindsay v. Miller (No 1) [1949] VR 13, the law of the place of administration was applied notwithstanding that the trustees were domiciled elsewhere. However, the trustees were not all resident in a single state where litigation would obviously occur. In today’s language, the place of administration was still the natural forum. 619 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 10. He refers specifically to Hutchinson v. Ross 262 NY 381. 620 And where the place of administration is designated by the settlor (e.g. as England), but has subsequently changed prior to proceedings being commenced (e.g. to Jersey), the subsequent change 618

220 Applicable Law in the Absence of Choice Chellaram621 itself shows that considerable importance can be attached to the place of administration in the absence of express stipulation where it can be inferred where that place is to be. In Chellaram, the domicile of the trustees was itself regarded as a clear indicator of the place of administration of the trust, at least where the trustees all have a permanent connection with a particular state. It could be said that such a case is one where the place of administration was impliedly designated by the settlor, thereby falling within Article 7(2)(a) after all. However, such an approach runs a serious risk of attributing an intention to the settlor which he never possessed. A case like Chellaram would today be one where the place of administration should be considered under Article 7(2)(c) in the context of the residence of the trustees. The fact that administration would inevitably be carried out in the place where the trustees are resident will make that factor especially potent.

(B) The Situs of the Trust Assets Turning to the situs of the assets, we have seen that, at common law, this factor exerted a considerable, but far from overwhelming, influence on the proper law.622 There is, it might be said, no reason to expect it to have any greater role under the Convention in determining the law of closest connection. Indeed, the character of the Convention is to treat the trust rather more as a matter of the law of obligations than the law of property. Moreover, the situs of assets at the time of the creation of the trust may be even more insignificant today, given the ease with which intangible property may be transferred. And where assets are scattered across the globe, the situs may fail to point to a single law of closest connection.623 Against that, however, it can be said that the common law cases treat the situs as far from irrelevant where all the assets are located in a single state and that, where a different law has been found to be the proper law, this has been in cases where the court has found factors indicative of an implied choice of law. If, ex hypothesi, there has been no choice of law, and particularly where no place of administration has been designated by the settlor, the situs becomes the most obviously important factor in the search for the law of closest connection.624 will not affect the determination of the applicable law and English law may be found to be applicable (unless, of course, the applicable law is found to have been changed under Art. 10). See ibid., 10. 621 [1985] Ch 409. 622 A-G v. Campbell, (1872) LR 5 HL 524; Lindsay v. Miller (No 1) [1949] VR 13; Iveagh v. IRC [1954] 1 Ch 364. For the position in America, see E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1031; Rose v. St Louis Union Trust Co 43 Ill 2d 312, 253 NE 2d 417 (1969). 623 J Schoenblum criticises the inclusion of the situs factor for this reason: (1994) 3 J Int Corp P 5, 10. However, where assets are situated in a single state, it seems eminently sensible to refer to this factor. 624 Furthermore, P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 43–4 clearly regard the situs as the most significant factor in ascertaining the law of closest connection.

The Four Factors Considered 221 The role of the situs might be especially significant where the trust property consists wholly or principally of immovables. The self-evident difficulties which can arise in enforcing an order relating to the trust property itself in that state mean that the temptation to do as that law would do is strong. The arguments for the application of the lex situs to trusts of land were considered above.625 However, it is likely that, even in the case of trusts of immovables, a teleological approach to the Convention means that the situs will not be a treated as a conclusive indicator of the law of closest connection. For one thing, the trust instrument may provide that the immovable property should be sold and the proceeds invested in movable property elsewhere. Moreover, if the Court of Justice’s view in Webb v. Webb626 that beneficial entitlement in land should be treated as an in personam interest is adopted in the context of other international Conventions, the apparent need to apply the lex situs diminishes. Finally, and perhaps most significantly, the whole nature of the Convention is one which emphasises the “obligation” aspects of the trust more than the proprietary characteristics. This manifests itself most clearly in the right to choose an applicable law. In the absence of choice, the proprietary aspects of the trust will be important, but are unlikely to be all conquering.

(C) The Place of Residence or Business of the Trustee The third factor mentioned, namely the place of residence or business of the trustee, is one whose importance in the common law cases varied from almost irrelevance627 to near decisiveness. That, of course, is not surprising, since where the original trustees are domiciled or habitually resident in different states, little can be gleaned about the law of closest connection to the trust. At the other extreme stand the common law cases where the trustees’ residence tell us something of the settlor’s intentions e.g. the setting up of a trust company in A-G v. Jewish Colonization Association.628 Indeed, where a particularly clear inference can be drawn from the trustees’ residence, it is more likely to produce an implied choice under Article 6. In Chellaram v. Chellaram,629 it is true that Scott J attached great importance to the substantial connections of the original trustees to England, but this is largely because the judge drew inferences from this as to the place of administration and the settlor’s intentions. In cases where no implied choice can be found, it is not obvious that the residence of the trustees is a factor to which too great importance should be attached. Moreover, residence is not a clear-cut concept and it may be unclear where a trustee resides. 625

In the context of Art. 6. Case C–294/92, [1994] ECR I–1717. The trustees’ residence was discounted in Harris Investments Ltd v. Smith [1934] 1 DLR 748 and Jewish National Fund Inc v. Royal Trust Co (1965) 53 DLR (2d) 577. See the comments of J-G Castel, Canadian Conflict of Laws, 4th edn., 541–2. 628 [1901] 1 KB 123. 629 [1985] Ch 409. 626 627

222 Applicable Law in the Absence of Choice That said, it is clear that a trust will most naturally be administered in a place where all the trustees are resident. If the place of administration is not stipulated, then Article 7(2)(a) is not triggered; but the place of administration ought certainly still to be an important factor and becomes so under Article 7(2)(c).

(D) The Objects of the Trust and their Place of Fulfilment The final factor mentioned in Article 7 is the objects of the trust and their place of fulfilment. This was the criterion “relegated” to position (d) from position (a) in the formulation of the Convention. As much as anything, this is because of the determination to include relatively fixed, clear cut factors in the hierarchy, largely for the benefit of civil law countries.630 Both the purpose of the trust and its place of performance may be difficult to ascertain, in the absence of express stipulation. The word “object” is perhaps rather unfortunate to a common lawyer, since it refers here to purposes of the trust, and not to the beneficiaries of the trust. Schoenblum criticises the object test, on the basis that most trusts are for beneficiaries and do not have a “purpose”.631 That is, however, merely an ambiguity as to the meaning of the word “object.” A trust may be for beneficiaries, rather than a purpose trust; but the trustee will still have a reason for creating it, such as protection of his family, and it is that reason which can and should be investigated under this fourth factor. Moreover, Schoenblum objects that there may be more than one object, or that a single object may need to be fulfilled in different states.632 This is not, however, a valid reason for objecting to the existence of the factor, which may be significant where there is one object and one place of fulfilment.633 How much relevance should be attached to the “object” of the trust? It will be recalled that, in relation to the implied choice of law under Article 6, the drafters took a positive decision not to include reference to the settlor’s purpose, which may be a “purely hypothetical intention or even an element of the objective connecting factor”.634 Such a remark is rather curious; one would have thought that the settlor’s purpose, where it can be clearly shown, would be an important indicator of the intended applicable law.635 Where the purpose reveals nothing about that intention, its relevance as an “objective” connecting factor is not self-evident. However, it seems that the reluctance to 630

See von Overbeck, para. 77, p. 387. J Schoenblum, (1994) 3 J Int Corp P 5, 11. 632 Ibid., 11. 633 Any more than it is a reason to object to the relevance of the situs that it will not assist where assets are situated in a number of jurisdictions. 634 Von Overbeck, para. 67, p. 384. 635 See the discussion of “the settlor’s purpose—implied choice of a law by which the trust is valid?” in the discussion of implied choice of law in relation to Art. 6, above. 631

The Four Factors Considered 223 consider purpose under Article 6 stems from the requirement that the choice be inferred primarily from the written terms of the trust instrument itself. That strict requirement means that extraneous factors are more likely to have their day under Article 7. Nevertheless, it is submitted that, under Article 7, a great deal of importance should not ordinarily be attached to a trust’s “object.” Article 6 is concerned with a choice which the parties made; the connecting factor in Article 7 is based simply on the law of closest connection; the search for intention has been abandoned at that point. Of course, those intentions may tell us something about the objective connections of the trust. It may be possible to infer something from the intentions as to e.g. where the assets should be invested, or the trust administered. But if it is not, then its importance to ascertaining the law of closest connection may not be substantial. One significant exception may arise in the case of charitable purpose trusts.636 Where the purposes are to be carried out wholly, or primarily overseas,637 the case for applying the law of the place of fulfilment is strong. The property will be held outside the jurisdiction and an administration order in relation to such a charity may be ineffective unless duly recognised in the courts of that country. There may also be sensitive questions as to the legality of the trust and its impact in the country in question. 638 In English domestic law, case law demonstrates a certain reluctance to uphold trusts to take effect abroad,639 largely due to the uncertainty as to the effects of such a trust overseas. Against this, it is clear that the application of the law of the place where the “object” is to be fulfilled would create difficulties of proof. An English court would have to determine whether such a trust would be valid by this law, as well as having to determine its likely effects in that state. Furthermore, the purpose of a charitable trust is very likely to be evident on the face of the trust instrument itself. If so, one might expect its relevance to be more significant in the search for an implied choice of law pursuant to Article 6. A final difficulty may arise in ascertaining the place where the objects “are to be fulfilled.” Should this place be ascertained as a matter of English law, or as a matter of the putative applicable law? The question is not easy to answer. A not dissimilar difficulty arises in the jurisdictional context under Article 5(1), Brussels Convention, which allocates jurisdiction in matters relating to contract to “the place of performance of the obligation in question.” That place of performance is 636 Considered more fully in the discussion of “the administration and validity of charitable purpose trusts” in relation to Art. 9, below. 637 See the Annual Report of the Charity Commissioners, 1992. 638 Compare the common law concern about ordering performance of a contract in a state where performance has become illegal: Ralli Bros v. Cia Naviera Sota y Aznar [1920] 2 KB 287. 639 Or, at least, trusts falling within the “fourth head” of other purposes beneficial to the community. No such reluctance appears in relation to the advancement of religion, the advancement of education and the relief of poverty: see further R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 471–2. However, such trusts have been upheld in e.g. Re Robinson [1931] 2 Ch 122; Re Levy Estate (1989) 58 DLR (4th) 375.

224 Applicable Law in the Absence of Choice to be determined by the law which governs the contract.640 The ECJ’s reasoning is that place of performance is a term of the contract and that terms of the contract are to be determined by the applicable law.641 Arguably, such impeccable logic should be applied here. The place of performance of the objects of the trust will, ultimately, be determined by the law applicable to the trust, as a term of the trust. But it is submitted that the ECJ’s approach should not be followed here. In the context of the law of jurisdiction, the ECJ was working with contracts which were assumed to be valid and to have a governing law. Here, the very question in issue is the applicable law. The place where the object of the trust is to be fulfilled is a connecting factor in identifying the law of closest connection. As such, it should be determined by the law of the forum.

3. THE RELEVANCE OF THE PLACE OF RESIDENCE OF THE SETTLOR AND BENEFICIARIES

It is perhaps rather curious that the residence of the beneficiaries, or the settlor, should have been omitted from the list of factors mentioned in Article 7, whilst that of the trustees is mentioned. We know that the list of Article 7 factors is not closed, so that they could be considered. Matthews and Sowden argue that much may depend upon the time when a court has to rule on the applicable law. If this occurs at the inception of the trust, there will as yet be no administration and “the domicile of the settlor is clearly a factor at that stage”.642 However, “Once the trust is ‘up and running’, the factor of the settlor’s domicile at the time of the creation is of little importance: what matters then is the administration of the trust. . .”643

Of course, this cannot be correct in principle, since the applicable law must be determined as it was at the time of inception of the trust;644 whether, in practice, it might influence a court is altogether harder to say. One might also expect that a trust for beneficiaries, all of whom were domiciled or resident in one state, would most naturally be governed by the law of 640 Case 12/76 Industrie Tessili Italiana Como v. Dunlop AG [1976] ECR 1473; Case C–288/93 Custom Made Commercial v. Stawa Metallbau [1994] ECR I–2913; Case C–440/97 Groupe Concorde v. Suhadiwarno Panjan, (1999) CLC 1976. Art. 5(1)(b), Brussels Regulation (on which see f/n 5 above) will amend the position by providing an autonomous definition of the place of performance (subject to contrary agreement) in the case of sale of goods (the place where under the contract the goods were delivered or should have been delivered) and in the case of provision of services (the place where under the contract the services were provided or should have been provided). For all other contracts, the place of performance will still need to be ascertained by reference to the applicable law of the contract. 641 Art. 8(1), Rome Convention. 642 P Matthews and T Sowden The Jersey Law of Trusts, 3rd edn., 43, citing Iveagh v. IRC [1954] Ch 364 in support. 643 Ibid., 43. 644 Unless, of course, it is sought to establish that the applicable law has been changed in accordance with Art. 10.

Testamentary Trusts 225 that state in the absence of choice and that the residence of beneficiaries is then as important as that of the trustees.645 However, this does not appear to be the case, since: (i) the place of residence of the beneficiaries tells the court little about the place of administration; and (ii) if the purpose of the trust was to protect beneficiaries all resident in a single state, this may become relevant by Article 7(2)(d).

4. TESTAMENTARY TRUSTS

It has been suggested that the Article 7 factors are too concerned with inter vivos trusts and, in the case of testamentary trusts, give insufficient attention to the law which governs succession, thereby making a considerable change to the common law position, of which testators who made their will pre-Convention may be unaware.646 O’Sullivan explains that the committee which reported to the Canadian Bar Association—Ontario felt that Article 7 would be applied more frequently to testamentary trusts than to inter vivos trusts, as the former were less likely to contain a choice of law clause. In this respect, the omission of the settlor’s domicile as a named factor is surprising and “While it may certainly be open to a court to look at domicile as a relevant factor, it is probable that a court would rank domicile behind the list of factors in determining ‘closest connection’ ”.647 Yet a settlor might assume that, in the absence of any express provision to the contrary, the law that governs succession under his will648 would also determine the validity of the trusts created thereby.649 Moreover, whilst domicile of the testator may not always be straightforward to determine, it is 645 Where no place of administration is designated by the settlor. See also Iveagh v. IRC [1954] Ch 364. 646 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 7–8 appears to use the fact that choice of law rules for testamentary trusts will be changed by the Convention, and the testator’s position potentially affected thereby, as an argument against altering a state’s choice of law rules. This, though, amounts to an argument for stagnation of the law and should be rejected, not least because it is doubtful whether the Convention must be given retroactive effect (see the discussion of Art. 22, below and of the “application of the Convention in the United Kingdom: the time factor”, above). 647 M O’Sullivan, “The Hague Convention on Trusts—Further Considerations” (1993) 2 J Int Corp P 65, 68. D Hayton, “The Hague Convention on Trusts: A Little is Better than Nothing but Why so Little?” (1994) J Int Corp P 23, 25 refutes suggestions that the law objectively applicable to testamentary trusts under the Convention might no longer be that of the testator’s domicile: “I cannot seriously believe that a common law court seised of the matter would so find.” He goes on to suggest that a ratifying state which is worried about this possibility could always enact a provision that where a testator’s will gives his address within the jurisdiction, this should be regarded as an implied choice of the law of that jurisdiction under Art. 6. This seems to stretch the concept of implied intention much too far, however, and would impute to the testator an intention which he probably never had. 648 The law of the testator’s domicile on death determines succession to his movable, but not immovable property, as a matter of English private international law: see section 14 of Part One of this book. 649 That is not to say that he impliedly intended this to happen; expectation and intention are not the same thing.

226 Applicable Law in the Absence of Choice considerably more clear-cut than factors such as the residence of the trustees or the situs of assets (which may be scattered across the globe and fail to provide a single point of contact). It is evident that this subordination of the settlor’s domicile or residence has proved a turn-off to practitioners in Ontario.650 Hayton has subsequently conceded that: “In retrospect, article 7 . . . should have expressly provided as one of the non-exclusive guiding factors ‘the law governing succession to the testator’s estate in the case of testamentary trusts’ ”.651

However, he points out that this may nonetheless be relevant under criterion (d), as relating to the objects of the trust, “because the purpose of the trust is validly to dispose of the testator’s estate on death, for which the testator’s lex successionis must be relevant, especially if it would validate the disposition”.652

This seems something of an artificial construction of the Convention and to stretch the notion of a trust’s “objects” beyond its limits. Had the testator’s domicile been intended to play such a key role, then express provision might have been made; in its absence, this factor cannot be anything like determinative, especially where the four listed factors clearly point to the law of another state.

5. OBJECTIVE FACTORS POINTING TO A NON - TRUST STATE

Hayton suggests that “A court is likely to take more account of factors connected with a trust-State than with a non-trust State”.653 The statement is not explained further. It can, presumably, only be explained on the basis that a finding that the law of closest connection is that of a non-trust state will sound the death knell for the trust under the Convention and that a court will want to 650 M O’Sullivan, (1993) 2 J Int Corp P 65. It appears to have been the main point of criticism by the Report to the Canadian Bar Association—Ontario. O’Sullivan even suggests the adoption of express legislation in Ontario to the effect that, in the absence of a choice of law, testamentary trusts created in Ontario shall be governed by the law of Ontario: ibid., 71–2. 651 D Hayton in A Borras, A Bucher, T Struycken and M Verwilghen (eds.) E Pluribus Unum, 121, 123. M Koppenol-Laforce, Het Haagse Trustverdrag, 267 remarks that “Strangely, the connection to the law of the deceased’s domicile in case of a testamentary trust has been left out. In English as well as in American p.i.l. this rule has been undisputed. I would, in this special case, plead for maintaining this common p.i.l. rule.” Conversely, it may be argued that a factor such as the situs of movable assets subject to a testamentary trust should be less significant under Art. 7: in England, the lex situs is not the law applied to determine succession to movable property. 652 D Hayton, ibid., 123. Hayton also argues elsewhere that the relative importance of the factors mentioned in Art. 7 can vary depending on whether one is dealing with “an inter vivos trust, a testamentary trust, a marriage settlement or a charitable trust”: see “The Hague Convention on Trusts: A Little is Better than Nothing but Why so Little?” (1994) J Int Corp P 23, 24. 653 D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 949.

Validity and the Law of Closest Connection 227 avoid that conclusion, if possible. However, in principle, the law of closest connection is to be determined entirely objectively and such a consideration ought not to be relevant.

6. VALIDITY AND THE LAW OF CLOSEST CONNECTION

A related question is whether, in ascertaining the law of closest connection, any relevance should be attached to the fact that by X law the trust is valid, whereas by Y law, it is not.654 Although not mentioned in the non-exhaustive list in Article 7, Nygh argues that “one important consideration, when other indicators are equal, will be the question of validity”;655 a view apparently shared by von Overbeck.656 Moreover, it was argued above that validity may be a relevant question in inferring a choice of law under Article 6.657 Nevertheless, it is submitted that its relevance should be rejected under Article 7. It may be legitimate to infer that a settlor would not have chosen a law by which the trust was invalid. But where there is, ex hypothesi, no choice of law, the law of closest connection remains the law of closest connection regardless of whether it leads to the upholding of the trust. It is also likely to be the law with the greatest interest in the matter, and in so far as it decrees that the trust fails, problems in its enforcement are more likely to arise.

7. REFLECTIONS ON THE APPLICATION OF ARTICLE 7

On the whole, the factors mentioned in Article 7 seem to be sensible, both in terms of those forming part of the list and in their hierarchical ranking. They do not represent a radical departure from the common law rules for inter vivos trusts, although there, of course, no distinction is drawn between inferred choice and the law of closest connection. It is arguable, however, that the factors are less well suited to testamentary trusts, as the domicile or residence of the settlor is not even mentioned expressly. Of course, the factors mentioned are not exhaustive. Confining the court to a consideration of the four factors specified alone would be unduly rigid and fail 654 Cf. National Shawmut Bank v. Cumming 325 Mass 457, summarised in M Rosenberg, P Hay and R Weintraub, Cases and Materials: Conflict of Laws, 10th edn., 803. 655 P Nygh, Conflict of Laws in Australia, 6th edn., 516, considering Augustus v. Permanent Trustee Company (Canberra) Ltd (1971) 124 CLR 245. The same view is expressed by A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., 757. See also A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 127, p. 71: “. . . courts in common law countries will avoid applying a law which would invalidate a trust if they can reasonably choose a law which upholds its validity.” 656 A von Overbeck, “Trusts: the Hague Convention and the Law of the Netherlands”, 9e Journée des Professeurs Suisses de Droit International Privé, available at http://www.isdc.ch/e/vO97.asp. 657 See the section “implied intention to create a valid trust?” in relation to Art. 6, above.

228 Applicable Law in the Absence of Choice to allow due weight to be accorded to other potentially relevant factors such as e.g. the settlor and beneficiaries of the trust and the language and form of the trust deed and the place of its execution. Nor does it seem that the law of closest connection must be ascertained solely by reference to the terms of the trust instrument itself. Whilst this seems desirable, it is perhaps regrettable that this ability to look beyond the trust instrument exists at the Article 7 stage, but is strictly confined in the search for an implied intention pursuant to Article 6.

8. TIME FOR ASSESSING THE LAW OF CLOSEST CONNECTION

No statement is made in Article 7 as to the time at which the law of closest connection should be ascertained. The problem is that certain factors relating to a trust may not be known at the time when the trust is created e.g. if there are further trustees to be appointed, whose domicile or residence may prove important, or if the assets of the trust are sold and the proceeds subsequently invested in another jurisdiction. Nevertheless, it is submitted that, in principle, the law of closest connection to an inter vivos trust should be determined as of the time when the trust was created.658 This accords with the common law approach.659 If a trust is to be validly created, it must be subject to a particular legal system from the moment of commencement. It is, of course, true that the law of closest connection may subsequently change, in the event of new factors, or changes in existing factors, coming to light.660 However, that only strengthens the argument that, where no change in the applicable law is alleged, it should be determined as it was when the trust came into existence. That is not, of course, to say that subsequent circumstances which come to light should not be considered by the court, provided that they are used to elucidate the situation prevailing at the time of creation of the trust e.g. by demonstrating the “objects” of the trust. In the case of testamentary trusts, this rule would be unsatisfactory, however, since the trust will not come into existence until the testator’s death, and not at the time of creation of the will. In the intervening period, the testator’s domicile may change, as may that of the trustees or beneficiaries.

658 This is also the view of Cheshire and North, 1037 and of P Nygh, Conflict of Laws in Australia, 6th edn., 517. Dicey and Morris state (at 1090), perhaps a touch too firmly, that it is “clear” that the law should be ascertained as at the point of creation. 659 At common law, subsequent changes clearly did not affect the law of closest connection (at least where it was not alleged that the applicable law had been changed): see Duke of Marlborough v. A-G (No 1) [1945] Ch 78; Iveagh v. IRC [1954] Ch 364; Devos v. Devos (1970) 10 DLR (3d) 603; Chellaram v. Chellaram [1985] Ch 409. A similar position prevailed at common law in contract (see Amin Rasheed Corp v. Kuwait Insurance Co [1984] AC 50, discussed in Dicey and Morris 1215–6), but it now appears that, under the Rome Convention, supervening factors may be considered: Giuliano and Lagarde Report, OJ 1980 C282/4, 20. 660 Art. 10.

Terms Invalid by the Law of Closest Connections 229 “If . . . the time for applying the Article 7 factors is at the date of creation of the trust . . . this may create results not intended nor able to be predicted by the testator, creating obstacles and uncertainty in the estate planning process. . . . This is to be contrasted with the case of a change in domicile where it would appear that most individuals do appear to appreciate the need for review and possible revision of their legal arrangements when they move to another jurisdiction, including revising their wills”.661

Accordingly, it is suggested that the Article 7 factors should, in the case of testamentary trusts, be determined as of the date of creation of a will. Although Schoenblum thinks that this view is “far-fetched”,662 it would be very much preferable for the protection of the testator’s expectations and (given the importance which the Convention attaches to the written document creating or evidencing the trust663 and, implicitly, to the factors as they stood at the time of the creation of that document) it is very hard to see why such an interpretation should not be adopted. It cannot be denied, however, that it would have been preferable for this to have been stated in the Convention, since the fear remains that a foreign state which is asked to recognise the testamentary trust might take a different view as to the relevance of factors supervening after the will is created but before the settlor’s death.

9. WHAT IF THE LAW OF CLOSEST CONNECTION STATES THAT CERTAIN TERMS OF THE TRUST ARE NOT VALID ?

Of course, ascertaining the law of closest connection requires a court to study the terms of the trust.664 This creates something of a logical paradox: for until the applicable law of the trust is known, it cannot be said what the terms of the trust are.665 In a case where one party claims that an alleged term of the trust was not intended by the settlor or agreed upon by the trustee, the law of closest connection might be expected to determine the trust’s terms; but what is that law of closest connection? Suppose that it is alleged that a Utopian administration clause was subsequently inserted into the trust instrument by the trustee, 661 M O’Sullivan, “The Hague Convention on Trusts—Further Considerations” (1993) 2 J Int Corp P 65, 69. See also Jewish National Fund Inc. v. Royal Trust Co (1965) 53 DLR (2d) 577, where the Supreme Court of Canada refused to consider supervening changes in determining the law of closest connection. 662 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 8. 663 Arts. 3 and 6. 664 And to be sure that the trust itself is valid: see the discussion of “formation and essential validity of the trust” in relation to Art. 8, below. 665 Lewin on Trusts, 17th edn., 290 states that Art. 7 is concerned with “. . . the law with which the trust would be most closely connected assuming that it was valid. That is what might be called the putative closest law . . .” (emphasis in original). However, it is very dangerous to assume that all the terms of a trust are valid, where the parties are disputing the validity of particular terms thereof, since this appears to be tantamount to taking the side of the person alleging validity.

230 Applicable Law in the Absence of Choice which the settlor claimed that he did not specify. The settlor and trustee are resident in Ruritania, where the trust assets are situated. Which law should determine whether the administration clause forms part of the trust, given that the answer to that question may well prove determinative as to what the law of closest connection actually is? The answer must be that, first and foremost, identification of the law of closest connection is a process of drawing together various connecting factors. The law of closest connection is determined by the terms of the trust and its centre of gravity. Since it is clearly a matter for the law of the forum to determine the law of closest connection, it must also be for the lex fori to decide what the terms of the trust appear to be, in order to make that decision. Of course, the English court cannot be certain at this stage what the terms of the trust may be. All it can do is to ask whether it is satisfied, to its civil standard of proof, that the term in question is part of the trust.666 But what it the English court satisfies itself (provisionally) that Utopian law is the law of closest connection, relying largely on the administration clause? It then transpires that, by Utopian law, the administration clause is not valid, because it was not agreed upon by the settlor. How should an English court react? One of the key functions of the applicable law is to determine the terms of a trust. Accordingly, on one view a term which is invalid by the applicable law of the trust does not exist, and must therefore be disregarded in the search for the law of closest connection. If the clause were disregarded, then an English court may then consider that Utopian law is not the law of closest connection. It may decide that the law of closest connection is Ruritanian law. Even that does not settle the matter, since it may then transpire that, by Ruritanian law, the Utopian administration clause is valid. Bounced as the applicable law is from pillar to post, it may be impossible sensibly to find any law of closest connection; in which case the English court may have to resort back to applying its own domestic law, on the basis that the application of no other law can be proved.667 Rather, it is submitted that the law of closest connection is to be determined entirely by the law of the forum. In an English court, English law should determine whether it appears that there is (i) a trust; and (ii) what the terms of the trust are and use these findings to ascertain the law of closest connection. So the fact that Utopian law does not regard the Utopian administration clause as having been agreed upon should be disregarded for the purpose of identifying the

666 In England, this should mean the balance of probabilities. Any lower standard would be unfair to one of the parties, where the terms of the trust are themselves disputed. The burden of proof would need also to be allocated. In principle, the person alleging that a term forms part of the trust should have the burden of proving that it does. See also the discussion of “consent to the choice of law clause” in relation to Art. 6, above. 667 On the proof of foreign law, see generally, R Fentiman, Foreign Law in English Courts, (Oxford, Clarendon Press, 1998).

Terms Invalid by the Law of Closest Connections 231 law of closest connection, which is, after all, a matter solely of identifying where the connecting factor leads. Two major objections might be raised to this position. One is the obvious point that once the applicable law is determined by the court, that applicable law should decide finally what the terms of the trust are. This is true but, whatever the answer to that question, the issue of which is the applicable law should not be revisited. It is the law of the forum which determines the applicable law and which gives it its lifeblood. Only after that process has been resolved finally does the applicable law “take over.” A second objection may be that it is inconsistent with the approach advocated for the express or implied choice of law, where it is argued that the choice of law clause or implied choice must itself be valid according to the law putatively chosen.668 However, pragmatism, if nothing else, justifies such a position. If it is alleged that the settlor stipulated that Utopian law should govern a trust, it would be distinctly curious for an English court to find that Utopian law governed the trust, notwithstanding that, by Utopian law, the choice of law clause is invalid. Although one could shut one’s eyes to this fact by saying that identification of the chosen law is a connecting factor, and thus a matter only for the law of the forum, the fact remains that the choice of law clause is, after all, a term of the trust and should be finally validated by the putatively applicable law, just as any other term must be. But, more to the point, the consequences of Utopian law finding that a Utopian choice of law clause is invalid can be much more easily accommodated by an English court. There is no further search for an express choice of law and, if there is no implied choice either,669 the only remaining connecting factor is the law of closest connection. That connecting factor is the only one which is itself not based on choice.670 If an English court finds that Utopian law is the law of closest connection, any further reference to Utopian law for the purpose of identifying the law of closest connection can easily lead to an English court being unable to find a law of closest connection, in an example such as that considered above. At that point, there is no other connecting factor to turn to and the entire trust may fall to be governed by the law of the forum, a law which may have no connection at all with the trust. Accordingly, it is submitted that, whilst an express or implied choice of law must be valid by the putative applicable law, the law of closest connection in the absence of choice should be identified solely by reference to the law of the forum.

668 See the discussion of “consent to the choice of law clause” and of “which law determines whether a choice can be inferred?” in relation to Art. 6 above. 669 By analogy to the express choice of law, the English court should require an implied choice of law to have been agreed upon by the putative applicable law. 670 Albeit that, of course, other terms stipulated by the settlor may prove crucial to ascertaining what the law of closest connection is.

232 Applicable Law in the Absence of Choice

10. LAW OF CLOSEST CONNECTION AND THE BRUSSELS CONVENTION

Article 5(6), Brussels Convention confers special jurisdiction671 for claims brought against the defendant: “in his capacity as a settlor, trustee or beneficiary of a trust created by the operation of statute, or by a written instrument, or created orally and evidenced in writing, . . . [on] the courts of the Contracting State in which the trust is domiciled.”

It is for each contracting state to formulate rules as to where the trust is domiciled. In the United Kingdom, section 45(3), Civil Jurisdiction and Judgments Act 1982 states that, “A trust is domiciled in a part of the United Kingdom if and only if the system of law of that part is the system of law with which the trust has its closest and most real connection”.672 This has the advantage that, for trusts falling within the scope of both Article 5(6), Brussels Convention and Articles 2 and 3, Hague Convention673 (as extended in the United Kingdom), the courts of the state of closest connection should be able to apply the law of the forum in respect of trusts containing no express or implied choice of law.674

671 I.e. an alternative place of jurisdiction to that of the defendant’s domicile under Art. 2, Brussels Convention. Art 5(6) is reproduced in the Brussels Regulation (on which, see f/n 5, above). 672 Emphasis added. 673 “A ‘blind trust’ where the trustees have no accounting obligations, would be a trust under the Brussels, Lugano and Rome Conventions; it would not be a trust under the Hague Convention [because the accountability of the trustee is a characteristic expressly mentioned in Art. 2, Hague Convention]”: M Lupoi, “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees 15, 18, f/n 16. See also P Matthews, “Trusts & Equity: How Dumb is the Blind Trust?” (1998) 3 Amicus Curiae 22. 674 At least insofar as the law of closest connection for Art. 5(6), Brussels Convention purposes is determined in accordance with Art. 7, Hague Convention. See also A Dyer, “International Recognition of the Trust Concept”, (1996) 2 Trusts and Trustees 5, 10.

ARTICLE 8—SCOPE OF THE APPLICABLE LAW The law specified by Article 6 or 7 shall govern the validity of the trust, its construction, its effects and the administration of the trust. In particular that law shall govern(a) the appointment, resignation and removal of trustees, the capacity to act as a trustee and the devolution of the office of trustee; (b) the rights and duties of trustees among themselves; (c) the right of trustees to delegate in whole or in part the discharge of their duties or the exercise of their powers; (d) the power of trustees to administer or to dispose of trust assets, to create security interests in the trust assets, or to acquire new assets; (e) the powers of investment of trustees; (f) restrictions upon the duration of the trust, and upon the power to accumulate the income of the trust; (g) the relationships between the trustees and the beneficiaries including the personal liability of the trustees to the beneficiaries; (h) the variation or termination of the trust; (i) the distribution of the trust assets (j) the duty of trustees to account for their administration.

1. ARTICLE 8(1)

The wide-ranging list of factors to be determined by the applicable law is nonexhaustive.675 Indeed, it is meant in part to assist non-trust states to understand the nature of the issues involved in dealing with trusts. Curiously, one of the biggest questions of all, namely the existence and validity of the terms of the trust, is not one of the matters specified in Article 8, though one would certainly expect the applicable law to determine this.676 Nor is there any mention of the formal validity of trusts.677 Moreover, it is, of course, quite possible that dépeçage will be applied to the trust, so that one or more of these issues might be governed by a different law. The applicable law governs the trust’s “construction”, “effects” and “administration”.678 In relation to construction, this reflects the common law position.679 675

Von Overbeck Report, para. 81, pp. 387–8. On the determining of the trust’s terms for the purpose of the different exercise of ascertaining the law of closest connection, see the discussion of “what if the law of closest connection states that certain terms of the trust are not valid?” in relation to Art. 7, above. 677 See the discussion of “formal validity” below. 678 The “validity” of the trust, which Art. 8(1) also mentions, will be considered separately below. 679 Re Pilkington’s Will Trusts [1937] Ch 574; Trustees Executors and Agency Co Ltd v. Margottini [1960] VR 417; Philipson-Stow v. IRC [1961] AC 727; Re Levick’s Will Trusts [1963] 1 WLR 311; Perpetual Executors and Trustees Association of Australia Ltd v. Roberts [1970] VR 732. 676

234 The Scope of the Applicable Law Indeed, since the sole purpose of construing the trust instrument is to determine what the settlor meant, it would seem bizarre to refer construction to any other law. The only reservation to this position is that it is, of course, possible that the settlor might expressly or impliedly choose a law to govern solely the construction of the trust.680 The law applicable to matters of administration may be less straightforward. It may be that the settlor will have selected a law to govern the administration of the trust which differs from that which governs the trust as a whole. If so, von Overbeck points out that some of the matters stated in Article 8(2) will be governed by the law applicable to the administration of the trust.681 It is also quite possible that, in the absence of choice, a court will determine that matters of administration should be governed by a separate law through the process of dépeçage.682 Above all, there remains the complexity of deciding what the word “administration” may mean and whether it may sensibly be distinguished from questions relating to the trust’s essential validity. This issue is considered in more detail below.683 The “effects” of the trust are also governed by the applicable law. This rather vague term does not reveal much.684 Presumably, the effects are primarily those stated in Article 8 (and also in Article 11). The word “effects” cannot, however, be taken to include all effects of the trust on third parties. Although the ability to recover trust assets from third parties is envisaged by Article 11(3)(d) to be subject to the law applicable to the trust, that sub-section goes on to state that “the rights and obligations of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum.” In particular, the question whether a third party purchaser in good faith takes free of the beneficiary’s interest is likely to be resolved by the “ordinary” rules regulating the transfer of property.685

680

Pursuant to Art. 9. Von Overbeck Report, para. 88, p. 389. 682 So much is clearly envisaged in Art. 9. At common law, see Re Wilks [1935] Ch 645; Re Kehr [1952] Ch 26. However, this conclusion may be less likely today: see Chellaram v. Chellaram [1985] Ch 409, 431–3. 683 In the discussion of Art. 9. 684 If anything. 685 See further the discussion of Art. 11(3)(d), below. 681

Article 8(2)(a) 235

2. ARTICLE 8(2)

Let us now look at the specific provisions of Article 8(2).

(A) The Appointment, Resignation and Removal of Trustees, the Capacity to Act as a Trustee and the Devolution of the Office of Trustee (i) Appointment and removal The inclusion of “the appointment, resignation and removal of trustees” is an important departure from the English common law position.686 In Chellaram v. Chellaram,687 beneficiaries under a trust sought an order directing the removal of the trustees and their replacement. Scott J was faced with an argument by counsel for the trustees that if the proper law of the trust was Indian, then arguably Indian law should govern this question.688 Counsel for the beneficiaries contended that, whatever the proper law of the trust, matters of administration were subject to their own choice of law rule, namely the law of the place of administration, which in the instant case appeared to be England.689 Scott J declined to follow either argument, treating the removal and replacement of trustees as akin to a matter of procedure, to be determined by “the law of the country whose courts have assumed jurisdiction to administer the trust in question”.690 He drew a distinction between “the identification and extent” of the beneficiary’s rights (which should be referred to the proper law) and the “machinery for enforcement” of those rights, which “depends upon the powers enjoyed by the English courts”.691 Appointment and removal did not affect the nature and extent of the beneficiary’s interest. At least on one view “the decision seems at face value retrogressive, as merely representing a simplistic application of the lex fori . . .”692 The (substantive) right to procure the removal of a trustee and his replacement by another must exist by reference to a legal system. Indeed, if the beneficiary cannot procure the removal of an incompetent or ineffective 686 For the position in Scotland, see A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 637–8. For the position in America, see E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1035–6. 687 [1985] Ch 409. 688 Which, in turn, would be a reason for granting a stay of proceedings. 689 The view that matters relating to administration should be governed by the law of the place of administration was supported by the (then current) edition of Dicey and Morris: see 11th edn., 1987, Rule 121, 683 (see further the discussion of Art. 9, below). 690 [1985] Ch 409, 433; save in the unlikely event that the trust deed conferred rights of appointment and removal on the beneficiaries, which would be subject to the proper law of the trust: ibid., 432–3. 691 Ibid., 432. 692 D Hayton, “Trusts in Private International Law” (1985) 135 NLJ 18, 19. Hayton does not say that this represents his own view. See also D Evans, “Administering Foreign Settlements” (1986) 102 LQR 28, 30; A Wallace, (1987) 36 ICLQ 454, 479–82.

236 The Scope of the Applicable Law trustee, the quality of his enjoyment of the trust property, not to mention its value, may be adversely affected. Use of the law of the forum also creates a reason to forum shop. A further difficulty not apparently seen by Scott J was the decision in re Kehr.693 In that case, an intestate died domiciled in Germany. Letters of administration of his English estate were granted to the widow’s attorneys. They sought an order declaring that they had the power to appoint trustees under section 42, Administration of Estates Act 1925 and that, once appointed, those trustees would have the powers of maintenance and advancement under sections 31 and 32, Trustee Act 1925 in respect of an infant beneficiary. Danckwerts J, drawing support from the judgment of Farwell J in re Wilks,694 ruled that the act of appointment concerned the administration of the estate, and not succession thereto, so that German law did not apply to the question and he was entitled to appoint trustees pursuant to section 42, Administration of Estates Act 1925. Moreover, Danckwerts J ruled that the trustees would have the powers of maintenance and advancement under sections 31 and 32, Trustee Act 1925, on the basis that: “it seems to me that trustees . . . whose powers and whose constitution depend upon the provisions of English law . . . should . . . have the powers of other trustees constituted according to English law and accordingly have the powers of maintaining infant beneficiaries conferred by section 31 of the Trustee Act 1925”.695

In other words, Chellaram suggests that the act of appointment is a procedural question to be determined by English law, irrespective of the law applicable to the trust; re Kehr suggests that a trustee so appointed by English law enjoys all the powers conferred on a trustee of a trust governed by English law. The Hague Convention sweeps away such difficulty and subjects the appointment and removal of trustees to the applicable law. Provided that the relevant court has jurisdiction to remove and replace trustees of a foreign settlement, it should use the applicable law to determine whether to do so.696 That being so, it would be quite inappropriate to confer the statutory powers existing in 693

[1952] Ch 26. [1935] Ch 645. 695 [1952] Ch 26, 30. 696 Where English law governs the trust, the question whether a court should appoint nonresident trustees is a complex question (though not strictly a private international law question). In summary, it appears that, as a matter of English law, a court will not appoint a trustee resident overseas to an existing trust (of which the trustees are presently English residents) unless the trust has subsequently acquired a very strong connection with an overseas state, such as where the beneficiaries have become resident there: Re Liddiard (1880) 14 Ch D 310; Re Whitehead’s Trusts [1971] 1 WLR 833 (discussed by J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français 226); see also In the Estate of Ardagh [1914] 1 IR 5. However, English court’s now take a more relaxed view when a trustee with an express power of appointment chooses to appoint foreign trustees and will not intervene unless no reasonable trustee could have made such a choice : Richard v. Mackay, judgment of 4 March 1987, Ch D (transcript included in (1997) 11 Trust Law International 22) and Re Beatty’s Will Trust (No 2), judgment of 28 February 1991, Ch D (transcript included in (1997) Trust Law International 77). See generally, P Matthews, Trusts: Migration and Change of Proper Law. 694

Article 8(2)(a) 237 English domestic law on trustees, save in the case where the trust is governed by English law.697 Schoenblum regards the subjection of the removal of trustees to the applicable law as unacceptable. He has in mind specifically testamentary trustees, whose removal is “carefully regulated by statute and generally treated as [a] matter . . . of procedure”.698 However, the objection is arguably over-stated, for two reasons. First, it is a recipe for distortion if the applicable law is used to determine the powers and duties of trustees, but not the circumstances in which they can be removed from office. Secondly, the forum has ample residual resources to apply its own international mandatory rules, or to disapply the governing law on public policy grounds.699 Article 8(2)(a) does not deal with the vesting of the property in the hands of new trustees. Strictly speaking, one might say that this is a property transfer and should attract the normal choice of law rule for inter vivos property transfers, namely the law of the place where the property is situated.700 However, this approach should not be followed. Property needs to pass from the settlor to the trustees in order to launch the rocket and create a valid trust in the first place. That initial vesting of legal title is properly subjected to the lex situs. However, where a trust has already been launched and the settlor has already divested himself of legal title, its passing as between trustees from time to time occurs within the context of a validly created trust and it is proper that the transfer should be determined by the law governing the trust.701 (ii) Capacity of the trustee The other important matter covered by Article 8(2)(a) is the capacity to act as a trustee. In many ways, it is surprising to find such a provision in the Convention. Capacity is normally regarded as a matter to be excluded from harmonised private 697 There is no reason to think that such powers should today be treated as mandatory rules of English law. Indeed, Danckwerts J doubted whether trustees constituted by the law of a foreign state would have the statutory powers conferred by English law: [1952] Ch 26, 29–30. 698 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 12. 699 See Arts. 16 and 18 respectively. It would, however, be contrary to the spirit of the Convention routinely to impose the forum’s law on removal of trustees, in the light of Art. 8(2)(a). 700 But cf. M Koppenol-Laforce, Het Haagse Trustverdrag, 264: “If such a decree results in the property directly vesting in the new trustee, this can only become effective with regard to property that is situated in the country of the court, or in such cases where the decision can be enforced on the basis of a convention.” 701 With the possible exception of trusts of immovable property. The fact that, as a matter of Utopian law, legal title to immovable trust property situated therein has not passed from the old to the new trustee is likely to make any purported dealings with the property in the new trustee’s capacity as legal owner ineffective. On the other hand, the policy of the Convention is not to apply the lex situs once a trust has been launched, even if the property in question is immovable. Hence a trustee of a trust governed by Ruritanian law who invests trust assets in immovable property situated in Utopia does not have to show that he has authority to do so by Utopian law. Accordingly, it is more likely in practice that a court would simply subject the question of whether legal title is vested in a new trustee entirely to the applicable law of the trust, whatever the nature of the property in question.

238 The Scope of the Applicable Law international law legislation.702 It is also a matter for which use of the applicable law703 is not thought appropriate. Capacity rules tend, by their nature, to have a mandatory characteristic. A person who, according to his personal law, lacks capacity to act as trustee due to his age or the fact that he has been declared bankrupt, should not obviously be able to have capacity conferred upon him by the settlor’s choice of an (objectively unconnected) law to govern the trust, by which the trustee has capacity.704 In fact, most opposition to use of the applicable law of the trust came from civil law delegations. It was pointed out that the law governing the trust might in fact impose greater capacity restrictions than any other law which might be applied.705 However, this seems unconvincing, since a settlor is most unlikely to choose a law by which the original trustee lacks capacity.706 Furthermore, presumably the settlor could split the trust, so that the issue of the trustee’s capacity would be governed by a different law to the rest of the trust.707 Given the inclusion of the trustee’s capacity within the Convention, it seems at first sight rather curious that the capacity of the settlor (and, it appears, that of the beneficiary) are outside the Convention, and appear to be treated as “rocketlauncher” questions.708 Why should the settlor be unable to confer capacity on himself and on the beneficiary by selection of the applicable law, whilst the settlor can confer capacity on the trustee by a choice of law?709 Nor is it obviously in the beneficiary’s interest that his beneficial interest should be managed by a trustee whose maturity, competence and reliability are not free from question. However, it is only the trustee’s capacity to act qua trustee which is covered by Article 8(2)(a). The questions whether he has “general” capacity to receive a property interest at all (as opposed to whether he can act as trustee of that interest) is not a trusts specific matter and ought to be governed by general property principles as a “rocket-launching” matter, which would tend to point to the lex situs.710 It is not at all obvious that a person who would not have capacity to 702 The most notable examples in the United Kingdom being the exclusion in the Brussels Convention (Art. 1(1)) and the Rome Convention (Art. 1(2)(a)). 703 I.e. the law governing the trust’s essential validity. 704 See further the discussion of the settlor’s capacity in Part One of this book. 705 Von Overbeck Report, para. 85, p. 388. See also Working Document No 25. 706 It may be that subsequently appointed trustees would fall foul of the applicable law. Of course, the settlor could have declared the trust without considering who might act as the (original) trustee of the settlement, but this is unlikely routinely to be the case. 707 Pursuant to Art. 9. 708 Von Overbeck Report, para. 59, p. 382. 709 One argument that might be made is that the settlor and beneficiary must have capacity to alienate and receive property at all under the “general” property law, whilst there is no question of the trustee receiving the benefit of any property. That does not hold water, however, since the trustee still receives legal title to the property and will normally be empowered to alienate that property to a third party. Nor can it be said that because the express trust is a voluntary creation, the autonomy to create it should extend to the issue of the trustee’s capacity (by choosing the law to govern capacity). This would not explain why the autonomy does not extend to the capacity of the settlor or beneficiary. 710 In the case of inter vivos transfers. In the case of testamentary transfers of movables, it would point to the law of the deceased’s last domicile.

Article 8(2)(b) 239 receive property outright by the lex situs should be able to receive property on trust, thereby bypassing that law.711

(B) The Rights and Duties of Trustees among Themselves This sub-section subjects the internal responsibility of the trustees to the applicable law. This might seem correct in principle. Those who have agreed to act as trustee will have done so pursuant to a trust governed by a particular law. In accepting the office, they may be deemed to have accepted that their relationship with one another will be governed by that same law. However, not every issue concerning the rights and duties of the trustees between themselves is obviously a “trusts” matter. As a matter of English law, where several trustees are liable for a breach of trust, their liability is joint and several.712 Suppose that the beneficiary sues for breach of trust and recovers the entire loss from one of the trustees. The trustee may then seek a contribution from the other trustees,713 pursuant to the Civil Liability (Contribution) Act 1978, section 1(1) of which provides that “any person liable in respect of any damage suffered by another person may recover contribution from any other person liable in respect of the same damage (whether jointly with him or otherwise).”

Such a claim might be considered to be part of the law of restitution.714 In fact, and for reasons which are not convincing, it has been held that the when the conditions of the Civil Liability (Contribution) Act 1978 are met, it applies in an English court regardless of the applicable law, and to the exclusion of that applicable law.715 This means that, notwithstanding Article 8(2)(b), contribution claims falling within the Act will be governed by the law of the forum.716 Where the 1978 Act does not apply, it would be possible to characterise the contribution claim as restitutionary and to subject it to the law governing such claims.717 Indeed, it could be argued that such a claim does not concern “the law applicable to trusts” and is outside the scope of Article 1, Hague Convention. But the tone of the Convention would suggest otherwise. Whilst it might be a 711

See further ch. 5 of Part One of this book. See further R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn., 649–50; J McGhee, Snell’s Equity, 30th edn. (London, Sweet & Maxwell, 2000), 333–5. 713 Wilson Moore (1833) 1 My & K 126, 146; re Harrison [1891] 2 Ch 349. 714 Dicey and Morris, 1488–90. 715 Arab Monetary Fund v. Hashim (No 9), The Times, 11 October 1994; A Briggs, “The International Dimension to Claims for Contribution” [1995] LMCLQ 437. See also The Kapetan Georgis [1988] 1 Lloyd’s Rep 352, 357–9; Petroleo Brasiliero SA v. Mellitus Shipping Inc and Others, [2001] 2 Lloyd’s Rep 203. See generally K Takahashi, Claims for Contribution and Reimbursement in an International Context: Conflict of Laws Dimensions to Third Party Procedure (Oxford, Clarendon Press, 2000). 716 Pursuant to Art. 16. 717 Dicey and Morris, Rule 200, 1485. 712

240 The Scope of the Applicable Law fiction to suggest that the trustees impliedly intended to subject their right of contribution inter se to the law applicable to the trust, such a law will probably meet their expectations. They can be viewed as having subjected their personal liability, be it joint, several or both, to the law applicable to the trust. In the situation where that liability arises, it makes sense for the consequences as between the trustees to be subjected to the same law.718

(C) The Right of Trustees to Delegate in Whole or in Part the Discharge of their Duties or the Exercise of their Powers The trustee’s right to delegate is determined by the applicable law. However, whilst this provision deals with the right to delegate, it does not deal with the consequences of such delegation. In particular, it does not deal with the primary and vicarious liability of the trustee upon delegation. As to primary liability of the trustee, this clearly arises pursuant to the trust instrument. The trustee’s personal liability to the beneficiaries is a matter for the applicable law of the trust.719 It would be wrong for the beneficiaries’ rights to be adversely affected by any acts of the trustee. Unless the law applicable to the trust states that the trustee may delegate their duties and states that such delegation may limit or exclude the personal liability of the trustee, then the trustee’s personal liability will remain unaffected. The question of the trustee’s vicarious liability for the acts and defaults of the delegate may be complicated if the delegate is employed pursuant to a contract720 which provides that the trustee shall bear no liability for the delegate’s actions. Once again, it might be said that if the trustee’s power to delegate derives from the law applicable to the trust, the consequences of that delegation should be governed by the same law. Some support for this view could be drawn from the sphere of torts, where it appears that vicarious liability is determined by the applicable law of the tort.721 It is suggested that, whilst the applicable law of the trust should accordingly determine the vicarious liability of the trustee for a properly appointed delegate, the nature and terms of the relationship formed between the trustee and the delegate is a matter for whatever law governs the relationship between them.722

718

This appears also to be the view of Dicey and Morris: see 1501. Art. 8(2)(g). 720 And if the law applicable to the contract differs to that which is applicable to the trust. 721 At common law, see the Halley (1968) LR 2 PC 193; the Waziristam [1953] 1 WLR 1446; Church of Scientology California v. Commissioner of Police (1976) 120 SJ 690. See also Dicey and Morris, 1525–6. See further Law Commission No 193 on Choice of Law in Tort and Delict (London, HMSO, 1990), para. 3.55. 722 Compare Dicey and Morris, 1526. 719

Article 8(2)(d) 241

(D) The Power of Trustees to Administer or to Dispose of Trust Assets, to Create Security Interests in the Trust Assets, or to Acquire New Assets This is one of the key provisions. It covers a number of issues which can be classed in two main categories: (i) the administration of the trust and; (ii) the trustee’s power to deal with and acquire trust property. These will be examined in turn. (i) Administration As to “administration”, more will be said in the context of Article 9, where the meaning of the word will be considered. It should be noted that the law applicable to the administration of a trust may well not be the law applicable to the rest of the trust, since it is the prime candidate to be the subject of dépeçage. In the absence of dépeçage, it is important to note that the vast body of law that is administration of trusts is subjected to the law applicable to the trust. Whilst this might seem not in the least surprising, there is English common law authority which applies English statutory provisions concerning trustees’ powers to trusts governed by a foreign law, where the trustee is appointed in an English court.723 It has been held that where a grant of administration is made in England, section 33(1), Administration of Estates Act 1925 (which grants statutory powers to the trustee to postpone sale) applies irrespective of the law applicable to the trust.724 The same is true of the maintenance and advancement powers contained in sections 31 and 32, Trustee Act 1925.725 In contrast, trustees of a foreign trust appointed overseas have been held not to have the statutory powers which exist in English law when litigating in an English court.726 Despite the apparently clear wording of Article 8(2)(d), Dicey and Morris suggest that such common law authorities “appear not to be affected by [the Recognition of Trust Act 1987’s] provisions”.727 Technically, this could be so. It could be said that these provisions were applied pursuant to the rules relating to the administration of estates and not the law of trusts, so that the Hague Convention has nothing to say on them. Certainly, they were so regarded by Farwell J in re Wilks,728 who treated the issue in question as one of characterising the powers of maintenance and advancement under the Administration of Estates Act 1925 as concerning administration of estates or trusts. He took the 723 Or, more precisely, where the trustee is appointed according to English law in an English court. See Re Wilks [1935] Ch 645; Re Kehr [1952] Ch 26; Re Ker’s Settlement Trusts [1963] Ch 553; Chellaram v. Chellaram [1985] Ch 409. 724 Re Wilks, ibid. 725 Re Kehr [1952] Ch 26. See also Dicey and Morris, 1094. 726 Re Kehr ibid.; Chellaram v. Chellaram, [1985] Ch 409. 727 Dicey and Morris, 1094. 728 [1935] Ch 645.

242 The Scope of the Applicable Law view that the act of administration of an estate did not cease until the estate could be distributed amongst the persons absolutely entitled, which, in the case of infants, meant not until they reached the age of majority. Hence the powers were exercised pursuant to the administration of the estate and were not affected by the law governing the trust. However, such a view is hard to maintain. They are powers which arise pursuant to a valid trust. In other words, these powers arise not as part of the “rocket-launching” process, but as part of the trust itself; it follows that their application should be governed by the Hague Convention. Alternatively, such rules could be treated as mandatory irrespective of the applicable law and applied pursuant to Article 16. However, it is desirable to keep the scope of Article 16 as narrow as possible, in order to maintain the integrity of the choice of law rules of the Convention, to accord with Article 8 and to apply the foreign law in something resembling the manner in which it would be applied in the relevant foreign court.729 It is difficult to see a pressing need for powers of sale or maintenance and advancement to be routinely subjected to English law in an English court. If the powers afforded by the governing law to a trustee to act in the interest of minor beneficiaries are wholly and unacceptably inadequate, that might be a reason for an English court to invoke the public policy provision of Article 18; if not, then the applicable law should simply be applied to determine the trustee’s powers. (ii) The trustee’s power to deal with and acquire trust property There is nothing striking in the Convention according to the trustee of a duly constituted trust the power to administer and dispose of trust assets, or to create security interests in them, in accordance with the law applicable to the trust. However, the power conferred in sub-para (d) self-evidently affects transactions in trust property. One might expect the lex situs to have something to say about the divesting or acquisition of property and security interests. Hayton argues that: “it will be the lex situs that determines what kinds of security interests may be granted, and in what manner, where a trustee has power under heading (d) to create them”.730

Put differently, it might be said that sub-paragraph (d) determines the trustee’s capacity to create such interests and nothing more. It is vital to define the respective limits of the lex situs and the law applicable to the trust in this area. The Convention does not accord any significant role to the law of the place where the property is situated in respect of trust assets and it cannot be fatal to the acquisition or disposition of trust property that the trust structure is not known in the situs. On the other hand, the applicable law of the 729 730

Although this aspiration does not extend to application of the doctrine of renvoi: Art. 17. D Hayton, (1987) 36 ICLQ 260, 273.

Article 8(2)(e) 243 trust does not apply to “rocket-launching” aspects of the trust. This includes the question of whether legal title to property has vested in the hands of the trustee. Accordingly, the position should be as follows: (α) If the trustee has the power, according to the law applicable to the trust, to alienate the trust property, the question whether title has passed to the transferee is to be governed by the lex situs. (β) If the trustee purports to create a security interest in the property, the question whether he can create such an interest is a matter for the law applicable to the trust. However, the question of whether such an interest has been created is a question of whether a property right has passed to the transferee and should be resolved by the lex situs. (χ) If the trustee purports to acquire new assets for the trust, there will have been an alienation of the property from a third party to the trustee. The question whether legal title in that property has passed to the trustee must be determined by the lex situs.731 But, consistent with the “rocket-launching” rules relating to assets transferred to a trustee on trust, the lex situs need not know the concept of equitable title, provided that the transferor has alienated the legal title to the property. Henceforth, the new assets will be subject to the trust and the law applicable thereto. (E) The Powers of Investment of Trustees The power of investment of trustees is one of the most essential aspects of the proper administration of the trust. It is, of course, possible that matters of administration will be the subject of a law separate from that applicable to the rest of the trust. But whichever law governs the question of administration, this law must also govern the investment powers of the trustee. Presumably, such law will also determine the consequences of an unauthorised investment or a failure to invest to the appropriate standard of care. It ought not to be the case that legislative provisions concerning investment be treated as international mandatory rules in an English court.732 If they were, they would be superimposed onto the applicable law. However, even before the decline of the Trustee Investments Act 1961,733 its provisions could be opted out 731 Note that this author did not advocate above that, where a new trustee of an existing settlement is appointed, legal title must be vested in him by the lex situs (see the discussion of Art. 8(2)(a)). The reason is that there, the settlor had already divested himself of legal title to property, which was being held on trust and it is not obvious that the lex situs should have a role to play in determining in which trustee legal title is now vested. However, in the present case, we are dealing with the transfer of a property interest from a third party to a trust, or vice-versa; it is a “general” alienation of property and the lex situs should be applied to determine whether legal title therein has passed. 732 On the position in Scotland, see A Anton and P Beaumont, Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 638. 733 The technical restrictions on what constitutes an authorised investment have been largely swept away by the Trustee Act 2000.

244 The Scope of the Applicable Law of in whole or in part and, where applicable, the rules on the proportion of riskier “wider-range” investments which could be made under the Act had been liberalised.734 In other words, the Trustee Investments Act 1961 contained neither international nor domestic mandatory rules. If, by the applicable law, exceptionally risky investments can be made without the need to diversify investments, and if the standard of care expected of the trustee is low, these are but consequences of use of the applicable law;735 and an English court should be very slow to reach the conclusion that application of the governing law in this area offends English public policy.736

(F) Restrictions upon the Duration of the Trust, and upon the Power to Accumulate the Income of the Trust This is one of the more awkward provisions, since it takes us into the area of perpetuities which, at least in some cases, is better regarded as part of the “rocket-launching” process.737 A trust which fails to comply with the law of perpetuities is arguably never launched in the first place.738 Moreover, in policy terms, it could be argued that: “. . . the primary interest in striking at the creation of rights in moveables [or immoveables] over an indefinite series of lives would seem to be that of the country where the assets are situated”.739

However, it was argued above740 that the key question is whether the accumulation or perpetuity restriction relates specifically to the creation of a trust (or trust-like structure) or is a more general restriction on the creation of property rights. If it is the former, it should be governed by the law applicable to the trust; if the latter, it should be treated as a “pre-trust” matter and outside the Convention’s scope. In England, section 164, Law of Property Act 1925 would appear on its wording to affect any purported transfer subject to excessive accumulation of income; in practice, however, it will normally be associated with the 734 From one-half of the fund (s. 2(1) Trustee Investments Act 1961) to three-quarters (Trustee Investments (Division of Trust Fund) Order 1996, SI 1996/845). See R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 538–9. 735 As would be a scenario where the investment powers of the trustee are tightly restricted. 736 Although it may be possible to vary a transnational trust in an English court: see the discussion of the variation of transnational trusts in relation to Art. 8(2)(h), below. 737 Although A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 639 points out that perpetuity issues were at common law generally regarded as an aspect of the essential validity of the trust itself: see e.g. Re Fitzgerald [1904] 1 Ch 573; Irving v. Snow 1956 SC 257. See also Saliba v. Falzon [1998] NSWSC 302. See further section 8 of Part One of this book. 738 See the discussion of Art. 4 above, Art. 15 below and section 8 of Part One of this book. See also D Hayton, (1987) 36 ICLQ 260, 269; Lewin on Trusts, 17th edn., 299–300. 739 A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 639. 740 See the discussion of Art. 4, above.

Article 8(2)(g) 245 creation of the trust structure and should, accordingly, be applied as a “rocket” provision only where English law is the governing law of the trust.741 It should not be applied under Article 8(2)(f) where English law is not applicable to a trust. That said, suppose that a settlor purports to create a trust which provides for the accumulation of income. The trust is governed by the law of Utopia (by whose accumulation rules it is valid) and concerns land situated in England (by whose accumulation rules it is invalid). In that scenario, how would an English court deal with such a trust, which manifestly infringes its own accumulation rule? The rules against accumulations and perpetuities are largely rooted in the desire to facilitate the alienation of property. Where property located in England has been rendered inalienable, this may lead an English court to conclude that application of the accumulation or perpetuity period of the governing law of the trust infringes English public policy.742 This might equally be true of an unreasonably short perpetuity period of the applicable law of the trust.743

(G) The Relationships between the Trustees and the Beneficiaries including the Personal Liability of the Trustees to the Beneficiaries (i) Personal and vicarious liability The language used in this sub-paragraph is extremely broad. The personal, compensatory liability of a trustee to a beneficiary for breach of trust is a matter which will clearly be subjected to the applicable law. Equally, it has been argued that vicarious liability for the acts of a delegate should be so determined.744 (ii) Who is a “beneficiary”? The meaning of the word “beneficiary” in Article 8(2)(g) may occasionally prove problematic. In the case of a discretionary trust, suppose that members of the qualifying class are not awarded a share of the trust fund by the trustee. If 741 Even if the property is located overseas. See also the Perpetuities and Accumulations Act 1964. See further Law Commission Report No 251 on the Rules against Perpetuities and Excessive Accumulations (1998). 742 Invoking the public policy provision of Art. 18, Hague Convention. It would not be appropriate to treat the periods specified by the English rules against accumulations and perpetuities as mandatory rules of international application. There may be perpetuity rules of another state which differ in the permissible duration of a trust, but which are well within the bounds of acceptability. 743 There may be an analogy to limitation periods. S. 1(1), Foreign Limitation Periods Act 1984 treats limitation periods as substantive and for the applicable law to determine. However s. 2(1) provides that if application of that limitation period “would to any extent conflict . . . with public policy, that section shall not apply to the extent that its application would so conflict.” See further P Carter, “The Foreign Limitation Periods Act 1984” (1985) 101 LQR 68; P Stone, “Time Limitation in the English Conflict of Laws”, [1985] LMCLQ 497. 744 See the discussion of Art. 8(2)(c) above.

246 The Scope of the Applicable Law they were to claim that the trustee had exercised his discretion improperly, should the matter be treated as concerning “the relationships between the trustees and the beneficiaries”? Arguably, the answer should be “yes”, since the claimants are745 putative beneficiaries. But on any view, the law governing the trust should determine this matter, since it goes to the obligations of the trustee under the trust. However, a further problem might arise if the trust is “split”, so that matters of administration are governed by a law separate from that which governs the other issues under the trust.746 The trustee’s exercise of discretion might be considered an exercise of his administrative power. However, the choice goes right to the heart of the question of the respective rights of the potential beneficiaries. As such, it is better not to be classified as a matter of administration, but should rather be determined by the law governing the essential validity of the trust. (iii) Fiduciary duties owed by the trustee to the beneficiary Difficulties arise when the obligation owed by the trustee to the beneficiary is one which is not trust specific. Those obligations which arise from the fiduciary position of a trustee may apply equally to the relationship between e.g. solicitor and client or agent and principal.747 Moreover, the duty of the fiduciary not to make an unauthorised profit from his position may lead to a claim by the beneficiary which is in substance restitutionary and which ought arguably to be subjected to the appropriate restitutionary choice of law rules.748 The question of whether a fiduciary relationship arises in a transnational context appears not to be the subject of an orthodox choice of law rule at all, at least where there is no other “specific relationship” between claimant and defendant.749 Reasons proffered for this approach, which are rooted in the Courts of Equity and their jurisdiction to prevail upon the conscience of the defendant through in personam orders, seem misguided and to provide no cogent reason in the present day for the marked failure to develop coherent choice of law rules for equitable claims.750

745

Or, perhaps, were. Pursuant to Art. 9 (on which, see below). 747 See R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., ch. 29. 748 On which, see Dicey and Morris, ch. 34 ; Cheshire and North, ch. 20; F Rose (ed.), Restitution and the Conflict of Laws; G Panagopoulos, Restitution in Private International Law; O Elias, Judicial Remedies in the Conflict of Laws (Oxford, Hart Publishing, 2001), ch. 10. 749 Paramasivam v. Flynn (1998–99) 160 ALR 203, considered below. See also El Ajou v. Dollar Land Holdings plc, [1993] 3 All ER 717 (reversed on different grounds: [1994] 2 All ER 685); Arab Monetary Fund v. Hashim (No 9), The Times, 11 October 1994; Trustor AB v. Smallbone, judgment of 9 May 2000 (CA); Kuwait Oil Tanker v. Al Bader [2000] 2 All ER (Comm) 271 (and see G Virgo, “Interest, Constructive Trusts and the Conflict of Laws” (2000) 8 RLR 122); Grupo Torras v. AlSabah [2001] Lloyd’s Rep Bank 36 (noted by J Garton, (2001) 15 Trust Law International 93). 750 R White, “Equitable Obligations in Private International Law: the Choice of Law” (1986) 11 Sydney LR 92; T Yeo, “Choice of Law for Fiduciary Duties” (1999) 115 LQR 571, 573. 746

Article 8(2)(g) 247 In Arab Monetary Fund v. Hashim (No 9),751 Chadwick J remarked that: “In the context of a claim to invoke its equitable jurisdiction it is for the English court to decide whether the necessary fiduciary relationship exists. Where the duties to which a relationship gives rise are determined by foreign law, the question for the foreign law is what is the nature of those duties. It is for the English court to decide whether duties of that nature are to be regarded as fiduciary.”

In other words, an English court may allow the governing law to determine the obligations of one party to the other; but English law will then finally determine whether the relationship, viewed in its context, is to be classified as a fiduciary one. English law will also decide, presumably, whether additional obligations should be imposed, such as it would require of any fiduciary in English domestic law (e.g. the requirement not to profit from the position), insofar as the governing law does not itself impose this obligation. The Federal Court of Australia in Paramasivam v. Flynn,752 in a rather anachronistic judgment, treated the law of the forum as the governing law for “general” breaches of fiduciary duty. There was authority in the shape of National Commercial Bank v. Wimborne753 for application of the law of the forum. In Paramasivam, the court cited Nygh, who states that: “The forum determines according to its own law whether a personal equity exists between the parties. It is not concerned with the question whether the lex situs recognises the equitable right or anything like it”.754

However, Nygh goes on severely to criticise the rule. Moreover, Scott on Trusts, also quoted by the Court, adopts a different line: “It is generally held that the applicable law is that of the state in which the fiduciary relationship is centred. . .”755 Miles, Lehane and Weinberg JJ were unmoved, preferring to apply the law of the forum: “Such an approach . . . has considerable advantages, both of convenience and of certainty over one which would seek the place where the relationship was “centred” or with which it had its closest connection.”

Suffice it to say that, if simplicity and predictability were the only relevant principles, the law of the forum would always be applied by a court. Moreover, the Federal Court’s ultimate choice of law rule was for: “the general application of the lex fori, subject, perhaps, to this: that where the circumstances giving rise to the asserted duty or the impugned conduct (or some of it) occurred outside the jurisdiction, the attitude of the law of the place where the circumstances arose 751 The Times, 11 October 1994; cited with approval in Dubai Aluminium v. Salaam [1999] Lloyd’s Rep 415 (appealed: (2000) 2 Lloyd’s Rep 168); Kuwait Oil Tanker v. Al Bader [2000] 2 All ER (Comm) 271; Grupo Torras v. Al-Sabah [2001] Lloyd’s Rep Bank 36. 752 (1998–99) 160 ALR 203. 753 (1978) 5 BPR [97323], at 11, 982. 754 P Nygh Conflict of Laws in Australia, 6th edn., 117. 755 A Scott and W Fratcher, The Law of Trusts, 4th edn. (Boston, Little, 1989), §665.

248 The Scope of the Applicable Law or the conduct was undertaken is likely to be an important aspect of the factual circumstances in which the Court determines whether a fiduciary relationship existed, and, if so, the scope and content of the duties to which it gave rise”.756

This “enlightened” law of the forum, if it is properly described as a choice of law rule at all, calls for the discretionary application of the objective proper law of the relationship. It is not easy to conceive of a rule which less demonstrates the characteristics of “convenience and certainty.” It “allows the court to diminish the influence of domestic law while retaining control over the types of foreign liabilities to admit—it can be worked as a public policy sluice-gate”.757

However, whatever law might apply to other (“general”) fiduciary relationships, should that which arises in the context of the trustee/ beneficiary relationship be treated differently and subjected to the applicable law of the trust?758 At first sight, it might appear that Article 8(2)(g) answers this question. It is not expressly confined to the personal liability of the trustee to the beneficiary qua trustee. Furthermore, that provision only “includes” personal liability as one aspect of “the relationships between the trustees and the beneficiaries.” Unfortunately, Article 8(2)(g) is not conclusive, at least in respect of the liability of the fiduciary to hold unauthorised profits on constructive trust. This is because this constructive trust may not fall within the scope of the Convention under Article 3,759 so that Article 8 can have no application to it. Accordingly, it is safer to determine the law applicable to a trustee’s fiduciary obligations from principle. An argument against applying the law governing the trust in this context would be that the office of a trustee creates not one set of obligations but two: the obligations under the trust and the obligations under the “general” law of fiduciary liability. It may be that a different, specific choice of law rule governs the latter.760 The prophylactic policy underlying the fiduciary liability, which forces a fiduciary who makes an unauthorised profit to reverse the enrichment which he receives from his wrongdoing, might be said to stand separately from the enforcement of the contract, trust etc. under which the fiduciary relationship arose. Fiduciary liability would then be characterised separately and governed by its own choice of law rules. There is an analogy with the case of concurrent actions in the law of contract and tort. If an employee is injured due to the failure of his employer to provide him with a safe working environment, then he 756

(1998) 160 ALR 203, 217. T Yeo, (1999) 115 LQR 571, 575. 758 G Panagopoulos, Restitution in Private International Law, 149–51. 759 See above. However, the United Kingdom has extended the scope of the Hague Convention’s application in s. 1(2), Recognition of Trusts Act 1987. A constructive trust imposed upon a fiduciary by the law of a part of the United Kingdom will be subject to the Convention regime, including Art. 8(2)(g). 760 Such a view is supported by L Barnard, “Choice of Law in Equitable Wrongs: a Comparative Analysis” [1992] CLJ 474, 503–7. 757

Article 8(2)(g) 249 may be able to sue for breach of contract or in negligence. Unless the two claims are characterised as mutually exclusive for private international law purposes, the claimant may have a choice how to frame his action.761 In some jurisdictions, the view has developed that the “general” law of tort should be subordinated to any specific arrangements that the parties have made, so that the contractual choice of law rule would be applied to the exclusion of the tortious choice of law rule.762 Although “the concept as such is unknown in the common law”,763 its main advantages are that it promotes clarity and is likely best to reflect the parties’ expectations.764 By analogy, it might be argued that, whatever law applies to fiduciary claims, the specific arrangement made by express trust, which will import as part of that specific arrangement a fiduciary relationship,765 should also prevail over whatever law governs “general” claims against a fiduciary.766 This author does not favour the application of the law applicable to the contract to the exclusion of the law applicable to the tort in the case of concurrent actions.767 Where rights arise, and obligations are imposed, under the “general law” of tort, it is far from obvious that they should be subordinated to the parties’ autonomy. Nonetheless, it might be possible to distinguish the contract/ tort scenario from the trustee’s obligations under a trust and his fiduciary obligations. In the above example, the tort claim clearly would not arise factually if the defendant had not been working on the defendant’s premises pursuant to a contract of employment. Legally, however, the claim is independent. If the claimant had been working for the employer free of charge, it would still have arisen. In contrast, the trust relationship legally creates fiduciary obligations which would not otherwise arise.768 Without the trust, there is no fiduciary relationship.769 This tends to justify the trustee’s fiduciary duties being subjected entirely to the law applicable to the trust.

761 See J Harris, “Choice of Law in Tort—Blending in with the Landscape of the Conflict of Laws?” (1998) 61 MLR 33. 762 P Nygh, Autonomy in International Contracts, 240–7. 763 Ibid., 243. However, Nygh goes on to argue that a number of more recent common law developments are consistent with this “accessory choice of law rule” approach. 764 Although it is not an approach that this author would favour: see the discussion of “contracts and trusts” in relation to Art. 6, above. 765 Or, at least, the common law trust will import this relationship. 766 However, where rights arise, and obligations are imposed, under the “general law”, it is far from obvious that this approach should be adopted. 767 Rather, he would advocate that, insofar as the process of characterisation leads to concurrent claims in contract and tort, the claimant should be able to choose how to frame his action. 768 See also G Panagopoulos, Restitution in Private International Law, 150, who comments that “One has to be careful to distinguish those cases where the existence of a particular relationship is an essential component of the issue in dispute and those where the particular relationship is a mere circumstantial fact.” 769 At least if there is no other fiduciary bond between the parties, as there might be if e.g. a solicitor is appointed a trustee for his client.

250 The Scope of the Applicable Law Although Barnard argues for a separate choice of law rule for claims against fiduciaries, centred around the law with which the claim has its closest and most real connection,770 she states that this: “will not subvert any established principles or international commitments owed under the Hague Convention on Trusts. The choice of law rule . . . applying to trustees . . . [is] consistent with, and indeed . . . [is] merely [a] specific application . . . of, the proper law approach proposed for the category of other fiduciaries and confidants. The proper law of the trust . . . [is] the legal system . . . governing the legal relationship . . . by which trustees . . . gain access to others’ assets”.771

This suggests that the law applicable to fiduciary liability would, in the express trust context, be the law applicable to the trust. Barnard’s general approach to equitable wrongs is not self-evidently correct. Stevens criticises her for treating liability for equitable wrongs as a category for choice of law purposes at all.772 Where the fiduciary is required to account for profits, they may be better classified as examples of restitution for wrongdoing and subject to the choice of law rules for such claims.773 However, even if Barnard’s general treatment of equitable wrongs for private international law purposes is questionable, the important point for present purposes is that, where trustees of an express trust are concerned, her approach leads to the application of the law applicable to that trust. Further support for use of the law applicable to the trust to determine the fiduciary obligations of a trustee is provided by Dicey and Morris. The authors argue that, in applying the restitutionary choice of law rules to equitable wrongs, “greater emphasis might be placed on the law under which the relationship between the relevant parties was created, or which governed the relationship between them, than on the fact that the enrichment occurred in a particular place”.774

Panagopoulos comments in connection with benefits which a trustee acquires for himself that: 770 L Barnard, [1992] CLJ 474, 504. She goes on to state (at 507) that, “The proper law would be the legal system governing the relationship in terms of which general access was gained to the beneficiary’s assets, e.g. the proper law of the relevant contract, or the proper law of the trust; or, failing such a relationship, the law of the place where the access was obtained, unless that place was fortuitous and another legal system has a more significant connection with the parties, in which case the other system should govern.” 771 Ibid., 506–7. 772 R Stevens, Restitution and the Conflict of Laws, ch. 5, 188–90. Dicey and Morris also reject the existence of a distinct private international law category of equitable wrongs: see 1499–501. See also Arab Monetary Fund v. Hashim [1993] 1 Lloyd’s Rep 543; Arab Monetary Fund v. Hashim (No 9), The Times 11 October 1994; Attorney General (UK) v. Heinemann Publishers Australia Pty Ltd (1988) 165 CLR 30. 773 In principle, Rule 200 of Dicey and Morris would lead to the application of the law of the place where the enrichment arose. However, Dicey and Morris appear to accept that the law of the underlying relationship might more appropriately govern the claim: 1499. 774 Dicey and Morris, 1499–500.

Article 8(2)(g) 251 “. . . the fiduciary duty breached is one imposed by the trust. Therefore, the law of the relevant relationship, namely the law of the trust, is the most appropriate to determine the trustee’s liability, including his liability to account for any wrongful gains”.775

Moreover, Waters points out that: “The case law in common law jurisdictions has always taken the position that the trustor can design the terms and the liabilities he chooses, and therefore the fiduciary obligations have become matters of trustor intent.”776 It ought equally to follow that settlor autonomy should be protected by application to his fiduciary duties of the law chosen by him to govern the trust. The court in Paramasivam v. Flynn777 expressly excluded from their rule fiduciary duties arising from agreement, suggesting that they might better be governed by the law applicable to the specific relationship from which they arise. This largely accords with the views of McHugh JA and Kirby J in AG for the United Kingdom v. Heinemann Publishers Pty Ltd.778 Support for the view that the law governing the specific relationship should also govern the “general” fiduciary duties owed by the trustee may also be found in the jurisdictional context. The Court of Appeal ruled in Maimann v. Maimann779 that an exclusive jurisdiction clause for the courts of Switzerland contained in a power of attorney must be taken to apply not only to a claim brought by the beneficiary under the power against its holder780 for breach of the terms of the power, but also to a “general” claim against the holder of the power for breach of fiduciary duty (in having allegedly wrongfully transferred assets subject to the power to a new bank account), and to a claim that the holder of the power was unjustly enriched by improper use of the power. All these claims hinged upon the terms of the power of attorney; and the power of attorney was subject to a jurisdiction clause. The fact that claims were brought for breach of “general” duties by the holder was immaterial. The jurisdiction clause regulating the specific relationship between the holder of the power and the beneficiary under the power should also be applied to such “general” claims. On balance, whether the fiduciary duties of a trustee are treated as a separate category for private international law purposes than his trusts liability or not, the choice of law rule appears to be one and the same, namely the law applicable to the trust. The fiduciary relationship may not be trusts specific but, in so far as the 775 G Panagopoulos, Restitution in Private International Law, 88 (and see ibid., 149–51.). “[L]ikewise, where a trustee, for example, overpays a beneficiary, it may be considered that the issue of the beneficiary’s liability to repay is to be characterised as an issue concerning the relationship between the trustee and the beneficiary and thus governed by the law applicable to the trust” (ibid., 150). 776 D Waters, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118, 124. 777 (1998–99) 160 ALR 203. 778 (1987) 19 NSWLR 86. See also United States Surgical Corp v. Hospital Products International Pty Ltd [1982] 2 NSWLR 766. 779 Judgment of 27 June 2001; [2001] EWCA Civ 1132. 780 In the instant case, the holder had been appointed pursuant to a sub-power of attorney. However, the appointment was on the same terms as the original power, including the Swiss jurisdiction clause.

252 The Scope of the Applicable Law rights and obligations arise because of the existence of a valid trust, then it makes sense for the law governing the trust also to regulate that relationship. Two final points need to be made on the interpretation of the trust. First, Yeo points out that “logically, it should be an issue of interpretation whether the parties, in choosing the proper law, also intended it to govern fiduciary obligations”.781 If it appears, as a matter of the law governing construction of the trust, that the settlor did not intend the trustee’s fiduciary duties to be subjected to the law applicable to the trust itself, then it should follow that they should be governed by whichever law he did intend to govern the fiduciary duties.782 If the settlor makes clear that the fiduciary liability of the trustee is not to be subjected to the law which he chose to govern the trust, but he does not stipulate which law is to govern the fiduciary liability, the default law applicable to fiduciaries must apply. Since it has been argued above that the choice of law rule for fiduciary trustees should be the law applicable to the trust, the law applicable to the trustee’s fiduciary liability will, in this scenario, be the law with which the trust has its closest connection.783 Secondly, if the trust instrument states that it shall be governed by the law of Utopia, but that the administration of the trust shall be governed by the law of Ruritania, it will be necessary to determine whether the fiduciary’s liability (which may be regarded as a failure to act in the best interest of the beneficiary) should be classified as a matter of administration. Barnard argues that it should not, as a matter of authority and principle. The cases which she cites do not provide any significant support for the proposition.784 As a matter of policy, she argues that if the law applicable to administration is that of the place of administration, this law may be prone to manipulation and alteration by the trustees and that fiduciary liability should not be vulnerable to such factors.785 Even if this were true at common law,786 it is clearly no easier under the Hague Convention to change the law governing administration than to change the law applicable to the entire trust.787 A stronger argument is that fiduciary duties 781

T Yeo, “Choice of Law for Fiduciary Duties” (1999) 115 LQR 571, 574. I.e. by way of dépeçage the fiduciary duties would be subject to a separate choice of law rule. If the law chosen by the settlor can be allowed to govern the trustee’s fiduciary liability where the settlor so intends, it would seem odd not to allow the settlor also to choose a separate law to govern the trustee’s fiduciary liability. 783 It is likely that this will be determined in accordance with Art. 7, Hague Convention, rather than by application of the common law approach to determining the law of closest connection. Even if the trustee’s fiduciary liability is not governed by the Convention, and even if the settlor has made clear that he does not want the law chosen by him to govern the trust also to govern the trustee’s fiduciary liability on the facts, it has been argued that the same choice of law rules which apply under the Convention to trusts matters should be applied to the trustee’s fiduciary liability. 784 Harris Investment v. Smith [1934] 1 DLR 748 (a breach of trust case where the proper law of the trust and the law of the place of administration were one and the same); Robertson v. Howden (1892) 10 NZLR 609 (where the law of the place of administration was itself unclear: see P Webb, “Trusts, Conflicts and New Zealand” [1975] NZLJ 128, 132). 785 L Barnard, [1992] CLJ 474, 483–4. 786 Which is itself a moot point: see ibid., 483–4. 787 Art. 10. 782

Article 8(2)(g) 253 relate not to “the manner and enforcement” of the beneficiaries’ rights, but to “the identification and extent of those rights”.788 The beneficiary has the right to the loyalty of the trustee and to recover any unauthorised profits or to be compensated for loss. Accordingly, it is submitted that fiduciary duties should not be regarded as part of the domain of the law governing the administration of the trust. (iv) Liability and remedies Whilst Article 8(2)(g) makes clear that the liability of the trustee to the beneficiary is a matter for the applicable law of the trust, it does not state whether the remedy to be granted pursuant to that liability should be so governed.789 If a trustee wrongfully misappropriates trust funds and uses them to invest in a profit making activity, which law should determine whether compensatory and/or restitutionary measures of recovery are available to the claimant? If profits can be recovered, which law should determine whether the trustee holds those profits on constructive trust or has a personal liability to account for their value? It is trite law that “the nature of the remedy is a matter of procedure to be determined by the lex fori”.790 However, application of the law of the forum is not unrestricted, since the remedy must “harmonise with the right according to its nature and extent as fixed by the foreign law”.791 Sensible limits have to be placed on the law of the forum. Nothing could be more non-sensical than e.g. for an English court to say that a beneficiary has a cause of action on the basis of breach of trust against the trustee for misappropriation of trust funds according to the governing law, only then to say that, as a matter of English law, the remedy which the governing law would impose (be it compensatory and/or restitutionary, personal and/or proprietary) is not available on the facts. The imposition of a constructive trust was apparently treated as a matter for the governing law in Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd 792 and in El Ajou v. Dollar Land Holdings Ltd.793 In the former case, Goulding J stated that “right and remedy are indissolubly connected” and noted the artificiality and arbitrariness of asking “whether the court vindicates the suitor’s substantive right or gives the suitor a procedural remedy”.794 The modern trend towards substantive remedial classification795 reflects the distorted consequences of dividing right from remedy. The remedy is the means by which the 788

To use the language of Scott J in Chellaram v. Chellaram [1985] Ch 409, 432. See generally, O Elias, Judicial Remedies in the Conflict of Laws. 790 Dicey and Morris, 159. See Liverpool Marine Credit Co v. Hunter (1868) LR 3 Ch App 479; Boys v. Chaplin [1971] AC 356; Phrantzes v. Argenti [1960] 2 QB 19. 791 Phrantzes v. Argenti, ibid., 35. 792 [1981] Ch 105. 793 [1993] 3 All ER 717. 794 [1981] Ch 105, 124. See also McKain v. RW Miller & Co (SA) (1991) 174 CLR 1, 24–5, 48–9. 795 Compare Art. 10(1)(c), Rome Convention. 789

254 The Scope of the Applicable Law right finds its practical expression. Accordingly, it is submitted that the law applicable to the trust should also be used to determine whether compensatory and/or restitutionary recovery is available and whether the appropriate response is personal or proprietary.796 Where the appropriate remedy is a personal claim for damages, however, it is likely that an English court will apply the law of the forum to the quantification of those damages.797 (v) Trustees’ liability to third parties Article 8(2)(g) does not deal with the liability of the trustees to third parties. Von Overbeck makes clear that this is quite deliberate.798 If a trustee harms a third party’s interest e.g. by wrongfully misappropriating their property and mixing it with the trust fund, there is no reason why the third party’s personal or proprietary right of recovery should in any way be affected by a trust to which he was not party. (vi) Third parties’ liability to trustees: personal and proprietary claims Conversely, if the trustee mistakenly pays money to a third party, Article 11(3)(d) in principle permits the trust property to be recovered,799 but subject to the choice of law rules of the forum.800 If the third party alleges that he is a purchaser in good faith and took free of the beneficial interest under the trust, the question from his perspective is whether he obtained a good title to property and this should be governed by the lex situs. If the claim is to make the third party personally liable for receipt of the trust property, the Hague Convention provides no answer. The trustee’s liability for breach of trust (where he has wrongfully paid the trust money to a third party) will clearly be determined by the applicable law of the trust; but the personal claim against the third party for knowing receipt of trust property is not a matter of the law of trusts.801 Nor does it seem to fit within the Convention. Matthews states that: 796 Even if the constructive trust imposed is not regarded as falling within the Hague Convention (on which, see Art. 3, above), it is argued that the law which determines whether it should be imposed should be that applicable to the express trust, the terms of which have been breached. 797 D’Alemida Araujo Ltd v. Becker & Co. Ltd [1953] 2 QB 329, 336; Boys v. Chaplin [1971] AC 356; Stevens v. Head (1993) 176 CLR 433. This principle has been subjected to criticism: see e.g. NV Handel Maatschappij J Smits Import-Export v. English Exporters Ltd [1955] 2 Lloyd’s Rep 69, 72; and see the comments of Lord Upjohn in the Court of Appeal in Boys v. Chaplin [1968] 2 QB 1, 31. As well as creating a difficult characterisation question of what exactly is a matter concerning the quantification of damages, the awarding of an amount substantially different to that which the governing law would grant may substantially distort the right in question. However, it is thought that, on the present state of authorities, the role of the lex fori in this matter is relatively secure. See also Art. 10(1)(c), Rome Convention. 798 Von Overbeck Report, para. 87, p. 389. 799 If the law applicable to the trust so provides. 800 See the discussion of Art. 11(3)(d), below. 801 Or, at least, it is not a matter of the law regulating the original trust.

Article 8(2)(g) 255 “A person who never has trust property is certainly not a ‘trustee’, within the meaning of Article 2. A third party who temporarily receives trust assets from the trustee but then disposes of them is also not such a ‘trustee’, unless they can be said to have been ‘placed under his control ‘for the benefit of a beneficiary or for a specified purpose’ (which must be doubtful). So the rules in the Convention relating to duties and liabilities of trustees will probably not apply to our (personal liability) ‘constructive trustees’ ”.802

This author would concur. Even if the assets are “placed under the control” of the third party, the personal obligation for which it is sought to make the third party liable is not breach of trust, but knowing receipt of trust property. The claim may be classified as one in unjust enrichment,803 and it might be argued that it would not be appropriate for the law applicable to the trust to determine the matter.804 Panagopoulos argues that: “The recipient’s liability arises independently of . . . [the trust] relationship, so therefore the law of such a relationship should not govern the recipient’s liability. . . . One must be careful not to allow the existence of circumstantial relationships to confuse the choice of law process”.805

In this writer’s view, Panagopoulos’ argument as to the independence of the trust is important, but not conclusive. If a third party has actual or constructive notice that he is receiving trust assets to which he has no entitlement, then it might be said that, by receiving those assets he agrees, or may be taken to have agreed, to subject any ensuing personal liability which he may incur to the law governing that trust and that application of that law would not defeat his expectations. Such an analysis must be used with care, since it is dangerously close to an “implied contract” approach to restitutionary claims. However, it is not without analogous support elsewhere806 and, insofar as the law applicable to the trust accords with the parties’ expectations, has much to commend it. It is not obvious that the third party should be able to improve his position by arranging to receive the assets in a jurisdiction by whose law he incurs no liability.

802 P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7, 12. 803 G Panagopoulos, Restitution in Private International Law, 150. See also L Smith, “Unjust Enrichment, Property and the Structure of Trusts” (2000) 116 LQR 412; A Burrows, “Proprietary Restitution: Unmasking Unjust Enrichment” (2001) 117 LQR 412. 804 See Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd [1981] Ch 105; El Ajou v. Dollar Land Holdings plc, [1993] 3 All ER 717 (reversed on different grounds: [1994] 2 All ER 685); Trustor AB v. Smallbone, judgment of 9 May 2000 (CA); Kuwait Oil Tanker v. Al Bader, [2000] 2 All ER (Comm) 271 (and see G Virgo, “Interest, Constructive Trusts and the Conflict of Laws” (2000) 8 RLR 122); Grupo Torras v. Al-Sabah, [2001] Lloyd’s Rep Bank 36 (noted by J Garton, (2001) 15 Trust Law International 93). 805 This argument should not be pushed too far, however, since the claim is for knowing receipt of trust property. 806 In Case C–214/89 Powell Duffryn v. Petereit [1992] ECR–I 1745, the ECJ ruled that a contracting party would be bound by a choice of law clause in a contract where he knew or had the means of knowledge of the existence of the clause. Of course, the case did not concern third parties.

256 The Scope of the Applicable Law Nevertheless, the reality is that, in the case of an unscrupulous third party,807 agreeing to be subject to the terms of the trust is probably the last thing that they would have done. On balance, it is suggested that personal claims against third parties808 should be not be subjected to the law applicable to the trust. Such claims may arise by virtue of the existence of a trust, but they are not trust claims. Nor should the danger of manipulation of choice of law rules be overstated. Such calculated conflicts manipulation is hardly likely to be attempted frequently; and where it is, its worst excesses can always be prevented by a finding that application of the governing law infringes English public policy.

(H) The Variation or Termination of the Trust (i) Variation of trusts (1) Introduction: the problem stated The inclusion of variation of trusts within Article 8 does not seem inherently controversial. In principle, variation should be subjected to the “ordinary” choice of law rules applicable to a trust, as determined for most express trusts by the Hague Trusts Convention. Once a trust has validly come into existence, its governing law must authorise any changes to the terms of that trust, just as, arguably, it should authorise any change in the applicable law. However, the variation of trusts is a matter of particular complexity in the conflict of laws, especially in English courts. It is far from clear in practice to what extent the applicable law on variation can be applied in an English court. The matter is complicated by statutory provisions of English law which purport to confer “jurisdiction” on an English court to vary a settlement.809 It is not clear what happens where that “jurisdiction” is lacking. Furthermore, pre-Convention case-law suggests that the English statutory variation provisions810 may be applied to trusts governed by foreign law. Today, it would appear that such a construction must either be considered to have been overruled, or the statutory provisions must be treated as being of mandatory application.811 807 Of course, not all third party recipients are unscrupulous, although it might be thought undesirable to devise a choice of law rule which assists those who are. 808 Including claims for knowing assistance, which have been treated for Brussels Convention purposes as matters relating to tort (Dexter Ltd (in Administration) v. Harley, The Times, 2 April 2001; Casio Computer Co Ltd v. Kaiser, judgment of 11 April 2001 ([2001] EWCA Civ 661) and which merit their own choice of law rule. The same cases also assumed that knowing receipt claims could be classified for Brussels Convention purposes as matters relating to tort. See further T Yeo, “Constructive Trustees and the Brussels Convention” (2001) 117 LQR 560. 809 This is not a reference to in personam jurisdiction of the court, which must, of course, still be established. The “jurisdiction” here is rather, as will be seen below, the conferral of competence on an English court with in personam jurisdiction to vary a trust. 810 On the Variation of Trusts Act 1958, see Re Ker’s Settlement Trusts [1963] Ch 553; Re Paget’s Settlement [1965] 1 WLR 1046. On the Matrimonial Causes Act 1973, see Nunneley v. Nunneley and Marrian (1890) 15 PD 186. 811 See the discussion of Arts. 15 and 16 below.

Article 8(2)(h) 257 Our concern in this section is with trusts whose applicable law is not English.812 Two particular schemes of variation in English law will be examined: the Variation of Trusts Act 1958 and the Matrimonial Causes Act 1973. The former provides a general basis of competence in the High Court to vary trusts; the latter permits a court to vary a trust upon a divorce, nullity or judicial separation. The latter, concerning as it does rules of family law and trusts, is especially problematic. (2) Divorce settlements: the Matrimonial Causes Act 1973 (α) General It is clear from section 24(1)(c), Matrimonial Causes Act 1973813 that an English court has the power to vary a trust governed by English law when it has jurisdiction over divorce, nullity or legal separation proceedings involving the same parties.814 Section 17, Matrimonial and Family Proceedings Act 1984 confers a similar power on the court in respect of foreign decrees. In both cases, the decision to vary is at the court’s discretion. The question arises as to whether variation can be made when a trust is governed by a foreign law. Two possibilities need to be considered: (1) that the trust can be varied using English law’s provisions on variation. (2) that the trust can be varied using the applicable law’s provisions on variation. These will now be considered (β) Substance or procedure: a role for English law? The Hague Trusts Convention indicates that variation is a matter for the applicable law of the trust. However, it could be argued that statutory variation provisions of English law are procedural, rather than substantive, in that they relate to the process by which an English court can vary a trust. Matters of procedure, as opposed to substance, will always be governed by English law in an English court.815 It follows that a procedural classification would allow an English court to vary a trust pursuant to English statutes irrespective of the law applicable to the trust. In Chellaram v. Chellaram,816 the beneficiaries sought an order directing the removal and replacement of the trustees. It was argued by counsel for the trustees that the trust was governed by Indian law. However, Scott J distinguished 812 See also the consideration of the court’s power to vary the governing law under the Variation of Trusts Act 1958 from English law to a foreign law in the discussion of Art. 10, Hague Trusts Convention. 813 S. 15, Family Law Act 1996 will introduce certain changes in this area, including allowing ancillary orders to be made prior to a divorce or legal separation. It is not presently in force. See also schedule 2 to the 1996 Act. 814 It may also adjust the interests of the parties, or extinguish the interest of one party: s. 24(1)(d). See also Cammell v. Cammell [1965] P 467; Tallack v. Tallack [1927] P 211. 815 See Dicey and Morris, Rule 17, 157. 816 [1985] 1 Ch 409.

258 The Scope of the Applicable Law between substantive questions of “the identification and extent” of the beneficiary’s rights and procedural issues concerning the “machinery for enforcement” of those rights.817 He ruled that the appointment and removal of a trustee were matters concerning the court’s procedure and hence to be determined by the law of the forum.818 Could the same be said of variation of trusts? In this regard, the 1973 Act is peculiarly complex. It might be thought procedural in that it lays down the circumstances when a English court may vary a trust and in that sense concerns the process of the court. However, it is substantive in that it goes on (in section 25) to lay down the factors to which English law should have regard in deciding how to exercise that power. Section 1(1), Variation of Trusts Act 1958 similarly confers a power on the court to vary, but contains very little in the way of substantive criteria for the exercise of the power, save to state that the court may vary “if it thinks fit.” It could be said that much depends on the variation in issue. Ordinarily, variation will affect the interests of the beneficiaries and in that sense clearly concerns the “identification and extent” of their rights. Occasionally, variation may simply be used for the purpose of “enlarging the powers of the trustees of managing or administering any of the property subject to the trusts” and may look less obviously a substantive process.819 Nevertheless, the consequence of altering the powers of trustees is to effect a substantive change in the administration of the trust. Indeed, trustees’ powers are expressly subjected to the applicable law of the trust under Article 8(2)(d), Hague Convention. More generally, Dicey and Morris observe that a mechanistic approach to the distinction between procedure and substance: “is now discredited . . . The primary object of this Rule is to obviate the inconvenience of conducting the trial of a case containing foreign elements in a manner with which the court is unfamiliar. If, therefore, it is possible to apply a foreign law . . . without causing any such inconvenience, those rules should not necessarily, for the purpose of this Rule, be classified as procedural”.820

It is suggested that variation is a paradigm case for the application of the Dicey and Morris view. There is no obvious reason why an English court should be less well equipped to apply foreign legislation on variation than other provisions of the applicable law. Accordingly, it is submitted that the applicable law’s provisions should be regarded as substantive and that hence, prima facie, its law of variation should be applied in an English court, in accordance with Article 8(2)(h), Hague Trusts Convention.

817

[1985] 1 Ch, 432. Ibid., 433; save in the unlikely event that the trust deed conferred rights of appointment and removal on the beneficiaries, which would be subject to the proper law of the trust: ibid. 432–3. 819 S. 1(1), Variation of Trusts Act 1958. 820 Dicey and Morris, 158. 818

Article 8(2)(h) 259 (γ) The role of the applicable law of the trust Dicey and Morris state that the jurisdiction of an English court to vary a settlement under the Matrimonial Causes Act 1973 may be exercised in respect of “property situated abroad and . . . governed by a foreign law and the trustees of which reside abroad”.821 This requires closer consideration. Certainly, section 24 provides a basis of competence to vary a trust governed by foreign law. However, whether a court with competence should actually vary a foreign trust raises a choice of law question for the applicable law to decide. The Hague Trusts Convention lays down choice of law rules for determining the governing law of a trust. Once it is established that a foreign law governs the trust, then Article 8 has the effect that the law to govern all questions of substance relating to the trust is the applicable law.822 One such question is stated in Article 8(2)(h) to be “the variation or termination of the trust.” It must follow that it is to the applicable law of the trust that an English court should look first and foremost when deciding whether to vary the trust. Dicey and Morris confirm that “the variation of trusts is one of the matters governed by the law identified by Articles 6 or 7 as the law governing the trust”.823 This is further supported by the words of the Matrimonial Causes Act 1973, which do not state that the jurisdiction to grant financial relief is only available where the governing law of a trust is English law. (δ) The exercise of jurisdiction by an English judge to vary a trust governed by a foreign law Even if an English court is empowered by the applicable law to vary, there is a logically separate question of whether, in its discretion, it should exercise its jurisdiction to do so under section 24, Matrimonial Causes Act 1973. Dicey and Morris correctly note that the fact that the variation of trusts is subjected to the governing law of the trust by the Hague Trusts Convention “is a choice of law rule and does not directly affect the jurisdiction of the English courts”.824 They go on to note in the context of variation generally that the application of a foreign law to a trust may “make it more likely that an English court would disclaim the exercise of jurisdiction, treating the courts of the relevant country as a more appropriate forum”.825 A particularly relevant point will be whether the English variation order is likely to be recognised abroad. Other factors include the connection with England of the parties to the divorce and of the trustees and the connection of the trust property with England.

821 Ibid., 772. They cite the pre-1973 Act cases of Nunneley v. Nunneley (1890) 15 PD 186; Forsyth v. Forsyth [1891] P 363; and Goff v. Goff [1934] P 107 in support. 822 Subject to the application of mandatory rules and public policy (on which, see Arts. 15, 16 and 18, below). 823 Dicey and Morris, 1093. 824 Ibid., 1093. 825 Ibid., 1093. A similar view is expressed by J-G Castel, Canadian Conflict of Laws, 4th edn., 552.

260 The Scope of the Applicable Law However, Dicey and Morris then argue that reluctance to exercise the jurisdiction to vary should not apply to variation under section 24, Matrimonial Causes Act 1973 since “it is important that this power should be exercised in accordance with the English lex fori as part of the whole range of powers exercisable in those contexts”.826 It appears that this passage is primarily concerned with the situation where the variation is to occur by English law827 but the principle that financial relief should, where practical, be awarded in composite proceedings together with the divorce itself, is a sound general rule. It follows that an English court should feel no diffidence in varying a trust using the applicable law, provided that the conditions for application of the foreign law are satisfied on the facts. Indeed, routine refusal by an English court to exercise its jurisdiction in a situation where it is permitted to vary by the applicable law risks nullifying the effects of Article 8(2)(h), Hague Convention. That said, there must be limits to this approach. In particular, it is evident that for an English court to vary trusts governed by foreign law and with no objective connection to England would risk incurring the displeasure of foreign courts and risk rendering any variation order unenforceable overseas. So much is evident in the Royal Court of Jersey’s decision in Re the Rabaiotti 1989 Settlement and Others.828 In that case, the English High Court had required a beneficiary of discretionary trusts administered in Jersey (two of which were governed by the law of Jersey and two by the law of the British Virgin Islands) to produce trust accounts and letters of wishes of the settlor pursuant to matrimonial property litigation involving the beneficiary in England. The trustees sought the guidance of the Jersey court as to whether this would be in the interests of the other beneficiaries and as to whether they should intervene in the English proceedings. The Royal Court ruled that disclosure of the documents should be allowed on the facts, but that the trustees should not intervene in the English proceedings. One concern of the trustees was that, if the documents were provided, the English court might feel more inclined to “exercise the jurisdiction, which it apparently may have, to vary the terms of the various trusts”.829 They also worried that submission by them to the English proceedings would make it more likely that they would be bound in Jersey by any English variation order. On the latter point, the Deputy Bailiff remarked:830 “The court regards it as unlikely that an English court would so exceed the normal bounds of comity as to purport to vary a settlement governed by Jersey or BVI law, 826

Dicey and Morris, 1093. Presumably pursuant to Art. 15, Hague Convention. 828 [2000] ITELR 763. For comment on the case, see: “Powers and Duties of Trustees” (2000) 4 Jersey Law Review 317; D Benest, “The Trustee’s Dilemma: to Disclose or not to Disclose?” (2000) 6 Trusts and Trustees 6; D Benest, “Re Rabaiotti Revisited: are Letters of Wishes Disclosable?” (2001) 7 Trusts and Trustees 16; C Davies, “Jersey: Disclosure of Trusts Documents” (2000) 6 Trusts and Trustees 33; B Labesse, “Why Litigate in Jersey?” (2001) 87 In House Lawyer 74; M Taylor, [2001] European Legal Business, 57. 829 Ibid., 783, per Birt (Deputy Bailiff). 830 Ibid., 786. 827

Article 8(2)(h) 261 administered in Jersey by Jersey trustees and which had no connection with England save that some of the beneficiaries resided there.”

Where does this leave the law? It is submitted that the sole test with which the English court needs to concern itself is whether, (assuming that it has jurisdiction), it is a natural forum in which to grant the particular variation order sought. In deciding this, it must, as Rabaiotti rightly shows, attach considerable weight to the law governing the trust and to the strength of objective connection of the trust to England.831 However, against this, it must weigh three main factors: (i) that the philosophy of Article 8(2)(h), Hague Convention makes clear that the mere fact that a foreign law governs the trust is nothing like a conclusive reason not to exercise the jurisdiction; (ii) that England may well be the natural forum on the basis that it is more appropriate for divorce proceedings and a variation order to be heard in the same court; and (iii) an English court should be especially reluctant to refuse to vary a trust if it appears that no other court overseas would be competent, or willing, to vary the trust. (3) Variation of Trusts Act 1958 Significantly, section 1, Variation of Trusts Act 1958 is headed “Jurisdiction of courts to vary trusts.” Immediately, this conveys an idea of competence to vary, rather than a substantive act of adjusting the rights and duties of the parties. Subsection (1) then states that: “Where property, real or personal, is held on trusts arising, whether before of after the passing of this Act, under any will, settlement or other disposition, the court may if it thinks fit by order approve . . . any arrangement . . . varying or revoking all or any of the trusts, or enlarging the powers of the trustees of managing or administering the property subject to the trusts:. . . .”

Subsection (3) provides that: “. . . the jurisdiction832 conferred by subsection (1) of this section shall be exercised by the High Court.”

These provisions contain no indication that an English court may only vary a trust governed by English law. “What section 1 in terms does is to confer on the court power, if it thinks fit, to approve a variation of the settlement: it confers jurisdiction on the court”.833

However, guidance on the scope of the section was provided in Re Paget’s Settlement.834 A trust was created by a settlor domiciled in England from funds provided to him by her father, an American. The trust was drawn up in New 831 Certainly, had the trusts in Rabaiotti been administered in England, it would not have “exceed[ed] the normal bounds of comity” to have varied the trusts pursuant to English matrimonial proceedings, even though the trusts were not governed by English law. 832 Emphasis added. 833 Per Ungoed-Thomas J in Re Ker’s Settlement Trusts [1963] Ch 553, 556. 834 [1965] 1 WLR 1046.

262 The Scope of the Applicable Law York, but executed in England. All the trustees were American and the trust assets were almost all situated in the United States. The trust was administered in New York, though there was at the time no equivalent variation legislation in that state. Upon an application by the trustees, the New York court ruled that the trust was governed by English law. However, Cross J proceeded in the Chancery Division on the assumption that the trust was governed by New York law, without conclusively deciding the point. It was at this stage that the scope of the Variation of Trusts Act 1958 became relevant. Cross J considered section 2 of the Act, which excludes the application of the Act to Scotland and Northern Ireland and suggested that: “. . . one might be inclined to think that what Parliament meant was that English settlements should be dealt with by the English courts, Scottish settlements by the Scottish courts and the Northern Irish settlements by the courts of Northern Ireland. The result of that would be that settlements which were neither English, Scottish nor Northern Irish could not be dealt with by any court in the United Kingdom”.835

However, he rejected such a view on three counts. First, it was inconsistent with the construction of the Act in re Ker’s Settlement Trusts,836 where UngoedThomas J ruled that a Northern Irish settlement could be varied and that the English court’s jurisdiction was unlimited by the law applicable to the trust or the foreign connections of the trust. Ungoed-Thomas J had, in turn, relied upon Forsyth v. Forsyth,837 a divorce settlement case where the court ruled that it could vary a Scottish settlement upon divorce pursuant to the Matrimonial Causes Act 1859 and had found that: “that decision suggests that, in the absence of indication to the contrary there is no reason for limiting to English settlements a power conferred on an English court to vary the trusts of a settlement”.838

Secondly, section 2, Variation of Trusts Act 1958 was concerned with the competence of the courts of Scotland and Northern Ireland. It did not limit the right of an English court to vary a foreign trust, which “jurisdiction is unlimited”.839 Thirdly, it would be undesirable if a trust governed by a foreign law could not be varied in England, if the trustees, beneficiaries and trust property were now located within the jurisdiction.840 That said, Cross J suggested that, whilst a court would be empowered to vary a foreign settlement, it might choose in its discretion not to do so. “Obviously, 835

[1965] 1 WLR, 1049. [1963] Ch 553. See F Mann, “The Variation of Trusts Act, 1958, and the Conflict of Laws”, (1964) 80 LQR 29. 837 [1891] P 363; see also Nunneley v. Nunneley (1890) 15 PD 186; Goff v. Goff [1891] P 363. 838 [1963] Ch 553, 556. As Cross J pointed out in Re Paget’s Settlement, the decision in Forsyth was not necessarily conclusive, since the Matrimonial Causes Act 1859 contained no equivalent to s. 2, Variation of Trusts Act 1958 expressly dealing with the jurisdiction of the courts of Scotland and Northern Ireland: [1965] 1 WLR 1046, 1049. 839 [1965] 1 WLR 1046, 1050. 840 Ibid., 1050. 836

Article 8(2)(h) 263 however, where there are substantial foreign elements in the case, the court must consider carefully whether it is proper for it to exercise the jurisdiction”.841 In such a scenario, the courts of the state with which the trust was most intimately connected might be the appropriate forum to vary. In this respect, it is perhaps instructive for the English court to ascertain whether a foreign court would be empowered to vary a trust. In Re Paget’s Settlement, a New York court would have been willing to do so, but only on the basis that it considered English law to be applicable and could accordingly apply the Variation of Trusts Act 1958 in New York. On the facts, the trustees, trust property and (arguably) the applicable law all pointed to New York. Nevertheless, the English domicile of the settlor and most of the beneficiaries persuaded Cross J that the jurisdiction to vary should be exercised in England. Nowhere is there a suggestion in Re Paget’s Settlement that the power to vary should be exercised in accordance with the applicable law of the trust. Instead, an English statute was applied to a foreign settlement. Today, such an approach would appear quite contrary to Article 8(2)(h), Hague Convention. Cheshire and North explain the apparent contradiction on the basis that “The 1958 Act . . . is concerned with the jurisdiction of the courts, whereas Article 8 of the Trusts Convention is concerned with choice of law matters.”842 However, even if one were to accept that the Variation of Trusts Act 1958 did concern jurisdiction, it is far from clear that it may be applied in a manner which renders Article 8(2)(h), Hague Convention meaningless. In Pearce v. Ove Arup Partnership Ltd,843 Lloyd J stated, when referring to the application of the Brussels Convention, that choice of law rules applicable in a contracting state must be adapted where jurisdiction is taken under the Convention so as not routinely to deprive a claim of substance and cause it to be struck out as bound to fail. Conversely, jurisdictional rules should not be adopted which deny a court jurisdiction simply because a trust is governed by a foreign law.844 It is suggested that the argument that the Variation of Trusts Act 1958 concerns solely jurisdiction does not withstand scrutiny. Properly speaking, the Variation of Trusts Act 1958 is not concerned with jurisdiction—it does not determine whether the court may entertain claims against a particular defendant, as do e.g. the rules of the Brussels Convention or the Civil Procedure Rules on service of a claim form out of the jurisdiction. It simply deals with the question whether a court with in personam jurisdiction over the defendant may vary a trust. That is a question of whether it may alter the rights of the parties; it is substantive and precisely the sort of issue which ought to be: (i) considered even in respect of trusts unconnected with England; and (ii) determined exclusively

841 842 843 844

Ibid., 1050. Cheshire and North, 1042. [1997] Ch. 293; the Court of Appeal implicitly accepted this view: [2000] Ch 403. See also J Harris, “Forum Shopping in International Libel”, (2000) 116 LQR 562.

264 The Scope of the Applicable Law by the applicable law of the trust, in accordance with the Hague Convention requirements.845 (4) Varying foreign trusts under English statutes: mandatory rules and public policy of the forum (α) Article 15, Hague Trusts Convention: domestic mandatory rules and variation under the Matrimonial Causes Act 1973 It might be argued that, notwithstanding the above arguments, an English court exercising its competence under the Matrimonial Causes Act 1973 should also decide whether to vary a foreign trust entirely by English law, without any reference to the applicable law. A provision which might assist in this respect is Article 15, Hague Convention.846 This provides that: “the Convention does not prevent the application of provisions of the law designated by the conflicts rules of the forum, in so far as these provisions cannot be derogated from by voluntary act.”

It applies to areas of law related to the trust.847 It preserves the role of rules of law which parties to a trust may not oust by prior agreement i.e. domestic mandatory rules. Article 15 is not concerned with the domestic mandatory rules of the forum as such, but with the domestic mandatory rules of the country whose law would be applied in an English court to a particular issue. Nevertheless, the question arises whether the provisions of section 24, Matrimonial Causes Act 1973 are of the type whose application may be permitted by Article 15 and allowed to override the applicable law, so that the variation issue is entirely governed by English law. A number of points may be made: (a) The effects of divorce on a trust are not specified as a relevant matter in Article 15. However, the von Overbeck stresses that those matters specified in Article 15 are merely “by way of example. Matters which are not listed may therefore override the trust’s rules”.848 It would be surprising if variation on divorce were excluded, since “the personal and proprietary effects of marriage” are mentioned in Article 15(1)(b) and “the transfer of title to property and security interests in the property” in Article 15(1)(d). That the effects of divorce are within Article 15 is also the view of Cheshire and North.849 Although the latter 845 Lewin on Trusts, 17th edn., 292 points out that the foreign applicable law might, however, only confer the power to vary on the courts of its own state. A similar point is made about variation in Scots courts by A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 639 (citing Montgomery v. Zarifi 1918 SC 128, 136). However, provided that an English court has in personam jurisdiction, arguably it should not be deterred from varying such a settlement, provided that the substantive conditions of the applicable law for varying the trust are met. 846 On which, see below. 847 It should be noted that Art. 15 will not obviously assist a variation claim under the Variation of Trusts Act 1958, since it deals with mandatory rules of the law applicable to a related area of law, not with mandatory trusts rules (which may be applicable, if at all, under Art. 16). 848 Von Overbeck Report, para. 139, p. 401. 849 Cheshire and North, 1043.

Article 8(2)(h) 265 suggest that Article 15(1)(b) could be used in the instant case,850 there is no need to bring this matter within a particular sub-rule; it suffices that the rule is of a domestic mandatory nature. (b) The question then becomes whether an English court is under a duty to apply the provisions of section 24, Matrimonial Causes Act 1973 and to override the applicable law. It is suggested that the answer must be in the negative. In this respect, it is necessary to consider the views of Hayton. He writes in relation to English mandatory provisions that “the answer is that the provisions must be applied”.851 He cites the von Overbeck Report,852 which gives the example of forced heirship rules of the forum, which might conflict with the rules of the governing law of the trust and states that “By virtue of Article 15, the law applicable to the trust must yield to the provisions for forced heirship.” With due respect, however, it is suggested that as both a matter of statutory construction and conflict of laws principle, this interpretation should not be followed, for the following reasons: (i)

Hayton states in relation to Article 16 (which deals with English rules designed to apply irrespective of the governing law) that the application of such rules is not at the court’s discretion, since it is in the very nature of those rules that they must apply even to a trust not governed by English law.853 This must be so in relation to rules that fall within Article 16. However, Article 15 does not deal with such powerful rules; rather, it allows an English court to apply rules of English law854 that bind the parties when the law applicable to the matter is English law. Domestic mandatory rules of English law do not necessarily need to be applied without discretion to all trusts governed by foreign law. Indeed, were an English court always to apply section 24, Matrimonial Causes Act 1973 to all foreign trusts varied upon divorce, the dividing line between domestic and international mandatory rules would become blurred for this type of trust. It should follow that domestic mandatory rules of the forum in divorce matters will only be imposed onto a trust governed by foreign law where the court considers the case to be a sufficiently strong one to merit so doing.855 (ii) Given that the English courts only varied trusts governed by foreign law prior to the entry into force of the Hague Trusts Convention as a matter of 850

As does G Miller, International Aspects of Succession (Ashgate, Aldershot, 2000), ch. 9, 251. D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 14. Von Overbeck Report, para. 54, p. 381; D Hayton ibid., 14. (Hayton’s emphasis). 853 D Hayton, ibid., 14–15. 854 Or the domestic mandatory rules of whichever law is designated by English choice of law rules as applicable to an area of law related to trusts. 855 Part of the difficulty is that Art. 15 applies to areas of law related to the trust. S. 24, Matrimonial Causes Act 1973 is at once both a role of divorce and a rule of trusts. To regard it wholly as a rule of divorce (and of mandatory application under Art. 15 in an English court) is routinely to apply it to trusts governed by foreign law and very much to undermine Art. 8(2)(h), Hague Convention. 851 852

266 The Scope of the Applicable Law discretion, it would be very surprising if the enactment of a Convention designed to achieve greater international harmony and co-operation on transnational trust issues should have introduced an obligation to apply English law in certain circumstances. The result would be the reverse of the Convention’s aims, in that it would compel the English court to apply provisions of English law to a trust governed by foreign law, despite the provisions of Article 8(2)(h) and even though it was never previously required to do so. (iii) The wording of Article 15 is in permissive terms only. It speaks of the Convention not preventing the application of English rules; it does not require their application. It is true that section 1(3), Recognition of Trusts Act 1987 states that “in accordance with Articles 15 and 16 such provisions of the law as are there mentioned shall, to the extent there specified, apply to the exclusion of other provisions of the Convention.” However, it is suggested that this does not compel the application of English law. Rather, it clarifies the effects of an English judge, in his/her discretion, choosing to invoke a domestic mandatory rule under Article 15. The effect of that exercise of discretion shall be to apply English law on that issue to the exclusion of the governing law. Indeed, the very existence of section 1(3) reminds us that the application of Articles 15 and 16 is an exception to the general application of the governing law of the trust to all substantive issues, including (on the basis of Article 8(2)(h)) variation. (iv) When addressing the issue of variation more specifically, Hayton argues that countries such as England: “in the light of Article 8(2)(h) . . . should decline jurisdiction if the forum’s law is applicable neither to validity nor administration of the trust except where public policy considerations reflected in Article 18 and 15(b) may justify the Divorce Court extending its jurisdiction so far as practical to a foreign trust”.856

The fact that an English court might stay variation proceedings, on the basis that the circumstances do not justify application of Article 15, tends to suggest that application of section 24 of the 1973 Act is discretionary, and that an English judge should feel diffidence towards exercising it in relation to a foreign trust. Indeed, the italicised part also tends rather to undermine the view that Hayton had previously expressed that English mandatory rules of variation must be super-imposed onto a foreign trust in all cases, since if an English court should always super-impose English variation rules falling within Article 15, then according to this latter passage, it would be justified in exercising its jurisdiction to vary so far as is practical. After all, if English law were considered to have such a strong interest in its law being applied, even to foreign trusts, it would presumably wish to exercise its jurisdiction and to apply that law to such trusts. Yet Hayton clearly contemplates in 856

D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 25 (emphasis added).

Article 8(2)(h) 267 the above passage that an English court should be diffident in varying by English law a trust governed by foreign law. Accordingly, it seems that the use of English mandatory rules on variation of a trust governed by foreign law should also be viewed with circumspection. For these reasons, it is suggested that a judge should not automatically conclude that variation under the Matrimonial Causes Act is a solely a matter for English law. The basic rule that variation is a matter for the applicable law should prevail. (β) The exercise of discretion by an English judge to apply English law under Article 15 Once it is established that a judge has a mere discretion whether to apply English substantive rules on variation to a foreign trust under Article 15, the question is whether that discretion should be exercised. It is clear that merely because Article 15 is a loosely worded provision, a judge should not be too creative in finding ways to vary a trust by English law. Von Overbeck observes that: “Not without reason, it was said that a hostile judge might always find in Article 15 a means of frustrating the trust”.857 “Frustrating” is a pertinent word, since it reemphasises that application of Article 15 is an impediment to the application of the general choice of law rule of the Hague Trusts Convention. In this respect, it should again be stressed that variation is primarily a matter for the applicable law. Even where a trust governed by foreign law is objectively connected to England, it is not obvious that the English variation provisions, rather than the applicable law’s variation provisions, should be applied by an English court.858 (γ) Article 16, Hague Trusts Convention: international mandatory rules of the forum Article 16859 preserves the application of the mandatory trusts rules of the forum, where such rules must be applied irrespective of the law applicable to the trust. Are there any such provisions of English law concerning variation of trusts? In re Kehr,860 Danckwerts J ruled that the powers of maintenance and advancement contained in sections 31 and 32, Trustee Act 1925 could be exercised by trustees appointed in England pursuant to the administration of an estate, notwithstanding that the trust was governed by foreign law. He could “see no reason why they should not have the powers of other trustees constituted according to English law”.861 Might it be said that the same principle applies to variation pursuant to English divorce proceedings? It is suggested that it should not. First, re Kehr would probably today be decided differently, since 857 858

Von Overbeck, para. 139, p. 401. An English court should be wary to intervene as a general principle: Tallack v. Tallack [1927]

P 211. 859 860 861

On which, see below. [1952] Ch 26. Ibid., 30.

268 The Scope of the Applicable Law the appointment of a trustee and his powers are both subjected to the applicable law of the trust under the Hague Convention.862 Secondly, Danckwerts J stated that “it may be doubtful whether trustees constituted by the law of a foreign country would have the powers conferred upon trustees regulated by English law. . .”863 In other words, he did not consider the English statutory provisions to be applicable irrespective of the governing law under which the trustee was appointed. More generally, it is submitted that Article 16 is inapplicable to variation. It has never been the case in English law that English variation rules have been applied without regard to the governing law of the trust or the degree of connection of the dispute with England. It seems clear from the pre-Convention cases that English statutory provisions on variation were never applied as a matter of due course; rather, the court exercised a discretion as to whether to apply them. In relation to the Variation of Trusts Act 1958, Cross J in Re Paget’s Settlement864 indicated that the discretion should not be exercised “simply because some, or even all, the trustees and beneficiaries were in this country.” Cheshire and North go as far as to say that the 1958 Act should not be applied at all when the governing law does not allow variation.865 (δ) Article 18: English public policy Article 18866 allows for the disapplication of provisions of the Hague Trusts Convention where their application would be manifestly contrary to English public policy. This provision ought not routinely to be applied for the following reasons: (a) it is inherently less attractive for a court, rather than super-imposing rules of English law onto a foreign trust (as Articles 15867 and 16 do) to have to state that foreign law offends public policy, since this is in the nature of a negative judgment as to the merits of foreign law; (b) the term public policy is inherently vague and capable of manipulation and hence not a desirable tool to be used unless absolutely necessary; (c) this is emphasised by the fact that use of the applicable law must be “manifestly” contrary to public policy. This provision must be restrictively applied, otherwise the international harmony for which the Hague Convention strives will be destroyed; (d) the concern of section 24 of the 1973 Act (and section 1(1), Variation of Trusts Act 1958) is not with an objection to the foreign law applicable to a trust per se, but with the need to apply English rules of variation to that 862 Arts. 8(2)(a) and (d) respectively. Technically, it could be said that re Kehr was concerned with the powers of the administrator of an estate, not a trustee, and remains unaffected by the Convention. 863 [1952] Ch 26, 29–30. 864 [1965] 1 WLR 1046, 1050. 865 Cheshire and North, 1043. 866 On which, see below. 867 Where English law is designated as applicable by the conflicts rules of the forum.

Article 8(2)(h) 269 trust. In other words, a court which varies under section 24 does not take issue with the law which governs the trust and seek to have it disapplied; rather, it super-imposes onto that law rules of variation of English law.868 Hence section 24 is by way of an addition to, not a subtraction from, the foreign law. Additions are the province of Articles 15 and 16. Article 18 deals with the disapplication of foreign law and is thus unsuitable to the possible super-imposition of English variation rules under section 24 of the 1973 Act (or section 1(1), Variation of Trusts Act 1958). (5) Application of English law by a foreign court If a foreign court is granting a divorce order and seeks to apply the English rules on variation to a trust governed by English law, there may be a problem with applying section 24, Matrimonial Causes Act 1973. That section applies where the divorce action takes place in England. It could be said that an overseas judge, when called upon to apply English law, should ask the question “what would an English judge do on the substance of this case, if the entire proceedings (including the divorce proceedings) were brought there?” Phrased in such a manner, the answer would be that he would look to section 24 of the 1973 Act. In other words, section 24 of the Act would still be applicable by an overseas judge. However, it is suggested that the above argument should be rejected on the basis that an overseas judge is applying English domestic law, but is not attempting to ask what an English judge would actually do on the facts if the case had instead been brought there (the exclusion of renvoi in Article 17, Hague Trusts Convention suggests that the judge should not attempt to reproduce the law on the substance which a judge of a foreign state would apply). It follows that section 24 of the 1973 Act is inapplicable in a foreign court. One needs instead to look at the English legislation on variation of a trust pursuant to a divorce in a foreign state (i.e. to Part III of the English Matrimonial and Family Proceedings Act 1984). Conversely, an English court seeking to vary a trust governed by foreign law pursuant to a divorce obtained in England must look to see if the foreign law has legislation which permits variation pursuant to a divorce obtained in a foreign country. For example, if it is sought to vary a Manx trust, it should apply section 17, [Manx] Matrimonial Proceedings Act 1986, which allows financial relief to be granted following an English divorce. (6) Conclusion The attitude of English courts to the variation of foreign trusts post-Hague Convention is not easy to predict.869 However, the position would appear to be in principle that: 868

As does s. 1(1), Variation of Trusts Act 1958. For this reason, Lewin on Trusts, 17th edn., 292 advises that “application to a court to vary trusts should preferably be made in the jurisdiction of their governing law.” 869

270 The Scope of the Applicable Law (a) variation is first and foremost a matter for the applicable law and that an English judge may (and arguably should) normally vary if the applicable law would so permit; (b) but an English court might feel duty bound to apply solely section 24, Matrimonial Causes Act 1973, even in a case governed by foreign law. Accordingly, it is necessary to look to Article 15 (or 16), to see if variation may occur by English law; (c) this would be by way of an exception to the application of the governing law. There is no reason for an English court routinely to super-impose the English rules on variation onto those of the governing law. Indeed, it has been argued above that neither section 24, Matrimonial Causes Act 1973 nor section 1(1), Variation of Trusts Act 1958 should normally be applied to a trust governed by a foreign law; (d) however, even if the foreign law permits variation, the English court does not have to exercise its jurisdiction to do so. The discretion to invoke the jurisdiction might not be exercised unless the connection with England is strong; and should probably not be exercised at all if the order will be unenforceable in the foreign court. That said, a routine refusal to vary trusts governed by a foreign law which permits variation would stifle the Hague Trusts Convention provisions and should not be encouraged; (e) the inability of a foreign court to vary the trust may be a material factor in the English judge’s decision whether to vary. (ii) Termination of trusts As to the termination of a trust, Article 8(2)(h) will cover the rights of both trustee and beneficiary870 to terminate. The vesting of the property in the beneficiaries upon termination may be considered an alienation of legal title which arguably ought to command the “general” lex situs rule for property transfers. However, the assets are conveyed to the trustee to hold on the terms of a trust and subject to the applicable law of the trust; those terms include the right to terminate. The lex situs must approve the alienation of the property by the settlor to the trustee to “launch” the trust, but it is not apparent that it should apply to the subsequent vesting of that property in the beneficiary. Where the right to terminate is exercised, the vesting of the property in the beneficiary ought to be determined by the applicable law of the trust. (iii) Variation, termination and administration A difficult question may arise if the trust is “split” so that its validity is determined by Utopian law, but its administration by Ruritanian law.871 How should 870 In England, the rule in Saunders v. Vautier (1841) 4 Beav 115 may allow the beneficiaries absolutely and present entitled to do this. 871 See further the discussion of Art. 9, below.

Article 8(2)(i) 271 the variation and termination of trusts be classified? On the one hand, the power to vary a trust looks as much like an administrative power of a trustee as e.g. the power of investment, maintenance or advancement. Indeed, the variation may relate solely to altering the administrative powers of the trustees. On the other hand, variation of a settlement will normally relate not just to the machinery for enforcing the beneficiaries’ rights, but to the rights themselves. It is a form of disposition of an equitable interest.872 In principle, it is suggested that the power to vary or terminate a trust should be regarded as a matter for the law governing administration; whereas the effect on the rights and interests of the beneficiaries should be governed by the law applicable to the validity of the trust.873

(I) Distribution of the Trust Assets This provision will cover the decision of a trustee of a discretionary trust as to whom should benefit from the trust and in what proportions. Presumably, it also covers the interpretation of the trust deed, in order to determine who are the permissible beneficiaries under the trust. Hayton points out that whilst distribution of the assets is a matter for the applicable law of the trust, reference may need to be made to another law. “Whilst the applicable law may apply to close a class of beneficiaries to include B’s legitimate and adopted children and their spouses living at the date of distribution of the trust assets . . ., it will be the personal law of B and the beneficiaries that will determine questions of legitimacy, adoption and marriage”.874

Presumably, if not already covered in Article 8(2)(d), the word “distribution” might be taken to include the use of trust assets for the purposes of e.g. maintenance and advancement.

(J) The Duty of Trustees to Account for their Administration The meaning to be attached to the word “account” is not apparent. Typically in English law, trustees “account” for their administration by being required to act unanimously,875 by exercising their discretion properly,876 by keeping accounts877 and by allowing inspection of the trust documents.878

872

IRC v. Holmden [1968] AC 685. Save, of course, if it appears that the parties intended something different by the word “administration.” 874 D Hayton, (1987) 36 ICLQ 260, 274 875 Tempest v. Lord Camoys (1882) 21 Ch D 571. 876 Ibid.; Re Manisty’s Settlement [1974] Ch 17. 877 Pearse v. Green (1819) 1 Jac & W 135. 878 O’Rourke v. Darbishire [1920] AC 581. 873

272 The Scope of the Applicable Law However, the word “account” might also cover something quite different and more substantive, namely the equitable remedy of account which might arise e.g. where a trustee makes an unauthorised profit for himself.879 In English domestic law, “whether the payment of the sum owing should be described as ‘damages’ or ‘compensation’ or ‘restitution’ is a matter of debate, and for most purposes the description of the obligation is a matter of no significance”.880

However, insofar as the remedy is restitutionary, then it is arguable that, for private international law purposes, it should not be subjected to trusts choice of law rules at all. Indeed, the liability to account for profits may be regarded as appertaining to a fiduciary duty and accordingly not trust specific.881 Nevertheless, where that liability to account arises from a breach of trust, the law that governs that trust ought also to determine the remedial consequences thereof; and where it arises from a breach of fiduciary duty, the duty is created by the trust relationship, so that the law applicable to the trust should still apply. In short, it is submitted that the liability of a trustee to account for profits made in breach of his obligations to the beneficiary should always be subject to the law applicable to the trust.882

3. FORMAL VALIDITY OF THE TRUST 883

(A) The Need for a Choice of Law Rule to Determine the Formal Validity of a Trust Von Overbeck comments that “Article 8 deals with substantive issues without expressly excluding issues of form”.884 The formal validity of the trust was at one stage to be included in Article 8 and expressly subjected to the applicable law. However, such a provision was subsequently removed.885 “It appeared in the course of discussion of the current Article 4 that problems of form concerned rather the preliminary instrument by which the goods were transferred to the trust, which instrument does not fall under the Convention’s coverage, than the trust itself”.886 879 In England, the view that the making of an unauthorised profit by a fiduciary rendered him personally liable to account only was at one time favoured: Lister & Co v. Stubbs (1890) 45 Ch D 1; A-G’s Reference (No 1 of 1985) [1986] 2 All ER 219. It now appears that such profits are to be held on constructive trust: A-G for Hong Kong v. Reid [1994] 1 All ER 1. 880 R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 81. 881 But see the discussion of “fiduciary duties” in relation to Art. 8(2)(g) above, where it is concluded that the law applicable to the trust should determine the trustee’s fiduciary duties. 882 Provided, of course, that the duty breached arises in connection with the trust and the fiduciary duties which it creates. 883 See also the discussion of Art. 3, above. 884 Von Overbeck Report, para. 82, p. 388. 885 Ibid., para. 82, p. 388; Report of the Special Commission, Nos. 69–71. 886 Ibid., para. 83, p. 388.

Formal Validity of the Trust 273 The Convention does impose the formal requirement that any voluntarily created trust must be evidenced in writing under Article 3887 and it is possible that this could be treated as the sole formal requirement, to which no additional national formalities may be added.888 It would be very surprising if this were the case, however, since the wording of Article 3 appears to provide minimal safeguards to ensure the genuine consent of the settlor.889 It is true that most restrictions on formal validity are likely to relate to the “rocket-launching” process of determining the formal validity of e.g. a will pursuant to which a trust is created. However, this will not always be the case. A fundamental distinction must be drawn between formality rules which affect the disposition of property at all (“rocket-launching” rules, falling outside the scope of the Convention and governed by the formality rules for “ordinary” property transfers) and formality rules which affect the disposition on trust (“rocket” rules, for which a Convention choice of law rule is needed). Let us assume for present purposes that the law applicable to the trust is to govern those formality rules that specifically relate to the trust. An example will illustrate the complexities which “trust specific” formality rules can create. In England, section 53(1)(b), Law of Property Act 1925, states that: “a declaration of trust respecting any land or any interest therein must be manifested and proved by some writing signed by some person who is able to declare such trust or by his will”.890

For testamentary trusts, it is sufficient that the “rocket-launcher”, that is, the will, is formally valid;891 but for inter vivos trusts, the formality requirement clearly relates to the trust itself and falls within the Convention’s ambit. Imagine that S orally declares that he wishes to transfer land situated in Utopia to T to hold on an inter vivos trust for B. T orally agrees to act as trustee. The trust is stated to be governed by English law. The property is vested in T, with no mention of the trust. Does the trust fail for want of formalities? In the United Kingdom, the Recognition of Trusts Act 1987 would apply, due to the extension of its application in section 1(2) to “any other trusts arising under the law of any part of the United Kingdom.” Strictly speaking, the alienation of legal title to T is a “rocket-launching” matter and would be valid provided 887 But, of course, this cannot apply in the United Kingdom to the broader categories of trusts to which the Convention’s rules have been extended by s. 1(2), Recognition of Trusts Act 1987. 888 Cf. the position concerning jurisdiction clauses under Art. 17, Brussels Convention, where the Court of Justice has ruled that the formalities provided in that Article are exclusive and the clause may not be impugned by additional formal requirements of national law: Case 150/80 Elefanten Schuh v. Jacqmain [1981] ECR 1671. 889 See further the discussion of Art. 3, above. 890 A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., 50 states this formality rule will apply even to land situated outside the jurisdiction, on the basis that it is a rule of evidence and treated as procedural (citing in support Rochefoucald v. Boustead [1897] 1 Ch 196, 207). However, this author would disagree. It would be extraordinary in a modern Convention to treat formality rules as anything but substantive (compare Art. 9, Rome Convention). 891 And the formality requirements are exceedingly liberal: see s. 1, Wills Act 1963 (considered in section 14 of Part One of this book).

274 The Scope of the Applicable Law it satisfies the formal requirements of the law of the place where the property is alienated,892 Utopia (which we will assume that it does). But it should also be necessary to comply with any formality requirements relating specifically to the creation of the trust structure. If the law applicable to the trust is applied to this question, the trust would fail, because English law will say that it does not comply with section 53(1)(b), Law of Property Act 1925,893 even if Utopian law would impose no formality requirements for the declaration of a trust of land. Conversely, an orally declared trust of land located in England, but subject to a trust governed by Utopian law, would be valid. English law, the lex situs, applies only to the question whether property may be alienated at all and not to the more specific question whether the creation of the trust structure is formally valid. Section 53(1)(b), Law of Property Act 1925 applies only to inter vivos transfers on trust and is accordingly inapplicable. The result would be that a trust of land, not evidenced in writing, would be effective in relation to land situated in England.894

(B) The Choice of Law Rule So, a choice of law rule to determine formal validity of the trust itself is required. Thus far, we have assumed that this would be the law applicable to the trust itself.895 It is suggested that this should indeed by the law to determine the formal validity of the creation of the trust structure. Other potentially applicable laws all have problems attached to them. The lex situs would not seem to accord with the general approach of the Convention, which is largely to shun that law in respect of “rocket” questions. It would also make little sense where property is located in non-trust states, which will obviously lack formality rules specific to the trust; and, of course, the disposition of legal title to the trustee must have been sanctioned already by the formality rules applicable to “ordinary” transfers of property.896 The law of closest connection suffers the familiar problem of inducing considerable uncertainty and “scaring” trust states by its lack of pragmatism. The personal law of the settlor, or of the trustee or beneficiary, do not seem appropriate for two reasons. First, formality rules are not intrinsically bound up with questions of status. Secondly, the use of personal law is not advocated by this author even in relation to the much more “personal” question of the capacity of a settlor to 892

Cheshire and North, 933. See further Part One of this book. At least if that sub-section is construed as intended to apply even to property located overseas. 894 Subject, of course, to English law determining that s. 53(1)(b), Law of Property Act 1925 is an international mandatory rule, applicable under Art. 16, Hague Convention. Although the subjectmatter is land, it is not obvious that a rule of formality needs to be treated as an international mandatory rule of the forum; it is not a rule which directly concerns rights to enjoy the land itself. 895 I.e. the law which governs the essential validity of the trust. 896 Which will mean, in the case of inter vivos transfers, the lex situs. 893

Formal Validity of the Trust 275 create a trust.897 The place of administration does not seem relevant to the matter, which does not appertain to the management of the trust assets. The place of execution of the trust instrument has an obvious connection to the process, but may equally be (i) arbitrary and (ii) of little enduring connection to the trust. Furthermore, it can be manipulated; and once one permits a law to govern the formal validity of the trust which is capable of manipulation, one might arguably just as well go the whole hog and permit the law chosen by the settlor to govern the trust also to apply to the question of formal validity. Article 9, Rome Convention adopts a very liberal approach to the formal validity of contracts, which goes beyond even the law applicable to the contract in order formally to validate the contract. Article 9(2) thereof states that: “a contract concluded between persons who are in different countries is formally valid if it satisfies the formal requirement of the law which governs it under this Convention or of the law of one of those countries.”

Hence a contract can be valid as to form, notwithstanding its formal invalidity by the applicable law. It is arguable that a similar rule should be applied to the trust itself as used in the Rome Convention. This view attracts the support of Sykes and Pryles, who comment on the common law position in Australia that: “A general agreement to create a trust is probably governed as to form either by the law of the place of the making of the agreement or the proper law”.898

Clearly, such a view, which imposes liberal formality requirements, is attractive for its pragmatism. However, this author would not favour this “alternative reference” test, for three main reasons. First, the approach of the Hague Convention is intrinsically less liberal than the Rome Convention in relation to matters of formality, since the former applies only to trusts evidenced in writing, whereas the latter imposes no such requirement for contracts. Secondly, if a trust can be formally valid if it satisfies the governing law of the trust, which the settlor is free to choose under Article 6, Hague Convention, it is far from obvious that the transfer should be saved if he chooses a law under which the trust itself is formally invalid. He has only himself to blame. Thirdly, if the trust is not formally valid by its applicable law, the applicable law (and, probably, a judge in the courts of the state whose law is applicable) would consider there to be no trust at all. Is it desirable for an English court to treat a trust governed by 897 See Part One of this book (especially sections 2–4). Admittedly, the argument for use of the settlor’s personal law is stronger in the case of will trusts, but the fact remains that, if a will is “launched”, the law which determines the formalities for creating the trust structure should not depend upon the origin of the transfer on trust. 898 E Sykes and M Pryles, Australian Private International Law, 3rd edn., 714; and see Van Grutten v. Digby (1862) 31 Beav 561; Guepratte v. Young (1851) 4 de G & Sm 217; Re Pilkington’s Will Trust [1937] 1 Ch 574. The correctness of this test is described as the “general view” by A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 134, p. 73 and is also favoured in Scotland by A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 635 and by E Crawford, International Private Law in Scotland, (Edinburgh, W Green, 1998), 382.

276 The Scope of the Applicable Law Utopian law as valid, when it would not be so regarded in Utopia, so that the trust “limps”? In short, it is submitted that, notwithstanding the absence of reference to formal validity in the Convention, the law applicable to the trust alone should be applied to determine if that law imposes any “trust specific” formality rules.

4. FORMATION AND ESSENTIAL VALIDITY OF THE TRUST

(A) Essential Validity of the Trust There can be no matters more central to the trust than its existence and validity. As to the latter, Article 8(1) states that the (putative) applicable law should be used. Although the term “validity” is a potentially broad one and might include formal validity as well as essential validity, von Overbeck suggests that matters of form are not covered by Article 8.899 Whenever the putative applicable law is used, the bootstraps objection can be made. Logically, it makes sense to ask: (i) is there a trust; and (ii) if so, what are its terms? A bootstraps900 approach requires the court (i) to assume901 that the terms of a trust are valid,902 including a choice of law clause; and (ii) to ascertain the putative applicable law from those terms and use it to determine whether the trust exists at all. Where there is found to be no trust, then there is no applicable law of the trust. One way around this bootstraps objection is to argue that where the trust is alleged to contain a choice of law clause, that clause is not just another term of the trust; rather, it stands separate from it and is insulated against the potential invalidity of the trust itself. A choice of law clause clearly agreed upon by the settlor and separate from the trust can more readily be used to determine the validity of the trust, even if the ultimate conclusion is that the trust is invalid. In the contractual context, such a view has recently been favoured by Nygh. He points out that as far as arbitration clauses are concerned, the principle of severability is enshrined in England in section 7, Arbitration Act 1996, which states that: 899

Von Overbeck Report, para. 83, p. 388. The matter is discussed in the preceding section. There are two sets of bootstraps here: those by which the putative applicable law is determined and those which use that putative applicable law to determine the validity of the trust itself. 901 To be fair, Art. 8 does not say that the terms of the trust must be assumed to be valid. However, since it advocates the application of a law specified by the Convention to a trust which may not be valid (and which may accordingly have no applicable law), it is implicit that the validity of the terms is assumed for the purpose of identifying a putative applicable law. 902 I.e. valid for the purposes of using them to ascertain whether a choice of law has been made, or otherwise to determine the law of closest connection. The correct approach to determining the terms of the trust for the purpose of ascertaining the law of closest connection is considered in the section “what if the law of closest connection states that certain terms of the trust are not valid?” in relation to Art. 7, above. 900

Formation and Essential Validity 277 “unless otherwise agreed by the parties, an arbitration agreement which forms or was intended to form part of another agreement (whether or not in writing) shall not be regarded as invalid, non-existent or ineffective because that other agreement is invalid, or did not come into existence or has become ineffective, and it shall for that purpose be treated as a distinct agreement”.903

Nygh also regards the choice of law clause as independent of the contract.904 If this is correct, then there is no cogent reason why the choice of law clause should not also be treated as independent in the trusts context too;905 in which case it could logically be used to determine whether the “trust” was valid, without falling foul of the bootstraps objection. However, it is submitted that such an approach cannot be accepted unconditionally.906 Bootstrap problems are not avoided, if the question whether the choice of law clause itself has been agreed upon must still be determined by the putative applicable law.907 Nor does the principle of severability convince. Why should the choice of law clause be treated other than as a disputed part of the main contract or trust, where it too is contested? It is even less apparent why this should be so when the law identified as putatively applicable regards the clause as part of the main contract or trust itself. Moreover, even if the clause can be treated as a distinct entity, it is parasitic on the trust908 itself and may be “infected” by the invalidity of the trust.909 Pragmatically, it is unobjectionable to use the putative applicable law where the validity of the trust is in issue, but both parties accept that, if it is valid, so too is the choice of law clause. But where one party disputes both the validity of the trust and its terms, use of the putative applicable law is theoretically difficult to defend.910 Where the choice of law is itself contested, the reason for according it a first amongst equals status is not self-evident. In the sphere of contract, a great deal of ink has been spilt on attempting to determine the correct way to resolve the issue of validity.911 There is little point in dwelling long on the matter in the trusts context, given that, for better or 903

P Nygh, Autonomy in International Contracts, 75. Ibid., 84. 905 True, the choice of law clause in a trust instrument will be the unilateral designation of the settlor and may be more difficult to construe as a separate agreement. However, it would still be possible for the choice of law clause to be treated as standing separate to the trust instrument. 906 See J Harris, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247, 252–5. 907 It is suggested that it should not be subjected solely to this law: see the discussion of “consent to the choice of law clause” in relation to Art. 6 above. 908 Or contract, as the case may be. 909 This, in turn, raises the question which law should determine a prima facie case of validity of the main contract or trust. In principle, this should be the lex fori, as it is a preliminary issue. 910 See e.g. the comments of Hoffmann LJ in Harbour Assurance Ltd v Kansas Ltd [1993] QB 701. 911 Discussion in English periodicals includes: A Jaffey, “Essential Validity of Contracts in the English Conflict of Laws” (1974) 23 ICLQ 1; A Jaffey, “Offer and Acceptance and Related Questions in the English Conflict of Laws” (1975) 24 ICLQ 603; D Libling, “Formation of International Contracts” (1979) 42 MLR 169; A Thomson, “A Different Approach to Choice of Law in Contract” (1980) 43 MLR 650; A Briggs, “The Formation of International Contracts” [1990] LMCLQ 192. 904

278 The Scope of the Applicable Law worse, the putative applicable law is clearly to be applied by virtue of Article 8.912 But for what it is worth, this author does not find the use of the putative applicable law wrong in principle. It was argued above in the context of considering whether a choice of law clause had been agreed upon, or, in the absence of choice, what the terms of the trust were from which the law of closest connection might be obtained, that it could not be simply assumed that the terms of the trust were valid.913 Where X and Y dispute the validity of a Utopian choice of law clause, use of Utopian law to determine whether that clause was agreed upon implicitly takes the side of the person so alleging.914 It was contended that an English court ought rather to satisfy itself, by applying the law of the forum, that there appears to be a trust.915 If so, it should then ask whether there appears to be a choice of law clause, or if not, what the terms of the trust appear to be. From that, an English court can form an even-handed view of the existence of the trust and its apparently applicable law. Having done this, then it is quite legitimate for the law identified as the prima facie applicable law to determine the validity of the entire trust. True, the conclusion may be that the trust was invalid, so that there was, after all, no applicable law. But that does not make the court’s approach wrong in principle. Selection of a connecting factor, namely the law which appears to govern a valid trust, is the best that a court can do. Accordingly, it is suggested that a prima facie determination of the trust’s terms should be conducted in the manner described above; and that the validity of the trust can then quite properly be subjected to the law thus identified. Such an approach appears quite consistent with the Convention.916 (B) Existence of the Trust On the question of the existence of a trust,917 the Convention is wholly silent. In part, this may be because most pre-conditions to the existence of a trust are “rocket-launcher” questions. However, not all should be so classified. If, for example, a dispute arises as to whether words used by a settlor show sufficient certainty of intention to declare himself a trustee,918 the question relates919 to 912 But it should be reiterated that the Convention nowhere deals with the means for determining the terms of the trust, which can accordingly be subjected to a more logically rigorous approach: see the discussion of “consent to the choice of law clause” in relation to Art. 6, above. 913 See the discussions of Arts. 6 and 7, above. 914 See the discussion of “consent to the choice of law clause” in relation to Art. 6, above. 915 On balance of probabilities. 916 Indeed, the Convention is silent on the question of how the validity of a choice of law clause is to be determined, or indeed any term of the trust. 917 As opposed to its validity. 918 It was argued above in the section “the general characteristics” in relation to Art. 2 that declarations by the settlor of himself as trustee are covered by the Convention. 919 Although a “rocket-launching” question of whether the settlor can alienate property at all still arises in relation to declarations of trust (see the discussion in Part One of this book), it is not the issue here.

Formation and Essential Validity 279 whether a trust comes into existence.920 Which country’s law should determine this question? On the one hand, the bootstraps objection to using the putative applicable law is as strong as ever: if no trust does in fact exist, then it is not obvious that the law applicable to that “trust” should be relevant. On the other hand, there is no obvious reason to apply different laws to questions of validity and existence, precisely because both raise the bootstraps objection. If Article 8(1) requires the validity of the trust to be determined by the putative applicable law, then the same should be true in relation to the existence of the trust.921 However, there may be one way in which existence should be treated differently from validity. The Rome Convention subjects the existence and validity of a contract to the putative applicable law. However, Article 8(2) then states that: “Nevertheless a party may rely upon the law of the country in which he has his habitual residence to establish that he did not consent 922 if it appears from the circumstances that it would not be reasonable to determine the effect of his conduct in accordance with the law specified in the preceding paragraph.”

If, for example, X makes an offer to Y to enter into a contract governed by Utopian law, and by Utopian law Y’s failure to respond to the offer after a specified time constitutes agreement to contract, Y may be able to rely on the law of his habitual residence to demonstrate that no contract has come into existence.923 By analogy,924 suppose that a settlor vests property in a defendant with the intention that the property be held by the defendant subject to a trust governed by Utopian law. The defendant claims that he did not consent to act as trustee;925 by Utopian law, the defendant’s silent acquiescence in the receipt of the property is sufficient to constitute an express trust. It is arguable that the defendant ought to be able to rely on his “home” law to show that he did not consent to act as trustee and that accordingly, the vesting of property in him created no trust at all. However, such an approach is inherently less attractive where there are multiple trustees of a settlement with different habitual residences. More importantly, no equivalent to Article 8(2), Rome Convention appears in the Hague Convention. It

920 The question is not a matter of construction of the trust (which is dealt with in Art. 8) since the issue is whether there is a trust at all. 921 And, as argued above, the putative applicable law should not be plucked out of the air. Rather, it should be determined from a studied process of ascertaining whether the law of the forum considers there to be a trust and what its terms appear to be. 922 Emphasis added. 923 See the Giuliano and Lagarde Report, OJ 1980 C282/4, p. 28; Egon Oldendorff v. Liberia Corp [1995] 2 Lloyd’s Rep 64; see also Cheshire and North, 588–9. 924 Obviously, the position is not entirely analogous, the contract being a more inherently bilateral process. However, precisely what was agreed between settlor and “trustee” may be in issue. 925 Of course, Art. 3 dictates that the trust should normally be evidenced in writing (save where the scope of the Convention has been extended, as it has been in the United Kingdom by s. 1(2), Recognition of Trusts Act 1987). However, the question may still arise whether the transferee agreed to act as trustee.

280 The Scope of the Applicable Law must follow that application of the putative applicable law can only be limited to the extent that its use would infringe the forum’s public policy.926

(C) The Terms of the Trust Curiously, the Convention is silent on the law which should be used to determine the validity of the terms of the trust.927 Some of those terms will be covered by more particular matters specified in Article 8(2). Intellectually, it is suggested that it is important to separate the two reasons why the terms may be relevant, namely (i) for the purpose of ascertaining the law of closest connection; or (ii) for the purpose of determining the rights and duties of the parties once the trust’s validity is established. As to the former, the approach advocated by this author has been discussed above.928 But where the validity of a trust has been established, its applicable law is settled;929 in other words, it is no longer merely putatively applicable. It follows that the conceptual objection to its use disappears and that the terms of the trust should be determined conclusively by the law applicable to the trust.

926

Pursuant to Art. 18. The word “construction” in Art. 8(1) does not appear to cover this; it apparently refers to the meaning of ambiguous terms not to the question of whether those terms exist at all. 928 See “what if the law of closest connection states that certain terms of the trust are not valid?” in relation to Art. 7, above. 929 It was argued above that the fact that certain “terms” are not valid by the applicable law should not cause the court to reopen the question of what that applicable law is. Of course, that applicable law can subsequently be varied pursuant to Art. 10. 927

ARTICLE 9—SPLITTING THE APPLICABLE LAW; THE ADMINISTRATION OF TRUSTS In applying this Chapter a severable aspect of the trust, particularly matters of administration, may be governed by a different law.

1. THE RIGHT TO SPLIT THE TRUST

The Convention permits the use of dépeçage,930 so that separate aspects of the trust may be subjected to different laws. It envisages that this separation may be especially appropriate in relation to matters of administration.931 Hence the settlor might provide that the validity of the trust be governed by Utopian law, whereas its administration be governed by Ruritanian law. However, the scope of Article 9 is wider than this. The drafters of the Convention expressly rejected proposals that only matters of administration could be subjected to a different law from that governing validity932 and/or that dépeçage could only be effected by express choice of the settlor pursuant to Article 6 and could not be achieved in the absence of choice under Article 7.933 Moreover, it appears that the trust may be split expressly by the settlor either by a choice of law to govern one aspect of the trust only934 or by specifying different laws to apply to different parts of the trust. However, trust splitting has its limits.935 930 P Nygh, Autonomy in International Contracts, 122–3 argues that the word dépeçage is a broader phenomenon than “splitting” the applicable law. Dépeçage may describe any transaction which is subjected to different choice of law rules (e.g. as to “rocket-launching” creation of the trust and its subsequent “rocket” effects.) Splitting covers only the subjection by the settlor of matters normally governed by a single law to different laws (such as validity and administration). See also F Juenger, Choice of Law and Multistate Justice (The Hague, Nijhoff, 1993), 138–9; W Reese “Dépeçage: a Common Phenomenon in Choice of Law” (1973) 73 Col L Rev 58. Following Nygh’s view, “splitting” would be the appropriate term in relation to Art. 9. However, von Overbeck consistently uses the word “dépeçage” to describe the subjecting of issues normally governed by a single laws to different law and the word will be used here for the same purpose. 931 The settlor may choose the law of any trust state to govern administration, but it is envisaged that the place of administration may provide the most appropriate law: von Overbeck Report, para. 91, pp. 389–90. For the common law position, see A Wallace, (1987) 36 ICLQ 454, 473–82. 932 Rather generally, von Overbeck gives the examples of “issues concerning property which is situated in different countries or beneficiaries domiciled in different countries”: ibid., para. 91, pp. 389–90. 933 Ibid., paras. 93–4, pp. 390–1. See also the Canadian proposal (Working Documents Nos 36–8). 934 In which case, the remainder of the trust will be governed by the law of closest connection. Art. 3(1), Rome Convention makes this clear, by providing that the parties can “select the law applicable to . . . any part of the contract” (emphasis added). See P Nygh, Autonomy in International Contracts, 128–30. 935 An American proposal by which, in the absence of choice, each and every issue would be subjected to the law with which it was most closely connected, was ultimately withdrawn in the face of criticism of its incompatibility with choice of law techniques in other states and of its uncertainty: Working Document No 12; von Overbeck Report, para. 94, p. 390.

282 Splitting the Applicable Law; Administration of Trusts The right to split the trust accords with modern trends in the law of contract.936 Article 3(1), Rome Convention states that “By their choice the parties can select the law applicable to the whole or any part of the contract.” The American Restatement (2d) also envisages that the applicable law of a contract may be split.937 The issue to be split must be “a severable aspect of the trust”.938 A similar requirement appears to apply to the Rome Convention, where the choice must “relate to elements in the contract which can be governed by different laws without giving rise to contradictions”.939 Hence, the same issue cannot be subjected to a different law depending upon which party commences litigation.940 If the matter in question cannot be split, then a choice in respect thereof will fail and it is likely that the entire trust will be subjected to a single law.941 It might be objected that the settlor’s freedom to choose different laws for every aspect of the trust may lead to something resembling a trust “sans loi”, in that it may consist of a set of rights and duties unrecognisable in any one legal system. There is an obvious fear that the trust will be split to avoid the application of otherwise applicable “mandatory” rules. However, if the choices of law do not fall foul of the mandatory rule provisions of Articles 15 and 16, and if the issue in question is logically severable, it is not obvious why the settlor should not be permitted so to split the trust.

2. SEVERABLE PARTS OF THE TRUST

Nonetheless, it is not clear what constitutes a “severable” part of the trust for Convention purposes.942 A helpful definition is provided by Matthews, who argues that: 936 See P Nygh, Autonomy in International Contracts, 128–33. L Maerten argues that use of dépeçage is “moins contestable” in the case of the trust, which involves a network of relationships between settlor, trustee and beneficiary and whose administration is a matter quite distinct from the question of its initial validity: “Le Régime International du Trust après la Convention de La Haye du 1er Juillet 1985” [1988] La Semaine Juridique 3319. However, we shall see below that distinguishing between administration and validity is far from straightforward. See also Y Loussouarn, “La Convention de la Haye d’octobre 1985 sur la Loi applicable aux Contrats de Vente Internationale de Marchandises” [1986] Revue Critique de Droit International Privé 271, 279–80. 937 1988 Revision, § 187, comment (i). 938 In Shannon v. Irving Trust 275 NY 95, a trust instrument otherwise governed by New Jersey law subjected the trustees’ right to expenses and commissions to New York law. The trust was upheld. 939 Giuliano and Lagarde Report 1980 C 282/4, 17. 940 Such as the right to repudiate a contract for non-performance. 941 Giuliano and Lagarde 1980 C282/4, 17. It could be said that the law governing the contract or trust’s essential validity should be applied to those issues which were unsuccessfully subjected to dépeçage. However, where a settlor has expressly chosen a law to govern essential validity, but limited its application, use of that law for matters unsuccessfully subjected to dépeçage may be the last thing that he intended and it may better to apply the law of closest connection to those issues. See further the discussion of “choice of the law of a non-trust state to govern part of the trust” below. 942 D Hayton, “Developing the Law of Trusts for the Twenty-First Century” (1990) 106 LQR 87, 94, f/n 14, argues that legislative clarification of the matter should be provided.

Distinguishing Validity from Administration 283 “. . . ‘severable’ must mean either a qualitative difference in the type of exercise being done, e.g. construction of the substantive provisions as opposed to matters of administration, or a physical division of the trust into two or more parts, each having separate assets and being governed by separate substantive provisions, for example like separate appointments constituting separate trusts for capital gains tax purposes”.943

He points out that a split may be “vertical” e.g. where a single instrument creates two trusts subjected to different laws944 or “horizontal” where a single trust subjects different issues to different laws. Both are permissible in principle, though the split specifically mentioned in Article 9 of administration from the rest of the trust is clearly an example of the latter.

3. DISTINGUISHING VALIDITY FROM ADMINISTRATION

It might be thought that there is nothing radical in the Convention’s sanctioning of different laws to govern validity and administration. Indeed, there is some support for subjecting the two issues to different choice of law rules in the ordinary course of events. The American Restatement (2d) deals separately with the administration of a trust. For will trusts of movables, it applies the law of the state designated by the testator to govern its administration, or: “if there is no such designation . . . the testator’s domicil at death, unless the trust is to be administered in some other state, in which case the local law of that latter state will govern”.945

For inter vivos trusts, administration will be governed “as to matters which can be controlled by the terms of the trust” by the chosen law, or in default, by “the local law of the state to which the administration of the trust is most substantially related”.946 It is suggested that such an approach should not be adopted in the Hague Convention context. The Restatement’s differential treatment of inter vivos and testamentary trust appears to be rooted in a desire to apply the same law to will trusts as governs succession. However, the law of succession is part of the “rocket-launching” process, by which the trust comes into effect. Once it has come into effect, the fact that it is of a testamentary nature is not significant, since the question concerns the prospective administrative powers of the trustee. At common law, the separation of validity from administration also had its advocates. Rule 121 of the 10th edition of Dicey and Morris provided that: “the administration of a trust is governed (semble) by the law of its place of 943

P Matthews, Trusts: Migration and Change of Proper Law 60 (see generally ibid., 59–66). A phenomenon common in “pour-over” trusts, where a trust of nominal value is created, into which substantial assets from another trust may then be transferred (see ibid., 62). 945 Restatement (2d) on Conflict of Laws, § 271. 946 Ibid., § 272. See also E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1034–9; A Scott and W Fratcher, the Law of Trusts, 4th edn., §604–24. 944

284 Splitting the Applicable Law; Administration of Trusts administration”.947 This place of administration “is likely to be the place where the trustee resides or carries on business, especially if a trustee is a trust corporation”.948 However, it has been pointed out that common law authority for applying the law of the place of administration is sparse.949 The relevant cases are consistent with application of the law of the place of residence of the trustees or with the law of the situs of the assets,950 or with a simple application of the proper law of the [whole] trust to questions of administration.951 In contrast, however, there is support for the view that administration should be governed simply by the proper law of the trust. In re Hewitt’s Settlement,952 Eve J was faced with a Scottish marriage settlement which was subsequently being administered in England by trustees resident in England on behalf of beneficiaries also resident in England. The trust property was also in England. Eve J nonetheless ruled that the Public Trustee could only be appointed under the Trustee Act 1906 if the trust was governed by English law and that, as the trust was governed by Scottish law, he could not be appointed.953 Furthermore, Dicey and Morris now state that “It is desirable that a trust should be treated as a unit and that the trusts of all the property comprised therein should be governed by a single law”.954 Whilst one can readily empathise with this view,955 Matthews brushes this statement aside as being “At its lowest . . . an irrelevant, pious wish, in view of Article 9 of the Hague Convention”.956 947

Dicey and Morris, 10th edn., 1980, 683. See A Wallace (1987) 36 ICLQ 454, 475–82. Dicey and Morris, ibid., 684; see also re Pollak’s Estate [1937] Transvaal Provincial Division 91. Such a view is preferable to the application of the law of the situs of the trust property (proposed by V Latham, “The Creation and Administration of a Trust in the Conflict of Laws”, (1953) 6 CLP 176), which may be located in several jurisdictions (at least where the property is not immovable). For other writers’ views on the place of administration, see A Wallace ibid., 476. 949 A Wallace, ibid., 476–79; E Sykes and M Pryles, Australian Private International Law, 3rd edn., 715–8 state that there is “considerable obscurity” on the issue (at 716), but ultimately favour application of the proper law (at 718). 950 Re Nanton Estate [1948] WWR 113; Re Tyndall [1913] SALR 39. 951 Re Pollak’s Estate [1937] Transvaal Provincial Division 91; Re Betts Brown Trust Fund Trustees 1968 SC 170. Both cases are considered by A Wallace, (1987) 36 ICLQ 454, 477. 952 [1915] 1 Ch 228. 953 P Croucher, “Trusts of Moveables in Private International Law” (1940) 4 MLR 111, 118–9 points out that the case may be of more limited significance and simply concern the issue of whether, as a matter of the English law of administration, the Public Trustee can be appointed to a trust whose validity was determined by a foreign law. In other words, English, not Scottish law was in effect applied to the administration of the trust. A Wallace accepts this as a possible interpretation, without expressly endorsing it: (1987) 36 ICLQ 454, 479. However, it is not obvious why the law initially governing the trust’s validity should affect English law’s determination whether the Public Trustee can be appointed to administer that trust. Had the administration of the trust been determined by English law, then it is arguable that the Public Trustee could have been appointed. There is no indication in the case that Eve J considered that the law governing the administration of the trust might be “split” from the law governing its validity. 954 Dicey and Morris, 1091. The same view is expressed (twice) by J-G Castel, Canadian Conflict of Laws, 4th edn., 540 and 541. 955 A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 184, p. 94 comment that “. . . the essential unity of the trust relationship and the administrative need for the regulatory role of the courts to be exercised in a coherent and consistent manner provides a strong pull towards unified application of a single law.” 956 P Matthews, Trusts: Migration and Change of Proper Law, 60. 948

Distinguishing Validity from Administration 285 By far the most forthright opinion on whether administration should be subjected to a separate law from validity in the ordinary course of events is contained in Chellaram v. Chellaram.957 Whilst Scott J ruled that the specific question which faced him, namely the removal and replacement of trustees, was a matter of procedure governed by the law of the forum,958 he expressly rejected the view that matters of administration should be governed by anything other than the proper law of the trust. He was particularly influenced by the profound difficulty of determining which matters relate to validity and which to administration: “I find myself unable to accept the distinction drawn . . . in Dicey and Morris between ‘validity, interpretation and effect’ on the one hand and ‘administration’ on the other hand. The rights and duties of trustees, for example, may be regarded as matters of administration but they also concern the effect of the settlement. The rights of the trustees are enjoyed as against the beneficiaries; the duties of the trustees are owed to the beneficiaries. If the rights of the beneficiaries are to be ascertained by applying the proper law of the settlement, I do not understand how the duties of the trustees can be ascertained by applying a different law, and vice-versa. In my judgment, a conclusion that the law of the place of administration of a settlement governs such matters as the rights and duties of the trustees, can only be right if that law is the proper law governing the settlement”.959

The administration of the trust and the beneficiaries’ rights are inextricably intertwined. After all, “one of the ‘rights’ of a beneficiary is to have the trust properly administered”.960 This is, of course, true only in part. Overlap between questions of validity and administration is sometimes unavoidable: so, a power of the trustee to appoint the beneficiaries of a discretionary trust may be administrative in nature, but the consequence of its exercise directly affects the interests of the objects961 of the trust. So too, a power to vary the trust may directly adjust the respective interests of the beneficiaries. However, it is submitted that it is possible to draw a sensible distinction between matters going to the beneficiary’s interest under the trust and his interest in seeing that the trust is properly administered.962 Some administrative powers or duties do not directly 957 [1985] Ch 409. See also the more recent Ontarion decision in Branco v. Veira (1995) 8 ETR (2d) 49. 958 See the discussion of Art. 8(2)(a), above. 959 [1985] Ch 409, 432; D Hayton, “Trusts in Private International Law” (1985) 135 NLJ 18. Scott J’s words were approved by the South Australian Full Court in re Estate of Webb (dec’d) (1992) 57 SASR 193. See also V Latham, “the Creation and Administration of a Trust in the Conflict of Laws” (1953) 6 CLP 176. Compare the words of Walsh J in Augustus v. Permanent Trustee Company (Canberra) Ltd (1971) 124 CLR 245, 253. 960 A Wallace, (1987) 36 ICLQ 454, 481. She goes on to support Scott J’s approach and to argue that a single choice of law should govern all aspects of the trust (at 481–2). It is not clear, however, whether she would wish an express choice by the settlor of different laws to govern validity and administration to be struck down. See also E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1036. 961 I.e. of the persons qualifying to be considered as beneficiaries under the trust. 962 But see E Sykes and M Pryles, Australian Private International Law, 3rd edn., 715, who complain of a “vast amount of confusion” as to the meaning of “administration.” See also A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., 754–6.

286 Splitting the Applicable Law; Administration of Trusts affect the respective interests of the beneficiaries. Hence, it is quite possible for e.g. the law governing the appointment or removal of trustees,963 or the duty to provide requisite written information concerning the administration of the trust or the rights and duties of the trustees inter se, to be subject to a law other than that which governs the trust’s validity. Equally, it is logically possible for some issues to be subject to two different laws. The power of a trustee to dispose of trust assets might be subject to the law governing administration; but whether the property has been disposed of is a matter for the law governing inter vivos property transfers.964 Likewise, a trustee’s power to vary a trust in principle may be subject to the law governing administration; whilst the effect of a purported exercise of that power on the beneficiaries’ rights could be subjected to the law governing validity, since it will affect the extent of their respective interests.965 Matthews takes the example of the duty to keep accounts in order to illustrate the difficulty of separating administration from the rest of the trust. The duty might appear administrative in nature; but what if the trustee, who is under no obligation to account, mismanages the fund, so that it is rendered worthless? “Have the rights of the beneficiaries under the settlement been reduced to zero by means . . . of mere ‘administration’?”966 The answer must be, for these purposes, “yes.” The fact that a fund may have vastly reduced in value due to maladministration clearly affects the value of the beneficiary’s interest, just as many or most acts of administration will affect the fund’s value; it does not, however, purport to affect the extent of the beneficial interest owned by the beneficiary in the property itself. The view that it is possible sensibly to demarcate matters of administration is supported by the American Restatement (2d) on the Conflict of Laws, which uses the word administration to mean: “those matters which relate to the management of the trust. Matters of administration include those relating to the duties owed by the trustee to the beneficiaries . . . They include the powers of a trustee, such as the power to lease, to sell and to pledge, the exercise of discretionary powers, the requirement of unanimity of the trustees in the exercise of powers, and the survival of powers . . . They include the liabilities which 963 Which, although regarded as procedural in Chellaram, is now treated by Art. 8(2)(a) as substantive. 964 See the discussion of Art. 8(2)(d), above. 965 See the discussion of Art. 8(2)(h), above. Powers of maintenance and advancement might appear administrative in nature, but D Hayton points out that they may substantially affect the nature of a beneficiary’s interest: Underhill and Hayton, Law of Trusts and Trustees 15th edn., 955. He gives the example of a foreign law which allows all of a beneficiary’s contingent interest to be advanced to him. Advancement of that interest might be said to change the contingent right into an immediate one. It might be thought preferable that the question whether all of the beneficiary’s interest may be advanced to him be determined by the law governing the trust’s validity and not the law governing administration. However, it is suggested that this should not be so and that powers of maintenance and advancement should be subjected to the law governing administration. They do not affect the validity of the trust itself and they do not alter the extent of the beneficial interest (even if that beneficial interest’s nature might be altered by advancement of the entire share, or by a refusal of a foreign law to allow the beneficiary to receive income from the trust at the age of 18). 966 P Matthews, Trusts Migration and Change of Proper Law, 65.

Distinguishing Validity from Administration 287 may be incurred by the trustee for breach of trust . . . They include questions as to what are proper trust investments . . . They include the trustee’s right to compensation . . . They include the trustee’s right to indemnity for expenses incurred by him in the administration of the trust . . . They include the removal of the trustee and the appointment of successor trustees . . . They include the terminability of the trust . . . On the other hand, where the question is as to who are beneficiaries of the trust and as to the extent of their interests, the question is one of construction rather than administration. . .”967

It might be said that the difference between validity and administration is essentially one between matters which affect the substantive rights and interests of the beneficiaries and matters which simply go to the obligations of the trustees to take due care of the trust property and manage and invest it properly.968 However, it still remains unclear how the demarcation is to be made for Hague Convention purposes. Unfortunately, the Convention does not provide any definition or clarification of the meaning of the word “administration”.969 It was decided “to eliminate all rules dealing with the characterisation of issues as belonging to validity or administration”.970 Despite this, one option would be to require an autonomous meaning of the word “administration.” This, it might be argued, would ensure the uniform application of Article 9 throughout the contracting states.971 Such a definition might draw inspiration from the American Restatement (2nd). However, it is submitted that the uniform approach should be rejected. For one thing, given that there is not a definition as such of even the word “trust” in the Convention,972 it is unlikely that a more specific term, such as “administration”, will be defined. More importantly, where the trust is expressly split by a settlor, the exercise of determining what is a matter of administration is simply a question of discerning

967 Restatement (2d) on the Conflict of Laws, § 271, comment a. The Restatement (2d) itself refers different aspects of the trust to different laws: § 268–73. This, in turn, is criticised by A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 181, p. 93, who “. . . see a need for only a single dichotomy, that between the law applicable to the validity of the trust and that applicable to its administration.” See also J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, 224. 968 A Dyer and H van Loon, ibid., para. 123, p. 70 would appear to favour an even wider definition of “administration”, which covers all matters save establishing the validity of the trust itself. This would include “. . . questions relating to investment policy, liability for breach of trust, the appointment and installation of successor trustees, of who will be entitled to current or ultimate distribution of the trust property or whether the interest of a beneficiary is alienable” (Emphasis added). The last two matters seem to go well beyond “pure” administration and affect the extent of the beneficial interest. See also ibid., paras 179–88, pp. 92–6. 969 The word’s meaning is important, since it is used both in Art. 9 and in ascertaining the law of closest connection under Art. 7. 970 Von Overbeck Report, para. 88, p. 389. 971 Just as the European Court of Justice has tended to insist on autonomous meanings to key phrases in the Brussels Convention, such as “matters relating to contract” and “matters relating to tort” to ensure uniformity. 972 There is only a description of its characteristics: Art. 2.

288 Splitting the Applicable Law; Administration of Trusts which issues the settlor intended to be subjected to this law.973 The trust instrument may expressly state which matters relate to administration. Likewise, where there is no choice of law, a court deciding whether to split the law applicable to administration from that governing validity is simply concerned with which issues ought objectively to be governed by a different law. However, if a settlor chooses Utopian law to govern validity and Ruritanian law to govern administration, that still leaves open the question of which law is to determine what he meant by “administration.” There is some support for the application of Utopian law in the above example: “I might add that it seems logical that it must be the proper law governing the validity and extent of substantive rights that must determine whether a matter pertains to substantive rights or to administration, and whether or not the law governing administration may be replaced by another law”.974

Von Overbeck explains that whilst this represented the English position, the American solution appeared to be that the law governing administration determines what the word “administration” means (that is, Ruritanian law in the above example).975 Where there are different laws governing validity and administration, the law governing validity does not ex hypothesi govern the administration of the trust and accordingly does not have a self-evident right to determine the meaning of the word “administration.” Arguably, it is more sensible for the law governing its administration to determine what “administration” means. However, it cannot be right that the meaning of the word “validity” is determined by the law governing validity, whilst the law governing administration determines what “administration” means. Clearly, the two terms need to be mutually exclusive.976 The problem can best be resolved when it is appreciated that the issue in question is essentially one of construction. Accordingly, whichever law governs the construction of the trust should determine what is meant respectively by “validity” and “administration”.977 It ought not to be possible for the question of the construction of the trust itself to be subjected to two different laws insofar as it concerns matters of administration and validity, precisely because there needs to be a 973 There is an analogy with the interpretation given to Art. 5(1), Brussels Convention, which accords jurisdiction in matters relating to contract to the courts of “the place of performance of the obligation in question.” That place of performance is not determined autonomously, because it is a term of the contract. Instead, the applicable law of the contract determines where it is to be performed: Case 12/76 Industrie Tessili Italiana Como v. Dunlop AG [1976] ECR 1473; Case C–288/93 Custom Made Commercial v. Stawa Metallbau [1994] ECR I–2913; Case C–440/97 Groupe Concorde v. Suhadiwarno Panjan, (1999) CLC 1976. 974 D Hayton, “Trusts in Private International Law” (1985) 135 NLJ 18, 19. Hayton also adopts this position in Hayton and Marshall, Commentary and Cases on the Law of Trusts and Equity, 11th edn. (Sweet & Maxwell, London, 2001), 1039–41. 975 Von Overbeck Report, para. 88, p. 389. 976 Otherwise, the possibility of overlap, or for an issue to be classified as relating to neither validity nor administration, is obvious. 977 Compare Lindsay v. Miller (No 1) [1949] VR 13.

Splitting the Law in Absence of Choice 289 single law to determine the boundary between “validity” and “administration”.978 Where the settlor does not specify a law to govern construction and only subjects the administration of the trust to a separate law, it may be assumed that construction is to be subjected to the same law as that which governs validity. Accordingly, in the above example, it would be Utopian law which would determine what the words “validity” and “administration” mean.

4. SPLITTING THE LAW IN THE ABSENCE OF CHOICE

Although the trust may be split in the absence of an express or implied choice of law,979 one might wonder whether a court should do so readily. It follows from the above discussion that, for a number of issues, trust splitting is perfectly possible and that a distinction between e.g. validity and administration can be drawn. That is not to say that it is desirable. Where the settlor provides expressly for the place of administration, it is likely that the law of that place will be found to be the law applicable to the whole trust, either by virtue of an implied choice of law under Article 6 or as the law of closest connection under Article 7.980 Where the place of administration is not specified, it is likely to coincide with the residence of the trustee and/or the situs of the trust assets, both of which are factors relevant to ascertaining the law applicable to the (whole) trust. Subjection of the trust to a single law does at least involve coherent application of the rules of a legal system.981 Whilst this advantage may be overridden in favour of the autonomy principle where a settlor has split the trust, where he has not, it is best avoided.982

5. CHOICE OF THE LAW OF A NON - TRUST STATE TO GOVERN PART OF THE TRUST

A difficult question arises if a settlor chooses the law of England to govern the validity of the trust, but the law of Civilopia to govern the administration thereof (the latter state being one where the trust concept is unknown). If the 978 If it were sought to split the issue of construction, the split would be invalid and one might expect the law governing the validity of the trust to determine the construction of the whole trust. 979 Re Wilks [1935] Ch 645; Re Kehr [1952] Ch 26; von Overbeck, para. 91, p. 390. 980 “In my opinion [Art. 9] . . . has lost much of its usefulness, because in many cases of searching for the law applicable to the trust as a whole, the formula of art. 7 will lead to the applicability of the law that would have been applied to the administration of the trust”: M Koppenol-Laforce, Het Haagse Trustverdrag, 267. 981 Although not necessarily to the application of that law as it would be applied in the relevant foreign court, since the doctrine of renvoi is excluded by Art. 17. 982 However, it is unlikely that the courts of all contracting states would agree. In particular, were the Convention to be ratified in the United States, it is likely that their Delegate’s preference to subject each and every issue to its law of closest connection would resurface to some extent (see further Working Document No 12; von Overbeck Report, para. 94, p. 390).

290 Splitting the Applicable Law; Administration of Trusts trust is valid by English law, what should be made of the severance provision? The options appear to be: (i) upholding the choice of Civilopian law and, since it involves a non-trust state, treating the administration of the trust as outside the Convention’s scope altogether; (ii) rejecting the choice and applying the law of closest connection to the administration of the trust, on the basis that Article 6(2) adopts such an approach; (iii) rejecting the choice and the severance of the trust, and applying the law governing validity to all aspects of the trust. Option (i) should be rejected: in the case of validity, the Convention’s approach983 is to ignore the choice of a non-trust state, not to regard it as taking the trust outside the Convention. Option (ii) is favoured by von Overbeck.984 It might be said that a settlor who stipulates for a specific law to govern administration clearly does not want the administration to be governed by the same law which governs validity and that his preference for dépeçage should be upheld. However, it is suggested that such a view should not be followed. A settlor does not choose dépeçage in the abstract. He chooses to subject the administration of the trust to the law of Civilopia. If that choice of law for Civilopia is invalid, then there is no choice of law clause to govern administration at all; in which case, the severance clause for Civilopia is invalid. Option (iii) is, it is suggested, more attractive in principle.985 If a settlor has not validly split the trust by his choice of law clause, then the court has no need to split the law applicable to validity and to administration and should not do so unless there are strong objective grounds for doing so.986

6. MATTERS WHICH CANNOT BE CONTROLLED BY THE TERMS OF THE TRUST

The American Restatement (2d) on Conflict of Laws envisages that there are some matters of administration which cannot be subjected to the settlor’s choice of law. It appears to be referring to matters which concern the application of mandatory rules and public policy. It gives the example of an attempt to exclude the trustee’s liability for failure to take reasonable care.987 It envisages that, in relation to testamentary trusts, it would be the mandatory rules and public policy of the testator’s domicile that would be applied;988 for inter vivos trusts, it 983 Art. 6(2) (which does not state that its approach only applies to the choice of the applicable law to govern validity). 984 Von Overbeck Report, para. 96, p. 391. 985 The Australian delegation had lobbied for such an approach to be expressly stated in the Convention’s text, but to no ultimate avail: Working Document No 39. 986 Whilst the settlor might not have wished to have the administration of the trust subjected to the law governing validity, he is unlikely to have preferred the law of closest connection to the administration of the trust. In any event, a settlor who is unwise enough to choose the law of a nontrust state to govern administration might be thought to run a very serious risk that the entire clause will prove ineffective. 987 Restatement (2d) on Conflict of Laws, § 271, comment h. 988 Ibid.

Charitable Purpose Trusts: Administration and Validity 291 would be the law of the place of administration.989 These laws might be said to have the greatest interest in matters of administration. In relation to inter vivos trusts, it may be difficult to act in a manner which infringes the law of the place of administration. However, the application of mandatory rules and public policy of any state other than that of the governing law or of the forum is severely limited by the Hague Convention. Article 15 does allow for the application of mandatory rules of a state whose law is not that applicable to the trust. However, this will be as a matter of the choice of law process and, in any event, applies not to trust law but to related areas of law. So, for example, the mandatory rules of the lex situs might be applied to “the transfer of title to property and security interests in property”,990 precisely because application of the lex situs is the choice of law rule for property transfers. Article 16(2) permits the application of international mandatory rules of third state of close connection, but Article 16(3) allows states to enter a reservation to the effect that they will not adopt Article 16(2), which the United Kingdom has duly done.991 In its absence, there appears to be no question of applying the mandatory trust rules of any state under the Convention simply because that state has a close connection to, and/or interest in, the matter. Hence, the mandatory rules of the testator’s domicile, or of the place of administration, should not be applied to a trust whose administration is governed by a different law.992

7. THE ADMINISTRATION AND VALIDITY OF CHARITABLE PURPOSE TRUSTS 993

(A) Validity Charitable purpose trusts are accorded no special treatment under the Convention.994 However, some important points emerged in pre-Convention cases as to the English courts’ attitude to such trusts in private international law. Clearly, a charitable purpose trust must be valid according to its governing law.995 Delany cites the old case of Habershon v. Vardon996 for the proposition 989

Ibid., § 272, comment f. Art. 15(1)(d). 991 See the discussion of Art. 16(2), below. 992 Unless it can be said to be a matter of English mandatory rules or public policy to give effect to the rules of a third state, such as the place of administration. Compare, in the contract sphere, Ralli Bros v. Cia Naviera Sota y Aznar [1920] 2 KB 287. 993 Common law discussion includes: G Keeton, “Trusts in the Conflict of Laws” (1951) 4 CLP 107, 117–8; V Latham, “the Creation and Administration of a Trust in the Conflict of Laws” (1953) 6 CLP 176, 186 and 190; V Delany, “Charitable Trusts and the Conflict of Laws”, (1961) 10 ICLQ 385, 393–5. 994 For a comparative study of charities and foundations, see K Neuhoff and U Pavel (eds.) Trusts and Foundations in Europe: a Comparative Survey (London, Bedford Square Press, 1971). 995 Re Hoyles [1911] 1 Ch 179. 996 (1851) 4 De G & Sun 467; cited by V Delany (1961) 10 ICLQ 385, 391–2. 990

292 Splitting the Applicable Law; Administration of Trusts that the purpose must also be charitable according to English domestic law.997 In that case, a trust to assist the restoration of Jews to Jerusalem was treated as ineffective in England, on the basis that it served an essential political purpose. As a matter of English (domestic) law, a charitable purpose trust may take effect wholly or partly overseas.998 However, an English court will not uphold a charitable disposition if the purpose is illegal in the place where it is to be effected.999 However, it would be extraordinary if today trusts governed by a foreign law had to comply also with the requirements for charitable status of the law of the forum, and quite contrary to the approach of the Hague Convention. A charitable purpose trust valid by its applicable law must be given effect in England, subject to any mandatory rule or public policy constraints.1000 The American Restatement (2d) on Conflict of Laws states that in the United States: “In the case of charitable trusts, the courts have been even more ready than in the case of private trusts to uphold the trust if valid under the local law of the state of administration . . . even though it would be invalid under the . . . [settlor’s local law]”.1001

(B) Administration However, matters may conceivably be different where the administration of charitable purpose trusts is concerned and there may legitimately be a greater 997 And see In re Moore dec’d, Moore v. His Holiness Pope Benedict XV [1919] IR 316, where Powell J remarked in the context of a disposition to the Pope to be used in his absolute discretion in the carrying out of his office, that “. . . the gift . . . being . . . possibly illegal according to the laws of this country, is void”; see also W Binchy, Irish Conflict of Laws, 507. 998 “It is absolutely clear that . . . the courts have been in the habit of treating the phrase ‘charitable purpose’ as not confined to charitable purposes within this realm”, per Maugham J in Re Robinson [1931] 2 Ch 122; see G Keeton, (1951) 4 CLP 107, 117–8; V Delany (1961) 10 ICLQ 385, 392. The case concerned a disposition to the German State for the benefit of soldiers injured in the Great War. See also the Report of the Charity Commissioners for England and Wales 1963, para. 72. Where the charity concerns the fourth category of trusts stated by Lord Macnaghten in Income Tax Special Purposes Commissioners v. Pemsel [1891] AC 531, it may be necessary to show some benefit accruing to the United Kingdom from the trust: Camille and Henry Dreyfus Foundation Inc v. IRC [1954] Ch 672, at least if the charity operates from overseas. 999 Thompson v. Thompson (1844) 1 Coll 381; Re Robinson [1931] 2 Ch 122; Re Jacobs (1970) 114 Sol Jo 515; McGovern v. A-G [1982] Ch 321; V Delany, ibid., 392. 1000 Historically, the Mortmain Acts formerly prohibited the devise of land situated in England for charitable purposes. In today’s parlance, it was a form of international mandatory rule. See e.g. Oliphant v. Hendrie (1784) 1 Bro CC 571; A-G v. Mill (1831) 2 Dow & Cl 393; Mayor of Canterbury v. Wyburn [1895] AC 89. See further, V Latham, (1953) 6 CLP 176, 186 and 190; V Delany, ibid., 393–5. 1001 American Restatement (2d) on Conflict of Laws, §269, comment (h), p. 157; see also Hope v. Brewer 136 NY 126, summarised in M Rosenberg, P Hay and R Weintraub, Cases and Materials: Conflict of Laws, 10th edn., 810–11 where the New York Court of Appeal stated that “Our law with respect to the creation and validity of [charitable] trusts . . . was designed only to regulate the holding of property under our laws, and in our state, and a trust intended to take effect in another state, or in a foreign country, would not seem to be within either its letter or spirit.” The trust in question had been created by a testator domiciled in New York “for the care and relief of the sick and infirm in Langholm, in Dumfriesshire, Scotland.”

Charitable Purpose Trusts: Administration and Validity 293 role for the law of the forum, at least where the trust is to be administered in that state. At common law, it would not suffice that a charitable purpose trust were valid by its applicable law, if it were need to be administered in whole or in part in England. The trust must be: “capable of adequate supervision . . . the methods available for supervision . . . by means of the cy-près doctrine and schemes, lend themselves to the requirement of continuing control, both through the instrumentality of bringing pressure to bear on the trustees and, when necessary, altering or modifying the purposes for which the trust property is to be applied”.1002

This may today appear contrary to the Convention’s spirit, which expressly subjects such trusts to the applicable law (subject only to mandatory rules and public policy provisions). Of course, insofar as such a foreign charitable purpose trust is genuinely incapable of being supervised by an English court, its enforcement should be deemed contrary to public policy. But simply to strike down a trust which is valid by its applicable law on the basis that it would not be charitable by English domestic law, but where it could still be administered in England, cannot be acceptable.1003

(C) Settling a Charitable Scheme; the Cy-près Doctrine It is said that an English court will not settle a scheme for a foreign charity.1004 It “has not the means of administering the charity abroad”.1005 One obvious difficulty is that the trust property may be overseas.1006 However, it is the trustee’s residence, and not the location of the trust assets, within the jurisdiction which determines whether a scheme will be ordered and whether property may be applied by an English court cy-près.1007 In Provost of Edinburgh v. Aubery,1008 a trust for poor labourers in Edinburgh was created. Lord Hardwicke refused to order a scheme in an English Court of Equity, considering that the matter was for the Scottish courts to consider. In Re Colonial Bishoprics Fund (1841),1009 an English trust was created for the benefit of a South African charity. Upon the trust’s failure, Luxmore J was asked to apply the funds cy-près. As the trustees 1002

V Delany, (1961) 10 ICLQ 385, 391. Although that trust may not benefit from the same tax privileges in England: Camille and Henry Dreyfus Foundation Inc v. Inland Revenue Commrs [1956] AC 39. 1004 Provost of Edinburgh v. Aubrey (1754) Ambler 256; A-G v. Lepine (1818) 2 Swan 181; Emery v. Hill (1826) 1 Russ 112; Minet v. Vulliamy (1826) 1 Russ 113; Martin v. Paxton (1826) 1 Russ 115. See also Re Oldfield (No 2) [1949] 2 DLR 175. 1005 A-G v. Sturge (1854) 18 Beav 597. 1006 Dicey and Morris, 1094. 1007 Re Colonial Bishoprics Fund (1841), [1935] 1 Ch 148. See V Latham, (1953) 6 CLP 176, 187; V Delany (1961) 10 ICLQ 385, 395–6. On the question whether the residence test applies more generally to the administration of a foreign charity, see V Latham, ibid., 188. 1008 (1753) Amb 236. 1009 [1935] 1 Ch 148. See also A-G v. Sturge (1854) 19 Beav 594. 1003

294 Splitting the Applicable Law; Administration of Trusts were within the jurisdiction, he ruled that he could do so.1010 If an English court is unable to administer the scheme, it may direct the claimant to apply to a foreign court for such an order.1011 It may also direct that the fund be paid to the overseas trustees.1012 Under the Hague Convention, it would appear that an English court (which otherwise has in personam jurisdiction) may now be called upon to administer a trust governed by a foreign law in respect of trustees outside the jurisdiction. Indeed, a settlor might choose Utopian law to govern a trust’s validity and English law to govern its administration. One escape route would be to argue that the unwillingness of an English court to seek to administer a charitable purpose trust operating wholly overseas is a subject-matter limitation on jurisdiction and not a choice of law rule, so that the Hague Convention has nothing to say on the matter.1013 The problem with such an argument is that a rule of jurisdiction ought not effectively to frustrate the application of the Convention’s choice of law rules.1014 An alternative argument would be that the Convention requires only the recognition of a foreign trust and not its enforcement; in other words, that a recognising court is not required to take active steps to give effect to the trust, such as settling a scheme for its administration. However, it is clear from the matters specified in Article 11 of the Convention that a recognising state may be required to take positive action in respect of a foreign trust.1015 A third possibility would be for the court to rule that, as a mandatory rule of English law, equity will not act in vain, and that to purport to control a trust operating wholly outside the jurisdiction would be to do just that. It is this last argument which is most convincing: pragmatism dictates that an English court should not seek to administer foreign charitable schemes as much today as it did before the Convention entered into force.1016

1010 Indeed, it seems that the courts, when applying the cy-près doctrine, will seek wherever possible to order a scheme to apply funds wholly within the jurisdiction, even where the initial disposition took effect overseas: V Delany, (1961) 10 ICLQ 385, 397–8; see also Re Mirrlees’ Charity [1910] 1 Ch 163; Re Joseph’s Will Trusts (1907) 26 NZLR 504. 1011 Re Fraser (1883) 22 Ch D 527. 1012 Lyons v. New Bonaker (1867) LR 4 Eq 655; see also A-G v. Stephens (1834) 3 My & K 347; Re Vagliano [1905] WN 179. See also V Delany, (1961) 10 ICLQ 385, 396–7. 1013 Dicey and Morris argue (at 1094) that the rule that an English court will not administer a foreign charity is “unaffected” by the Hague Convention. Although they do not say why, they may implicitly be treating the rule as jurisdictional. 1014 Pearce v. Ove Arup Partnership Ltd [1997] Ch 293; [2000] Ch 403; J Harris, “Justiciability, Choice of Law and the Brussels Convention”, [1999] LMCLQ 360. 1015 Such as ordering recovery of the trust assets where the trustee has wrongfully mixed them with his own assets (Art. 11(3)(d)). 1016 At least where the trustees are resident in a state which would not recognise an English judgment settling a charitable scheme.

Non-Charitable Purpose Trusts 295

8. NON - CHARITABLE PURPOSE TRUSTS

Whilst it might be envisaged that the introduction of the trust into the private international law of non-trust states might provide an impetus for its introduction into their domestic law, application of the Hague Convention’s private international law provisions might even be used as a basis for reform of English trust law. A good example is provided by non-charitable purpose trusts. It is well known that, in English domestic law, non-charitable purpose trusts will not ordinarily be enforceable.1017 The reason lies in the difficulty of administration, enforcement and supervision. “Just as a car needs an engine, so a trust needs a beneficiary”.1018 The trust needs a person to whom the trustee is accountable and who is competent to enforce the trust. There must be “somebody in whose favour the court can decree performance”.1019 Hayton argues that, provided that a person is identified by the settlor as responsible for enforcing a non-charitable purpose trust, to whom the trustee owes obligations, it is far from obvious why such trusts should not be permitted in English law. The concept of an “enforcer” of the trust is well known in offshore jurisdictions, and especially in the Cayman Island’s STAR trust regime.1020 Hayton draws support for his view from the Hague Convention: “After all, if the English court gives effect to a trust expressly governed by ‘Suntopian’ law which allows a settlor to enforce the trust, although he is not a beneficiary, and which permits non-charitable purpose trusts to be valid where an enforcer has been appointed in the trust instrument, why cannot the same result be achieved under English law if the trust provisions are appropriately drafted?”1021

He continues: “Is it seriously suggested that the English court would invoke the orthodox beneficiary principles to refuse to give effect to a foreign non-charitable purpose trust of English assets where under the foreign governing law as a result of legislation (like that of Jersey or Bermuda or the Isle of Man or the Cayman Islands or the British Virgin Islands) such a trust is valid where the trust instrument expressly provides for an enforcer? Would the 1017 See J Martin, Hanbury and Martin, Modern Equity, 16th edn. (London, Sweet and Maxwell, 2001), ch. 14. 1018 D Hayton, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96, 97. 1019 Morice v. Bishop of Durham (1804) 9 Ves 399, 405. See also Re Astor’s Settlement Trusts [1952] Ch 534; Re Denley’s Trust Deed [1969] 1 Ch 373. 1020 On which, see A Duckworth, STAR Trusts (Saffron Walden, Gostick Hall, 1998); D Hayton, “STAR Trusts” (1998) 4 Amicus Curiae 13; D Hayton, “STAR Trusts”, (1998) 8 Offshore Taxation Review 43; P Matthews, “Shooting STAR: the New Special Trusts Regime from the Cayman Islands” (1997) 11 Trust Law International 67; A Duckworth, “STAR Wars: the Colony Strikes Back” (1998) 12 Trust Law International 16; P Matthews, “STAR: Big Bang or Red Dwarf?” (1998) 12 Trust Law International 98; A Duckworth “STAR Wars: Smiting the Bull” (1999) 13 Trust Law International 158. See also A Duckworth, “The New Frontier of Purpose Trusts”: http://www.iltrust-in-italia.it/webtutti/@Congress%201999/Duckworth%20New%20frontier.htm. See further the discussion of “STAR trusts” in relation to Art. 18, where it is concluded that the upholding of such trusts in England should not be deemed contrary to public policy. 1021 D Hayton, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96.

296 Splitting the Applicable Law; Administration of Trusts English court really hold the assets to be subject to a resulting trust for the settlor? It is submitted that, because the beneficiary principle should be regarded as the “enforcer principle” the English court would regard the foreign trust concept neither as repugnant to the English trust concept nor contrary to English public policy, the trust being one that the English court can ‘both enforce and control’ ’.1022

That being so, an English court shows that it can give effect to non-charitable purpose trusts;1023 and if that is the case, there is no obvious policy justification for refusing to give effect to such trusts if they are governed by English law. The argument is attractive, but not without difficulty. First, it is a dangerous line of argument that, because X is sanctioned in private international law, it should equally be sanctioned in domestic law. It is inherent in the application of the choice of law process that foreign law will be applied in an English court and that a claimant will sometimes be able successfully to vindicate a right which English domestic law does not recognise. The same is true in, for example, tort, where an English court will apply the law of the place where the tort occurred1024 and may find D liable for the commission of a tort not even existing in domestic law. That does not argue for the introduction of that tort into English domestic law. Indeed, the justifications for giving effect to foreign rights are partly dictated by other concerns not present in domestic law, such as those of comity and reciprocity (in the sense of encouraging foreign courts to give effect to English rights) and by the expectations of the parties as to which law will govern their relationship. Secondly, Hayton himself suggests elsewhere in his writing that a non-charitable purpose trust valid by its applicable law is a paradigm example where the English courts might invoke the public policy provision of the Convention.1025 That said, it is difficult to see what offends English public policy in giving effect to a trust where a party has locus standi to enforce it. Moreover, whilst applying concepts drawn from private international law in the context of domestic law is dangerous, it is true that there may sometimes be cases where it is difficult to see what policy justifications motivate the sanctioning of a legal relationship when governed by foreign law, but not when governed by English law. Non-charitable purpose trusts do seem to belong in that group. A trust which can be enforced by a named party does not impose overly-onerous obligations on the court. It is much more appealing for a court to uphold the express trust, which reflects the intentions of the settlor, than to rule that the trust fails and that a resulting trust for the settlor arises. Accordingly, applying the analogy of private international law to domestic law in this case has much to commend it and may lead to a greater willingness in future to uphold English domestic non-charitable purpose trusts. 1022

D Hayton, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 100. The fact that it is doing so only for private international law purposes is immaterial for present purposes; Hayton’s point is that the willingness to do so seriously undermines the argument that non-charitable purpose trusts cannot be enforced by an English court. 1024 Pursuant to the Private International Law (Miscellaneous Provisions) Act 1995, Part III (especially, s. 11). 1025 D Hayton, in J Glasson (ed.) International Trust Laws, ch. C3, 15. 1023

ARTICLE 10—CHANGING THE APPLICABLE LAW The law applicable to the validity of the trust shall determine whether that law or the law governing a severable aspect of the trust may be replaced by another.1026

1. MOTIVATION FOR , AND PERMISSIBILITY OF A CHANGE OF GOVERNING LAW

“. . . [W]hy should anyone want to change the proper law of a trust? Sometimes there are fiscal reasons . . . Or exchange control might apply to trusts governed by one law, but not to trusts governed by another . . . These days, the common reason given for wanting to change the proper law is that the trust in question is being ‘exported’ to a different jurisdiction, and, it is said, it will be easier if it is governed by the new local law; it will make legal advice cheaper and easier to obtain, and administration less onerous”.1027

If it is purported to change from the law of England to the law of Utopia, English law will decide if the change is permissible. In principle, it would seem that a right to change the governing law could be expressly contained in the trust instrument.1028 “If it is possible to change the proper law by transferring trust assets to another settlement governed by a different law, there seems no reason in principle why it should not be possible to include an express power to change the proper law of the settlement itself”.1029

It is less clear whether it could be implied into the trust instrument, though it is not obvious why this should be prohibited in principle.1030 However, in some 1026 See generally the very helpful and detailed account in P Matthews, Trusts: Migration and Change of Proper Law, Part II. The evolution of Art. 10 is described by von Overbeck, paras. 100–3, pp. 392–3. 1027 P Matthews, ibid., 67. See also R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7(5) Trusts and Trustees 24, 27, who points out that aligning the residence of the trustees with the applicable law supports the general view that “the courts of the country, whose laws govern the trust, ought to have jurisdiction over the trustee.” 1028 “. . . [W]e think it is clear that the law can be changed under an express provision in the trust instrument”: Lewin on Trusts, 17th edn., 293. See also Richard v. Mackay, judgment of 4 March 1987, Ch D, (transcript included in (1997) 11 Trust Law International 22). 1029 R Vos, (2001) 7(5) Trusts and Trustees 24, 28. 1030 But the permissibility of implied change is not certain. In Duke of Marlborough v. A-G (No 1) [1945] Ch 78, the Court of Appeal took the view that, in the absence of an express provision in the trust instrument, the applicable law could be changed only with the concurrence of all the beneficiaries, and that this would result in a new settlement being created. However, it is difficult to see why an implied power to change the governing law of an existing settlement should not be permitted, providing it is clear that the settlor did intend to make the change. That is a question of degree of evidence, not of substance. See also P Nygh, Conflict of Laws in Australia, 6th edn., 517 who tentatively supports implied change, citing the case of Duyvewaardt v. Barber (1992) 43 RFL (Can) (3rd) 139 in support. J-G Castel, Canadian Conflict of Laws, 4th edn., 540 suggests that a change can be effected at common law only by an express power reserved to the settlor, by agreement between all the beneficiaries or by court order.

298 Changing the Applicable Law states, including Jersey, change of governing law is apparently permitted only if the trust expressly provides therefor.1031 Despite the freedom contained in Article 10, the provision for change is significantly less liberal than that adopted in the Rome Convention context. There, it is simply provided in Article 3(2) that: “the parties may at any time agree to subject the contract to a law other than that which previously governed it.” In other words, the Rome Convention lays down a substantive rule of law to the effect that such changes are permissible. In contrast, the Hague Convention simply provides a choice of law rule to determine the validity of the change. In contract, the extent to which change of governing law should be permitted in principle has been much discussed.1032 The matter is much less widely considered in trusts. In English law, and in the law of most trust states, there seems little doubt that the beneficiaries can collectively agree to change the governing law1033 and there is no reason why the trustees should not be empowered to do likewise. More difficult is the “floating” choice of law, where the settlor provides that he, or the trustee, may choose the law applicable to the trust at a time subsequent to its creation. As a matter of English common law, such clauses may well have been invalid in contract law.1034 It might be said that a valid trust cannot exist unless it has a governing law at its inception. However it is suggested that such floating choices should be permitted as matter of English law.1035 Until a governing law is selected, the trust may simply be governed by the law of closest connection in the absence of choice, as determined by Article 7.1036

2. A NEW LAW OF CLOSEST CONNECTION

What if it is argued that the objective applicable law has changed?1037 If, for example, the trust assets are vested in the hands of new trustees who are resident in a different state, it may be that the law of closest connection changes. It seems then that the relevant question is whether, according to the law of closest connection at the time of the trust’s creation, the applicable law may be changed in 1031

P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 46. See A Briggs, “The Validity of ‘Floating’ Choice of Law and Jurisdiction Clauses” [1986] LMCLQ 508; A Beck, “Floating Choice of Law Clauses” [1987] LMCLQ 523; D Pierce, “PostFormation Choice of Law in Contract” (1987) 50 MLR 176. 1033 Duke of Marlborough v. A-G (No 1) [1945] Ch 78, 85. 1034 Armar Shipping Co Ltd v. Caisse Algérienne [1981] 1 WLR 207; EI Du Pont de Nemours & CO v. Agnew and Kerr [1987] 2 Lloyd’s Rep 585; A Briggs, [1986] LMCLQ 508; A Beck, [1987] LMCLQ 523. 1035 P Nygh, Autonomy in International Contracts, 100–3 argues strongly for the validity of such “floating” clauses in contract law. 1036 Indeed, it will be that law of closest connection which will determine whether the change of law is permissible (Art. 10). 1037 Such “objective” change of law is also covered by Art. 10: von Overbeck Report, para. 99, p. 392. 1032

Change of Law and Variation of Trusts 299 this manner.1038 If that law does not allow change save by express choice pursuant to a power in the trust deed, then the objective applicable law cannot be altered by changes in the centre of gravity of the trust.1039

3. CHANGE OF LAW AND VARIATION OF TRUSTS

The law governing a trust may also be changed by the court, upon an application pursuant to the Variation of Trusts Act 1958.1040 A trust which had substantial English connections at the time of creation may now be almost entirely connected to a foreign country, if e.g. the trustees and beneficiaries now reside overseas. In such a case, the English court may revoke an English trust and create a new settlement governed by a foreign law, with foreign trustees.1041 However the court will require an enduring connection of the parties with the foreign state before it will authorise variation.1042 The mere emigration of the beneficiaries abroad will not suffice. It appears that the parties need to show both an intention permanently to leave England and a substantial period of residence overseas. The court will be especially wary of those who seek a change for apparent tax avoidance purposes.1043

4. THE ROLE OF THE LAW INITIALLY GOVERNING THE TRUST

The application of the law presently governing the trust1044 to the permissibility of change may be difficult in practice, since it may not be clear exactly what needs to be permitted by that law. A good example arose in the Cayman Islands in Bridge Trust Company Ltd and Slatter v. A-G, Wahr Hansen and Compass 1038 In this respect, the English proposed text would have required the power to change the applicable law to be contained within the trust instrument itself, thereby precluding “objective” changes of the applicable law. Such a view was not, however, supported by the Australian and Canadian delegates; see von Overbeck Report, para. 103, p. 393. See also Working Documents Nos. 36 and 38. In any event, where a trust is initially governed by English law, it would appear that any change of law must be pursuant to an express provision or an application under the Variation of Trusts Act 1958: see D Hayton, Underhill and Hayton, Law of Trusts and Trustees 15th edn., 948; Duke of Marlborough v. A-G (No1), [1945] Ch 78. But see above, where this author argues that an implied power to change the applicable law should be sanctioned. 1039 This is also the view of D Hayton, (1987) 36 ICLQ 260, 272. 1040 On which, see the discussion of the variation of transnational trusts in respect of Art. 8(2)(h), above. 1041 Re Seale’s Marriage Settlement [1961] Ch 574. 1042 Ibid.; Re Roberts [1983] 70 FLR 158. See also Re Renshaw’s Trust Deed [1928] NZLR 460; Georges-Picot v. Henderson [1963] NZLR 950. See further R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7 Trusts and Trustees 24, 25. 1043 Re Weston’s Settlements [1969] 1 Ch 223; Richard v. Mackay, judgment of 4 March 1987, Ch D, (transcript included in (1997) 11 Trust Law International 22); Re Beatty’s Will Trust (No 2), judgment of 28 February 1991, Ch D (transcript included in (1997) Trust Law International 77). 1044 Which Art. 10 requires.

300 Changing the Applicable Law Trust Company Ltd.1045 A trust governed by the law of the Bahamas contained a clause which provided that the trust should have its situs and be governed by the law of whichever state the trustees should from time to time elect in writing. The trustees elected to subject the trust to the law of the Cayman Islands. A question arose as to the validity of a purported disposition for a charitable purpose and as to the respective roles of the law of the Bahamas and the Cayman Islands. Section 4(4), [Cayman Island’s] Trusts (Foreign Element) Law 1987 stated1046 that: “If the terms of a trust so provide the governing law of the trust may be changed to or from the laws of the Islands provided that: (i) in the case of a change to the law of the Islands, such change is recognised by the governing law of the trust previously in effect; (ii) in the case of a change from the law of the Islands, the new governing law would recognise the validity of the trust and the respective interests of the beneficiaries”.1047

The statute manifestly gave a role to the law of the Bahamas, but what exactly was that role? Clearly, the principle of changing the applicable law had to be recognised by Bahamian law;1048 but did the particular charitable disposition have to have be valid initially by Bahamian law? The point was of great significance, since it appeared that the trust itself was not charitable, and accordingly invalid, by Bahamian law.1049 Harre CJ concluded that Bahamian law had only to “recognise” the principle that a change of law could be made, and the fact that a purported change had been made by the settlor; it did not matter whether the particular change on the facts was valid by Bahamian law. It would be a matter for Cayman Islands law to determine whether the trust itself was valid; and sure enough, by that law the trust was valid. Such an approach appears ostensibly liberal and practical. If it is sought to escape the clutches of Bahamian law and instead to instate Cayman Islands law as the applicable law, then it may be as well to refer as little as possible to Bahamian law. The court’s approach effectively treats a clause authorising a change of law as separate from the trust itself and capable of surviving the trust’s invalidity. However, whilst Bahamian law recognised the principle of change, the fact remained that, as a matter of Bahamian law, the trust was invalid and, accordingly, so too was the purported change of law. It followed from the failure of the trust that, as a matter of Bahamian law, the settlor immediately 1045 See J Goldsworth, “Validity of Change of Law from the Bahamas to Cayman Islands” (1997) 3 Trusts and Trustees 7. 1046 It has now been incorporated into s. 89(4) of the Trusts Law (1998 Revision). 1047 Offshore states have tended to include express provision to the effect that any change of law must not invalidate the trust or affect the respective interests of the beneficiaries: see further M Lupoi, Trusts: A Comparative Study, 241. 1048 The principle is now incorporated into Bahamian law by s. 5, Trusts (Choice of Governing Law) Act 1989. 1049 In which case, it is not obvious that Bahamian law should have been used at all; however, the parties did not dispute that the original law regulating the “trust” was Bahamian law.

Particular Issues Relating to Change 301 became entitled under a resulting trust.1050 In principle, it is submitted that the law initially governing the trust must determine both whether change is possible in principle and whether it has been achieved in practice. It is only at the point that the change is effected that the “old” law ceases to have a legitimate say in the matter.1051

5. PARTICULAR ISSUES RELATING TO CHANGE OF LAW

(A) Retrospective Change Can a change of law be retrospective? The difficulty is that such a change may prejudice the interest of third parties, or of beneficiaries to the trust.1052 However, provided that the rights of third parties to the trust or other vested rights are not prejudiced, e.g. if the change in law relates only to an internal matter such as the law regulating the trustee’s powers of investment or delegation, it is not obvious why retrospective choice should not be allowed in principle. If parties are free to choose an applicable law, then arguably they should also be allowed to change it where no prejudicial consequences result.1053

(B) Other Relevant Laws Does any other law have a legitimate say under the Convention on the question of whether a change can be, and has been, made? In contract, North argues that the law of the forum should determine if change can be made, on the basis that whether the forum accepts the change concerns the connecting factor of freedom of choice, by determining exactly which choices are permitted.1054 However, Nygh convincingly argues that the real question is one of validity.1055 In any event, in trusts the connecting factors are determined solely by the Hague Convention for matters within its scope and the Convention allows a change of law in principle, providing that it is permitted by the presently applicable law. 1050 P Matthews, Trusts: Migration and Change of Proper Law, 71 suggests that the court must impliedly have reasoned that the “. . . resulting trust included all the terms of the attempted express trust (including the shifting law clause) consistent with that trust.” 1051 If Bahamian law states that there is no trust, it will presumably state that the law applicable to that trust cannot be changed. However, it might provide that the “change” to the law of the Cayman Islands represents the valid creation of a new charitable purpose trust to be subject to Cayman Islands law. 1052 The Rome Convention provides express protection in Art. 3(2), which states that any subsequent change of the contract’s applicable law “shall not prejudice its formal validity . . . or adversely affect the rights of third parties.” 1053 See P North, Essays in Private International Law (Oxford, Clarendon Press, 1993), 64–7. 1054 Ibid., 57. 1055 P Nygh, Autonomy in International Contracts, 101; J Harris, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247, 255–6.

302 Changing the Applicable Law The law of the forum will accordingly have no say on the permissibility of change. Nygh argues that the law to which it is sought to change should determine whether such a change is permissible and valid. He rejects the role of the law from which it is sought to change, which “may be the very law that the parties are trying to escape”.1056 Pragmatically, this is not unattractive. If it is sought to change the applicable law from Utopian to Ruritanian law, application of Ruritanian law involves subjection to the putative applicable law. If the putative applicable law can be used to determine the validity of an initial choice,1057 why not also a subsequent one? Is it desirable for an English court to apply Ruritanian law to a trust, when a Ruritanian court would itself not consider the trust’s law to have changed to Ruritanian law? Indeed, Schoenblum forcibly rejects the need for a change of law to be sanctioned by the law initially governing the trust at all.1058 He points out that the initially governing law may not permit change or may itself change over time. If a settlor is allowed to choose the law in the first place, why should he not also be allowed freely to change that law if it does not meet his requirements? A theoretical objection to Schoenblum’s view might be that once the trust has been created, it is regulated by the legal framework of the law initially applicable, including its laws on whether change is permissible. A theoretical endorsement of his view might be that the change is in any event best viewed as the creation of a wholly new trust,1059 so that the stranglehold of the law governing the “first” trust is broken. Pragmatically, the force of Schoenblum’s argument is undeniable and the straight jacket of the initial choice of law might prove a disincentive to ratification of the Convention.1060 In any event, Article 10 makes clear that the change must be sanctioned by the law from which it is sought to change. But it is silent as to whether the new law must also sanction the change. In other words, in the above example, must both Utopian law and Ruritanian law regard the change of law as valid? Logically, it is suggested that they both have a role to play. When a law is found to govern 1056

P Nygh, ibid., 101. As it certainly does in contract law: Art. 3(4), Rome Convention. It will be recalled that this author does not favour this approach in trusts: see the discussion of “the consent to the choice of law clause” in relation to Art. 6, Hague Convention, above. 1058 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 13–14. See also J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 239. 1059 See “one trust or two?” below. 1060 But even offshore, the change may need to be sanctioned by the law initially governing the trust: see s. 89(4), Cayman Island’s Trusts Law (1998 Revision). This requires a change from the law of the Islands to be recognised by the new governing law, but also (in subsection (a)) that a change to the law of the Islands “is recognised by the governing law of the trust previously in effect.” Contrast the British Virgin Islands’ Trustee (Amendment) Act 1993, s. 81 of which provides for a change from or to the law of the Islands and requires only that, where a change is made from the Islands’ law “the new proper law would recognize the validity of the trust and the respective interests of the beneficiaries.” (Emphasis added). A change to the law of the Islands would apparently not need to be recognised by the law previously governing the trust. 1057

Particular Issues Relating to Change 303 the trust, it governs all questions relating to the validity thereof. If the law for the time being governing the trust does not permit change, then that change should not be allowed. If it does allow a change to Ruritanian law, but the change is not permitted by Ruritanian law, it is difficult to see how the trust can be governed by the latter law, which is henceforth to govern the validity of the trust and its terms. Worse still, the “new” law may regard the entire trust as void. If Ruritanian law is not given a say as to whether it now governs the trust, the change to Ruritanian law may be permitted, only for that law immediately to declare that the trust is invalid. Such difficulty is avoided if a change has also to be valid by Ruritanian law. This is because Ruritanian law is extremely unlikely to regard the choice of its law as permissible, where it is to be used to govern a trust which it regards as invalid. If so, the trust remains governed by Utopian law, under which the trust is valid. Accordingly, it is submitted that a change needs to be valid both by the “new” law chosen and by the law from which it is sought to change.1061

(C) Severable Elements of the Trust The question of changing the applicable law is complicated where an element of the trust is severable. Suppose that a trust is stated by the settlor to be governed by Utopian law, but that matters of administration are stated to be governed by Ruritanian law. Suppose it is argued that the law governing administration has changed to Arcadian law. According to the Convention, it is the law governing validity (that is, Utopian law) which determines whether the law governing administration has changed.1062 Why should this be so? One answer is that it may be assumed that a settlor who “splits” the applicable law intends that all questions of validity are subject to a single law, including the validity of a change in the law governing the administration of the trust. However, this may be a fiction. The choice of law rule in Article 6 is designed to give effect to the intentions of the settlor and it may well be that, properly considered, he intended the validity of all matters concerning administration to be subjected to the law which governs the administration of the trust, (that is, Ruritanian law), including the question whether the law governing the administration of the trust can be changed. Nevertheless, the Convention’s approach appears to be that the law which determines whether the trust itself is valid (that is, Utopian law) should still govern the question. Such an approach is unattractive in principle, since it may defeat the settlor’s intentions and may mean that a change from Ruritanian to Arcadian law to govern administration is valid, notwithstanding 1061 This is not renvoi, it is simply the application of two different laws to the validity of change issue. See further J Harris, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247, 255–6. 1062 Von Overbeck, paras. 99 and 101, p. 392.

304 Changing the Applicable Law that the law initially governing administration, Ruritanian law, said that it was not. Effectively, the law governing validity is treated as the dominant law under the trust.1063

(D) Changing the Law Applicable to a Charitable Purpose Trust Matthews points out that there are great risks in permitting a change in the law applicable to a charitable purpose trust. The trust might be subjected to a new law by which the definition of charity is more restrictive than that of the old law and the purpose trust might then fail. He suggests that a power in a trust instrument to vary the law should be expressly excluded in relation to any disposition for a charitable purpose, or alternatively that a settlor should consider creating a trust for a charitable institution with enduring legal status rather than a purpose trust.1064 Similar sentiments are expressed by Duckworth in the Cayman Islands, who argues that: “. . . the mere possibility of a new governing law whose concept of charity differed from that of the Cayman Islands may be sufficient to make the disposition noncharitable at the outset and therefore void”.1065

6. ONE TRUST OR TWO ?

A point of great theoretical and practical importance is raised by Matthews: namely, if it is purported to change the governing law from English to Utopian law, does this amount to the amendment of a subsisting trust, or the creation of a new trust altogether? After all, “How can a particular trust, which exists only as a viewpoint of certain facts by a given legal system, pass out of that legal system and into another, whilst remaining the same trust?”1066 In favour of the “new trust” argument, he cites the Supreme Court of Delaware’s decision in Wilmington Trust Co v. Wilmington Trust Co,1067 where the court spoke of “re-creating” a trust by the law of another jurisdiction. Moreover, it was made clear by the English Court of Appeal in Duke of

1063 Even if the law initially selected to govern the administration of the trust (Ruritanian law) considers that any purported change in the law governing administration is invalid and that, accordingly, Ruritanian law still governs the trust. Provided that the law which governs the validity of the trust itself (Utopian law) finds that a change in the law governing administration to Arcadian law can be made, that will be conclusive. 1064 The example he gives being a trust for the University of Oxford. See P Matthews, Trusts: Migration and Change of Proper Law, 83. 1065 A Duckworth, “National Digest for Cayman Islands”, in J Glasson (ed.) International Trust Laws, ch. A7, 26. 1066 P Matthews, Trusts: Migration and Change of Proper Law, 67. 1067 26 Del Ch 397; P Matthews, ibid., 68.

One Trust or Two? 305 Marlborough v. AG (No 1)1068 that where the law applicable to a trust is varied by consent of all the beneficiaries, this is effectively to be treated as a new settlement. On the other hand, it might be said that a change from English to Utopian law is only permitted because English law so decrees.1069 In that sense, English law remains the governing law. If so, the essential legal framework of the trust has not changed and the trust should be treated as a subsisting one when the applicable law is changed. However, it is suggested that the limitations of this latter argument must be appreciated: English law may be the law which sanctions a change to Utopian law; but Utopian law is in truth being applied not pursuant to English domestic law, but pursuant to English private international law. Whenever a foreign law is applied in an English court, it will be because English law so decrees. The fact that English law gave the change to Utopian law its lifeblood does not hide the fact that the applicable law has changed and that the trust is now set against a new legal framework.1070 That might naturally be construed as the creation of a new trust altogether. After all, “a trust does not have any independent existence. . . . If the rules which give the beneficiaries . . . [their] rights are changed (because the proper law of the trust is changed) can it be said that the trustees continue to operate on the basis of the same agreement?”1071 But against that, it may be said to be implicit in Article 10, Hague Convention that the “one trust” approach is adopted for Convention purposes; it would not seem nearly as important for a change in the applicable law to be sanctioned by the law for the time being governing the trust, if what is sought to be created is a new trust altogether. Whatever the answer, the outcome may be unattractive. If a change in the applicable law involves the creation of a new trust, there may be very unattractive tax implications for the trust.1072 If it involves the variation of a subsisting trust, Matthews sees another problem. He imagines a trust governed by English law containing a power to vary the applicable law. If shortly before the expiry of the perpetuity period, the proper law is changed to that of Jersey, it would be for Jersey law to determine whether the trust could continue past the English perpetuity period. However, this would amount to:

1068

[1945] Ch 78. P Matthews, Trusts: Migration and Change of Proper Law, 68–9. 1070 It might be argued that the trust instrument and its terms have remained factually unchanged, and that therefore the trust should be regarded as continuing. However, those terms must be construed, and the validity of the terms of the trust determined, by reference to a legal system. When that legal system changes, it is difficult to see how the trust can be regarded as the same. In any event, a change in applicable law will often be motivated by the fact that the trust has been exported and the residence of the trustees and place of administration changed. 1071 R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7 Trusts and Trustees 24, 28. 1072 P Matthews, Trusts: Migration and Change of Proper Law, 76–7; R Vos, ibid., 28. D Waters, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118, 127 points out that a variation of a trust is normally treated as the creation of a new trust for taxation purposes. 1069

306 Changing the Applicable Law “a power (contained in the original [English] settlement) to permit excessive duration of the trust and excessive accumulations of income . . . Under the English common law, a special power (such as this would be) is invalid since it could be exercised in a way which would contravene the English perpetuity rules”.1073

In other words, the “subsisting trust” argument might lead to the striking down of the power to vary by English law ab initio. It is not surprising then that Matthews suggests that a power to change the applicable law should not be included in a trust instrument as a matter of course and that, where used, very careful drafting is required.1074 Otherwise, the apparently attractive flexibility of the clause can easily create a tax trap for the unwary.1075 If a power to change the applicable law is nonetheless included, he suggests that this should not be exercisable by the settlor, since this incurs the risk that the trust will be treated either as a sham, or that the settlor will be deemed to have kept control of the trust assets for taxation purposes.1076

7. CHANGES TO THE CONTENT OF THE LAW APPLICABLE TO THE TRUST

What should happen if the content of the applicable law should change during the lifetime of the trust, whether by legislative or case law development? Do such changes affect the parties to the trust? The question is not answered by the Convention.1077 In contract, Nygh argues that: “. . . if parties to an international agreement were truly autonomous, they should be able to select any existing, past or future rules (or none) for their contract and no changes could be made to their relationship without their consent”.1078

In other words, the parties would be free to state that they wish the law of Utopia to be applied, as it stood on a particular date. However, such a view has found little favour, since: 1073 P Matthews, ibid., 76; re Watson’s Settlement Trusts [1959] 1 WLR 732. Matthews suggests that perpetuity problems might be countered by the insertion of words such as “Subject to the application (if any) of the rule [or rules] against perpetuities and excessive accumulations”: ibid., 83. 1074 Ibid., 80–1. P Matthews and T Sowden, The Jersey Law of Trusts, 3rd edn., 46–9 also point out that the implications for the trust of a change in law can be unexpectedly unappealing and that the tax implications need careful consideration and conclude (at 48) that “It is little wonder that, amid this complexity, some draftsmen play safe, and make clear that the proper law cannot be changed at all.” 1075 P Matthews (in Trusts: Migration and Change of Proper Law, 73–4) suggests that a preferable strategy might be to appoint assets of an English trust to an existing foreign trust having the same beneficiaries. He advises that the trust instrument might specify a particular foreign trust to which assets may be appointed, allowing the assets to be moved from their existing location in times of economic or political difficulty (a “pilot” trust). 1076 Ibid., 87–8. 1077 Von Overbeck, para. 97, p. 392. 1078 P Nygh, Autonomy in International Contracts, 63. He noted that some French and Belgian decisions have favoured the view that the law can be frozen in time, but that elsewhere the view has gained very little support.

Changes to the Content of the Applicable Law 307 “the prevailing view is that the selection of the law by the parties supplies the connecting factor under the law of the forum, which thereupon will apply the law, as it stands at the relevant time, to determine the rights and liability of the contract”.1079

In re Helbert Wagg & Co Ltd,1080 Upjohn J rejected an argument that a choice of German law might refer to that law as of the date of contracting. However, that was on the basis that the parties intended German law to be applied as it stood at the time of litigation. Nygh argues that: “At most one can say that the authorities show that, in the absence of an express intention to ‘freeze’ the law, the proper law, as it exists from time to time, will apply.”1081

A court may be reluctant to apply the governing law to the extent that that law has been retrospectively changed.1082 More generally, however, it appears that the argument against freezing the applicable law in time is akin to that which prevents choice of anational law, be it in the private international law of contract or of trusts.1083 The connecting factor employed under the Hague Convention allows the settlor to choose to subject the trust to a particular legal system;1084 but he must choose a system which exists in a particular trust state and a court must also apply that state’s domestic law as it would be applied from time to time in the selected legal system.1085 Indeed, Article 23, which considers the problem of a choice of the law of a state which has several territorial units, provides that “any reference to the law of that State is to be construed as referring to the law in force in the territorial unit”.1086 Against this, it may be noted that although the Convention allows the settlor to choose the applicable law, it makes no attempt to apply it as it would be applied in the courts of any given foreign state. The settlor may, for example, carve up the trust through the process of dépeçage so that it represents the law of no one legal system;1087 and the doctrine of renvoi is excluded from the Convention.1088 True, freezing in time might allow the settlor to evade the 1079 Ibid., 63. See also F Mann, “The Time Element in the Conflict of Laws” (1954) 32 BYIL 217; J Morris, “The Time Factor in the Conflict of Laws” (1966) 15 ICLQ 422; J Grodecki, “Conflict of Laws in Time” (1959) 35 BYIL 58; E Spiro, “The Incidence of Time in the Conflict of Laws” (1969) 19 ICLQ 357. 1080 [1956] 1 Ch 323. 1081 P Nygh, Autonomy in International Contracts, 65. 1082 Von Overbeck, para. 99, p. 392. 1083 See the discussion of “the need to choose the law of a recognised legal system” in relation to Art. 6, above. 1084 Art. 6. 1085 Although it will not necessarily apply the designated law as a judge in the courts of that state would apply it, since the choice of law rules of the chosen state are disregarded and only its domestic law applied: Art. 17. 1086 Emphasis added. Art. 23 appears to lay down only a rule of construction. It does not directly prohibit the settlor from expressly providing that a law should be frozen in time. Moreover, the primary purpose of Art. 23 is clearly not to address the problem of freezing a choice of law in time (not least because the law chosen may not be that of a state comprising several territorial units). See further the discussion of Art. 23, below. 1087 Art. 9. 1088 Art. 17.

308 Changing the Applicable Law clutches of subsequent legislation, so that the trust might “limp” from state to state. However, the worst excesses of such evasion may be controlled by the mandatory rules of the forum and by use of the public policy doctrine.1089 In principle, then, it is suggested that freezing in time ought to be permitted where this is the express will of the settlor. Where, however, he is silent, the law should be considered to apply as it stands from time to time; and in the absence of any choice of law, the law of closest connection should be applied as it currently stands.1090 So much for the theory. In practice, however, if it is true to say that the settlor may not choose an anational law to govern the trust under the Convention, it is unlikely that he will be able to freeze a legal system in time either. Both scenarios would result in the application of the law of no current legal system of the world.1091 Any attempt to freeze the law of the designated state risks invalidating the choice of law altogether and is best avoided by the settlor.

8. DETERMINING AND CHANGING THE PLACE OF ADMINISTRATION

Nothing is said in the Convention about the moving of the place of administration1092 of a trust,1093 as opposed to changing the law governing its administration.1094 Express provision1095 may have been made in the trust instrument to permit a change in the place of administration: “Such provisions . . . are often termed ‘flee’ clauses, and they are designed for use in cases where changing

1089

See Arts. 15, 16 and 18. This is also the view, in the contract sphere, of P Nygh, Autonomy in International Contracts, 66. 1091 At least the doctrine of dépeçage requires the application of the current legal rules of various systems to the respective aspects of the trust, even if the sum total is a trust governed by the law of no one current legal system. 1092 This will normally be the place of residence of the trustees and there is obviously a very close link between the moving of the trust and the appointment of trustees resident in a foreign jurisdiction (which is, according to Art. 8(2)(a), a matter for the law applicable to the trust). 1093 The matter is considered by A Dyer and H van Loon, Report on Trusts and Analogous Institutions, pp. 100–1, 195–9. See also R Venables, Non-Resident Trusts, 8th edn. (London, Key Haven, 2001), ch. 8. In America, see R Hendrickson and N Silverman, Changing the Situs of a Trust (New York, Law Journal Seminars Press, 1982–(loose-leaf)); E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1038–9. 1094 The latter being dealt with in Art. 10. Art. 9 makes clear that the law applicable to administration may be different from the law which governs validity (see above). 1095 As a matter of English domestic law, a trustee with an express power of appointment who appoints foreign trustees will be given a wide margin of discretion and “the test is simply whether the proposed transaction is not so inappropriate that no reasonable trustee could entertain it”: J Martin, Hanbury and Martin, Modern Equity, 16th edn., 520, considering Richard v. Mackay, judgment of 4 March 1987, Ch D (transcript included in (1997) 11 Trust Law International 22) and Re Beatty’s Will Trust (No 2), judgment of 28 February 1991, Ch D (transcript included in (1997) Trust Law International 77). Prior to these decisions, a much stricter approach had been taken: Re Whitehead’s Trusts [1971] 1 WLR 833. 1090

Determining and Changing the Place of Administration 309 political conditions require rapid action”.1096 Layton CJ extolled the virtues of permitting such a change in the Delaware decision in Wilmington Trust Co v. Wilmington Trust Co:1097 “There is no substantial reason why a donor, in dealing with that which is his own, may not provide for a change in the location of his trust with a consequent shifting of the controlling law. In an era of economic uncertainty . . . such a provision would seem to amount to no more than common foresight and prudence.”

A further, related motivation may be the avoidance of taxation.1098 Schoenblum describes the Convention’s lack of provisions on trust mobility as “a remarkable oversight”.1099 However, it is not obvious that it is either “remarkable” or an “oversight.” One would not expect a private international law Convention autonomously to determine whether and when a trust can be moved, since this would amount to the creation of a uniform law on the question. Moreover, von Overbeck states that it might have been provided in Article 8 that the issue of changing the place of administration should be subjected to the applicable law,1100 but that the Conference decided not to make express provision.1101 However, whether the place of administration may be changed may well be covered by Article 8(2)(h), which concerns the variation of trusts. In any event, it seems very much an issue central to the trust’s administration and one which should be subjected to whichever law governs its administration. Cheshire and North point out that the place of administration may be particularly difficult to determine where a settlor subjects his property in different states to separate administrative regimes.1102 Whilst it might be said that there are two places of administration, and that each could be subject to a 1096 P Matthews and T Sowden The Jersey Law of Trusts, 3rd edn., 46. For a very useful, detailed study, see P Matthews, Trusts: Migration and Change of Proper, Part I. See also R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7 Trusts and Trustees 24, 24–7. Vos considers the Bahamian Court of Appeal decision in The Private Trust Corporation v. Grupo Torras SA, judgment of 16 April 1999 (on which see P Fletcher, “Grupo Torras in the Bahamas Court of Appeal” (1998) 4(8) Trusts and Trustees 28) and the US decision in Federal Trade Commission v. Affordable Media LLC (1999) 179 F 2d 13130 (US), (on which see P McNair, “Anderson in the Court of Appeals—Contempt, Impossibility and Foreign Trusts” (1999) 5(9) Trusts and Trustees 27), both of which show a degree of antipathy to “flee” clauses. 1097 26 Del Ch 397, 24 A 2d 309 (1942); see M Rosenberg, P Hay and R Weintraub, Cases and Materials: Conflict of Laws, 10th edn., 804–6; A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 198, pp. 99–100. For a discussion of changing of the place of administration in American law, see J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, pp. 230–3. 1098 Re Whitehead’s Trusts [1971] 1 WLR 833. 1099 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little”, (1994) 3 J Int Corp P 5, 13. He goes on to state that this, and the provisions which do exist on changing the applicable law, are “symptomatic of a general detachment from or an apparent lack of sophistication about the real circumstances under which internal trusts operate today” (at 14). He argues that it may be of little use to be able to change the applicable law if the situs of administration cannot also be changed. 1100 I.e. the law governing the administration of the trust. 1101 Von Overbeck, para. 89, p. 389. 1102 Cheshire and North, 1038–9; Re Tyndall [1913] SASR 39.

310 Changing the Applicable Law different law, they point out that some issues will be common to both regimes, such as the appointment of trustees. In that scenario, they favour the place of administration being identified by the place of residence of the majority of the trustees.1103 However, it is submitted that, where the settlor does not sever the trust expressly, and where the trust is to be administered in two different states, such a case would not in any event be an appropriate one for the court to split the trust, there being no one obvious law which should govern the trust’s administration in the absence of a choice of law by the settlor.

1103 Cheshire and North, ibid., 1039; Re Smyth [1898] 1 Ch 89, 94.; compare Permanent Trustee Co (Canberra) Ltd v. Permanent Trustee Co of New South Wales Ltd (1969) 14 FLR 246, 252–3.

ARTICLE 11—THE RECOGNITION OF TRUSTS CHAPTER III—RECOGNITION A trust created in accordance with the law specified by the preceding Chapter shall be recognised as a trust. Such recognition shall imply, as a minimum, that the trust property constitutes a separate fund, that the trustee may sue and be sued in his capacity as trustee, and that he may appear or act in this capacity before a notary or any person acting in an official capacity. In so far as the law applicable to the trust requires or provides, such recognition shall imply in particular(a) that personal creditors of the trustee shall have no recourse against the trust assets; (b) that the trust assets shall not form part of the trustee’s estate upon his insolvency or bankruptcy; (c) that the trust assets shall not form part of the matrimonial property of the trustee or his spouse nor part of the trustee’s estate upon his death; (d) that the trust assets may be recovered when the trustee, in breach of trust, has mingled trust assets with his own property or has alienated trust assets. However, the rights and obligations of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum.

1. THE NEED FOR ARTICLE 11

Article 11 is at once one of the most important and self-evident Articles of the Convention. One of the Hague Convention’s primary functions is to facilitate the recognition by non-trust states of the trust structure. It does this first by explaining the characteristics of a trust in Article 2, then by laying the ground for that recognition by a set of harmonised choice of law rules. However, the word “recognition” means very little in this context. A contracting state to the Hague Convention will apply its choice of law rules to a trust falling within the scope of the Convention. That trust must show the characteristics specified in Article 2 and meet the requirement of Article 3.1104 If the trust is found to be governed by English law, then English law will determine the nature and effects of the trust, precisely because it is the governing law. 1104 Unless the contracting state has extended the operation of the Convention pursuant to Art. 20 beyond such trusts, as the United Kingdom has done in s. 1(2), Recognition of Trusts Act 1987.

312 The Recognition of Trusts If a contract is found to be governed by French law, an English court does not ask itself whether to “recognise” that contract; it simply determines its validity and effects by French law. “The so-called ‘recognition’ of the trust merely concerns the fact that the trust has to be given in principle all those effects which the foreign law governing the trust attaches to it”.1105

So, if according to the applicable law of the trust, the trust has the effects specified in Article 11, it must be given those effects in each contracting state; if it does not, then it should not be.1106 Von Overbeck argues that the recognition principle in Article 11: “did not absolutely go without saying: it would have been possible to recognise . . . without regard to whether the law applied by the foreign court was the same law as that which would have been applicable in the recognising country . . . certain effects of trusts created under other laws”.1107

This point is hard to comprehend. The principle of recognition of trusts governed by a law specified in accordance with Articles 6 and 7 is self-evident. It is inconceivable that the Convention might also require1108 recognition of trusts whose governing laws have not been determined in accordance with the Convention.1109 There can no question of the word “recognition” in Article 11 being interpreted so as to have a more significant meaning. In particular, the word “recognition” has nothing to do with the effects to be given to foreign judgments in trust matters. A contracting state to the Hague Convention will not be compelled to recognise an English trust judgment unless the English court is jurisdictionally competent in the eyes of the state of recognition. Likewise, a non-contracting, non-trust state will be compelled to recognise an English judgment if its criteria for the recognition of foreign judgments are met. In that sense, the obligation on non-trust states to “recognise” the trust exists quite independently of the Hague Convention.1110 1105

H Kötz, in D Hayton (ed.) Modern International Developments in Trust, ch. 3, 44. Art. 11(3) is not a rule of uniform law. It merely sets out typical effects of the trust in a given legal system. Art. 11(3) makes clear that these apply “in so far as the law applicable to the trust requires or provides. . . .” Only Art. 11(2) appears to be applicable regardless of what the governing law of the trust might say. 1107 Von Overbeck, para. 106, p. 394. 1108 Although it does, of course, permit such recognition in Art. 14. 1109 It is true that a foreign judgment may be recognised, even if decided by application of choice of law rules different to those of the state of recognition. However, the word “recognition” in Art. 11 does not refer to judgments; rather, it refers to giving a trust certain effects where it is valid according to the Convention’s choice of law rules. 1110 In the Brussels Convention context, this equates to almost automatic recognition of English trust judgments throughout Europe (Art. 25) in civil or commercial matters. (Art. 25, Brussels Convention will be replaced by Art. 33(1), Brussels Regulation; see f/n 5, above). Whilst almost all states will have a public policy defence to the recognition of judgments, it is not clear that it should be readily invoked simply because the state does not know the trust structure. There is, after all, nothing repugnant in itself in a typical trust judgment. In the Brussels Convention context, it would 1106

Article 11(2)—A Uniform Law? 313 It is clear that the law governing the trust need not be that of a contracting state. This follows from the fact that Articles 6 and 7 do not require the applicable law to be that of a contracting state. However, if the applicable law is that of a non-trust state, then the matter will be taken outside the scope of the Convention by Article 5 and it follows that the recognition principle of Article 11 does not apply.1111

2. ARTICLE 11(2)— A UNIFORM LAW ?

Article 11(2) is not a terribly ambitious provision. It makes clear that the trust assets are separate from the personal assets of the trustee. It does not state what the consequences of that fund being separate are, but it is very strongly implied that the trust fund is not available for the personal use of the trustee or to his creditors, a fact reinforced by Article 11(3)(a) and (b). The trustee’s capacity to sue and be sued is also covered by Article 11(2). It makes clear that this extends beyond lawsuits and covers any relevant appearances before an authority or public notary. In specifying certain minimum requirements of recognition, Article 11(2) to some extent lays down a uniform substantive law.1112 As Lupoi notes,1113 whilst the separate fund requirement will be common to all trusts falling within the Convention (because they must show the characteristics listed in Article 2 and this is one of the stated characteristics), the applicable law of a “trust” satisfying the Article 2 characteristics may not recognise the trustee’s right to sue, be sued, or to appear before notaries and public authorities; yet those requirements must still be recognised under Article 11. One might have thought that Article 11(2) was intended to be an illustration of the typical nature of a trust so as to assist non-trust states, but that the matter would ultimately be for the governing law to determine. However, this does not seem to be the case; it follows that all Convention trusts qualifying for recognition must be accorded these effects.

seem distinctly odd to include jurisdiction provisions in relation to trusts (Art. 5(6)), only then to find that the decision of the court with jurisdiction could not be recognised or enforced elsewhere. (Art. 5(6), Brussels Convention is reproduced as Art. 5(6), Brussels Regulation). 1111 Although recognition would still be possible outside the Hague Convention rules, it is difficult to see how this could be achieved in practice. 1112 See also K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 27. 1113 M Lupoi, Trusts: A Comparative Study, 354.

314 The Recognition of Trusts

3. THE SPECIFIC REQUIREMENTS OF RECOGNITION : ARTICLE 11(3)( A )–( C )

(A) Uniform Law? Article 11(3) specifies four key consequences of the recognition of a trust.1114 It might be thought that these lay down autonomous substantive rules independent of the law applicable to the trust. But as Gaillard and Trautman note,1115 these provisions lay down only the most basic of requirements for the existence of a trust, in order to clarify matters for non-trust states. Article 11(3) provides illustrations of the effects of the trust, but these effects are to be recognised only “in so far as the law applicable to the trust requires or provides.” It is the governing law which is the ultimate arbiter. Likewise, any additional consequences which follow according to the trust’s governing law should also be “recognised” in a contracting state, save insofar as the provisions on mandatory rules and public policy are infringed. Von Overbeck comments that sub-paragraphs (a), (b) and (c) are easy to understand.1116 To an extent, this is true. They are all natural consequences of treating the trust fund as separate from the trustee’s personal assets. Accordingly, those trust assets are not available to satisfy claims brought against the trustee by his personal creditors, or upon the trustee’s bankruptcy or insolvency, nor do they form part of his matrimonial property or estate. However, these sub-paragraphs do given rise to a certain controversy.

(B) Trusts Assets and Claims against the Settlor One difficulty is that Article 11(3) makes clear that the assets are not available to satisfy personal claims brought against the trustee. It does not say the same of the settlor.1117 Whilst one might think this self-evident where the settlor has divested himself of legal and equitable ownership, it cannot be so in all cases. As Lupoi notes, Article 2 includes within the Convention cases where assets have 1114 But note that it leaves some questions for the choice of law rules of the forum to determine: “Art. 11 does not treat the question what form the identification of trust property should take . . . The law of the country where the goods are situated will have to decide whether the trustee should identify himself as trustee when he gets the goods transferred to him. This goes for the original transfer from the settlor to the trustee as well as for the transfer from third persons to the trustee. The applicable trust law may have rules on how the trustee should comply with his duty to earmark the trust property. This law is applicable on the basis of art. 8 HTC. The law of the country where the goods are situated will decide to what extent these rules can be followed, and whether extra requirements are to be fulfilled”: M Koppenol-Laforce, Het Haagse Trustverdrag, 268–9. 1115 E Gaillard and D Trautman, “Trusts in Non-Trust Countries: Conflict of Laws and the Hague Convention on Trusts” (1987) 35 Am J Comp Law 307, 332–3. 1116 Von Overbeck, para. 112, pp. 394–5. 1117 See also the Paris Court of Appeal decision of 18 April 1929, [1935] Revue Critique de Droit International Privé 149; K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 20–1, 28.

Inter-Relationship of Articles 11, 15 and 16 315 been placed by the settlor only under the “control” of the trustee, where the settlor has reserved rights and powers in respect thereof. In such a case, the assets may technically be available to satisfy claims of the settlor’s creditors.1118 It is inconceivable that a common law court would so construe the Convention;1119 but the risk of it happening elsewhere1120 cannot be wholly discounted.

4. THE INTER - RELATIONSHIP OF ARTICLES 11 , 15 AND 16

More fundamentally still, Article 11 is subject to the provisions of Articles 15 and 16,1121 which throws some doubt on the question whether the trust assets must be treated as a separate fund from the trustee’s personal wealth. Article 15, in particular, preserves the application of the mandatory rules designated by the conflict rules of the forum insofar as they concern matters such as title to property, security rights, creditor protection in matters of insolvency and the protection of third parties acting in good faith.1122 This raises the question of what happens if the requirement in Article 11 that the trust property constitutes a separate fund from the trustee’s personal assets itself infringes the mandatory rules of the law designated by the forum. Could a trust still function without the separate fund element, or would it be stripped of any meaning or function?1123

5. ITALIAN COURTS AND THE SEPARATE FUND

It is this concept of a division of property ownership inherent in the common law trust which can be so problematic in non-trust states.1124 Paton and Grosso, for example, state that: “Italian property law contains two fundamental principles: (1) the principle of the indivisibility of ownership in property, considered as fundamental to the Roman legal system and faithfully reflected in the present Italian Civil Code;1125 1118 M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law 222, 230; M Lupoi, Trusts: A Comparative Study, 339. 1119 Assuming that the trust is not a sham. 1120 In states which have no institution homologous to the English-model trust. 1121 Von Overbeck, para. 105, pp. 393–4. 1122 See the discussion of Art. 15, below. 1123 At least in the eyes of a common lawyer. 1124 See also the discussion of “‘translating’ the trust into civil law systems” below. 1125 Although the idea of a fiduciary relationship is known in Italy. F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, 75 explains that such an arrangement is seen as “. . . the interaction of two agreements. The first agreement is external in nature: a voluntary act which is fully valid and actionable with respect to third parties. The second agreement is internal and personal: it modifies the ‘external’ agreement to the extent that it can compel the transferee to respect the obligations assumed with respect to the transferor.” However, this is very different from the common law position, since “. . . the transferor’s remedies for breach of this second ‘internal’ agreement are limited to actions against the ‘fiduciario.’ ” (ibid., 76).

316 The Recognition of Trusts (2) the principle that rights in rem form a numerus clausus, limited to the categories contained in the Civil Code. These do not include fiduciary relationships which, to the limited extent recognised, form part of the category of rights in personam”.1126

They go on to observe that: “as equitable rights such as are found in a trust are considered by scholars to be closer to the Italian law of obligations, at best in the nature of enforceable extra-contractual rights, one can see the difficulty that beneficiaries would have in protecting their interest were, for example, a trustee to become bankrupt”.1127

Nevertheless, they point out that, even prior to ratification of the Convention, Italian courts had shown ingenuity in giving effect to English trusts.1128 In Piercy v. EFTAS1129 the testator, a British national and domiciliary, left movable and immovable property on trust for sale in his will. He appointed his wife and children as trustees, with power to postpone sale. The three children would take equal beneficial shares of the proceeds of sale of the property. Part of the estate was land situated in Sardinia, which was the subject of a Sardinian governmental expropriation decree. A governmental Commission sued the widow to obtain the property. The children objected that the widow had no beneficial interest in the land and that the court could not give effect to the decree. The court was unwilling to give direct effect to a trust in Italy, but instead regarded the beneficiaries as absolute owners of the property, so that expropriation orders needed to be obtained against the three children. Such a solution reflected the spirit of the testator’s intentions. Although not accepting the dual ownership of property, the court was able to look beyond legal title to observe that, in substance, it was the children who had the benefit of the property. Of more recent note is a decision of the Corte di Cassazione,1130 where the concept of separation of a fund was acknowledged for the purpose of recognising that certain securities which had not been duly registered, so that they were apparently owned by a fiduciary company in administration, were in fact owned by clients thereof. “The Court of Cassazione therefore expressly recognised the ‘separation’ of assets, confirming the clients as ‘effective owners’ of the securities”.1131 The Tribunale di Roma also held (on 6 December 1989) that in determining liability 1126 A Paton and R Grosso, “The Hague Convention on the Law Applicable to Trusts and on their Recognition: Implementation in Italy”, (1994) 43 ICLQ 654, 656. 1127 Ibid., 656. 1128 See also the discussion of “contacts between the trust and the Italian legal system” in F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, 12–8 and by F Albisinni and R Gambino “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 76–9. 1129 Tribunal of Oristano, judgment of 15 March 1956, discussed by A Paton and R Grosso, (1994) 43 ICLQ 654, at 656–7. 1130 Corte di Cassazione, sez I civile, 14 ottobre 1997 n 10031, discussed by F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, pp. 10–11; F Albisinni and R Gambino, (1993) 2 J Int Corp P 73, 79. See also C Reymond, “Réflexions de droit comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé 7, 16–7. 1131 F Albisinni, ibid., 11.

Separate Status of Trust Assets and Trustee’s Personal Assets 317 to pay commission to a broker, one should look beyond the trust company in whose name a villa had been registered to the individual who was, in substance, the owner.

6. DOES THE CONVENTION REQUIRE A CONTRACTING STATE TO ACKNOWLEDGE THE SEPARATE STATUS OF THE TRUST ASSETS FROM THE TRUSTEE ’ S PERSONAL ASSETS ?

(A) The Issue Curiously, Nygh states that Article 11 “. . . imposes an obligation, particularly on States parties which do not know the institution of the trust such as Italy, to give effect to the separation between beneficial and legal ownership which is peculiar to the trust.”1132 This is incorrect: the Convention does not require a contracting state to introduce the concept of equitable ownership into its legal system, even as a matter of private international law and does not mention legal and equitable title.1133 More fundamentally, it could be said that the preservation in Article 15 of mandatory rules of transfer of title to property and security interests in property1134 means that the Convention does not necessarily even: “introduce the concept of a person having a fiduciary patrimony separate from his private patrimony (so as to be unavailable to claims of the person’s private creditors, spouses or heirs) into a State having no such distinction in its domestic law”.1135

On this view, the Convention “takes away with one hand (Art. 15) what it gives with the other (Art. 11)”.1136 It would follow that recognition of the trust would require only that: “the trustee may sue and be sued in his capacity as trustee and that he may appear or act in this capacity before . . . any person acting in an official capacity.”1137

Thus construed, it would be undeniably correct to regard the Hague Convention as “Much ado about very little”.1138 It is an interpretation of the Convention which both fundamentally restricts its impact and yet ought to make its ratification by non-trust states a relatively sweet pill to swallow. However, such an interpretation, which would not recognise the separate status of the trust assets from the trustee’s private assets, might be thought to 1132

P Nygh, Conflict of Laws in Australia, 6th edn., 521. “It is unlikely that the Convention . . . wished to introduce dual ownership”: K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 28. 1134 Art. 15(1)(d) (on which, see below). 1135 D Hayton, “The Netherlands Implementation of the Hague Trusts Convention”, (1996) 5 J Int Corp P 127, 128. 1136 M Koppenol-Laforce, Het Haagse Trustverdrag, 276. 1137 Art. 11(2). 1138 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5. 1133

318 The Recognition of Trusts undermine the whole nature of the English-model trust.1139 Article 2 stresses that the separate nature of the assets is a fundamental characteristic of a trust; a view strongly reinforced by Article 11, which also stresses the protection of the trust assets from the claims of the trustee’s personal creditors. Article 15 should not derogate from these central characteristics, without which “the trust, as developed in courts of equity in common law jurisdictions”1140 is stripped of its very essence.1141 A far better construction is that, “Article 15 applies to any beneficial interest that a beneficiary may have as a part of his private patrimony (e.g. as life tenant or remainderman) but not to the fiduciary patrimony of the trustee”.1142

In other words, Article 15 preserves the application of mandatory rules in relation to matters of transfer of property and security interests in property insofar as they affect the beneficiary’s interest under the trust. They have no application to the trustee’s interest.1143 It follows that Article 15 does not detract from the separate status of the trust assets from the trustee’s personal wealth and that this separation of the trust assets is required by the Convention.1144

1139 M Koppenol-Laforce, Het Haagse Trustverdrag, 276 prefers the interpretation that holds that recognition of the separate nature of the trust fund is required by the Convention and that the Convention bridges the gap between trust and non-trust states. Thus viewed, “ratification of the convention will smooth the path for an introduction of a trustlike device of national origin, which will even more stimulate the conciliation of the two legal systems.” D Waters, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118, 122 comments that “. . . the essence of a trust is an asset or assets that are dedicated to the sole benefit of a ‘beneficiary’ or the sole furtherance of a purpose.” 1140 Preamble to the Convention. 1141 The separate status of the trust assets “is an essential element of a trust, without which its recognition would have no meaning”: von Overbeck, para. 108, p. 394. 1142 D Hayton, “The Netherlands Implementation of the Hague Trusts Convention”, (1996) 5 J Int Corp P 127, 128. Hayton notes that there is an exception to this principle in English law. Where a trust is created to defraud creditors, creditors will be able to claim the trust assets themselves under the Insolvency Act 1986: (1987) 36 ICLQ 260, 275. 1143 Unless, of course, the trustee is also a beneficiary of the fund. Art. 15 might also be used to determine the proprietary effects of a transfer by the settlor to the trustee of assets in order to defeat the claim of creditors or heirs (at least if this is not considered to be a “rocket-launching” matter). See D Hayton, ibid., (1996) 5 J Int Corp P 127, 129. See also the discussion of Art. 4, above and section 17 of Part One of this book. 1144 For the likely interpretation of Art. 15(1)(d) and (e) were Switzerland to ratify the Convention, see A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6, 9. A recently commissioned report for the Swiss government advocates the implementation of the Hague Convention and the widespread extension of the fiducie, which would recognise the separation of a person’s personal and fiduciary patrimony. Against this backdrop, it seems highly unlikely that the Swiss would adopt a narrow approach to the scope of the Hague Convention. See further, L Thévenoz, Trusts in Switzerland. Ratification of the Hague Convention on Trusts and Codification of Fiduciary Transfers; see also D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. The Luxembourg Bill to implement the Hague Convention, introduced on 14 December 2000, also recognises the separation of a person’s personal and fiduciary patrimony (the text is available at http://www.cc.lu).

Separate Status of Trust Assets and Trustee’s Personal Assets 319 (B) Interpretation in Italy This latter interpretation has been adopted in Italy and in the Netherlands. In Italy, the Hague Convention was simply incorporated into domestic law.1145 Some writers have certainly not seen the Convention as having a minor impact: “Viewed in the context of Italian domestic law, it becomes clear how revolutionary the application of the Convention will be. It does three things. It introduces into the Italian conflict of laws rules a new category, the trust, notwithstanding that there is no corresponding internal law. It provides the Italian judge with a choice of law rule so that he may identify which foreign law should be applied to the particular trust. It sets down a minimum recognition requirement (a kind of ‘minimum uniform law’)1146 for the purposes of the application of the trust to a fact situation in Italy”.1147

Subsequently, trusts which are objectively connected solely with Italy have been created by subjecting them to the law of a trust state and are becoming ever more commonplace.1148 Great interest in the trust has been demonstrated by the legislature, practitioners and academics alike.1149

(C) The Netherlands1150 In the Netherlands, the Wet Conflichtenrecht Trusts 1995 went further.1151 Article 84, section (3) of Book 3, Dutch Civil Code lays down a general rule that

1145 The ratification text can be found in M Lupoi, Trust Laws of the World (Rome, ETI, 1996), 26. All of the key provisions of the Hague Convention are enacted; see also P Messina, “Trusts: the Italian Approach” in N Vogt (ed.) Disputes Involving Trusts, 295. One writer has commented that “Its relatively rapid approval may, in part, be attributed to the apparent lack of understanding of its potential impact and the absence of serious consideration of any provisions for its implementation with respect to tax laws”: F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, 19. See also F Albisinni and R Gambino, “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 80. 1146 In fact, Art. 11(3) is not, strictly speaking, a rule of uniform law. It states the consequences of trust recognition “in so far as the law applicable to the trust requires or provides . . .” (Emphasis added). Art. 11(2) may, however, be construed as a rule of uniform law, as it is not subjected to the law applicable to the trust. 1147 A Paton and R Grosso, (1994) 43 ICLQ 654, 658. See also L Hessey, “When in Rome . . .” (2000) 82 In-House Lawyer 31. 1148 See the discussion of “‘international’ trusts” in relation to Art. 6, above. 1149 See M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–1999, ch. 4, 46–9 for a summary of developments in Italy. See also P Matthews, “Un Trust per l’Italiano” (1998) 12 Trust Law International 104; P Matthews, “New Draft Italian Trust Law” (2000) 14 Trust Law International 33. 1150 On the reception of the trust in the Netherlands, see further K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 24–5. 1151 On whether the Dutch ratification might provide guidance for the Swiss on whether and how to ratify the Convention, see A von Overbeck, “National Digest for Switzerland”, in J Glasson (ed.) International Trust Laws, ch. A52, 5–6.

320 The Recognition of Trusts upon a transfer of property from A to B, B becomes absolute owner of the property.1152 In particular, Article 84, section 3(3) provides: “A juridical act which is intended to transfer property for purposes of security or which does not have the purpose of bringing the property into the patrimony of the acquirer, after transfer, does not constitute valid title for transfer of that property”.1153

The 1995 implementing legislation makes clear that clause 84(3) should not impede application of the Convention by taking the rather radical step of effectively excluding Article 15(1)(d) and (e), Hague Convention altogether. Article 4 of the Wet Conflichtenrecht Trusts 1995 provides that Dutch laws on transfer of property, security interests and creditor protection shall not prevent the recognition of a trust otherwise satisfying the Hague Convention’s requirements.1154 It follows that conflicts rules of the forum which restrict the transfer to the trustee of fiduciary patrimony or bare legal title, or which allow creditors of the trustee to claim assets held by him on trust, may not be applied. Nor could the application of clause 84(3) be preserved by arguing that it relates to the vesting of the property in the hands of the trustee on trust and is a preliminary issue that is excluded from the Convention by Article 4. “If this were allowed no foreign trust of Dutch assets could ever be created and there would be no scope for clause 4 of the 1995 Act to apply to Dutch assets”.1155

Of course, the express recognition by a contracting state of the concept of a separate fiduciary patrimony raises the question of when assets received by a trustee should be regarded as part of that patrimony, as opposed to his private patrimony. Hayton points out that states such as the Netherlands may require legislation to make clear that the trust survives the trustee’s death or insolvency and that nothing vests in a deceased trustee’s heirs upon his death.1156 They will also 1152 D Hayton, “The Netherlands Implementation of the Hague Trusts Convention”, (1996) 5 J Int Corp P 127, 129. This is subject to a provision in the case of transfers into bank accounts to be held for the settlor’s benefit, where “the chose in action is not regarded as owned by B upon B’s insolvency but is available only for A and not for B’s creditors generally.” (ibid.). See further M Koppenol-Laforce, Het Haagse Trustverdrag, 271–2. 1153 M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” 137, 141. The authors quote with concurrence the statement of H van Mens, in F Sonneveldt and H van Mens (eds.) The Trust—Bridge or Abyss between Common and Civil Law Jurisdictions (Deventer, Kluwer, 1992) 53–4 that this provision is “very disappointing because it is to be considered an alienation between civil-law and common-law systems at a time when the open economy of our global village requires unification.” 1154 M Koppenol-Laforce and R Kottenhagen, ibid., 144. See also M Koppenol-Laforce, Het Haagse Trustverdrag, 271: “The explanatory memorandum to the WCT does not provide a satisfying explanation for the fact that in 1992, at the time the new Dutch Civil Code was introduced, art. 3:84, s. 3 stood in the way of ratification of the Trusts Convention, but that this was no longer the case with ratification in 1996. The transfer to the trustee has been saved, which looks rather off if the trustee is domiciled in the Netherlands.” 1155 D Hayton, (1996) 5 J Int Corp P 127, 129. See also M Koppenol-Laforce and R Kottenhagen, ibid., 144–5. The authors also cite with approval a similar statement from D Hayton, “Developing the Use of Trusts in the Netherlands” in 128 Weekblad voor Privaatrecht, Notariat en Registratie 1997, no. 6281, 544. Note that M Koppenol-Laforce and R Kottenhagen do not regard the Dutch legislation as conferring upon a beneficiary a property right binding upon third parties: ibid., 145. 1156 Unless, of course, he is also a beneficiary. See D Hayton ibid., 129.

Following and Tracing Trust Assets: Article 11(3)(d) 321 need to deal with acquisitions of property by the trustee and, in particular, with property improperly acquired by the trustee, to which beneficiaries may wish to show priority over the claims of the trustee’s own creditors.1157 In some ways, the effective exclusion of Article 15(1)(d) and (e) might be thought a step too far, since it might be construed as preventing the application of Dutch mandatory rules of property and insolvency even in respect of claims to the beneficiary’s interest. Imagine that A transfers property to B to be held on trust for C. A states that the trust is to be governed by Utopian law. Subsequently, C becomes insolvent. It would appear to follow that questions concerning the right of creditors to claim C’s entitlement under the trust would be determined by Utopian law and that the mandatory rules designated by Articles 15(1)(d) and (e) would not be applicable in the Netherlands. Such an approach takes settlor autonomy to its very limits and gives to the chosen law a greater impact than it would have even in a trust state.1158

7. FOLLOWING AND TRACING THE TRUST ASSETS : ARTICLE 11(3)( D )

(A) The Dominance of the Law Governing the Trust This is not the place for a detailed study of the role of tracing in private international law.1159 Our present concern is only with the following and tracing of assets or their value which originate from a Hague Convention trust. Suffice it to say here that, in principle, the tracing of assets or their value and the recovery of them, is to be treated as a consequence of recognition of the trust, insofar as the law governing the trust so provides.1160 The provision covers both recovery of the trust property itself and the mixing of trust property with the trustee’s own property.

1157 As a matter of English law, such improperly acquired property forms part of the trust fund. However, Hayton points out that in South Africa only lawfully acquired property is regarded as part of the fiduciary patrimony: ibid., 130. 1158 It is true that Art. 15(1)(f) preserves the application of mandatory rules protecting third parties acting in good faith. But given that the Netherlands has not given effect to Art. 15(1)(e), which deals with creditor protection, it would seem peculiar for it to preserve the application of mandatory rules of insolvency by applying Art. 15(1)(f) to creditors “acting in good faith.” 1159 See G Panagopoulos, “Cross Border Tracing; Conflict of Laws Issues” (1998) 6 RLR 73; G Panagopoulos, Restitution in Private International Law, 94. It may be doubted whether Panagopoulos affords sufficient consideration to Art. 11(3)(d) in his discussion. 1160 A provision which “greatly alarmed civilian delegate and absolutely terrified the representative of the Bank for International Settlements . . .”: M Lupoi, Trusts: A Comparative Study, 355. It is somewhat curious that “the Convention has refrained from describing the legal position of the beneficiaries, while recognising their right to trace the trust assets in the hands of the trustee”: K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 31.

322 The Recognition of Trusts (B) Third Parties and the Residual Role of the Conflicts Rules Designated by the Law of the Forum That said, Article 11(3)(d) is not, in reality, as far reaching as it may first appear. It mentions expressly the mixing of trust assets with the trustee’s own property; it does not expressly deal with the mixing of the trust assets with a third party’s assets. Although it does also deal with the alienation of trust assets (presumably covering the following of them into the hands of a third party), it does not make clear what should happen if those trust assets are subsequently mixed by a third party with his own assets, or if a clean substitution for the trust property is made.1161 More fundamentally still, where the property or its value is to be recovered from a third party holder of the assets, that third party’s rights and obligations “shall remain subject to the law determined by the choice of law rules of the forum.” It should be noted that this provision extends significantly further than the provisions of Article 15, since it is not limited to the application of mandatory rules of whichever law governs the third party’s property rights.1162 In formulating the protection contained in the second sentence of Article 11(3)(d), the drafters of the Convention had in mind specifically claims to recover trust property from banks, although the provision is not so limited.1163 Nor is the provision expressly confined to purchases by third parties.1164 Hence it might appear possible for a contracting state to recognise a trust governed by Utopian law, by which law trust property or its value may be recovered from a third party, but to decree that the assets may nonetheless not be recovered, on the basis that, according to its property choice of law rules, title to the trust property has passed to the third party by purchase or mixing. However, von Overbeck has subsequently commented that: “the meaning of [the final sentence of Article 11(3)(d)] has to be ascertained from its history rather than from its wording . . . It refers not to third persons to whom the trustee has transferred assets in breach of trust, but to third persons, in a contractual relationship with the trustee, who hold assets of the trust”.1165 1161 The recognition of the clean substitution of trust assets is not expressly provided for, but it would fatally undermine the Convention if recognition was not required in this case. Moreover, M Koppenol-Laforce, Het Haagse Trustverdrag, 269 concludes that “Most countries recognise in their own law the possibility of substitution if it concerns a fund with a specific purpose. Therefore, recognition will not be a problem, although it is not completely clear where the basis for this can be found in the Convention.” 1162 See the proposals of the Bank for International Settlements, Preliminary Document No 10, No 6; von Overbeck, para. 115, p. 395. 1163 See further von Overbeck, ibid., paras. 113–7, pp. 395–6. 1164 The United Kingdom had submitted a proposal to make clear that Art. 11(3)(d) did not affect the acquisition of trust assets by third parties (Working Document No 47, von Overbeck, para. 116, pp. 395–6).This was not adopted. Curiously, it was objected that it would be improper to refer to acquisitions here, which were dealt with by Art. 15. However, that latter provision concerns only the mandatory rules concerning transfers of property. 1165 A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention”, (1996) 2 Trusts and Trustees 6, 8. See also the remark of von

Following and Tracing Trust Assets: Article 11(3)(d) 323 If this were correct, it would very much limit the scope of “the choice of law rules of the forum” in this area. However, it is difficult to see why a third party purchaser is not a “holder” of the assets, regardless of whether he is in a contractual relationship with the trustee. It may be, for example, that the trust property has been misappropriated by another party and then sold to the third party, who purchased it in good faith. It not obvious why, from the third party’s perspective, the question whether he has acquired good title in that scenario should not be governed by ordinary property choice of law rules, rather than by the law applicable to the trust, especially where he has no actual or constructive notice of the existence of the trust. To limit his protection to the mandatory rules of the law governing property transfers pursuant to Article 15(1)(d), or the protection of third parties acting in good faith under Article 15(1)(f), is likely to defeat the third party’s expectations.1166 It is suggested that it would be better to interpret the final sentence of Article 11(3)(d) as preserving the conflict rules of the forum relating to any third party purchasers of trust assets, regardless of whether they were in a contractual relationship with the trustee.

(C) Tracing Trust Assets or their Value into a Mixed Fund Held by a Third Party That said, there is a strong argument that the final sentence of Article 11(3)(d) should only be invoked in England where the issue concerns acquisition by the third party and whether he, as a bona fide purchaser, takes free of a subsisting equitable interest. It would not be relevant to the question whether the trust assets or their value could be traced into the third party’s hands, which is determined under the first sentence of Article 11(3)(d) by the law applicable to the trust. It is very far from obvious that the question whether trust assets or their value can be traced into a mixed fund in the hands of a third party should be subjected to a different law to that which governs the trust. This doubt is reinforced by the natural wording of Article 11(3)(d) second sentence, which refers to the “rights and obligations” of third party holders. Yet, “In truth, tracing is a process of identifying assets . . . It tells us nothing about the legal or equitable rights to the Overbeck “Trusts: the Hague Convention and the Law of the Netherlands”, at http://wwwisdc.ch/e/vO97.asp, that the provision “refers only to third persons in a contractual relationship with the trustee who hold assets of the trust.” If this is so, it is extremely unfortunate that the point was not made clear in the second sentence of Art. 11(3)(d). 1166 But see contra, M Koppenol-Laforce, Het Haagse Trustverdrag, 268: “In my opinion, many commentators give an incorrect interpretation of art 11, s.3, sub d, second sentence, when they say that it also covers the transfer of trust property to a third person by the trustee acting in breach of trust. The Convention makes it completely clear that the position of the third party is covered by art. 15, s.1, sub f.” However, it is not at all self-evident that, insofar as there is a tension between Art. 11(3)(d), second sentence, and Art. 15(1)(f), the latter should prevail. Indeed, Art. 15(1)(f) is to some extent a residuary provision, applying as it does “in other respects” to the protection of third parties acting in good faith (emphasis added).

324 The Recognition of Trusts assets traced”.1167 Accordingly, it may be that the second sentence of Article 11(3)(d) has no application to the “mere” process of identifying value in a mixed fund held by a third party. At the same time, it would be most unfortunate to classify the tracing process as procedural and to subject it to the law of the forum.1168 Although tracing is not determinative of the claimant’s rights, it is a pre-condition to establishing some form of entitlement, be it personal or proprietary, to value received by the defendant. In that sense, it is a step in determining the rights of the parties, not simply in enforcing those rights. Furthermore, it is very clear from the first sentence of Article 11(3)(d) that, in principle, the ability to trace value into a mixed fund is treated for Convention purposes as substantive, and referred to the governing law of the trust. Accordingly, where the Hague Convention is applicable, the process of identifying trust property or its value should be a matter for the law applicable to the trust.

(D) The Position of Third Party Purchasers of Trust Property Acting in Good Faith Accordingly, it is submitted that, in England at least, the only matter which should “remain subject to the law determined by the choice of law rules of the forum” is whether a third party purchaser has taken free of the beneficiary’s interest under the trust (even if the law applicable to the trust states that trust assets or their value survive in the third party’s hands). From the third party’s perspective, a purchase of trust property is an “ordinary” property acquisition and the lex situs should determine whether he takes free of the beneficial interest under the trust. Panagopoulos describes this as an “ancillary proprietary issue”.1169 The defence, that D is a bona fide purchaser, is a matter which attracts its own choice of law rule, irrespective of the law which governs the claimant’s action and ability to trace. The application of the lex situs to determine whether a third party purchaser in good faith takes free of the interest of the claimant is very much supported by 1167 Per Lord Steyn in Foskett v. McKeown [2001] AC 102, 113. See also Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd [1981] Ch 105; El Ajou v. Dollar Land Holdings plc, [1993] 3 All ER 717 reversed (but not on this issue) [1994] 2 All ER 685; Arab Monetary Fund v. Hashim [1996] 1 Lloyd’s Rep 589; A-G (UK) v. Heinemann Publishers Australia Pty Ltd (1988) 165 CLR 30; Thahir v. Pertamina [1993] 1 SLR 735. See further R White “Equitable Obligations in Private International Law: the Choice of Law” (1986) 11 Sydney LR 92, 109–10; J Bird, “Choice of Law” in F Rose (ed.) Restitution and the Conflict of Laws, ch. 3, 86; R Stevens, “The Choice of Law Rules of Restitutionary Obligations”, ibid.. ch. 5, 186; L Smith, The Law of Tracing (Oxford, Oxford University Press, 1997), esp. 275–7. 1168 See G Panagopoulos, Restitution in Private International Law, 96–7. Nor does a procedural classification sit easily with Goulding J’s view in Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd, ibid., 127, who appeared to favour a substantive approach to determining whether the defendant held on constructive trust, observing that “the plaintiff’s equitable interest was given . . . by a rule of substantive law and not as the mere result of a remedial or procedural rule.” 1169 G Panagopoulos, “Cross Border Tracing: Conflict of Laws Issues” (1998) 6 RLR 73, 81–2.

Following and Tracing Trust Assets: Article 11(3)(d) 325 Macmillan Inc v. Bishopsgate Investment Trust Plc (no 3).1170 The case arose from the collapse of the Maxwell business empire. The claimant, a subsidiary of one of the Maxwell group of companies, was a Delaware corporation which owned shares in a New York company. These shares were transferred to the Bishopsgate Investment Trust as nominee and were deposited in the Depository Trust Co in New York. Bishopsgate then used them (without the claimant’s knowledge) as security for financial assistance granted by certain defendant lenders to the Maxwell companies. The security was created by transfer through the Depository Trust Co to the lenders in New York, or by deposit of share certificates in England. All the shares were ultimately registered in New York. The claimant alleged that the defendants held the shares on constructive trust for it. Under New York law, the claim would fail (because the defendants would be considered bona fide purchasers for value without notice); however, by English law it would succeed, because the defendant would be considered to have had constructive notice of the claimant’s claim. The Court of Appeal applied the lex situs, New York law. In doing so, it rejected the view that the claimant’s action should be treated as restitutionary. The only live issue was of who had the better title to the shares. That, in turn, depended solely on whether the defendants took good title to the property by acting in good faith. The question whether the defendants had taken title was a proprietary matter; as such, it should be resolved by the lex situs. It is important to stress that the only question for the lex situs will be whether the purchase in good faith gives the third party good title. The lex situs should certainly not be used by a third party simply to impugn the trust itself and to argue that the trust interest does not exist, on the basis that the lex situs does not know the trust concept. That would drive a coach and horses through the Convention;1171 the trust “rocket” should not be brought back down to earth by the vesting of “trust” assets in a non-trust jurisdiction. In short, there should be a two-stage process to determine the ability to recover trust assets or their value from the hands of a third party under the Convention: 1172 1170 [1996] 1 WLR 387 (on which, see A Briggs, “From Complexity to Anticlimax: Restitution and Choice of Law” (1996) 4 RLR 88; J Stevens, “Restitution or Property? Priority and Title to Shares in the Conflict of Laws” (1996) 59 MLR 741; R Stevens, “The Law Applicable to Priority in Shares” (1996) 112 LQR 198; J Bird, “Choice of Law Rule of Priority Disputes in Relation to Shares” [1996] LMCLQ 57; W Swadling, “A Claim in Restitution?” [1996] LMCLQ 63; C Forsyth, “Characterisation Revisited: an Essay in the Theory and Practice of the English Conflict of Laws” (1998) 114 LQR 141). 1171 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 14–15 takes an example of an unauthorised transfer by a trustee of trust assets to his personal creditors in a jurisdiction which does not recognise the separateness of the trust fund from his own private assets as a situation where “the trust could be successfully emptied.” This should not happen, however, since the example concerns the impugning of the trust itself and not an argument that the creditor acquired title free of the trust on the basis of his own acquisition in good faith. 1172 On the law applicable to personal claims against third parties to the trust, see the discussion of Art. 8(2)(g), above.

326 The Recognition of Trusts (i) The law governing the trust determines whether the assets or their value can be followed or traced into the hands of the third party. If they may not be, then they cannot be recovered from the third party; (ii) If they can, the lex situs shall determine whether a third party purchaser acting in good faith has taken free of the beneficial interest under the trust. If he has, the assets or their value cannot be recovered; if he has not, then the assets or their value can be recovered.1173

(E) Consequences in the United Kingdom It should be remembered that Article 11(3)(d) deals only with tracing and following of assets of a trust validly created and falling within the scope of the Hague Convention. Hence it would cover the tracing of assets of an express, written trust which were allegedly received by a third party, X, who it is claimed now holds that property on constructive trust. However, where the initial trust pursuant to which the claim was brought falls outside the scope of the Convention, e.g. if was created by oral agreement not evidenced in writing, the Convention has nothing to say on the following or tracing of the value of that trust asset. However, the United Kingdom has extended the scope of the Hague Convention in section 1(2), Recognition of Trusts Act 1987 to a much broader range of trusts.1174 It follows that the following and tracing provisions of the Hague Convention are equally applicable to such trusts. Accordingly, the role of the law applicable to the trust will be a dominant one in the tracing process in the United Kingdom.1175

(F) Constructive Trusts and Third Parties; Proprietary Effects of the Convention; Application in the Netherlands Opinions may differ as to whether the ability to recover trust property or its value from a third party is an essential characteristic of the trust. Waters argues that it is an integral feature of the trust that “the beneficiary can recover the trust property from a knowing third party when the trustee has wrongly made away the property to that person”.1176 However, he concedes that “civil law systems would be unlikely to permit any in specie recovery of wrongly withheld or alien1173

See further Art. 15(1)(d), below. See the discussion of “the statutory extension of the Convention rules in the United Kingdom” in relation to Art. 3, above. 1175 Whether Panagopoulos would agree is not easy to say. He makes curiously little reference to the Hague Convention in discussing the appropriate choice of law rules for tracing. 1176 D Waters, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118, 120. 1174

Following and Tracing Trust Assets: Article 11(3)(d) 327 ated trust property”.1177 Lupoi rejects the ability to trace and recover trust property as a constituent ingredient of the trust. In civil law systems, the remedy against a third party recipient will normally be personal only.1178 But as Waters convincingly argues: “If the law requires that a trust fund be segregated from the trustee’s other assets, and then permits a wrongful alienation of any asset by the trustee either to himself or to a third party, e.g. a personal creditor, to terminate the legal consequences of segregation, the most essential element in the trust concept, so far as that asset is concerned, will be missing when the beneficiary seeks relief.”1179

Whatever one’s views on this matter, however, the important question for present purposes is whether the Hague Convention treats the right to proprietary recovery as an essential ingredient of its “trust.” More specifically, does it require civil law states to recognise the constructive trust which arises as a matter of English law where the property is in the hands of a third party?1180 It would seem that Article 11(3)(d) covers the constructive trust imposed where trust property is sought to be recovered from a third party, since otherwise the ability to recover trust property would mean little in common law systems.1181 “Thus, if a trustee disposes of trust assets to a third party, not a bona fide purchaser for value, who thereby necessarily becomes a trustee (though not intending to), then there is still a trust for the purposes of the Convention”.1182

Despite this, Koppenol-Laforce and Kottenhagen argue that, in the Netherlands, the beneficiary may be confined to a personal claim in respect of trust property situated within that state. Whether Dutch law would permit an assertion of proprietary rights against a third person is harder to say: “An argument against this characterisation would be that the Trusts Convention does not have the intention of introducing the trust concept into non-trust countries. An argument in support of this characterisation would be that it does more justice to the position of the beneficiary”.1183

1177

Ibid., 120. M Lupoi, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–1999, 35 (esp. 42–3). 1179 D Waters, (1999) 53 Bulletin for International Fiscal Documentation 118, 126. 1180 Not being a bona fide purchaser. 1181 See the discussion of “constructive trusts” in the section “which types of trust are within the Convention?” in relation to Art. 3, above. 1182 P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7, 12. 1183 M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” 137, 148. They point out that, even within the Netherlands, there is a difference of opinion as to whether Dutch property law (and the protection that it affords to transferees of property) is compromised by the Wet Conflictenrecht Trusts 1995 (esp. Art. 4 thereof). See also M Koppenol-Laforce, Het Haagse Trustverdrag, 273–5. 1178

328 The Recognition of Trusts They conclude that: “Foreign writers are unanimously of the opinion that the proprietary effects of the breach of trust will not be recognised in a non-trust country”.1184 They go on to state that: “In our opinion the question whether the third party has a defence against recovery by the beneficiary will be governed by Dutch law, as the law of the place where the asset was situated at the time of transfer”.1185

Whilst this author would not presume to comment on the finer points of Dutch law and would accord considerable weight to the views of Dutch scholars on the matter,1186 it may be observed that, if the proprietary effects of the trust are not recognised, this should be on the basis of any relevant Dutch mandatory rules on the protection of third parties who purchase property in good faith.1187 It should not be based on the non-recognition of the concept of a constructive trust over property in the hands of any third party, since it is a key principle of the Hague Convention that contracting states should not refuse to recognise a trust on the basis that they do not know the trust (or a category thereof) in their legal system;1188 and the principle that a trust may have proprietary effects on third parties, enshrined in the first sentence of Article 11(3)(d), means that it should not refuse to recognise a trust simply on the basis that it does not know the category of the constructive trust either.1189 Nevertheless, Koppenol-Laforce contends that giving effect to the constructive trust will not be practically possible in the Netherlands, at least in respect of Dutch property: 1184 M Koppenol-Laforce and R Kottenhagen, ibid., 150—a statement which sits rather curiously with the authors’ prior discussion of the proprietary rights of a beneficiary. Compare the Luxembourg Court of Appeal’s decision in case no 15430 ABN—Amro Bank Luxembourg SA v. Trustees of the C (Jersey) Foundation (27 April 1994) (noted by P Kinsch “Trusts in a Civil Law Environment: Jersey Trustees may Sue in Luxembourg” (1996) 5 J Int Corp P 120, 122) which “makes it clear that the lex situs rule of the private international law of property cannot be used by third parties who have dealt with trustees to contest, in the event of proceedings being brought, the trustees’ right to sue.” However, the case under discussion was concerned only with the capacity of trustees to sue in Luxembourg. The author goes on to suggest (at 123) that “tracing remedies available to the beneficiary of an English trust would not be enforced . . . if they conflicted with the provisions of the lex situs.” 1185 M Koppenol-Laforce and R Kottenhagen, ibid., 148. See also H Verhagen and N Faber, “A Trace of Chase Manhattan in the Netherlands?” (1998) 6 RLR 165, discussing the unreported Dutch decision in Ontvanger v. Hamm qq. 1186 The matter is also complicated by the interpretation of the Dutch Wet Conflichtenrecht Trusts, and especially Art. 84(3) thereof (see the discussion of the Netherlands implementation of the Hague Convention in the section “does the Convention require a contracting state to acknowledge the separate status of the trust assets from the trustee’s personal assets?”, above). See further M Koppenol-Laforce, Het Haagse Trustverdrag, 270–6, who observes (at 271) that, in enacting Art. 84(3) “In my opinion the legislator has not gone so far that the beneficiary has been given some sort of proprietary right, which is valid against third persons, as in trust law.” 1187 If the proprietary effects are recognised, some rules will be required in the Netherlands on the degree of enquiry expected by third parties into the trustee’s powers of disposition: ibid., 277. 1188 Unless the governing law of that trust is a state which does not know the trust or the category in question (Art. 5), or the conditions for exercising the discretion in Art. 13 are met. 1189 But M Koppenol-Laforce, Het Haagse Trustverdrag, 273 concludes that “The view that the beneficiary is left with a personal right if the trustee transfers, in breach of trust, assets situated in a non-trust country, seems, however, the most realistic one.”

Following and Tracing Trust Assets: Article 11(3)(d) 329 “Regardless whether one is of the opinion that Article 11 forces a country to recognise the constructive trust, the fact remains that such a trust cannot be imposed or established by a court in a non-trust country for property that is situated in a non-trust country. The recognition of a decision originating from a foreign court whereby a constructive trust has been imposed for property situated in a non-trust country will be denied because of mandatory rules of local property law, which can be applied as a consequence of Article 15. . . .”1190

It is not evident that the second part of this quotation is correct. True, the Convention does not introduce concepts such as the constructive trust into the domestic law of states which do not presently have them. Equally, however, Koppenol-Laforce acknowledges that “Contracting States did not have the freedom to deny recognition of the trust on the grounds of general principles of their law of property”.1191 Arguably, if a non-trust state routinely refuses to recognise a constructive trust imposed by a common law court in a transnational context over Dutch property, that is exactly what it is doing. A far more convincing argument to support Koppenol-Laforce’s position is that states which have ratified Article 13 need not recognise a trust having its closest connection1192 “with States which do not have . . . the category of trust involved.” However, this does not excuse a state such as the Netherlands from recognising constructive trusts having their closest connection to a state which does have this category of trusts, even if the trust assets are located in the Netherlands. On balance, this author would contend that the constructive trust imposed on a third party holder of trust assets should (a) fall within the Convention and (b) in principle be recognised in non-trust states; (c) subject only to the discretion of the courts of states which have enacted Article 13 to refuse to recognise such trusts where the conditions of that Article are met. It would be undesirable for Article 15 to be invoked to deny the effectiveness of an entire category of trusts and contrary to the intention of Article 11(3)(d), first sentence.

(G) Reflection on Article 11(3)(d) The principle of Article 11(3)(d) should be supported, provided that it is properly confined. Although Schoenblum criticises the second sentence thereof for rendering the trust unduly vulnerable,1193 one can hardly expect a third party who purchases property in good faith in a given situs, believing the transfer’s validity to be determined by the lex situs, to welcome being informed that he 1190

Ibid., 268. Ibid., 268. 1192 Except for the choice of applicable law, the place of administration and the habitual residence of the trustee: Art. 13 (on which, see below). 1193 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 15. He later comments (at 22) that, “If investments by trusts in the United States can no longer be protected from third country claims, the trustees will eventually find other stable markets in which to invest.” 1191

330 The Recognition of Trusts holds on constructive trust, on the basis that the law applicable to the trust so states. This would place an intolerable burden on purchasers and dramatically affect the alienability of property. Hayton rightly points out that: “. . . it is an axiomatic principle of private international law that the rights and obligations of a transferee are subject to the law determined by the choice of law rules of the forum and such law will normally be the lex situs”.1194

8. “ TRANSLATING ” THE TRUST INTO CIVIL LAW SYSTEMS 1195

(A) Scope of Discussion Non-trust states may experience great practical difficulty, or at least need to show considerable ingenuity, in giving effect to a trust within the jurisdiction. Most civil law states have traditionally had no category of trusts in domestic or in private international law. They have sometimes attempted to “translate” the trust and its effects into the closest domestic equivalent.1196 To compare the trust with functionally similar institutions in non-trust states and to consider how it has been, and might be, received into such legal systems would merit a book in its own right.1197 It is not proposed to attempt a detailed study here,1198 not least because the Hague Convention now requires contracting states not to “translate” the trust, but to recognise it in its own right. It must, however, be observed that attempts to accommodate the trust have frequently been insightful, imaginative and effective in civil law states and that English lawyers may sometimes tend to underestimate the understanding of the trust in those states (and the degree of consequential irritation that such a view can create overseas).1199 1194 See D Hayton, “The Hague Convention on Trusts: a Little is Better than Nothing but Why so Little?” (1994) J Int Corp P 23, 26. 1195 See also the discussion of “Italian courts and the separate fund”, above. 1196 A survey of civil law responses to the trust can be found in A Dyer and H van Loon, Report on Trusts and Analogous Institutions, paras. 147–78, 77–92 and in K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23. 1197 And is an exercise in comparative law much more than one in private international law. 1198 See e.g. C de Wulf, The Trust and Corresponding Institutions in the Civil Law; D Hayton (ed.), Modern International Developments in Trust Law; D Hayton (ed.) Principles of European Trust Law; M Koppenol-Laforce, “The Trust, the Hague Trusts Convention and Civil Law Countries: a Mission Impossible?” (1998) 3 Notarius International 27; M Lupoi, Trusts: A Comparative Study; F Sonneveldt and H van Mens, The Trust—Bridge or Abyss Between Common and Civil Law Jurisdictions?; D Waters, The Trust in Civil Law Jurisdictions—the Dutch Experience (1999) 7 J Int Corp P 131; D Waters, “The Concept Called ‘the Trust’” (1999) 53 Bulletin for International Fiscal Documentation, 118; W Wilson (ed.), Trusts and Trust-Like Devices (United Kingdom Comparative Law Series, Vol. 5). A very useful, short summary by D Hayton of attempts to translate the trust into civil law jurisdictions may be found in J Glasson (ed.) International Trust Laws, ch. C3, 17–23. 1199 See e.g. M Lupoi, Trusts: A Comparative Study, 157, f/n 304, who is particularly critical of a passage in J Morris and P North, Cases and Materials on Private International Law (London, Butterworths, 1984), 96 where the authors remark (in the context of the domicile of trusts for the purposes of Art. 5(6), Brussels Convention) that “. . . continental lawyers seem to think that a trust is some kind of unincorporated association: and it seemed best to go along with them when it did no obvious harm.”

“Translating” the Trust into Civil Law Systems 331 (B) Possible Approaches Hayton comments that: “A bare trust may be regarded as creating an agency or mandate. A fixed trust may be regarded as a type of contract, or, if concerning local assets, may be regarded by the lex situs either as conferring ownership on the life tenant subject to an obligation to pass ownership on to the remainderman or as conferring ownership forthwith on the remainderman subject to personal obligations owed to the life tenant. A discretionary trust should normally be regarded as a contract. A charitable trust may be regarded as a contract or, possibly, as if it were a foundation or Stiftung with separate personality. An inter vivos trust where the settlor in his lifetime retains extensive powers of appointment or revocation may be regarded as a testamentary trust with the trustees being exécuteurs testamentaires. In testamentary trusts the trustees will normally be regarded as special exécuteurs testamentaires with more powers than those of executors or of mandataires. Where the issue concerns powers of trustees the courts of the lex situs of the relevant assets favour facilitating matters by holding that the trustees have the necessary ownership powers to enable a transaction to be effected in the course of the trustees’ management of the assets”.1200

(C) Contractual Approaches The contract is frequently used to give effect in some form to the trust.1201 The trust can be understood as an agreement to confer benefits on third parties. A celebrated example of the translation of the trust is the Swiss Federal Tribunal’s decision in Harrison v. Credit Suisse.1202 An American settlor transferred assets inter vivos to Credit Suisse on trust to provide an annuity for his first wife and 1200 D Hayton, “International Recognition of Trusts”, in J Glasson (ed.) International Trust Laws, ch. C3, 17–8; and see A Dyer and H van Loon, Report on Trusts and Analogous Institutions, paras. 148–9, 78. F Albisinni and R Gambino, “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 76 state, that “Critical comparisons, however, demonstrate that the apparent similarities between the trust and these instruments such as foundation, contracts of ‘mandato’, contracts with third parties and usufruct, are for the most part superficial.” See also F Albisinni and V Sinisi “Trusts from the Italian Perspective” in Trusts for Europe (Legal Studies and Services Ltd, 1988), 53. 1201 P Mayer, Droit International Privé, 4th edn. (Paris, Montchrestien, 1991), 447, (para. 712) gives the trust as an example of “. . . un type de contrat spécifique d’un système juridique déterminé” (Emphasis in original). 1202 ATF 96.II.79 JT 197.I.329, discussed by A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 150, p. 80; E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 315–6; C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé, 7, 10–1; A Burnand, “Laws of Succession and Testamentary Dispositions—Conflicts of Laws and Recognition of Foreign Trust— a Swiss Overview” (1995) 1(10) Trusts and Trustees 15, 17. See also A von Overbeck, “National Digest for Switzerland” in J Glasson (ed.) International Trust Laws, ch. 52, 4; P Supino and A Limburg, “A Swiss Perspective on Trusts” in A Kaplan (ed.) Trusts in Prime Jurisdictions, ch. 20, 383, pp. 394–6; K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, 21–2. The contract approach was taken much more recently in France by the Cour de Cassation: judgment of 20 February 1996, [1996] Dalloz-Sirey Jurisprudence 390.

332 The Recognition of Trusts their children for life. He died domiciled in New York and his second wife, who was sole legatee under his will, contested the inter vivos transfer in Switzerland. The Swiss Federal Tribunal considered which institution would best give effect to the trust and decided that it shared characteristics of property law and gifts, agency and the stipulation pour autrui. Essentially, it was a complex form of contract.1203 The court found that its governing law was Swiss, as the characteristic performance1204 of the trustee’s obligations was to be effected in Switzerland.1205

(D) The Trust as a Company Even this approach, progressive as it seemed at the time, is now considered “outdated” in Switzerland.1206 A new Swiss Private International Law Statute came into force on 1 January 1989 (the “Loi Federale sur le Droit International Privé”). On 1 February 1994, in OD-Bank in Liquidation v. Bankrupt’s Estate WKR1207 the District Court of Zurich applied that statute in the context of a purported trust to be governed by the law of Guernsey. Although it ruled that the instant arrangement was a sham because the “settlor” acted as if the assets continued to be owned by it absolutely,1208 the wider importance of the case is 1203 For another contractual approach to the trust, see also the Paris Court of Appeal decision of 10 January 1970 in Courtois v. De Ganay [1971] Revue Critique de Droit International Privé 518, where “Assimilant abusivement le trust à un contrat, les juges . . . ont naturellement emprunté au droit international privé des obligations contractuelles, les facteurs de contact”: L Maerten, “Le Régime International du Trust après la Convention de La Haye du 1er juillet 1985” [1988] La Semaine Juridique 3319. See also D Hayton, “International Recognition of Trusts”, in J Glasson (ed.) International Trust Laws ch. C3, p. 18; C Reymond (1991) LXVIII Revue de Droit International et de Droit Comparé 7, 10; A Dyer and H van Loon, ibid., paras. 152, 155, 79–82. In non-contracting states to the Hague Convention, it will continue to be necessary to determine a law applicable to the “trust”: C Deneuville, “French Attitude Towards Trusts”, in R Atherton (ed.) International Encyclopaedia of Estate and Trust Law: Selected Papers 1997–1999, ch. 2, 15, 17. Deneuville also remarks more generally (at 20) that “the trust has been and still is today held in suspicion [in France]”. 1204 A concept now enshrined in Art. 4(2), Rome Convention on choice of law in contract. On its application in the trust context, see A Dyer and H van Loon, ibid., para. 157, 82–3. 1205 Where a trust is treated as functionally equivalent to a contract, the choice of law rules contained in the Rome Convention will not determine the applicable law even in contracting states to that Convention, since Art. 1(2)(g), excludes “the constitution of trusts and the relationship between settlors, trustees and beneficiaries” from the Rome Convention. 1206 E Paltzer, “Trusts in Switzerland: some Recent Decisions of the Swiss Courts”, (2000) 8 J Int Corp P 137, 139; see also D Forbes-Jaeger and E Stormann, “The Trust in Switzerland—Revisited” (2000) 8 J Int Corp P 141 for detailed commentary and extracts (in English) from the case. See further, D Forbes-Jaeger and E Stormann, “The Trust in Switzerland” (1996) 4 J Int Corp P 75; F Guillaume, “ Incompatibilité du Trust avec le Droit Suisse? Un Mythe s’effrite” [2000] Revue Suisse de Droit International et de Droit Européen, 1; P Stibbard, “Swiss Courts Declare Guernsey Trust a Sham” [2001] Private Client Business 111. 1207 Case No FB920075, 98 Blätter für Zürcherische Rechtssprechung (1999) No 52, 225ss; noted E Paltzer, ibid., 137. 1208 This is itself significant, since it demonstrates that a Swiss court is prepared to investigate the validity of a trust by its governing law. “As a matter of prudence, it is therefore recommended that

“Translating” the Trust into Civil Law Systems 333 that the court applied Article 150 of the Private International Law Statute, which states that “All organised associations of persons and all organised economic units shall be considered to be companies within the meaning of this Statute.” The court found that the trust structure was not a mere contractual arrangement but should be treated as a company within the meaning of Article 150 “since its main activity was the administration and holding of participations”.1209 Whilst this analogy to the company will rather surprise common lawyers,1210 the case indicates that the express trust exists in Swiss law for private international law purposes1211 and that its applicable law will be determined, as for other “companies”, by Article 154(1) of the Statute. Accordingly, the law by which the company is organised will govern (even if there is no suggestion that a trust may be created subject to Swiss law). The decision has been warmly received.1212 No longer is it necessary to adopt the Harrison approach of “painful dissection of the common law trust into an array of Swiss legal institutions and acts for the purposes of recognition”, since there has been “a revolution in the reception by Swiss private international law of foreign law”.1213 The reasoning has also been confirmed by the Swiss Federal Court.1214 Nevertheless, a note of caution should be sounded: it may be that the company provisions of the Swiss Statute cannot be applied in their entirety to the trust;1215 and more significantly, it has been pointed out that the trustee’s capacity may be recognised in Switzerland, but that: “his powers will be those of a fiduciaire in the Swiss meaning of the word. This means that the trustee will be regarded as the sole and complete owner of the assets”.1216

trustees of foreign law trusts should be aware of circumstances in which a cause of action may arise against the trustees in Switzerland”: P Stibbard, [2001] Private Client Business 111, 113. 1209 E Paltzer, (2000) 8 J Int Corp P 137, 138. See also the Jersey decision in Clothilde Rahman v. Chase Bank (CI) Trust Co Ltd and Others [1991] JLR 103; and the litigation in the Bahamas, Jersey and England in Private Trust Corp v. Grupo Torras [1997/98] 1 OFLR 443. See further W Goodman, “Vulnerability of Trusts to Attack—Retention of Powers and Interests by Settlor” in J Glasson (ed.), International Trust Laws, ch. B8; A Saker and J McKeown, “Reservation of Powers to Settlors” [2001] Private Client Business 154. 1210 A Dyer, “International Recognition of the Trust Concept”, (1996) 2 Trusts and Trustees 5, 10 comments that “Trusts . . . even though they are sometimes referred to as if they were legal entities, are not such in law, but rather consist of a set of legal relationships.” 1211 Albeit not in its own right as a distinct private international law category. 1212 D Forbes-Jaeger and E Stormann, (2000) 8 J Int Corp P 141, 156 describe it as “ a brightline leading case, by its degree of analysis in comparative law and a truly international, open minded approach to the interplay between common and civil law systems.” 1213 Ibid., 158. 1214 Judgment of 3 September 1999: (2000) I Semaine Judiciare, 269; noted E Paltzer, (2000) 8 J Int Corp P 137, 139; D Forbes-Jaeger and E Stormann ibid., 157. 1215 J Thorens, “The Trustee in a Civil Law Country: the Case in Switzerland” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 3, 23, 30. 1216 Ibid., 31.

334 The Recognition of Trusts It also means that the trustee could alienate property absolutely1217 to a third party.1218 (E) Agency and Mandate Elsewhere, sometimes trusts may be treated as a form of agency or mandate. In Faillité Four Seasons Overseas NV v. SA Finimtrust,1219 a settlor purported to transfer title to Eurobonds to a trustee. However, the Luxembourg court treated the arrangement as a mandate under which the trustee held the bonds as agent for the original owner and ruled that it was the latter only who could bring any relevant legal proceedings. However, this somewhat strained mandate analysis would no longer appear necessary in Luxembourg, at least where a trust is governed by the law of a trust state. In the Luxembourg Court of Appeal’s decision in ABN Amro Bank SA v. Trustees of the C (Jersey) Foundation,1220 a settlor of a Jersey trust had purported to pledge assets to the banks in respect of his personal indebtedness. When the banks sought to set off the debt against those assets, the trustees sued the banks in Luxembourg. This raised the question of their locus standi. In view of Four Seasons, the banks argued that the powers of trustees over trust assets situated in Luxembourg had to be compatible with Luxembourg law and that the trustees were mere agents with no title to vindicate. The Court of Appeal ruled that a trust valid by Jersey law could be recognised notwithstanding that the trust did not exist in Luxembourg domestic law and that the trustees could sue. The decision is of limited significance, however, in that “The only effects recognised in the instant cases were the trustees’ right to sue and the founder’s inability to deal with the trust assets without the trustee’s consent”.1221 1217

Without the need to ask whether the third party takes free of the interest of the beneficiary. Certain writers have argued that Switzerland should go on to ratify the Hague Convention: see e.g. C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust, at http://www-isdc.ch/e/Reymond97.asp; A von Overbeck “Trusts: the Hague Convention and the Law of the Netherlands, at http://www-isdc.ch/e/vO97.asp; A von Overbeck, “La Ratification de la Convention de la Haye sur le Trust par les Pays Bas: un Exemple pour la Suisse?” in Collisio Legum, Mélanges offerts à G Broggini. See also P Supino and A Limburg, “A Swiss Perspective on Trusts” in A Kaplan (ed.) Trusts in Prime Jurisdictions, ch. 20, 383, 394. A recently commissioned report for the Swiss government advocates the implementation of the Convention: see L Thévenoz, Trusts in Switzerland. Ratification of the Hague Convention on Trusts and Codifiction of Fiduciary Transfers; see also D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. 1219 Trib d’arrondisesment de Luxembourg, decision of 21 January 1971, [1972] Revue Critique de Droit International Privé 51, discussed by D Hayton in J Glasson (ed.) International Trust Laws, ch. C3 p. 19; C Reymond, (1991) LXVIII Revue de Droit International et de Droit Comparé 7, 9; A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 150, 79. 1220 Case no 15430 (27 April 1994) and case no 17370 Trustees of the C (Jersey) Foundation v. Internationale Nederlanden Bank (Luxembourg) SA (22 May 1996), both discussed by P Kinsch, “Trusts in a Civil Law Environment: Jersey Trustees may Sue in Luxembourg”, (1996) 4 J Int Corp P 120, 121–3. 1221 P Kinsch, ibid., 122. However, of much greater significance is the Bill presented to the Luxembourg Parliament on 14 December 2000 to implement the Hague Convention: see http://cc.lu; see also D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. 1218

The Hague Convention—the Death of Translation 335 (F) Testamentary Trusts and Powers of Revocation Where testamentary trusts are created, Hayton observes that some civil law states may experience difficulty in separating the “rocket-launching” aspects of the trust from the trust itself.1222 They may also encounter problems with inter vivos trusts, where the settlor retains effective control of the trust assets and maintains a power of revocation during his lifetime. Such a trust may be treated as testamentary in nature. The trustee may be regarded as an exécuteur testamentaire and as the owner of the property, or at least as possessing substantive administrative powers over property owned by the beneficiaries.1223 It follows that a settlor is most unlikely to be able successfully to prevent a civil law state applying its forced heirship rules simply by creating a trust of his assets (whether testamentary or inter vivos).1224 9. THE HAGUE CONVENTION — THE DEATH OF TRANSLATION

The Convention offers a partial resolution of the difficulty of translation by introducing the trust into all contracting states for private international law purposes only. The Convention does not introduce the concept of the trust into domestic legal systems; but the trust must be recognised qua trust and not by reinvention in the guise of the nearest civil law analogue. In one sense, this may magnify the difficulties for non-trust contracting states, which must recognise the trust and will have to find a way to give effect to it, rather than tailoring it to fit into their pre-existing domestic legal frameworks.1225 In determining how to do this, it remains likely that such states will be forced to turn to analogues in their domestic law.1226 Gaillard and Trautman’s view that the “the extensive discussions in civil law countries of the appropriate characterisation of trust 1222

D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 21. See D Hayton, ibid., 22 for a consideration of the position in France, Germany, Italy and Switzerland. 1224 “Civil law states are astute to prevent attempts to avoid the application of forced heirship rules”: D Hayton, ibid., 23. He goes on to consider the French decision in Caron v. Odell, 2 March 1985, [1986] Revue Critique de Droit International Privé 66. In that case, a French national became resident in America, so that succession to his movable property would be governed by common law rules (which did not provide for forced heirship). He sought to acquire land in France, but feared that succession to this land would be governed by French law and subject to a forced heirship claim by his daughter. Ingeniously, he sought to make the immovable property movable, so as to avoid the application of French law. He accordingly formed an American company, consisting of shares held on trust for himself and for Odell. In his will, he left none of his estate to his daughter. The Cour de Cassation was suitably unimpressed by this manipulation of choice of law rules and ruled that the land should be available to satisfy his daughter’s claim to a forced share. “However, French notaries consider that the case can be restricted to circumstances where there is a blatant scheme to defraud an heir and that the purchase of an immovable by a company should not be presumed fraudulent especially when a person is merely trying in a fair way to arrange estate planning for the benefit of his family.” (ibid., 23). 1225 See M Lupoi, “Effects of the Hague Convention in a Civil Law Country” (1998) 4 Trusts and Trustees 15. 1226 Or, of course, to homologues thereto (if any). 1223

336 The Recognition of Trusts [sic] will come to an end . . .”1227 may be true for private international law purposes, but it is unlikely to be true when it comes to giving effect to the trust in some meaningful way in practice in a state which simply does not have the trust. One is bound to agree with Schoenblum that, at least in some states, “Even a good faith effort to carry out the purposes of the Convention is likely to result in a hybrid construct that more closely resembles civil law institutions than the trust”.1228

There remain serious questions as to: “how the trustee is to function day-to-day in a civil law jurisdiction, where individuals, entities and governmental agencies have virtually no experience or guidance as to how to deal with a trust”.1229

However, this may be a necessary price to pay for the introduction of the trust into the private international law of non-trust states. Doubtless, they will find ways to accommodate the trust qua trust; and once they do, the path to introduction of the trust into their domestic legal systems (should they desire this), will be smoothed.

1227 E Gaillard and D Trautman, “Trusts in Non-Trust Countries: Conflict of Laws and the Hague Convention on Trusts” (1987) 35 Am J Comp Law 307, 329. 1228 J Schoenblum, “The Hague Convention in Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 13. 1229 Ibid. However, it is suggested that J Schoenblum is wrong to criticise the Convention for failing to provide “ a detailed uniform law . . . which could have been adopted on a unilateral basis and which would have assured a far greater unity of outcome.” That is not the function of Hague private international law conventions. It is also such a monumental task that it is arguably more sensible to start with the more modest private international law aims of the Hague Trusts Convention.

ARTICLE 12—REGISTRATION OF THE TRUST Where the trustee desires to register assets, movable or immovable, or documents of title to them, he shall be entitled, in so far as this is not prohibited by or inconsistent with the law of the State where registration is sought, to do so in his capacity as trustee or in such other way that the existence of the trust is disclosed.

1. REGISTRATION PERMISSIBLE

Article 12 is primarily concerned with the difficulty of reflecting on the public register of non-trust states the existence of a trust. It also applies to “private registers having a public function, such as a shareholders’ register”.1230 Where a trust is entitled to recognition, the trustee may ask for the trust to appear on the register, or, if this cannot be done, for it otherwise to be reflected thereon.1231 He will need to produce any relevant documentation and comply with the requisite formalities of the place of purported registration.1232

2. PROHIBITION OF REGISTRATION

Somewhat enigmatically, Article 12 states that registration may be refused to the extent that it is prohibited by or inconsistent within the law of the state addressed.1233 Clearly, registration will be prohibited if the trust is not entitled to recognition under Article 11.1234 This will equally be the case if the trust infringes Article 13, on the basis that it has its closest connection with a nontrust state. However, if the trust is, in principle, entitled to recognition, it would be contrary to the spirit of the Convention for a contracting state to refuse registration simply on the basis that it does not know the concept of the trust in its domestic law. It follows that registration should be refused only to the extent that it infringes the mandatory rules identified in Articles 15 and 16 or is manifestly incompatible with the public policy of the forum pursuant to Article 18.

1230

Von Overbeck Report, para. 121, p. 396. Ibid., para. 119, p. 396. 1232 Ibid., para. 119, p. 396. 1233 Lewin on Trusts, 17th edn., 295 gives the example in English law of beneficial ownership of shares, whose appearance on the share register is precluded by s. 360, Companies Act 1985. See also M Lupoi, Trusts: A Comparative Study, 356 1234 On the problems which might be experienced in civil law states from registration of the trust, see A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 177, p. 91. 1231

338 Registration of the Trust One might wonder then why it was found necessary to include a specific reference to prohibition on registration in Article 12.1235 The Netherlands has not included this clause in its Wet Conflichtenrecht Trusts 1995, Article 3, which otherwise restates Article 12 of the Convention. There is much to be said for this approach in non-trust states. As Lupoi points out, Article 11 requires states to recognise that trust assets are safe from the trustee’s creditors. “How could systems where registration is widespread permit this result without at the same time at least permitting—and I am not saying compelling—the forms of registration to which they are accustomed?”1236

Moreover, he argues that where the state of registration has mandatory rules1237 which normally require rights in a type of asset to be registered in order to bind third parties, refusal to register a trust of such assets would fatally undermine the trust’s recognition. He therefore argues that civil law states should strongly encourage, or even require, that trusts are registered. One can readily sympathise with the view: non-recognition in a civil law state of the effects of the trust on third parties is, generally speaking, significantly less attractive for the settlor and the beneficiary than recognition with the sacrifice of secrecy as to the existence and location of the beneficial interest. 3. ENGLISH REACTION

The principle of registration of a Hague Convention trust in non-trust states might seem rather curious to an English lawyer, given that the details of the trust do not normally appear on the register in common law states. Instead, the trustee appear on the register in England.1237a This ensures the protection of the beneficiary’s anonymity. In the case of immovable property, a purchaser who pays the purchase money to two or more trustees will overreach the beneficiary’s interest and become absolute owner of the property.1238 Indeed, excluding the details of the trust from the register is considered positively desirable as a way of facilitating the alienability of land. So integral a part of the English land law system is it that it is unlikely that a trustee of a trust governed by a foreign law, which fulfils the recognition conditions of Article 11, would be permitted to register the details of the trust on the English land register. In other words, the non-registration of trusts would, in England, be treated as a mandatory 1235 The von Overbeck Report, para. 119, p. 396 simply states without elucidation that it was “preferred” to make direct reference to the prohibition on registration; see also Report of the Special Commissioners, Nos. 96–104. 1236 M Lupoi, Trusts: A Comparative Study, 357. 1237 Which can be applied under Art. 15 or 16. 1237a Where registered land is concerned, the trust does not appear independently on the register, but may be registered as a minor interest against the legal estate affected by the trust. However, this will act only as a notification to a purchaser of the trust’s existence and the details of the beneficial interest will not appear. The general principle stated in s. 74, Land Registration Act 1925 is that “. . . trusts shall, so far as possible, be excluded from the register. 1238 Ss. 2 and 27, Law of Property Act 1925.

Registration in Italy 339 rule.1239 Similarly, it may not be stated on the share register that shares are owned by a trustee.1240 It is no surprise then that English reaction to the proposed registration of trusts was one of concern. The Law Society of England and Wales said so much in its observations to the Hague Conference and remarked of the provision that it did: “not understand its purpose. It is a fundamental principle of English law that trusts are ‘veiled’. . . . In fact, it is regarded as desirable or even necessary for the protection of persons dealing with the trustees that they should not be concerned whether he is a trustee or the true owner of assets. If a purchaser were to be obliged to enquire about true ownership and whether a trustee has authority to sell, this would often constitute a serious inconvenience. It would make trust assets less readily marketable than their non-trust equivalents, which would inevitably be reflected in a reduction in their value”.1241

The response must be two-fold: first, that registration may be the only means in certain civil law states of ensuring that the trust binds third parties; and secondly, that Article 12 does not seek to introduce a registration requirement into common law states where the trust is not habitually registered. An English court can simply and legitimately say that routine registration of the trust is inconsistent with the law of England.1242

4. REGISTRATION IN ITALY

In Italy, it appears that registration will be far from straightforward.1243 Albisinni explains1244 that documents evidencing transfers of land, mortgages and attachment of assets must be registered with the Land Registry1245 and that the necessary documentation must itself comply with certain formalities. “With respect to a trust, the notary would be required to verify the powers of the trustee and the validity of his or her nomination in circumstances involving a purchase or sale of trust property. These issues can be particularly complex and are not covered by the Convention nor by the Italian law ratifying the Convention”.1246 1239 So that, if Art. 12 had required registration (which it does not), an English court would have invoked English mandatory rules to prevent this, either qua lex situs under Art. 15 or qua lex fori under Art. 16. 1240 S. 360, Companies Act 1985. 1241 Actes et Documents, 160; see also M Lupoi, Trusts: A Comparative Study, 173. 1242 It need not even rely on the mandatory rule provisions of Arts. 15 and/ or 16, since Art. 12 itself permits the English court to deny registration on the grounds that it is prohibited in the state of registration. 1243 F Albisinni and R Gambino, “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 82: “In Italy, the registration of documents related to real property serves two purposes, to record the transfers and for tax purposes.” 1244 F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, 21–2. 1245 Art. 2643, Italian Civil Code. 1246 F Albisinni, in J Glasson (ed.) International Trust Laws, ch. A51, 21–2.

340 Registration of the Trust In contrast, the position in the Netherlands appears more relaxed in relation to the registration of trusts.1247 It is less clear whether it would suffice to register the existence of the trust, or whether the beneficiaries’ identities must also be disclosed.1248 Albisinni argues the latter view would support Italian tax legislation targeted at the “real” owner of the property. However, it would affect confidentiality of the trust and would be difficult in the event of a trust for a class of persons, or for a purpose.1249 In practice, he argues that it will probably be permissible in Italy to register the name of the trustee only.

5. FACILITY OR OBLIGATION ?

Article 12 is merely intended as a facilitator. The trustee may choose not to take advantage of it and may register as owner of the property1250 if this would facilitate the alienability of the assets “on the basis that foreign purchasers would not want the aggravation of worrying about trustees’ powers of disposition and beneficiaries’ rights”.1251 However, in states such as Italy where domestic mandatory law requires registration, the trustee will be left with little effective choice but to register.

1247 M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law”, 137, 146–7 argue that the trust itself need not even be entered in the Dutch public register for the separateness of trust property to be achieved; nor need the share register state that the owner holds shares on trust. However, “The other side of this coin is that anyone who could have noted the trust in the register, but has not (yet) done so, must accept that a third party acting in good faith is not bound by the factual situation . . . One could think of making a note in the register to the effect that the owner holds the assets as trustee. In this way the spirit of the Dutch system of publication and proprietary rights is upheld.” See also M Koppenol-Laforce, Het Haagse Trustverdrag, 272–3. 1248 For a suggestion that the beneficiaries’ identities would also need registration were Switzerland to ratify the Convention, see C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, at http://www-isdc.ch/e/Reymond97.asp. 1249 F Albisinni, in J Glasson (ed.) International Trust Laws, ch. A51, p. 22. A fortiori if there are beneficiaries not yet in existence. See also J Schoenblum, (1994) 3 J Int Corp P 5, 15. 1250 Or, in the case of several trustees, as co-owner. 1251 D Hayton, (1987) 36 ICLQ 260, 277.

ARTICLE 13—RIGHT TO REFUSE RECOGNITION TO TRUSTS OBJECTIVELY CONNECTED TO NON-TRUST STATES No State shall be bound to recognise a trust the significant elements of which, except for the choice of the applicable law, the place of administration and the habitual residence of the trustee, are more closely connected with States which do not have the institution of the trust or the category of trust involved. 1. THE PURPOSE OF ARTICLE 13 : A NECESSARY COMPROMISE ?

Article 13 is concerned with cases where the settlor has selected the law of a trust state to govern the trust, or has designated that the trust be administered in such a state, or has chosen trustees who are habitually resident in that state, but where the other elements point objectively to the law of one or more non-trust states.1252 A “non-trust” state is one which does not know the institution of the trust in its domestic law.1253 This provision allows a Civilopian judge to refuse to recognise a trust otherwise qualifying for recognition under Article 11.1254 It reinforces the fact that the trust is not introduced by the Convention into non-trust states’ domestic legal systems by preventing backdoor trust creation1255 e.g. by a Civilopian settlor transferring property located in Civilopia to Civilopian trustees to hold for Civilopian beneficiaries subject to a trust stated to be governed by English law. “[I]n this way civilians will become familiar with the express trust, both its modus operandi and the scope of its applicability, without a feeling that the common law trust has invaded their internal law”.1256 1252 Art. 13 refers only to “states” in the plural, but it would be extraordinary if it did not apply when all objective elements pointed to a single non-trust state: see von Overbeck, para. 122, p. 397. 1253 But see M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in P Jackson and D Wilde (eds.) The Reform of Property Law, 222, 224. Lupoi argues that the “shapeless” description of a trust in Art. 2 means that there are very few, if any, non-trust states in the world (see the discussion of “the characteristics considered: the ‘shapeless’ trust” in relation to Art. 2, above). However, a teleological construction of the Convention would suggest that Utopia is a “non-trust” state if it does not allow the trust “as developed in the courts of equity in common law jurisdictions” (Preamble to the Convention) or a homologous or analogous institution thereto, to be governed by Utopian domestic law, even though it recognises the trust for private international law purposes. 1254 Rather than relying on wholesale use of mandatory rules and pubic policy to prevent recognition, Art. 13 was adopted in Italy: see A Paton and R Grosso, “The Hague Convention on the Law Applicable to Trusts and on their Recognition: Implementation in Italy” (1994) 43 ICLQ 654, 655. We shall see below that Italian courts have not, however, readily invoked Art. 13. See also R Lafer and A Siegel, “Trusts of Movables in the Conflict of Laws”, (1961) 36 NYULR 713, 715–6. 1255 See M Lupoi, Trusts: A Comparative Study, 328, who argues that an unfortunate attitude emerged in the Conference whereby it was thought necessary “for civil law states to defend themselves from trusts” (emphasis in original). 1256 D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation—Kollision und Vereinheitlichung—Conflicts and

342 Trusts Objectively Connected to Non-Trust States Were such a trust to be recognised in Civilopia, Civilopian residents would effectively be at liberty to create trusts in purely domestic situations.1257 Yet Schoenblum lambastes Article 13: “No serious-minded and responsible practitioner could allow a civil law client with assets in a civil law jurisdiction to resort to a trust for those assets on the basis of the Convention”.1258 He fears that Article 13 will just force such clients to place their assets in trusts offshore and will make contracting trust states look unattractive in the international marketplace. However, the objection is somewhat overstated. Contracting trust states need not enact Article 13. If they do not, this will not affect the decision of contracting non-trust states on whether to invoke Article 13, as the Convention does not work on a reciprocal basis.1259 A state minded to invoke Article 13 in respect of a trust objectively connected to Civilopia will do so regardless of whether the applicable law thereof is English law or Suntopian1260 law. In non-trust states, some formalised recognition of trusts by Convention must be better than the preexisting situation. Without Article 13, prospects of ratification of the Convention in non-trust states would be substantially reduced.1261

2. THE UNITED KINGDOM

Article 13 was considered to have no useful application in the United Kingdom and is omitted from the schedule to the Recognition of Trusts Act 1987. Indeed, common law states strongly resisted earlier proposals that all contracting states, be they civilian or common law, must refuse to recognise trusts which are objectively connected to a non-trust state.1262 English law seeks to facilitate, not to

Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème Anniversaire, 591, 609. 1257 See also F Sonneveldt in F Sonneveldt and H van Mens (ed.) “The Trust—Bridge or Abyss between Common and Civil Law Jurisdictions?” ch. 1, 16. 1258 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little”, (1994) 3 J Int Corp P 5, 21. 1259 The Convention is open-ended. 1260 To adopt D Hayton’s terminology. 1261 The strong criticism of J Schoenblum, (1994) 3 J Int Corp P 5, 16 that Art. 13 “leaves a gaping hole in the Convention, permitting any state and its courts so inclined to refuse to recognise the trust” is overstated, since Art. 13 circumscribes the circumstances in which it can be invoked and since there is no reason to expect Art. 13 to be invoked arbitrarily by non-trust states. In any event, it is a price well worth paying if it facilitates recognition of the Convention in non-trust states. Some recognition in such states must be better than none; and Schoenblum’s point has absolutely no baring on the question whether a trust state should ratify the Convention, because Art. 13 is not applied reciprocally and recognition is not confined to trusts governed by the law of a contracting state (unless Art. 21 has been invoked). For further defence of Art. 13 from an Italian perspective, see M Lupoi, “Avoiding Creditors and Family Claimants with a Trust: Can It Be Done?” in International Academy of Estate and Trust Law, Papers—Italy, 3.1 (1992) (considered in Schoenblum’s article, 16, f/n 64). 1262 See the proposal of Argentina, Egypt, France, Greece and Spain in Working Document No 33; von Overbeck, paras. 128–9, p. 398.

Application of Article 13 343 restrict, the opportunities to create a trust subject to English law, or for a Civilopian settlor to place assets on trust in England. Nevertheless, the non-enactment of Article 13 may potentially cause considerable complications and have significant ramifications. Lupoi makes the point that the broad scope of Article 2 of the Convention means that an English court might be faced with a number of institutions qualifying for recognition under Article 11 and that: “. . . pure purpose trusts, variant types of trusts, and so on, can now be formed by Englishmen under the law of, say, Belize or Cyprus, even if all the elements of the trust are connected with England”.1263

Of course, an English court could always invoke its public policy1264 to refuse to recognise such trusts; but in the light of the refusal to enact Article 13, that would scarcely seem appropriate as a routine response. Accordingly, the Recognition of Trusts Act 1987 appears to have the surprising effect of authorising an English settlor in an otherwise wholly domestic context to create a trust1265 unknown or unauthorised in English domestic trust law, by the simple expedient of choosing a foreign law to govern it. That would drive a coach and horses through the traditional doctrinal limitations of English trust law.1266 3. APPLICATION OF ARTICLE 13

(A) General The application of Article 13 is discretionary and it should not be applied so as to undermine the approach enshrined in other articles of the Convention.1267 Suggestions that contracting states should be required to refuse recognition to certain trusts1268 met with strong objection by delegates. It was felt that the scope of Article 11 would be severely compromised by such a provision. Indeed, 1263 M Lupoi, “Effects of the Hague Convention in a Civil Law Country” in P Jackson and D Wilde (eds.) The Reform of Property Law 222, 225; see also M Lupoi, Trusts: A Comparative Study, 362. 1264 Under Art. 18. 1265 I.e. a “trust” as that term is understood in Art. 2. 1266 And might also undermine the argument that non-charitable purpose trusts cannot be given effect in English domestic law: see the discussion of “non-charitable purpose trusts” in relation to Art. 9, above. 1267 When deciding how to exercise the discretion, a court “. . . will note that the Convention ignores the notion of ‘foreign’ trusts, that one of its unfailing principles is the freedom of choice of the law of the trust, that the trust with which it deals is the shapeless trust and not the English-model trust, that nowhere does it impose criteria based on the citizenship, residence or domicile of the settlor or the beneficiaries, that its aim is to facilitate recognition of trusts, and, above all, that it harmonizes the effects of each specific trust with the national law, thanks to the basic provision in article 15. Finally, the court will note that every possible objection to trusts with close links to legal systems which do not recognize them as a generally applicable legal structure was raised during the course of the works, but that no proposal to refuse recognition or to limit the freedom of choice of applicable law of the trust was accepted:” M Lupoi, Trusts: A Comparative Study, 366. 1268 Working Document No 33 (proposal of Argentina, Egypt, France, Greece and Spain); supported by Italy (Actes et Documents 291). See also the Dutch proposal (Actes et Documents 213).

344 Trusts Objectively Connected to Non-Trust States the Convention was intended to facilitate recognition, not to circumscribe the situations where it may occur, as Article 14 demonstrates.1269 Article 13 is not intended to be a general means by which non-trust states may avoid recognising trusts.1270 Rather, the unfortunate idea developed that nontrust states may use Article 13 to protect themselves where a settlor improperly (or even immorally) seeks to evade the domestic law of a non-trust jurisdiction by creating a trust subject to the law of a trust state.1271 “The clause will be used above all by judges who think that the situation has been improperly removed from under the application of their own laws”.1272 One is reminded of Lord Wright’s requirement in Vita Foods Products Inc v. Unus Shipping Co Ltd that a choice of law in contract be “bona fide and legal”.1273 In the trusts context, no restriction is placed on the choice of the law of a trust state1274 under the Hague Convention,1275 so that, for example, a settlor may choose English law to govern a trust otherwise entirely connected with the law of Utopia; instead, the consequences of a choice of law are circumscribed, although even then only when the law of closest connection is that of a non-trust state. However, it must be the case that a mere choice of an applicable law which has no objective connection with a trust state is insufficient to trigger the application of Article 13, since there would otherwise have been little point in allowing the settlor the freedom to choose that law under Article 6 in the first place. More generally, it is much more helpful to think in terms of the key question whether a non-trust state wishes to go as far as to recognise trusts in an otherwise essentially internal context, rather than to ask whether there is any degree of mala fides in the trust’s creation. Such bad faith is (a) likely to be missing; and (b) does not seem relevant to the question whether Civilopia wants to admit “domestic” trusts into its legal system.1276 Gaillard and Trautman comment that in some civil law states choice of law is only normally permitted where a transaction is genuinely international in nature.1277 However, such a requirement would not have been appropriate for 1269

See below. M Lupoi, Trusts: A Comparative Study, 360 remarks: “A State which holds that the trust subverts its system of real rights . . . should simply not ratify the Convention. . . . Adherence to the Convention . . . means precisely that trusts do not subvert anything (or, if we prefer, that the level of subversion they produce is acceptable).” 1271 Ibid., 328, 359–60. 1272 Von Overbeck, para. 123, p. 397. 1273 [1939] AC 277, 290. See also United Services Fund v. Richardson Greenshields of Canada Ltd (1987) 40 DLR (4th) 94. 1274 A non-trust state’s law may not be chosen: Art. 5. 1275 In Switzerland, “there is always the risk of ‘fraus legis’ (abuse in the choice of an applicable law). If a trust infringed a ‘forced heirship’ provision of the law applicable to an estate, a Swiss court might consider it void either as a construction meant to avoid forced heirship or as an ‘inter vivos’ gift”: A Burnand, “Laws of Succession and Testamentary Dispositions—Conflicts of Laws and Recognition of Foreign Trust—a Swiss Overview” (1995) 1(10) Trusts and Trustees 15, 17. 1276 M Lupoi, Trusts: A Comparative Study, 328. 1277 E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 319. There is no provision in Art. 13 to deal with what is essentially a domestic trust of a trust state (e.g. England), where the settlor has chosen the law of a different trust state to govern the trust (e.g. New South Wales), because “those States do not seem to feel the need for it” (von Overbeck, para. 124, p. 397). Hence there is 1270

Application of Article 13 345 the trust, since what may begin life as a purely domestic trust may subsequently need to be recognised overseas e.g. if the trustees seek to buy property abroad. Nonetheless, given that the Convention did not seek to introduce the trust into domestic legal systems, it would have been inappropriate to require recognition where a settlor habitually resident in Civilopia transfers property situated in Civilopia to Civilopian trustees to hold for Civilopian beneficiaries and states that the trust shall be governed by English law, since this would effectively blur the boundary between the domestic and private international law of trusts. Moreover, “the reservation for mandatory provisions does not suffice, because the result—if no domestic mandatory rule was offended—could be the use of the trust device, for example by civil-law people, with respect to assets located in a civil-law country, in a setting wholly domestic to that country”.1278

Article 13 goes beyond the purely domestic situation and also covers the case where the trust is entirely connected objectively to more than one non-trust state and has no significant connection to a trust state save one conferred by choice1279 of the settlor. Hence, the concern of Article 13 goes beyond introducing the trust into essentially domestic transactions and extends to ensuring that the settlor has grounds for imposing a trust structure in the first place. In this respect, it goes further than Article 3(3), Rome Convention, which superimposes the mandatory rules of a state other than that chosen to govern the contract, but only where “all the other elements . . . are connected with one country only”.1280 It is not necessary that the non-trust state with which the trust is objectively connected be that where recognition is sought. A judge of one non-trust state might refuse to recognise a trust “as a matter of solidarity with another state”.1281 Moreover, the judge of a trust state could refuse to recognise a trust objectively connected with a non-trust state.1282 The restriction also applies to categories of trust. The meaning of the word “category” was considered in the discussion of Article 5, above. Hayton gives the example of a state which has the equivalent to the charitable purpose trust, but not the private trust for beneficiaries as one where Article 13 might be invoked1283 and it would indeed appear that these will be treated as separate categories. More generally, however, it is most unlikely that different non-trust no general requirement that a trust be sufficiently “international” in nature for it to be recognised. See the discussion of “‘international’ trusts” in relation to Art. 6, above. 1278 E Gaillard and D Trautman, ibid., 319. 1279 I.e. choice of law, place of administration or trustee (although the latter two may not, of course, have been expressly chosen by the settlor). 1280 Emphasis added. 1281 Von Overbeck, para. 122, p. 397. Earlier drafts of the Convention had suggested that only the state of objective connection could refuse recognition; see ibid., paras. 125–7, pp. 397–8. 1282 Although, “it is clear that it is in fact an escape clause in favour of States which do not have trusts”: ibid., para. 123, p. 397. 1283 D Hayton, (1987) 36 ICLQ 260, 274.

346 Trusts Objectively Connected to Non-Trust States states will adopt a consistent approach to the word “category”.1284 There is an obvious risk that, if categories are very narrowly drawn, the courts of the state with which the trust is most closely connected will readily reach the conclusion that it does not have the category in question and may refuse recognition.

(B) “Significant Elements” Exactly what are the “significant elements” to be considered is not specified,1285 save that they do not include the choice of the applicable law, the place of administration and the habitual residence of the trustee.1286 At one stage, a proposal had been submitted by which the relevant factors were expressly stated, so that a trust should not be recognised if the trust assets were primarily located in non-trust states and/or the settlor and beneficiaries were nationals and habitual residents of such states.1287 However, such a proposal risked seriously undermining the positive work of the Convention in facilitating widespread trust recognition. It also risked uncertainty as to what was meant by assets being “primarily located” in a non-trust state and as to whether it was appropriate to attach such significance to the situs, which may well be arbitrary.1288 As to habitual residence and nationality, the English delegation objected that it ought to be possible for a French settlor to create a trust for his family of a holiday residence in England.1289 In the event, the fixed criteria were abandoned. Hayton argues that it will be left to each state to form its own view as to what the “significant elements” are for the purposes of Article 13. However, this is a recipe for inconsistency and uncertainty. Von Overbeck comments that the “habitual residences of or the nationalities of the persons involved or the location of the property” will remain especially important.1290 Presumably, the 1284 It is unlikely that trust states would either, but the concern of Art. 13 is primarily with nonrecognition of trusts in non-trust states. 1285 This is one reason why A Anton (with P Beaumont) Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 645 describes Art. 13 as “singularly unfortunate.” The criticism of the Article as a whole is rather too harsh, however, since its existence removes one important disincentive in non-trust states to ratification of the Convention. Those states might otherwise fear indirect introduction of the trust into their domestic legal systems. 1286 It is accordingly somewhat curious for M Koppenol-Laforce, Het Haagse Trustverdrag, 275 to state that a Dutch court should apply Art. 13 only “if the only trustee has his habitual residence in the Netherlands and if it concerns trust property situated in the Netherlands” (Emphasis added). Art. 13 expressly precludes reference to the habitual residence of the trustee. 1287 See the proposal of Argentina, Egypt, France, Greece and Spain, Working Document No 33; von Overbeck, paras. 128–33, pp. 397–400. It would not have been necessary that both the situs and the nationality and residence of the parties be in a non-trust state; either would have sufficed. Compare the Italian proposal: Working Document No 34. 1288 Von Overbeck, para. 130, p. 399. In the case of immovable property situated in a non-trust state, however, the case for non-recognition is much stronger. (See the discussion of “trusts of land” in relation to Art. 6, above). 1289 Ibid., para. 130, p. 399. 1290 Ibid., para. 122, p. 397.

Application of Article 13 347 “persons involved” for this purpose are the settlor and the beneficiaries.1291 However, it is difficult to see why personal connecting factors of these parties should be considered important, given that they do not feature at all in the list of factors by which the law of closest connection to a trust is determined under Article 7.1292 Rather, one might expect all contracting states to have regard to those factors specified for determining the applicable law in the absence of choice under Article 7 (and whose relevance is not expressly discounted in Article 13),1293 and to weigh them as they appear in the hierarchy of Article 7.1294 Hence it would be the situs of the assets which would be of primary importance, followed by the objects of the trust (and the places where those objects are to be fulfilled). However, it should be borne in mind that the hierarchy in Article 7 is not rigid and that any other relevant factors can also be considered in that context.1295 The importance attributed to the situs ought also to allay fears that may persist in offshore states that a trust created by a Civilopian settlor of Civilopian assets for the benefit of Civilopian beneficiaries might be refused recognition, if the trust assets are transferred to a trust state and subjected to the law of that latter state. In such a case, not all of the “significant elements” are located in Civilopia and recognition should not be refused in a non-trust contracting state.1296 The dominance of the situs under Article 13 would be felt especially where the trust property consisted wholly or partly of land. The arguments in favour of the use of the lex situs for trusts of land were considered above.1297 Those arguments apply a fortiori where the situs does not know the trust concept at all. The difficulties facing a non-trust state in subjecting land within its jurisdiction to a trust make the case for refusal to recognise a trust very strong. Where the trust consists wholly or principally of movable property, the role of the situs may be weaker. Indeed, the Convention’s philosophy would be very much weakened were states habitually to refuse to recognise trusts solely on the 1291 These being the persons specified in the proposal of Argentina, Egypt, France, Greece and Spain (Working Document No 33), mentioned above. 1292 Where, in terms of personal connecting factors, only the place of residence or business of the trustee is mentioned. 1293 So that the place of administration of the trust and the place of residence of the trustee (both listed in Art. 7) could not be considered. 1294 Support for this view can be found in M Lupoi, Trusts: A Comparative Study, 361–2, where the author remarks that “An interpretative connection between article 13 and 7 is, therefore, necessary . . . Given that the first and the third [factors stated in Art. 7] . . . are declared by article 13 to be irrelevant, it must be concluded that the circumstances (‘significant elements’) which may impede recognition of a trust are limited to the second and fourth listed . . . [in Art. 7].” 1295 M Koppenol-Laforce, Het Haagse Trustverdrag, 269 observes that trust states are unlikely to regard a trust as valid in any event if the trustee is domiciled in a non-trust state and the trust is administered from a non-trust state. However, Koppenol-Laforce readily admits that there is no conclusive evidence on this. It is not obvious that such trusts should be refused recognition under the Convention. 1296 See further D Hayton in A Borras, A Bucher, T Struycken and M Verwilghen (eds.), E Pluribus Unum, 121, 126. 1297 See the discussion of “trusts of land” in relation to Art. 6.

348 Trusts Objectively Connected to Non-Trust States basis that assets were situated in a non-trust jurisdiction. Against this, the importance of the situs in matters proprietary cannot be denied and, in common law systems, the lex situs continues to be the dominant choice of law rule for most inter vivos outright property transfers.1298 It is unlikely that the situs will be determinative of whether a state invokes Article 13; equally clearly, it is likely to be a highly relevant factor. Where the “objects” of a trust are to be fulfilled in a non-trust state, difficulties might arise. In the context of Article 7 above, it was argued that the “objects” of the trust is not ordinarily a factor to which great importance should be attached.1299 However, the position may be different in the case of a charitable purpose trust whose terms are to be fulfilled in a non-trust state. So, for example, a trust which seeks to improve the conditions of a religious group1300 in Civilopia might both prove controversial and need the help of the relevant authorities in Civilopia to enforce the trust.1301 At one stage, the draft Article 13 required that all the significant elements of the trust be connected to non-trust states. This was subsequently dropped, so that it appears that the Article may still be invoked even where some minor elements of the trust relate to a trust state.1302

(C) The Time Factor Article 13 is silent as to the time at which the objective factors should be assessed. Von Overbeck’s view is that the relevant time is likely to be that when recognition is sought, rather than when the trust was created.1303 Indeed, the alternative view might lead to abuse. Suppose that the settlor of an essentially domestic transaction in Civilopia temporarily located assets in a trust state at the time of the trust’s creation, with a direction that they be sold and the proceeds remitted to Civilopia. It ought not to be possible for him to argue that the courts of Civilopia must recognise the trust, on the basis that it had substantial connections with a trust state at the time of creation. On the other hand, von Overbeck’s view means that a settlor could not be sure at the time of creation of the trust that it will subsequently be recognised overseas. A trust which had significant connections with a trust state at the time of its creation may cease to be so connected at the time of recognition. The rights, duties and expectations of the parties to the trust risk being severely compromised. Moreover, a trustee would have to make sure that the trust maintained substantial connections with a trust state, notwithstanding that it might be beneficial e.g. to invest those 1298

See further section 7 of Part One of this book. See “the objects of the trust and their place of fulfilment”, above. 1300 Such as that in A-G v. Jewish Colonization Association [1901] 1 KB 123. 1301 See the discussion of “the administration and validity of charitable purpose trusts” in relation to Art. 9, above. 1302 Von Overbeck, para. 133, p. 400. 1303 Ibid., paras. 124 and 126, p. 397. The view is shared by D Hayton, (1987) 36 ICLQ 260, 274. 1299

Application of Article 13 349 assets from time to time in a non-trust state. Furthermore, insofar as there is an inter-relationship between Articles 7 and 13, it will be recalled that the time for assessing the law of closest connection for the purposes of the former is apparently that at which the trust was created.1304 On balance, it is suggested that the better view is that a court of recognition should have regard to the situation at the time of the trust’s creation1305 and consider subsequent developments only if: (i) a party contesting recognition can demonstrate that the settlor acted mala fides at the time of creation by artificially setting the trust up so as to escape the clutches of Article 13; or (ii) the objective connections with a non-trust state are now so strong and enduring that recognition of the trust would create an unacceptable erosion into the domestic law of a non-trust state and into the principle that the Convention does not introduce the trust into the domestic legal systems of such states.1306

(D) Partial Recognition It is not clear whether a court may choose to recognise a trust in part only. Such a response might be desirable e.g. if the settlor and beneficiaries are Civilopian nationals and habitual residents, and the trust property consists partly of assets located in Civilopia and partly of assets in England. Obviously, the part of the trust in question must be logically severable. If it is, one would think that any degree of recognition is better than none and that partial recognition should be permitted. In the above example, the trust would be valid only in respect of the assets located in England.

(E) The Consequences of Non-recognition Nor is it self-evident what should happen where a state does refuse recognition under Article 13. Koppenol-Laforce asks, “Is it the very transfer to the trustee that is not recognised (thus there is no trust), or is it the effects of the trust that are not recognised, so that the property is not a separate mass and the beneficiary lacks any protection whatsoever[?]”1307 Of course, each state’s approach may differ. In principle, however, it is difficult to see how the transfer of legal 1304 It was also argued that, in the case of testamentary trusts, the relevant time should be that at which the will was drawn up, rather than when it came into force. But it was suggested that for neither inter vivos nor testamentary trusts should the law of closest connection be assessed as of the time at which the trust is before the court. See the discussion of “time for assessing the law of closest connection” in relation to Art. 7 above. 1305 Or, in the case of testamentary trusts, at the time when the will was drawn up. 1306 A non-trust state should be slow to reach this conclusion in respect of a trust which had, at the time of its inception, its closest connection with a trust state and should only do so where the connection with the non-trust state is now overwhelming and permanent. 1307 M Koppenol-Laforce, Het Haagse Trustverdrag, 270.

350 Trusts Objectively Connected to Non-Trust States title cannot be recognised, since that is a “rocket-launching” issue separate from the Convention itself.1308 This would suggest that it is only the trust itself that is not recognised. However, for the state of recognition to treat the property as unsegregated and owned absolutely by the transferee will clearly frustrate the basis of the transfer. But a state which refuses to recognise an express trust under Article 13 is hardly likely to view the equitable title as reverting to the settlor on the basis of an automatic resulting trust, if, ex hypothesi, it does not recognise the trust or the separation of ownership. It is suggested that, on balance, non-recognition normally entails treating the transferee as absolute owner of the property, (unless the “rocket-launching” law which regulated the transfer of legal title to him determines that the entire transfer is void; in which case the property remains absolutely with the settlor).1309 This unfortunate result only goes to show that the scope of Article 13 should be kept as narrow as possible.

(F) Application in Italian Courts The Italian delegate to the Conference “did not wish to give Italian citizens a new legal tool with which to reach results achievable under no presently existing institutions”.1310 However, it is notable that Italian courts have not subsequently reached for Article 13 to refuse recognition to trusts governed by the law of a trust state, but with strong objective connections to Italy. In Casani v. Mattei,1311 the Tribunale di Lucca accepted the validity of a testamentary trust created by a testator of dual Italian and American nationality. The trust instrument was duly executed in America, where the testator was resident at the time, on discretionary trust for his daughter and her children. Italian law, which regulated succession to the testator’s estate as a matter of Italian private international law, gave her an absolute entitlement to half of the estate. The daughter did not seek to assert her forced share, however, seeking solely to contest the validity of the trust. The court refused to impugn the trust itself and indicated 1308 Unless the lex situs (or, in the case of testamentary transfers of movables, the law of the deceased’s last domicile) determines that the failure of the trust vitiates the basis of the transfer of legal title. 1309 See Part One of this book. 1310 Actes et Documents, 288; see also M Lupoi, Trusts: A Comparative Study, 328–9, who cogently points out the strength of the Italian delegate’s view, which stresses that Art. 13 should not be regarded as a way of a civil law state protecting itself from subversive, or even immoral, domestic trusts (for there may be absolutely no hint of mala fides on the part of the settlor); rather, the issue is solely one of whether it is desired to introduce the trust concept in the domestic setting. The “defensive” approach to subversive trusts was, however, more prevalent in the Conference proceedings. It was also fuelled by the common lawyers, who Lupoi describes (at 329) as having taken “and kept up a condescending attitude towards the civilians” which assumed that civil law judges would not be able to supervise the trust and that when they had acquired some sophistication in the matter, they might feel less insular and desirous of protection from the trust. (He has in mind, in particular, a statement by the United Kingdom delegate: Actes et Documents 232). 1311 Reported in English in (1998/99) 1 ITELR 925. The case is considered further in the discussion of Art. 15(1)(c), below.

Application of Article 13 351 that any forced heirship claim that the daughter might bring would merely reduce the proportion of the estate which could be left on trust. Commentators have seen the decision as a landmark: “To summarise, the conclusion of the Tribunal is the first and very clear judicial confirmation [that] . . . the Hague Convention, even if not intended to introduce the trust into domestic law of countries who do not have this institution, in fact eliminated the barriers to . . . [its] entrance in Italian law. . . .”1312

Others question whether: “the Convention [has] effectively introduced the creation of trusts (and not merely their recognition) into civil law jurisdictions which ratify the Convention?”1313 Important as the decision is, the answer to that question must be “no”.1314 This was a trust governed by the law of a trust state recognised as valid by an Italian court, albeit pursuant to an Italian will. There is no suggestion that, had the settlor chosen Italian law to govern the trust, the trust would have been upheld; and nothing in the Convention to require that the trust should be upheld in that context. It follows that the decision is quite consistent with the introduction of the trust for private international law purposes only and represents a shrewd, restrained application of Article 15.1315 Such recognition will necessarily impact on related areas of Italian substantive law1316 but it will not itself introduce a purely domestic trust governed by Italian law into Italy.1317 Other cases where trusts governed by a foreign law have been accepted show an almost overwhelming objective connection of the trust to Italy.1318 The Milan Commercial Court1319 approved a trust governed by Jersey law created 1312

F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51,

18. 1313 (1998/99) 1 ITELR 925, 926. See also P Matthews, “Un Trust per l’Italiano” (1998) 12 Trust Law International 104, 105. 1314 A similar view to the present author’s is expressed by G Broggini “Folgen der Ratifikation des Trusts- Übereinkommens in Italien”, at http://www-isdc.ch/e/Broggini97.asp. He points out that although solicitors and notaries will inevitably seek to introduce a trust into Italian law, introduction of a domestic law trust is not a necessary requirement for civil law states which ratify the Convention. 1315 On the question of whether a trust of which an Italian citizen is trustee would be valid in Italy, see F Albisinni, National Digest for Italy”, in J Glasson (ed.) International Trust Laws, ch. A51, 23–4. 1316 However, insofar as those related areas of law contain mandatory rules, it is quite clear from Art. 15 that they shall prevail. F Albisinni, ibid., 24 appears to think that this substantial impact of the trust on related areas of law (insofar as those related areas do not contain mandatory rules) undermines von Overbeck’s assertions as to the Convention’s impact (i.e. that it is limited to introducing the trust for private international law purposes only). This is somewhat curious, since it is inevitable that the recognition of a private international law concept will have an impact on domestic and private international law in related areas such as succession; a trust governed by foreign law, once recognised in Italy, does not sit in splendid isolation and has to interact with other areas of Italian law. 1317 Unless it is said that Italy is a trust state for Convention purposes, (as M Lupoi, (1995) 1(3) Trusts and Trustees 15, 17 suggests), so that even a choice of Italian law to govern a trust entirely connected with Italy could not be struck down under either Art. 5 or Art. 13. See also P Matthews, “Un Trust per l’Italiano” (1998) 12 Trust Law International 104; P Matthews, “New Draft Italian Trust Law” (2000) 14 Trust Law International 33. 1318 M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 46–9.

352 Trusts Objectively Connected to Non-Trust States by an Italian company with an Italian corporate trustee. The trust consisted of shares in a United Kingdom company, but the settlor company had transferred Italian land to that company. The land had been transferred as security for the debenture holders, so that the “settlor, trustee, beneficiaries and the underlying property (though not the actual trust property . . .) were Italian.”1320 The court upheld the trust and did not invoke Article 13. Similarly, the Genoese Commercial Court1321 did not consider invalid a trust governed by Maltese law and with a Maltese trust company, where the settlor, beneficiaries and trust assets were all located in Italy. Matthews points out that since both cases were decided as part of the court’s non-contentious jurisdiction without bilateral argument, they do not have the authoritative status of Casani. Nevertheless, together these cases come very close indeed to permitting a settlor in a wholly domestic situation to create a trust, provided only that it is governed by the law of a trust state, and represent a substantial step towards the introduction of a domestic law of trusts.1322 Could something even more fundamental be read into the cases? The Italian cases often do not seem to indicate a positive decision not to apply Article 13. It may rather be that the Italian courts consider Article 13 to be inapplicable, on the basis that Italy is not a “non-trust” state for Convention purposes, given the broad, “shapeless” description of the trust under Article 2.1323 If that is true,1324 then presumably even a choice of Italian law to govern the trust is not a choice of a state which “does not provide for trusts” under Article 5, so that Italian courts1325 are bound to recognise even purely internal trusts governed by Italian law and the Convention has indeed introduced the trust into a domestic legal system. If Lupoi’s view of the “shapeless” trust of Article 2 is correct, very few states will be non-trust states and the scope for use of Article 13 anywhere will be extremely limited. However, it was argued above1326 that Article 2 should in practice be interpreted so as to be limited to transfers of ownership to a trustee of assets forming a separate fund from his personal wealth.1327 Otherwise, Article 13 is almost stripped of any practical relevance and Article 5, which prohibits the choice of a law of a state which “does not provide for trusts” would scarcely restrict the settlor in his choice of a civilian law to govern a trust. 1319

On 27 December 1996. See further P Matthews, (1998) 12 Trust Law International 104. P Matthews, ibid., 104. Emphasis in original. 1321 On 24 March 1997; see ibid., 104. 1322 And see P Matthews, “New Draft Italian Trust Law” (2000) 14 Trust Law International 33. 1323 M Lupoi argues that Italy is a trust state for Convention purposes: M Lupoi, (1995) 1(3) Trusts and Trustees 15, 17. See the discussion of “the characteristics considered: the ‘shapeless’ trust” in relation to Art. 2, above. 1324 It was argued in the discussion of Art. 2 above that such a broad “shapeless” interpretation should not be adopted in an English court (even if it is literally what Art. 2 seems to require). 1325 And the courts of other contracting states, if they adopt the same interpretation. 1326 See the discussion of Art. 2, above. 1327 One might draw support from the words of the Preamble, which make express reference to the common law trust developed in the courts of equity as “a unique legal institution.” 1320

Application of Article 13 353 Moreover, application of Article 13 is preferable to other, more negative routes of non-recognition which civil law contracting states might otherwise resort to, such as application of the forum’s public policy,1328 which are otherwise likely to be invoked to prevent the creation of essentially common law trusts within civilian states.1329 On balance then, the Italian courts’ approach might be taken as encouragement for other civil law states not to invoke Article 13 too readily where the trust has a strong objective connection to a civil law state (and the law of a trust state is chosen). However, it should not be used to fuel the argument that a trust governed by the law of a civil law state must also be given effect.

1328

Pursuant to Art. 18. H Verhagen, “Trusts in the Civil Law: Making Use of the Experience of ‘Mixed’ Jurisdictions”, (2000) 8 European Review of Private Law 477, 498 states that Lupoi’s views “do not seem to have found much acceptance.” 1329

ARTICLE 14—FREEDOM TO ADOPT MORE LIBERAL RECOGNITION RULES The Convention shall not prevent the application of rules of law more favourable to the recognition of trusts. It would be a retrograde step if only trusts which fell within the Convention’s scope could be recognised.1330 The Convention aims to facilitate the recognition of trusts, especially in non-trust states, and not to circumscribe their recognition. Of course, contracting states might choose to extend the operation of the Convention’s rules, so that recognition of a broader range of trusts would automatically follow.1331 This is, of course, what the United Kingdom has done in section 1(2), Recognition of Trusts Act 1987.1332 However, a court can “recognise” trusts for private international law purposes even if it has not extended the Convention’s application. Closer inspection shows that Article 14 does not add much to the Convention, at least in England. The word “recognition” does not refer to foreign trust judgments;1333 rather, it means that, in practice, an English court will have choice of law rules for categories of trust not covered by the Convention or by its statutory extension, e.g. for many types of constructive trust governed by a foreign law.1334 A trust created in accordance with those English choice of law rules will, axiomatically, be given effect in England.1335 Conversely, one which does not will not be given effect in England.1336

1330 Had proposals been adopted to require states to refuse recognition in certain cases, then Art. 14 would have been eliminated: see von Overbeck, paras. 128–9, p. 135 and the discussion of Art. 13, above. 1331 Art. 20 (on which, see below). 1332 See the discussion of “the statutory extension of the Convention rules in the United Kingdom” in relation to Art. 3, above. 1333 See the discussion of “the ‘problem’ of recognition of foreign trusts judgments” in relation to Art. 3, above. 1334 See the discussion of “constructive trusts” and of “trusts ‘arising’ by judicial decision” in relation to Art. 3, above. Dicey and Morris, 1096 comment that “if constructive trusts are regarded, as seems best, as within the subject of Restitution, Rule 200 will apply to indicate the proper law of the obligation represented by the constructive trust.” Rule 200 is set out at 1485. Rule 200(1) states that “the obligation to restore the benefit of an enrichment obtained at another person’s expense is governed by the proper law of the obligation.” Their view is almost certainly too widely stated, since we saw above (in the discussion of Art. 3 and of its statutory extension in the United Kingdom) that some constructive trusts will be subject to the Hague Convention’s choice of law rules. 1335 Save insofar as it infringes any relevant mandatory rules or public policy under Art. 15, 16 or 18. 1336 A category of trusts not covered by the Hague Convention and not known in a particular contracting state for private international law purposes would not be “recognised” in that contracting state.

ARTICLE 15—PRESERVATION OF MANDATORY RULES IN RELATED AREAS OF LAW CHAPTER IV—GENERAL CLAUSES The Convention does not prevent the application of provisions of the law designated by the conflicts rules of the forum, in so far as those provisions cannot be derogated from by voluntary act, relating in particular to the following matters— (a) the protection of minors and incapable parties; (b) the personal and proprietary effects of marriage; (c) succession rights, testate and intestate, especially the indefeasible shares of spouses and relatives; (d) the transfer of title to property and security interests in property; (e) the protection of creditors in matters of insolvency; (f) the protection, in other respects, of third parties acting in good faith. If recognition of a trust is prevented by application of the preceding paragraph, the court shall try to give effect to the objects of the trust by other means.

1. FUNCTION

This key Article preserves the application of mandatory rules in areas related to, and likely to interact with, the trust. In case of conflict between rules of succession, transfer of property, security, creditor protection, protection of minors etc. on the one hand, and the law of trusts on the other, the Convention gives way, but only insofar as the rules in those related areas are mandatory. Article 15 is concerned not with the mandatory rules of the forum in related areas, but with the mandatory rules of the state whose law is designated by the forum’s choice of law rules as applicable to those areas. Put differently, Article 15 deals with the interaction of a state’s private international law of trusts with its other private international law rules, not with its domestic law.

2. NATURE OF MANDATORY RULES

Mandatory rules are ones which the settlor is not free to derogate from in a given legal system. They are “laws of a strictly positive, imperative nature”1337 protecting a public interest in matters moral and political and in the protection 1337

76.

F von Savigny, A Treatise on the Conflict of Laws (W Guthrie translation, 2nd edn., 1880),

356 Preservation of Mandatory Rules of vulnerable groups.1338 Mandatory rules do not invalidate a choice of law by the parties, but bind them to certain rules of another legal system. They involve an essentially positive process of application, of superimposition onto the governing law; whereas the finding that a foreign law infringes public policy1339 involves a negative statement on the governing law in question and its disapplication in whole or in part. Article 15 speaks of mandatory rules as ones which “cannot be derogated from by voluntary act.” As von Overbeck notes, the terminology may be slightly confusing in this context. He gives the example of forced heirship provisions: they are rules with which a testator must comply; however, the heirs may choose to waive their rights “and it would perhaps have been more exact to say . . . ‘derogated in advance’ ”.1340 3. DOMESTIC AND INTERNATIONAL MANDATORY RULES

In any discussion of mandatory rules, it is important to determine whether a provision concerns a “domestic” mandatory rule or an “international” mandatory rule. In trusts litigation occurring in England, a domestic mandatory rule would be one which could not be derogated from in a purely domestic case1341 where English law governs the trust, such as e.g. the certainty requirements for creation of a trust.1342 However, such rules would not obviously be applied in an English court if the trust was governed by a foreign law which had different certainty rules.1343 By contrast, an international mandatory rule of English law would be applied in English courts whatever the law otherwise applicable to the trust. An example might be the rules on occupation rights of co-owners of land, which may fall to be applied to any land situated within the jurisdiction, even if that land is subjected to a trust governed by a foreign law.1344 In contract, a 1338

See further P Nygh, Autonomy in International Contracts, ch. 9. Under Art. 18. 1340 Von Overbeck, para. 137, pp. 400–1; see also the Austrian proposal, Working Document No 49. 1341 Or one which would be purely domestic were it not for the inclusion of a foreign choice of law clause. 1342 On the position in contract, see P Nygh, Autonomy in International Contracts, 200. 1343 Although they might be to the extent that the trustee, beneficiary and the court need sufficient certainty to enforce the trust and its terms. However, it is not obvious that English rules e.g. on what forms of words constitute sufficient certainty of intention to create a trust need to be superimposed onto a trust governed by a foreign law. To the extent that a foreign law imposes unacceptably liberal or restrictive certainty requirements, its application can always be restricted on public policy grounds under Art. 18. 1344 See ss. 12 and 13, Trusts of Land and Appointment of Trustees Act 1996. See also the discussion of “trusts of land” in relation to Art. 6, above. Another example might be the law against perpetuities, which may be applied to land within the jurisdiction, even if subjected to a foreign law trust. However, this is a weaker example, since the application of the governing law of the trust’s perpetuity rules may or may not be acceptable to English law. Accordingly, it would be better to disapply them only to the extent that they infringe English public policy, pursuant to Art. 18, Hague Convention. (This, of course, assumes that the perpetuity rules are “rocket” matters within the Convention and not “rocket-launching” matters outside the Convention’s scope (on which, see section 8 of Part One of this book, Art. 4, above and Art. 15, below)). 1339

Characterisation, the Trust and Article 15 357 paradigm example of an international mandatory rule is found in the Unfair Contract Terms Act 1977, certain parts of which will, by virtue of section 27(2), be superimposed onto a contract in an English court, even if the parties have chosen a foreign law to govern the contract.1345 The reference in Article 15 is to the broader category of domestic mandatory rules. This can be deduced from the much stronger wording of Article 16, which refers to “rules of that State which must be applied even to international situations, irrespective of rules of conflict of laws.” It is also consistent with Article 15’s purpose, which is to preserve the application of rules of a designated1346 domestic legal system which are applicable to matters of succession, property, bankruptcy etc.1347

4. CHARACTERISATION , THE TRUST AND ARTICLE 15 : IS MUTUAL EXCLUSIVITY IMPOSSIBLE ?

A particular difficulty with the trust is its intimate relationship to other areas of law which are subject to independent choice of law rules, such as those concerning property, marriage and divorce, succession and insolvency. There may be a conflict between the law applicable to the trust and the law applicable to the related area of law on a particular issue, such as e.g. where a testamentary trust is validly created by the law applicable to the trust, but infringes a requirement of the law governing succession that a testator must leave a proportion of his estate to certain relatives. It becomes necessary then to determine a hierarchy of choice of law rules and to determine to what extent, if at all, the law applicable to the trust should “override” the law applicable to succession, or viceversa. One approach would be to attempt to solve the problem through the process of characterisation, so that there is mutual exclusivity between the categories of trust, property, succession, insolvency, matrimonial property etc. It would be necessary to determine what the essence of the issue is. For example, if X disputes the validity of a testamentary trust, on the basis that, as the testator’s husband, he should have been left a certain proportion of his wife’s estate, then it could be said that the primary legal issue is the existence and effect of forced heirship rules and the law governing succession matters should apply, rather than the law governing the trust. On this view, there is no overlap between the private international law of trusts and succession and each applies to matters within their respective scope. 1345 S. 27(2), Unfair Contract Terms Act 1977 commences with the words: “This Act has effect notwithstanding any contract term which applies or purports to apply the law of some country outside the United Kingdom, where. . . .” It should be noted that this provision does not apply where the parties have not chosen a law and the applicable law is determined in the absence of choice. 1346 I.e. designated by the choice of law rules of the forum. 1347 And which would have to be applied if there was no trust.

358 Preservation of Mandatory Rules Such an approach of mutual exclusivity is adopted for jurisdictional purposes when applying the Brussels Convention. A good example is Re Hayward (dec’d).1348 Mr Hayward, a debtor, and Mr Hulse jointly purchased a property in Minorca, which was registered as held by them in equal shares. Subsequently, the debtor became bankrupt and his assets were vested in his trustee in bankruptcy. The debtor died intestate and his widow claimed entitlement to his share in the property, which she then purported to transfer to Hulse. Hulse became sole registered owner in the Minorcan property register. The trustee sought an order entitling him to a half-share in the property and to rectification of the Minorcan land register. Such a claim could be seen as a matter of the law of property, trusts, bankruptcy or succession. Rattee J ruled that the claim “had as its objects rights in rem in immovable property” within the meaning of Article 16(1)(a), Brussels Convention. The claim was not best classified as one in trusts, because the trustee in bankruptcy asserted legal title.1349 It did not concern succession,1350 because the question of whether Hayward’s widow inherited the property had nothing to do with the peculiarities of succession law and everything to do with whether the property had duly vested in the trustee in bankruptcy. Nor did the claim concern bankruptcy, because, having previously been appointed trustee in bankruptcy, the right which the trustee now asserted against the Minorcan property did not depend on powers peculiar to the law of bankruptcy, but was simply an assertion of a property right such as any legal owner might bring.1351 Whether the classification decision reached by Rattee J was correct is not selfevident;1352 however, the more important point for present purposes is that the case illustrates the avoidance of overlap between categories which the Brussels Convention necessitates. At the choice of law stage, if it were similarly possible throughout the process of characterisation to create mutually exclusive categories, disputes should in theory not arise as to which of a number of different laws designated by different choice of law rules should take priority, as they would cover different ground. The idea may be sound, but it lacks pragmatism. Mutual exclusivity works in the Brussels Convention context because almost all key terms have an autonomous definition common to all contracting states. However, at the choice of law stage under the Hague Convention, there is not even an autonomous definition as such of the trust, such is the difficulty in formulating one;1353 certainly, the prospect of autonomous, mutually exclusive definitions being formulated by 1348

[1996] 3 WLR 674. See also Pollard v. Ashurst [2001] 2 WLR 722. That is not to say that a claim by a trustee in bankruptcy is not a trusts matter; it is just to say that, when forced to make a hard choice between a trusts classification and other competing classifications, the trust classification should not prevail. 1350 Both succession and bankruptcy fall outside the Brussels Convention: Art. 1(2). 1351 See also Pollard v. Ashurst [2001] 2 WLR 722. 1352 Discussed further in J Harris, “Rights in Rem and the Brussels Convention”, (1997) 22 EL Rev 179. 1353 There is merely a description of its characteristics in Art. 2. 1349

Characterisation, the Trust and Article 15 359 contracting states of what are matters of trusts, bankruptcy, succession, family law etc. is wholly unrealistic. It might be possible otherwise to preserve mutual exclusivity of trusts and other matters if each state were to characterise the issue according to its own private international law categories, rather than to attempt an autonomous classification common to all contracting states. However, this would mean that different states would almost certainly draw the boundaries between trusts and other areas differently from each other. Nor is it apparent that this approach would work. Characterisation by the law of the forum inevitably risks distorting the foreign law which is to be applied, if no regard is had as to how the issue would be classified by the governing law and as to its intended application in that law. However, on an issue such as e.g. whether a testamentary trust complies with forced heirship rules of a given state, the law applicable to succession will have a legitimate interest in its law being applied to protect the testator’s family; and the law applicable to the trust will have a legitimate interest in being applied to the validity of the trust.1354 There is a natural overlap between the trust’s validity and rules of forced heirship. Ultimately, it is very difficult to see mutually exclusive characterisation working effectively in this area, precisely because the relationship between the trust and other areas of law is so very intimate. For example, if the question arises whether a creditor has a legitimate claim to the trust assets upon the bankruptcy of the beneficiary, then the issue appears at once to be one in which both the law governing the trust and the law governing bankruptcy have a legitimate interest and it is somewhat artificial to say that the issue is either wholly one of trusts or wholly one of bankruptcy. In fact, there is an inter-action of both sets of rules and it would be preferable for a scheme of private international law to accommodate that interaction, rather than to strive for mutual exclusivity. However, as soon as one accepts that the trust overlaps with other areas of law for private international law purposes, the hierarchical question of how much weight to attach to the laws respectively governing each area arises. For the Convention to have required the trust to prevail in its entirety over the law governing succession, property, bankruptcy etc. would have (a) risked seriously undermining the interests and legitimate expectations of third parties; and (b) been too bitter a pill for many would-be contracting states1355 to swallow. It is one thing to introduce the trust into one’s system of private international law; it is quite another to render it all powerful within that system. Effectively, such an approach would treat such related matters as incidental questions which may arise in applying the law applicable to the trust. On the other hand, if the trust had been entirely subordinated to the other relevant laws governing related matters, a great deal of the impact of the Hague Convention would have been lost. The 1354

And may have an interest in “insulating” the trust from attack from forced heirship and other

rules. 1355 Or, at least, many non-trust states. Offshore states would have positively welcomed the protection of their trusts from attack.

360 Preservation of Mandatory Rules result is a compromise in Article 15, in that the law governing the trust will determine most issues which might arise under that trust, but will yield to those rules of law of property, succession, bankruptcy etc. which are mandatory in nature. That is not to say that there will not be difficulties of characterisation, or areas where it is appropriate to apply the law of the related area in its entirety to the exclusion of the law governing the trust. There will be some issues which, although related to the trust are, properly considered, not trust matters at all. One such area has already been considered, namely the provision of Article 11(3)(d) on recovery of trust assets or their value, which provides that the rights and obligations of a third party holder of assets “shall remain subject to [all] the law designated by the choice of law rules of the forum”.1356 It is suggested that the question whether a third party purchaser acting in good faith has taken free of the beneficiary’s interest under a trust, for example, should be regarded as a pure property question. Although it does concern the durability of an interest under a trust, looked at from a purchaser’s point of view it is a question of title, which he might legitimately expect to be subject to the lex situs rule and not to the law governing a trust to which he was not party and of whose existence he may have been wholly ignorant.

5. ARTICLE 15 AND THE “ ROCKET - LAUNCHER ”

It is also true that a number of issues which at first sight would appear to fall within Article 15 will in fact not be caught by the Convention at all. They may rather be “rocket-launching” questions1357 and the law designated by the forum’s conflict rules should be applied to such matters in its entirety. Questions of forced heirship may be better seen as affecting the validity of the transfer to the trustee and hence excluded from the Convention.1358 Rules regulating the transfer of property to the trustee to hold on trust and the constitution of the trust will likewise be excluded. Accordingly, it may be that the application of Article 15 will be rather narrower than might appear at first sight, since the interaction of the trust with genuine “rocket” questions may be limited. Hayton gives the following three examples of matters falling within Article 15: a trust set up shortly prior to divorce or death to defeat a spouse’s rights; a 1356 See “following and tracing the trust assets: Article 11(3)(d)”, above. Compare Art. 15(1)(d), which applies only to mandatory rules, but which (unlike Art. 11(3)(d)), is limited neither to third parties nor to transfers in breach of trust. 1357 Excluded from the Convention by Art. 4. These issues are discussed in detail in Part One of this book. 1358 Although, of course, everything will depend upon the legislation in question and whether it affects the transfer of property, or specifically affects the creation of a trust structure with that property. See the discussion in section 17 of Part One of this book, and the discussion of Art. 4 (above) and of “forced heirship” (below). It may be rare that a rule of forced heirship would render the transfer to the trustee void in its entirety: A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6, 7. The von Overbeck Report favours the view that forced heirship questions fall within Art. 15: para. 54, p. 381.

Obligation to Apply the Mandatory Rules of the Designated Law? 361 trust created to defeat a creditor’s claims to trust assets; and a trust whose applicable law is chosen to avoid English rules against perpetuities applicable to land.1359 The first example looks rather more like a claim that the trust has not been validly created at all, at least where the allegation is that the transfer to the trustee is void. As such, it is difficult to see how it is other than a preliminary matter falling outside the Convention. The same is true of the second example. The third example concerns rules of perpetuity. It is not certain whether these should be regarded as “rocket” or “rocket-launching” matters. In so far as the perpetuity rules in question affect the transfer of assets at all, rather than specifically regulating the creation of the trust structure, they should be regarded as “rocket-launcher” questions.1360 The striking conclusion is that “general” laws which render the transfer of property to be held on trust void ab initio relating to e.g. forced heirship, creditors, perpetuities and the like should be regarded as outside the Convention, because they regulate the alienation of legal title to the transferee and affect “general” principles of property law. The obvious criticism of such an approach is that it takes so much outside the Convention that it reduces the impact of the law applicable to the trust greatly. It also leaves relatively little to be subject to Article 15, whose main application will be to transactions which render the trust itself void or voidable1361 or to the variation of the trust.1362 But insofar as a law prevents the transfer on trust in the first place, it is very hard to see how it can be classified other than as a preliminary matter.

6. MUST A JUDGE APPLY THE MANDATORY RULES OF THE LAW DESIGNATED BY THE CONFLICTS RULES OF THE FORUM ?

The wording of Article 15 is in permissive and not obligatory terms. The Convention “does not prevent” the application of the relevant mandatory 1359

D Hayton, (1987) 36 ICLQ 260, 277. But see the decision of the Supreme Court of New South Wales in Saliba v. Falzon [1998] NSWSC 302, which supports a “rocket” classification of the rule against accumulations of English law (on which, see ss. 164–6, Law of Property Act 1925; see also the Perpetuities and Accumulations Act 1964; see further Law Commission Report No 251 on the Rules against Perpetuities and Excessive Accumulations (1998). The case is considered in the discussion of “common law authorities on implied choice of law” in respect of Art. 6, above. 1361 Hayton’s view, that the difference between claims that a trust is void and that it is voidable may determine whether a matter falls outside the Convention, or within it but subject to Art. 15, is endorsed by M Koppenol-Laforce, Het Haagse Trustverdrag, 270. If the claim is that the relevant law renders the trust voidable, but not void, it is easier to argue that the trust was launched in the first place, and to see any claim to set it aside as falling within Art. 15. However, this author favours the view that the key question is still whether legal title has vested in the trustee. If the claim is that the trust itself is voidable, this is a “rocket” matter within the Convention and is subject to Art. 15; but if the claim is that the transfer of legal title to the trustee is voidable, this is still naturally to be construed as affecting the “ordinary” rules on transfer of property and should still be treated as a “rocket-launching” matter falling outside the scope of the Convention. 1362 Insofar as a state has domestic mandatory rules on the variation of trusts: see Art. 8(2)(h), above. 1360

362 Preservation of Mandatory Rules rules.1363 This might suggest that a judge is not obliged to apply those rules. However, Hayton takes the view that relevant mandatory rules must be applied.1364 Koppenol-Laforce agrees, and argues that a court must apply Article 15 of its own motion and that “In no specific case does the court have the option of not applying art. 15”.1365 In contrast, Gaillard and Trautman state that “the Convention calls for . . . reference, on an optional basis, to the domestic mandatory rules of the jurisdiction dictated by the forum’s choice of law rules”.1366 Section 1(3), Recognition of Trusts Act 1987 might appear to clarify matters in the United Kingdom. It states that: “in accordance with Articles 15 and 16 such provisions of the law as are there mentioned shall, to the extent there specified, apply to the exclusion of other provisions of the Convention”.1367

This seems, at first sight, to impose a duty on an English court. But it may be argued that section 1(3) is inconclusive, because the very question in issue is the extent to which the application of mandatory rules is specified in Article 15. It could be said that section 1(3) merely explains what effects are given to a domestic mandatory rule where the judge elects to invoke it, and makes clear that such a rule will then be superimposed onto the trust.1368 On balance, it is hard to see how an Article which “does not prevent” the application of mandatory rules can be construed to mean that a judge is required to apply such rules. Moreover, since Article 15 is designed to limit the unwanted excesses of applying the law applicable to a trust, it is difficult to see why a state which is happy to apply the law applicable to a trust in its entirety on the facts of a given case should be prevented from so doing. It is the forum which selects the law applicable to related matters of succession, property, insolvency etc. and if it does not wish to apply that law on the facts, that is the forum’s prerogative. Accordingly, it is suggested that the application of domestic mandatory rules under Article 15 should be regarded as discretionary.

1363

Dicey and Morris, 1092 describe Art. 15 as “remarkable for the looseness of its drafting.” D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 14. The same view is expressed by J Schoenblum, (1994) 3 J Int Corp P 5, 18, f/n 74. 1365 M Koppenol-Laforce, Het Haagse Trustverdrag, 270. 1366 (1987) 35 Am J Comp Law 307, 337. Emphasis added. 1367 Emphasis added. One construction of this section is that, in the United Kingdom, Arts. 15 and 16 override all Convention provisions, including Art. 4, so that even “rocket-launching” questions are subjected to the mandatory rule provisions of the Convention. This would, however, be an extraordinary result. In any event, Arts. 15 and 16 apply only “to the extent there specified” and they are specified under the Convention to apply only to matters falling ex hypothesi within the Convention’s scope. There is no indication that s. 1(3), Recognition of Trusts Act 1987 purports to affect the scope of application of Arts. 15 and 16. Accordingly, those Articles should not be applied to “rocket-launching” matters. 1368 This issue is discussed further in the discussion of the variation of trusts in respect of Art. 8(2)(h), above. 1364

Application of Article 15 363

7. APPLICATION OF ARTICLE 15

Although Article 15 applies the mandatory rules of the law designated by the conflicts rules of the forum, von Overbeck remarks that “the intent was to preserve above all the forum’s substantive law in cases where its conflicts rules designated its own law”.1369 In other words, would-be contracting states wanted the freedom to apply their mandatory rules when the applicable law of the related issue and the law of the forum were one and the same. It might be thought that there is no need for such an attitude, given that Article 16 preserves the role of the forum’s mandatory rules. However, that latter Article is concerned only with international mandatory rules of the forum. Article 15 appears to cover the broader category of domestic mandatory rules.1370 The matters mentioned in Article 15 are intended as examples only and mandatory rules designated by the choice of law rules in other related areas may be applied.1371 Such residual flexibility might appear inevitable, given the unpredictability of inserting the trust into non-trust systems. But the flexibility presents a real danger: “Not without reason, it was said that a hostile judge might always find in Article 15 a means of frustrating the trust”.1372 The inclination to do so might be particularly strong for a non-trust state suspicious of the trust concept.1373 However, it must be remembered that Article 15 can be invoked only to the extent that application of the Convention rules, or recognition of a trust, infringes the mandatory rules of the relevant state. It should not be used to undermine altogether a trust conforming to the Convention rules.1374

8. PARTICULAR MATTERS SPECIFIED IN ARTICLE 15

(A) The Protection of Minors and Incapable Parties Article 15(1)(a) deals with rules for the protection of minors and incapable parties. This will often manifest itself in rules on capacity. Its scope of application 1369

Von Overbeck Report, para. 138, p. 401. And, of course, will only apply the mandatory rules of the forum if that law is also the lex causae. 1371 Indeed, some states (including Germany) wanted to apply Art. 15 to a broader range of matters than those expressly stated in the final version of Art. 15 and to incorporate such matters as property rights (as opposed to the more limited proprietary protection ultimately specified) and company law. See further M Lupoi, Trusts: A Comparative Study, 363. 1372 Von Overbeck Report, para. 139, p. 401. 1373 K Williams, “Trusts and the Hague Convention” (1986) 7 Company Lawyer 95 suggests that Art. 15 will probably be invoked less frequently “as ‘non-trust’ states come to understand the trust.” For many civil law states at least, this view may tend to underestimate their ability to deal with the trust and their experience with it: see the discussion of “‘translating’” the trust into civil law systems” in relation to Art. 11, above. 1374 See also M Lupoi, Trusts: A Comparative Study, 365. 1370

364 Preservation of Mandatory Rules may be rather narrow. The capacity of a settlor to create a trust is in any event excluded from the Convention; 1375 and it may be that the capacity of a beneficiary to receive a benefit is also excluded.1376 The capacity to act as a trustee is, however, within the Convention.1377 A state might e.g. designate a personal connecting factor, or the lex situs, to determine the capacity of a party to manage property. The mandatory rules of that state will be superimposed onto the trust.1378

(B) The Personal and Proprietary Effects of Marriage Article 15(1)(b) deals with the effects of marriage on the trust.1379 It is likely to concern especially regimes having a form of community of property and the effects to be given to an ante-nuptial contract.1380 Community of property is a common phenomenon in a number of states.1381 Where there is no contract, an English court will apply the law of the matrimonial domicile1382 to determine the effects of marriage on title to movable property.1383 Where there is a marriage contract, its effect is to be governed by the proper law of the contract.1384 The parties may expressly or impliedly agree on this law;1385 in the absence of choice, the law of the matrimonial domicile will govern.1386 It might appear that the impact of Article 15(1)(b) will be limited. Dicey and Morris comment that “nearly all systems of community of property allow the spouses to contract out of the system if they so desire . . .”1387 If so, such community of property rules are not apparently ones which “cannot be derogated 1375

See Part One of this book (esp. sections 2–4). See Art. 8(2)(a), above and section 6 of Part One of this book. 1377 Art. 8(2)(a). More accurately, the specific capacity to act as a trustee is included; the “general” capacity to receive property at all is excluded. See also section 5 of Part One of this book. 1378 But a state should be slow to reach this conclusion. It is the contention of this author in Part One of this book that “general” rules of property law should be applied only to the “rocket-launching” matter of vesting legal title in the trustee and not to the “rocket” matter of the creation of the trust structure with that property. Accordingly, mandatory rules should be applied under Art. 15(1)(a) only where the restriction on the trustee’s capacity relates specifically to his competence to hold and manage trust property. 1379 See generally Dicey and Morris, ch. 28; Cheshire and North, ch. 34. 1380 See section 20 of Part One of this book. 1381 See the brief survey in Dicey and Morris, 1067–8. They divide the various systems of community of property into three categories: full community (giving the examples of the Netherlands and South Africa), community of gains (Spain) and community of chattels and gains. 1382 Dicey and Morris, Rule 148, 1066. The rule is expressly limited to movable property. On immovables, see Re De Nicols (No 2) [1900] 2 Ch 410. 1383 Where the spouses have different domiciles at the time of marriage, historically the husband’s domicile prevails. However, such a rule has very little to commend it today and Dicey and Morris (at 1069) prefer to apply the “law of the country with which the parties and the marriage have the closest connection . . .” 1384 Ibid., Rule 149, 1073–4. 1385 E.g. Re Hernando (1884) 27 Ch D 284; Re Hewitt’s Settlement [1915] 1 Ch 228. 1386 Duke of Marlborough v. AG (No 1) [1945] Ch 78; Dicey and Morris, Rule 149, 1073–4. 1387 Ibid., 1068. 1376

Particular Matters Specified in Article 15 365 from by voluntary act” and will not be applied under Article 15. However, this is the case only if the spouses bilaterally agree to a different arrangement. If one spouse unilaterally purports to create a trust for the benefit of a third party of property which is subject to community of property, the attempt must fail. Put differently, the law applicable to matrimonial property regimes will almost certainly mandatorily require its application unless the spouses agree a different arrangement between each other. Article 15(1)(b) may be applied to property which was held on trust for a spouse at the time of marriage. The trust will already have been “launched” prior to the marriage, but may be compromised by the mandatory rules of a matrimonial property regime. However, a matrimonial property regime may also prevent one spouse from creating a trust of property acquired during the marriage, at least if that person does not obtain the consent of the other spouse. This is a restriction on the alienation of matrimonial property at all and not on the creation of the trust structure with the property. As such, it should be regarded as a “rocket-launching” matter lying outside the scope of the Convention. It follows that, in this scenario, the matrimonial property rules may be applied in their entirety to the trust, since Article 15(1)(b) will be inapplicable.

(C) Succession Rights, Testate and Intestate, especially the Indefeasible Shares of Spouses and Relatives (i) Forced heirship and the law applicable to the trust Mandatory rules of succession may concern, in particular, forced heirship.1388 Testamentary trusts which are valid by their applicable law might nonetheless infringe rules of the law applicable to succession which require that a proportion of the testator’s estate be left to his immediate family. Kötz takes the examples of Article 25 of the German EGBGB and Article 26 of the Japanese Horei, both of which require the application to matters of succession of the law of the state to which the testator belonged at the time of his death. “In the case of a Japanese national, domiciled in Japan, but resident in the US, who establishes a testamentary trust and subjects it to Californian law, general questions concerning the law of succession would have to be determined by Japanese law, according to articles 26 and 28(3) of the Japanese Law on International Private Law (Horei) whereas trust-related questions would have to be determined, according to Article 6 of the Convention, by Californian law”.1389 1388 Indeed, where rules of forced heirship do exist in a given legal system, they are almost certain to be mandatory in nature. See D Hayton, “Trusts and Forced Heirship Problems”, (1993) 2 J Int Corp P 3. 1389 H Kötz, “The Hague Convention on the Law Applicable to Trusts and their Recognition”, in D Hayton (ed.) Modern International Developments in Trust Law ch. 3, 43.

366 Preservation of Mandatory Rules Accordingly, it becomes necessary to determine whether forced heirship issues should be classified as matters of succession or of trusts. In this author’s view, it should not be for the law applicable to the trust to determine whether forced heirship rules apply to testamentary trusts. Forced heirship provisions should be characterised as rules of succession, not trusts, since they are normally concerned with the alienation of a testator’s estate on death and not specifically with the alienation on trust. Accordingly, it is suggested that the statement in Lewin on Trusts,1390 that “if the applicable law [of the trust] includes forced heirship rights, the English court will enforce them” should be treated with caution, since, in the case of testamentary trusts, the question of forced heirship is not to be classified for private international law purposes as a trusts matter. In the case of inter vivos trusts, the question whether rules of forced heirship are infringed appertains to the validity of an alienation of property by the settlor and should be determined by the lex situs (because it will normally concern the ability to alienate property inter vivos at all, not the question whether property can be alienated inter vivos by trust).1391 If the law applicable to succession decrees that a portion of a testator’s assets may not be freely disposed of by will, an English court will not normally strike down the entire trust. Rather, it will subject the residue to the trust, assuming also that the trust is valid according to the law applicable to the trust.1392 However, it is possible that the scope of Article 15(1)(c) is narrower than might first appear. “The heirs are just as likely, perhaps more likely, to question the launch of the trust—and in that respect the Convention is inapplicable”.1393 It was argued above that provisions of succession law may fall outside the Convention altogether, insofar as they impugn the transfer of assets to the trustee and appertain to the settlor’s ability to alienate his property at all.1394 Pursuant to Article 4, the forum will apply its own choice of law rules to determine whether the settlor can and has alienated his property at all.1395 The law applicable to succession may accordingly fall to be applied in its entirety1396 to forced heirship claims concerning testamentary dispositions.

1390

Lewin on Trusts, 17th edn., 301. The law applicable to succession should not be applied to impugn an inter vivos trust on forced heirship grounds, because the forum will apply the lex successionis only to assets which it regards as forming part of the testator’s estate on death. Where those assets have already been validly dissipated inter vivos, they do not form part of the testator’s estate on death. 1392 Re Annesley [1926] Ch 692. A Duckworth, “An Offshore View of Forced Heirship—Global Conflict and its Planning Implications: Part 2” [1995] Private Client Business 334, 336 argues that common law states are likely to be more sympathetic to claims by a surviving spouse than to those of other relatives. This can only be of relevance if (as argued above), an English judge has an element of discretion whether to apply the lex causae’s mandatory rules under Art. 15. 1393 A Duckworth, in J Glasson (ed.) International Trust Laws, ch. B1, 24. 1394 See Art. 4 above and section 17 of Part One of this book. 1395 As opposed to whether he may create a trust with that property. 1396 And not just its mandatory rules under Art. 15. 1391

Particular Matters Specified in Article 15 367 (ii) Types of forced heirship rules Duckworth argues that there are three types of forced heirship rule:1397 “strict forced heirship” whereby the individual may dispose of only part of his estate;1398 “forced heirship by indivisible shares” whereby the testator has power over all his estate, but certain family members have a minimum entitlement which they can enforce if the testator fails to leave them the requisite amount;1399 and “judicial adjustment”, whereby certain family members may apply to the court, in its discretion, to make some provision for them.1400 The first category of rule seems clearly to affect the validity of the transfer of assets to the trustee and should be considered a “rocket-launching” matter. Equally clearly, category three does not affect the testator’s ability to create the trust structure, so that any relevant mandatory rules of this type should be relied upon under Article 15. However, it is in the nature of this category that it is unlikely to contain mandatory rules. The second category is more uncertain, since it does not prevent the testator from disposing of his assets on trust, provided that he makes adequate provision for family members with that trust. On balance, and in order to give the Convention as wide a scope as possible, such rules may be deemed to fall within the Convention and capable of application by virtue of Article 15. Given that such rules would impose a mandatory obligation on the settlor to provide for family members, the likelihood of them being classed as mandatory rules is high. 1397 A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, 5. A short summary of the types of forced heirship rules existing in a number of jurisdictions is given by Duckworth (at 8–14). The author also provides a very useful discussion of the private international law solutions to forced heirship issues in offshore centres (at 42–55) and elsewhere (at 55–60). See generally D Hayton (ed.), European Succession Laws (Bristol, Jordans, 1998). See also E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., 1021. 1398 Such rules exist in e.g. France, Scotland and Islamic law systems: see A Duckworth, ibid., pp. 8–12. For a fuller discussion of forced heirship in France, see A Duckworth, [1995] Private Client Business 334; M Oldham, “Financial Obligations within the Family—Aspects of Intergenerational Maintenance and Succession in England and France” [2001] CLJ 128, (esp. 152–61). For a fuller discussion of forced heirship in Scotland, see S Mackintosh, “An Onshore View of Forced Heirship— Domestic Conflict and Its Planning Implications” [1996] Private Client Business 161 and 197. Mackintosh illustrates that the Scottish approach is somewhat unusual in conferring forced heirship rights, but giving a settlor almost total freedom of disposition during his lifetime. On the position in Switzerland, see A Burnand, “Laws of Succession and Testamentary Dispositions—Conflicts of Laws and Recognition of Foreign Trust—a Swiss Overview” (1995) 1(10) Trusts and Trustees 15, 16–17. 1399 Germany and a number of the United States of America are given as examples of such systems by A Duckworth in J Glasson (ed.) International Trust Laws, ch. B1, pp. 12–13. 1400 England is, of course, a paradigm example. The Inheritance (Provision for Family and Dependants) Act 1975 gives spouses, children and dependants of the testator the right to apply to the court for assistance, should the testator have failed to make reasonable financial provision for them. It applies only if the deceased died domiciled in England or Wales, but may be invoked even in respect of property situated outside the jurisdiction (see Hamlin v. Hamlin [1985] 2 All ER 1037). Duckworth also gives the examples of Ireland, New Zealand and China: ibid., pp. 13–14. However, D Hayton points out that it will be difficult for a foreign state to apply the lex successionis where that law provides for discretionary judicial adjustment and that the foreign state might even refuse to do so, regarding the discretion as a matter for the home courts of the state having such provision: “Cross Border Estates: Part 2: the Problems of Diversity” [1994] Private Client Business 250, 256–7.

368 Preservation of Mandatory Rules (iii) Claw-back claims and inter vivos trusts The most serious aspect of forced heirship from an English lawyer’s viewpoint is the fear that even an inter vivos trust might fall prey to “claw-back” claims of the law applicable to succession.1401 Even if an English court would be unlikely to invoke such rules, the same can certainly not be said of a judge in a civil law jurisdiction faced with an English trust.1401a Dyer and van Loon comment that: “The influence of the law of succession is so strong that even a living trust may fail because it infringes on mandatory rules governing succession, notably those concerning the reserved share of children”.1402

The trustees might find themselves liable in an overseas jurisdiction beyond the extent of the trust assets which they now hold.1403 However, it is very unlikely that an English court will permit a validly created inter vivos trust subsequently to be impugned by foreign forced heirship rules.1404 Hayton comments that most rules of forced heirship will only affect assets of inter vivos trusts when those assets are situated in the state where the settlor’s forced heir is habitually resident.1405 Where the property remains outside a state having forced heirship rules, such as England, the problem may be solved through the process of characterisation. The only relevant question is whether, at the time when the settlor purports to create an inter vivos trust, the settlor can and has 1401 See further Part One of this book (esp. section 18). Note that ss. 10 and 12, Inheritance (Provision for Family and Dependants) Act 1975 include in the deceased’s estate for the purpose of the exercise of the court’s discretion property alienated by the deceased within six years of his death if made with the intention of defeating claims under the Act. The trustee himself cannot, however, be liable for more than the value of the trust assets remaining in his hands at the date of the claim (s. 13). 1401a Furthermore, it may be that payments made to a person during their lifetime will be taken into account when ascertaining their entitlement under the deceased’s estate. See, for example, the decision of the Royal Court of Guernsey in Funk v. Krombach (2001) 15 Trust Law International 103, where it was held that there was a presumption in Guernsey law that lifetime payments should be considered when determining entitlement under an estate (see also s. 11A, Trusts (Guernsey) Law 1989, as amended). 1402 A Dyer and H van Loon, Report on Trusts and Analogous Institutions, para. 122, 70. Compare Lewin on Trusts, 17th edn., 301 which states that “We understand that such succession rights do not generally (according to the foreign laws that confer them) invalidate or limit lifetime gifts, but at most give rise to claims, usually money claims, at the settlor’s death” with D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3, 3–4 who remarks that, “On the death of a person subject to a lex successionis with a forced heirship regime the general rule (except for Islamic systems) is that the deceased’s actual estate is notionally increased to include gifts made by him during his lifetime, whether without a time limit or within some period . . . before death.” 1403 See also D Hayton, ibid., 4. He points out that the problem may be compounded by the difficulty of foreseeing the law applicable to succession at the time of creation of an inter vivos trust, since (in the case of movables) this will depend on the testator’s domicile at the time of his death. 1404 Even if the arguments considered in the following paragraphs do not avail, an English court would be likely to treat retrospective claw-back as contrary to English public policy under Art. 18. 1405 D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 14. Hayton gives the examples of Holzberg v. Sasson (1986) 75 Revue Critique de Droit International Privé 685 and Caron v. Odell (1986) 75 Revue Critique de Droit International Privé 66. See also C Deneuville, “French Attitude Towards Trusts”, in R Atherton (ed.) International Encyclopaedia of Estate and Trust Law: Selected Papers 1997–1999, ch. 2, 15, 18.

Particular Matters Specified in Article 15 369 alienated his property to the trustee according to the lex situs.1406 If the lex situs states that he cannot, because it must be left for his children, then the transfer is invalid. But if the transfer is valid by the lex situs (and the trust itself is valid by its applicable law), it has validly come into effect and cannot later by impugned on the testator’s death by the law applicable to succession. Further cogent support for the protection of inter vivos trusts from claw-back claims of the law applicable to succession is provided by Lewin on Trusts. The authors points out that an English court will only apply rules of the lex successionis to matters which English law characterises as ones of succession: “But if they [that is, forced heirship rights] are succession rights there are no assets [of a validly created inter vivos trust] to claw back. At the settlor’s death these are no longer in the estate, which alone is subject to rules of succession in English eyes . . . For that reason, it is thought that if any transfer has passed a good title to the trustee as a matter of the property law governing the transaction, and if the trusts are valid by the law designated by the Convention to govern the validity, the English court will not give effect to any right of forced heirs under the settlor’s personal law to claw back any of the settled assets after the settlor’s death”.1407

Put differently, it is not for the law governing succession to determine, in an English court, which assets form part of the settlor’s estate. Assets which are validly disposed of inter vivos according to English private international law will not be subjected to the law applicable to succession. The authors draw support from Pouey v. Hordern,1408 where a claim was brought by the daughter of a French national to a share of assets which had been the subject of a special power of appointment. Farwell J ruled that this appointed property did not form part of the testator’s estate, suggesting that “once the property had gone into a settlement there is nothing in the settlor’s estate for the succession law of the settlor’s nationality or domicile to claw back”.1409 It would appear to follow that an English court will not allow an inter vivos trust validly created in its eyes to be subject to claw-back claims.1410 However, that might be scant comfort if the assets in question are situated in a civil law

1406 “English private international law rules should lead an English court to characterise the transfer of title to property merely as an inter vivos transfer governed by the lex situs”: D Hayton, ibid., 14–5. See also D Hayton, “Cross Border Estates: Part 3: the Hague Conventions” [1994] Private Client Business 329; G Miller, International Aspects of Succession, 228–9. 1407 Lewin on Trusts, 17th edn., 302. See also D Hayton, “Trusts and Forced Heirship Problems” (1993) 2 J Int Corp P 3, 7. 1408 [1900] 1 Ch 492. 1409 Lewin on Trusts, 17th edn., 302. The same view is expressed by D Hayton, “The Hague Convention on Trusts: a Little is Better than Nothing but why so Little?” (1994) 3 J Int Corp P 23, 26–7. See also D Hayton, “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159, 163; and K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, ch. 23, p. 31. See further: Re Mégret [1901] 1 Ch 547; and the Bermudan Court of Appeal decision in Garner v. Bank of Bermuda (noted (1992) 1 J Int Corp P 133). 1410 But see the limited claw-back provision in England of s. 10, Inheritance (Provision for Family and Dependants) Act 1975.

370 Preservation of Mandatory Rules state which does have forced heirship rights1411 and in which the heirs might successfully invoke those rights.1412 (iv) Civilian perspective From a civilian perspective, trusts and succession law may not prove happy companions. Droz expresses the view that reconciling the trust with the law applicable to succession may prove almost impossible in practice.1413 Conflict may arise in relation to common law testamentary trusts which create beneficial interests that cannot be claimed until the age of majority is attained, or which ride roughshod over rules of forced heirship of the law applicable to succession. This, in turn, raises the spectre of civil law would-be contracting states making wholesale use of mandatory succession rules and public policy. Droz laments that: “Le trustee ne peut être assimilé à un simple exécuteur testamentaire. It est plus que cela puisqu’il est propriétaire des biens . . . Nous en avons deduit qu’un trust soumis à une loi étrangère ne peut fonctionner dans le cadre de lois successorales qui l’ignorent”.1414

That said, the chasm between rules of succession and of trusts appears to have been bridged in Italy. The Tribunale di Lucca adopted a progressive interpretation of the Convention and the effect of forced heirship rules in Casani v. Mattei.1415 Casani died in Italy 1991. Although born in Italy, he was a dual American and Italian national. He executed a will in Kentucky written in English, which left his estate to the defendant (a friend) on trust for Casani’s daughter and her children. The claimant sought a declaration that the will was void, since it did not comply with Italian forced heirship legislation.1416 The court found that succession to the estate was governed by Italian law. However, it went on to rule that Article 15 of the Convention did not entitle an heir to claim that the will trust itself was void; it merely allowed the heir to require that her entitlement be received, so that only the remaining disposable share could be subject to the declared trust. Since the instant claim was solely for a declara1411 See e.g. Holzberg v. Sasson, Cour de Cassation, decision of 4 February 1986, [1986] Revue Critique de Droit International Privé 685; see also D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 21. 1412 Not least because the judgment of that foreign court may qualify for recognition and enforcement in England. 1413 G Droz, “Regards sur le Droit International Privé”, (1991–IV) 229 Recueil des Cours 9, 248. But compare C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé 7, 15 who suggests that “le principal bénéfice de la Convention se fera sentir dans le domaine successoral.” 1414 G Droz, ibid., 248. 1415 Reported in English in (1998/99) 1 ITELR 925. See M Lupoi, “The Domestic Trust Theory is Upheld in Italy”, (1998) 4 Trusts and Trustees 24. See also the decision of the Tribunale of Milan 1996 discussed by P Matthews, Trusts: Migration and Change of Proper Law, 57–8; P Matthews, “Italian Trust for Debenture Holders” (1997) 11 Trust Law International 20. See also the discussion of Art. 13, above. 1416 Namely Art. 602 of the Italian Civil Code.

Particular Matters Specified in Article 15 371 tion of invalidity, and not for a reduction in the disposable share of the testator’s estate to allow the daughter to claim her forced share, the claim failed. The court observed that: “the result is that the forced heir.. cannot argue that the will is void (either totally or partially), merely because it has a trust at its heart. But he can easily take proceedings to reduce any provisions in a will which, by putting the inheritance under the exclusive control of the trustee, have been identified as prejudicial to the reserved share”.1417

Such an approach is highly constructive. It allows the law of succession to make inroads into, but not to defeat, the trust altogether. It is hoped that other states will follow this approach when applying Article 15(1)(c), Hague Convention, since the legitimate concern of the mandatory rules of the law applicable to succession is to ensure that the testator’s relatives receive their entitlement, not to prevent the creation of a trust altogether. (v) The offshore dimension Whilst in many states the concern will be to ensure that the trust does not override existing mandatory provisions of related areas of law, in some states, particularly states situated offshore, the position will be quite the converse. They may wish to ensure that a choice of the local law to govern the trust is not undermined by e.g. a provision of forced heirship. Some offshore jurisdictions have taken steps which greatly limit the application of Article 15. Take, for example, the Isle of Man. The (Manx) Recognition of Trusts Act 1988 enacted the Hague Convention into its law. Nonetheless, section 5 of the (Manx) Trusts Act 1995 states that: “. . . no trust governed by the law of the Island and no disposition of property to be held upon the terms of such a trust is void, voidable, liable to be set aside or defective in any fashion . . . by reason that— . . . (b) the trust or disposition— (i) avoids or defeats any right, claim or interest conferred by foreign law upon any person by reason of a personal relationship to the settlor or by way of heirship rights; or (ii) contravenes any rule of foreign law or any foreign judicial or administrative order or action intended to recognise, protect, enforce or give effect to such a right, claim or interest.”

Such discriminating legislation prevents the application of e.g. forced heirship rules to undermine a trust governed by Manx law, but does not apply to a trust governed by a foreign law. The demands of legal certainty and the need to ensure that those who choose Manx law to govern a trust can be sure that the trust is insulated from claims arising by a foreign law may dictate such a policy. 1417 (1998/99) 1 ITELR 925, 947 (the law reporter’s unofficial English translation). J-P Béraudo explains that in France “un testament qui serait soumis à la loi française puisse créer un trust sur la quotité disponsible de la succession”: Les Trusts Anglo-Saxons et le Droit Français, 221 (referring to a judgment of 9 March 1895; 1896 Clunet, 628).

372 Preservation of Mandatory Rules However, it scarcely accords with the philosophy of the Convention to adopt such a differential approach1418 and leaves no significant role for the mandatory rules of the law applicable to succession to be applied under Article 15(1)(c) in relation to Manx trusts. It also allows a party to choose Manx law specifically to avoid the application of a foreign law, including its mandatory rules, which might apply in a related area. Yet the Isle of Man is far from alone amongst offshore centres in adopting such legislation.1419 Obviously, the nature of the market in offshore centres and the issues confronting them can be very specific and the importance of protecting the integrity of local law from attack by foreign law paramount in a highly competitive market. The result may, however, be that present or future contracting offshore states may use Article 15 very rarely, if ever, in respect of trusts governed by local law. Indeed, they are much more likely to say that application of a law which does have forced heirship claims infringes the forum’s public policy. Conversely, offshore states may have legislation which prevents a trust governed by a foreign law from impinging on any rules of local law. In Jersey, for example, Article 45(2), Trusts (Jersey) Law 1984 renders a trust governed by a foreign law unenforceable: “(a) to the extent that it purports; (i) to do anything the doing of which is contrary to the law of Jersey; or (ii) to confer any right or power or impose any obligation the exercise or carrying out of which is contrary to the law of Jersey; or (iii) to apply directly to immoveable property situated in Jersey”.1420

Sub-sections (i) and (ii) are remarkably broad and seem to suggest that wherever the foreign law differs from the law of Jersey, it cannot be given effect.1421 Any act sanctioned by application of a foreign trust law which is not a feature of Jersey trust law could be said to be “contrary to the law of Jersey.” If that is so, then Article 45(2) cannot sit today alongside the Hague Convention,1422 since it strips the Convention of all effect and removes the right which the Convention 1418 See also A Duckworth, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, ch. B1, p. 61. In this author’s view, the Manx law is also of highly dubious legality, since it purports to apply also to foreign judgments, notwithstanding the Isle of Man’s existing obligations to recognise such judgments: see the discussion in section 21(D) of Part One of this book. 1419 Almost identical legislation to that in the Isle of Man appears in s. 91, [Cayman Islands] Trusts Law 1967 (1998 Revision). In Bermuda, the legislation prevents a trust validly created by Bermudan law from being set aside or varied by a foreign law of succession, marriage or insolvency, unless the law of Bermuda has a corresponding law or rule of public policy: see s. 11, [Bermudan] Trusts (Special Provisions) Act 1989. 1420 Art. 45(3), Trusts (Jersey) Law 1984 goes on to state that in this scenario “any person in whom the title to such immoveable property is vested shall not be, and shall not be deemed to be, a trustee of such immoveable property.” 1421 It is not said that the act in question must be contrary to Jersey’s criminal law; merely that it is “contrary to the law of Jersey.” 1422 Even though K Baker and E Devenport, “National Digest for Jersey” in J Glasson (ed.) International Trust Laws, ch. A15, 30 comment that the extension of the Hague Convention to Jersey meant that the Trusts (Jersey) Law 1984 “required only minor amendment.” In this author’s view, Art. 45(2) of that Law should also have been amended.

Particular Matters Specified in Article 15 373 gives the settlor to choose the law governing the trust. It is tantamount to saying that any law can be chosen by the settlor, so long as it is Jersey law, or is identical to Jersey law! Sub-section (iii) is also very difficult to reconcile with the approach of the Hague Convention, which clearly requires contracting states to subject the validity of all trusts falling within its scope, including trusts of land, to the law governing the trust, and not to the lex situs.1423 It is difficult to believe that all provisions of a foreign trust law which apply directly to Jersey land can be said to offend Jersey public policy,1424 or that all rules of Jersey land law are mandatory, without straining these concepts beyond their intended limits. (vi) Administration of estates The administration of estates is not mentioned in Article 15. Presumably, this is because administration is a separate matter prior to the creation of the trust; in other words, it is a “rocket-launcher” issue. However, it should be noted that the very broad characterisation of the term “administration of estates” in re Wilks1425 and re Kehr,1426 if followed today, may have an impact on will trusts and lead to the English statutory powers of maintenance and advancement being applied to trusts governed by a foreign law.1427 (D) The Transfer of Title to Property and Security Interests in Property This important provision preserves the application of mandatory rules concerning transfer of title to property and security interests in property. During the drafting of the Convention, the Bank for International Settlements had sought to exclude from the Convention altogether business trusts and those which were intended to create security interests. Its observer expressed strong concerns about the effects of recognition of trusts on third party rights and on security interests. Ultimately, the delegates decided to include such trusts, whose exclusion would very much have weakened the impact of the Convention and have created very considerable difficulty in defining what was meant by a “business trust.”1428 Nonetheless, the dangers perceived by the Bank were recognised and Article 15(1)(d) (together with Article 11(3)(d) and Article 15(1)(f)) is the result. The term “transfer of title to property” is intended to emphasise that not every property issue falls within the sub-paragraph.1429 Such a broad provision would have had a devastating impact on the recognition of trusts. 1423 Notwithstanding this author’s sympathies as to the importance of the lex situs to trusts of land (see Art. 6, above). 1424 See Art. 18, below. 1425 [1935] Ch 645. 1426 [1952] Ch 26. 1427 See further Art. 8(2)(a) and (d), above and section 13 of Part One of this book. 1428 See Preliminary Document No 10; Working Document No 19; A Dyer and H van Loon, Report on Trusts and Analogous Institutions, 102; von Overbeck Report, para. 27, p. 376. 1429 Cf. the German proposal in Working Document No 44 (discussed by von Overbeck, para. 143, p. 402) to refer simply to “property”; see also A von Overbeck, “Trusts: the Hague Convention and the Law of the Netherlands”, at http://www-isdc.ch/e/vO97.asp.

374 Preservation of Mandatory Rules Article 15(1)(d) (together with Article 15(1)(e)) could, on one view, strip the trust of its essence. We saw in the discussion of Article 11 above1430 that it might be argued that the Convention does not require a contracting state even to accept the principle that trust assets constitute a fiduciary patrimony separate from the trustee’s personal patrimony and that a state might argue that its mandatory rules of property and insolvency dictate that the assets be available to satisfy claims of the trustee’s personal creditors. This issue was discussed in connection with the Dutch ratification of the Convention. Article 84, section (3) of Book 3, Dutch Civil Code, which precludes the valid transfer of assets “which does not have the purpose of bringing the property into the patrimony of the acquirer”, was circumvented by Article 4, Wet Conflichtenrecht Trusts 1995. The latter provision effectively excludes Article 15(1)(d) and (e), Hague Convention altogether, so as to forestall any argument that the Convention does not require the trust assets to be regarded as a fund separate from the trustee’s private wealth. It was argued that such an interpretation is to be welcomed and that, more generally, the “better view”1431 is that the trust assets should be treated under the Convention as separate from the trustee’s personal patrimony, since a “trust” shorn of the notion of a fund separate from the trustee’s personal assets is not in any meaningful sense a trust at all.1432 Article 15(1)(d) cannot be concerned solely with the transfer of property to the trustee, since this is a “rocket-launching” matter and outside the scope of the Convention. For common law trusts, it is more naturally to be applied to questions of transfer of equitable title to the beneficiary.1433 Questions such as whether equitable title has vested in the beneficiary and at what stage, whether the beneficiary’s interest in the property transferred is immediate, contingent or determinable and whether a beneficiary absolutely entitled can terminate the trust and call for the trust property to be transferred to him might be covered by this provision. In the case of transfers of beneficial title to land, the lex situs may well have something to say.1434 However, since the concern is only with questions appertaining to the transfer and the interest that it creates, not with property more generally, it would appear that many questions which one might naturally wish to refer to the lex situs, such as whether the beneficiary has a right to occupy the land, are not within the scope of Article 15(1)(d). Sub-paragraph (d) makes no reference to the position of third parties, which 1430 See the discussion “does the Convention require a contracting state to acknowledge the separate status of the trust assets from the trustee’s personal assets?” in relation to Art. 11, above. 1431 D Hayton in J Glasson (ed.) International Trust Laws, ch. C3, 12. 1432 See also A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6, 9. 1433 D Hayton, “The Netherlands Implementation of the Hague Trusts Convention”, (1996) 5 J Int Corp P 127, 128. One exception might be where a settlor declares himself trustee of his property, in order to defraud his personal creditors. Hayton notes that, in English law, the creditors will be able to claim the trust assets under the Insolvency Act 1986: (1987) 36 ICLQ 260, 275. 1434 As we saw above that Art. 45(2), Trusts (Jersey) Law 1984 does (see the discussion of “the offshore dimension” with respect to Art. 15(1)(c)). See also the discussion of “trusts of land” in relation to Art. 6, above.

Particular Matters Specified in Article 15 375 is expressly covered by sub-paragraph (f). However, Article 15(1)(f) is to some extent a mop-up provision1435 and it is likely that there will be some overlap with Article 15(1)(d). A transfer of property rights to the beneficiary will inevitably affect and bind all third parties (save the bona fide purchaser for value without notice) and the beneficiary may need to assert those rights against a third party. There will likewise be an overlap with Article 11(3)(d), second sentence, which preserves the application of the choice of law rules of the forum in respect of third party holders of trust assets.1436 However, that provision is both broader (in that it is not limited to the mandatory rules of the designated state) and narrower (in that it concerns the specific issue of whether a third party takes free of the beneficial interest under a trust). More fundamentally, could it be argued that Article 15(1)(d) applies to mandatory rules of property by which the third party asserts that title to trust property has passed to him, even though he did not act in good faith? After all, from the third party’s viewpoint the question is one of the “ordinary” acquisition of property and, in the case of purported outright transfers, English private international law does not seem to require a third party purchaser to act in good faith. In Winkworth v. Christie, Manson & Woods Ltd,1437 Slade J suggested that, as far as purported outright inter vivos transfers of tangible movables are concerned, application of the lex situs rule is subject to the requirement that the purchaser must have acted bona fide; in other words, the bona fide purchaser rule would be treated as an international mandatory rule of English law. However, Cheshire and North repudiate this view: “It is suggested that this exception can only be justified, if at all, as an example of the broader public policy exception and that would mean that it would not apply in every case where the English concept of good faith had not been satisfied; but only in the rare case where the application of the law of the situs in the particular circumstances was quite unacceptable to public policy”.1438

If the trustee wrongfully disposes of trust property to a third party, Kötz explains that, in Japan, the third party must return the asset to the beneficiary, if the trust is duly registered, or, failing that, if the third party knew or was grossly negligent in failing to ascertain that the property was transferred in breach of trust.1439 However, in Germany, the third party is very likely to take good title, as he is not apparently required to consider whether the trustee was acting in breach of trust.1440 Kötz doubts whether this is a mandatory rule of the forum, but it cannot be said with certainty that it would not continue to be applied were Germany to ratify the Convention. Nevertheless, it is contended that Article 15(1)(d) should not be interpreted so 1435

It applies “in other respects” to the protection of third parties acting in good faith. See “following and tracing the trust assets: Art. 11(3)(d)”, above. [1980] Ch 496 1438 Cheshire and North, 945. 1439 Art. 31, Japanese Trust Act 1922; H Kötz in D Hayton (ed.) Modern International Developments in Trust Law, ch. 3, 46. 1440 Ibid., 46; Bundesgerichtshof 4 April 1968, Juristenzeitung 1968, 428. 1436 1437

376 Preservation of Mandatory Rules as to allow third parties to rely on mandatory rules of property law to assert title to trust assets where they have not acted in good faith. The position of trusts is treated as materially different from outright property transfers to the extent that Article 11(3)(d) in principle allows the law governing the trust to determine whether trust assets may be recovered where the trustee has alienated them; whereas for “ordinary” property transfers, English private international law is only interested in the law of the place of the last transaction in the property, not the property’s origin. Moreover, a purchaser who does not act in good faith and who has actual or constructive notice of the existence of a trust can hardly object to use of the law applicable to the trust, rather than the lex situs, to determine whether the trust property may be recovered from him. It was argued above that Article 11(3)(d), second sentence, should be construed as preserving the role of the lex situs solely in respect of purchasers acting in good faith;1441 Article 15(1)(d) should not undermine that by permitting third parties who have not acted in good faith to rely on mandatory rules of property law by which they acquire title.

(E) The Protection of Creditors in Matters of Insolvency Article 15(1)(e) at one stage referred to “insolvency” without any qualification. The enormous scope of such a provision, and the enormous scope that it created for uncertainty, led to its narrowing specifically to deal with creditor protection. It is not, however, necessary that a creditor should have acted in good faith.1442 One area which might cause particular difficulties in non-trust states is the protective trust. Such a trust allows the beneficiary to receive income from the trust, but provides that, in the event of the beneficiary’s insolvency, his right to the income ceases. The income becomes subject to a discretionary trust, of which the bankrupt is but one object. Only if the trustees choose to appoint the bankrupt as an object of this trust will the trustee in bankruptcy have a claim thereto.1443 The protective trust may be viewed as a means of frustrating the claims of a creditor and may be vulnerable to the mandatory rules of the designated foreign law.1444 As with Article 15(1)(d), it has been argued that Article 15(1)(e) should be taken to concern the protection of creditors upon the beneficiary’s insolvency, not the trustee’s insolvency.1445 Otherwise, the fundamental requirement that 1441 See the discussion of “the position of third party purchasers of trust property in good faith” in relation to Art. 11(3)(d), above. 1442 Von Overbeck, para. 145, p. 403; Working Document No 58. 1443 See further s. 33(1), Trustee Act 1925; Billson v. Crofts (1873) LR 15 Eq 314; Re Ashby [1892] 1 QB 873; Re Burroughs-Fowler [1916] 2 Ch 251. See generally, R Pearce and J Stevens, The Law of Trusts and Equitable Obligations, 2nd edn., 134–5. 1444 Or, indeed, be seen as contrary to the public policy of a state asked to recognise a protective trust governed by English law under Art. 18.

Particular Matters Specified in Article 15 377 trust assets be treated as a fund separate from the trustee’s personal wealth could be undermined. Despite this, Kötz considers certain problems that might arise were Germany or Japan to ratify the Convention.1446 He explains that in Japan, Article 16(2), Japanese Trust Act 1922 would allow a beneficiary to object to any attempt by a personal creditor of the trustee to seize trust assets. In Germany, the trustee’s positions might be translated into the fiduziarische Treuhänder, so that, by virtue of section 43, German Bankruptcy Act 1877, the trust assets will not be susceptible to claims by the trustee’s personal creditors. However, he notes that problems may arise in Germany where the assets in the trustee’s hands are not the original trust property, but are assets subsequently acquired from a third party. In such a case, it appears that the assets may not be treated as separate from the trustee’s personal property and may be susceptible to attack. It may be that such a provision will be treated as a mandatory rule of its creditor protection law pursuant to Article 15(1)(e) and that the beneficiary might find that the trust assets are successfully claimed by the trustee’s creditors. Likewise, Kötz explains that problems may arise in Germany if the trustee has mixed trust property with his own in a bank account. The specificity principle of German law (“Bestimmtheitsgrundsatz”) would appear to prevent the beneficiary from tracing into the fund, even if the law applicable to the trust so permits and would put the mixed fund at risk of claims by the trustee’s personal creditors. That principle is likely to be treated as a mandatory rule of German law and could be superimposed onto the applicable law of the trust.1447 Needless to say, this author would not support such an interpretation, which would compromise the separate status of the trust assets. (F) The Protection, in other Respects, of Third Parties Acting in Good Faith This preserves the application of mandatory rules for the protection of third parties acting bona fide. It is not limited to purchasers and might extend to other third party recipients of trust property.1448 Those who hold trust property, or who have mixed the trust property with their own assets, will be subject to 1445 See the discussion “does the Convention require a contracting state to acknowledge the separate status of the trust assets from the trustee’s personal assets?” in relation to Art. 11, above; and see Art. 15(1)(d), above. 1446 H Kötz, in D Hayton (ed.) Modern International Developments in Trust Law, ch. 3, 44–6. 1447 Ibid., pp. 46–7. 1448 A third party who has dissipated the trust property may be personally liable for knowing receipt of trust property: but there is no indication that this personal claim is covered by the Convention in any event and it may be that it is subject to the choice of law rules governing claims in unjust enrichment. See e.g. Chase Manhattan Bank NA v. Israel-British Bank (London) Ltd [1981] Ch 105; El Ajou v. Dollar Land Holdings plc, [1993] 3 All ER 717 (reversed on different grounds: [1994] 2 All ER 685); Arab Monetary Fund v. Hashim (No 9), The Times, 11 October 1994; Trustor AB v. Smallbone, judgment of 9 May 2000 (CA); Kuwait Oil Tanker v. Al Bader [2000] 2 All ER (Comm) 271 (and see G Virgo, “Interest, Constructive Trusts and the Conflict of Laws” (2000) 8 RLR 122); Grupo Torras v. Al-Sabah, [2001] Lloyd’s Rep Bank 36 (noted by J Garton, (2001) 15

378 Preservation of Mandatory Rules Article 11(3)(d), first sentence, which subjects the ability to recover those assets or their value to the law applicable to the trust.1449 However, we saw above that the Netherlands may regard the according of proprietary effects to a trust, so that it may bind third parties, as contrary to its fundamental principles of property law. To the extent that such mandatory rules may not be derogated from, and to the extent that they apply to third parties acting in good faith,1450 they may be applied under Article 15(1)(f).1451 However, such mandatory rules should be invoked with caution, since the intention of Article 11(3)(d), first sentence to allow trust property to be recovered from third parties, if the law applicable to the trust so provides, is reasonably clear. Moreover, third party purchasers of trust property in good faith are accorded a measure of protection elsewhere under the Convention. The question whether traceable assets may in fact be recovered from a third party purchaser of those assets in good faith will be subject to the lex situs pursuant to Article 11(3)(d), second sentence.1452 Article 15(1)(f) only preserves mandatory rules dealing with third parties who act in good faith. In so far as a state has rules which protect a third party irrespective of his state of mind, its mandatory rules should not be applied under this sub-section. It is to be hoped that third parties who do not purchase property in good faith will not otherwise be able successfully to rely on Article 15(1)(d) to show that title to the trust property has passed to them.1453

9. ARTICLE 15(2) — OTHERWISE GIVING EFFECT TO THE OBJECTS OF THE TRUST

Somewhat enigmatically, Article 15(2) then urges the judge who finds himself unable to give full effect1454 to the trust to seek other ways to give effect to the settlor’s intentions. “This is an appeal to goodwill rather than a rule of strict Trust Law International 93). If the personal claim were within the Convention, Art. 15(1)(f) would not avail a third party recipient, since this only applies to third parties acting in good faith (who would not normally be personally liable in any event). 1449 More accurately, Art. 11(3)(d) does not expressly mention property mixed in a third party’s hands; but it was argued that the law applicable to the trust should determine whether the trust assets or their value can be traced into such a mixed fund (subject to the beneficiary demonstrating by the lex situs that he was a bona fide purchaser and that, even if the assets are traceable, he has taken free of the beneficial interest under the trust). See the discussion of “following and tracing the trust assets: Article 11(3)(d)”, above. 1450 But only to this extent. Note that Art. 15(1)(f) does not require the third party to be a purchaser. 1451 See the discussion of “constructive trusts and third parties; proprietary effects of the convention; application in the Netherlands” in relation to Art. 11(3)(d) above; see also M KoppenolLaforce, Het Haagse Trustverdrag, 270–6. 1452 See “the position of third party purchasers of trust property acting in good faith” in relation to Art. 11(3)(d), above. 1453 See the discussion of that sub-section above. 1454 Slightly curiously, the word “recognition” is used here (given that Art. 15 is a provision concerning application of choice of law rules).

Article 15(2)—Giving Effect to the Objects of the Trust 379 law”1455 and is, as such, distinctly curious. It simply urges states to “get into the spirit” of the Convention and respect the settlor’s wishes. “This is a very vague exhortation and is likely to prove particularly difficult to operate in civil law countries.”1456 It is also difficult to see its value. Contracting states remain free to find ways to give effect to the settlor’s intentions without needing further express authorisation.1457 But to the extent that a mandatory rule prevents the trust from having a certain effect e.g. on the property rights of third parties or the claims of creditors, it is very hard to see how the spirit of the trust can be preserved. Precisely because the rule with which the trust conflicts is mandatory, the trust rule gives way. Presumably, Article 15(2) refers to the fact that the Convention requires recognition of the trust qua trust. If it cannot be recognised qua trust, a nontrust state may still be able to give some effect to it by “translating” it into the nearest domestic law analogue.1458 “Thus the Convention gives priority to recognition of the trust as a sui generis institution but allows analogy of the trust to a contract of mandate or agency, for example, if its recognition as a trust is not feasible”.1459

But to a considerable extent, Article 15(2) looks like little more than further reminder to states of the need to interpret exceptions to the dominance of the trust and its applicable law narrowly; it is not obvious that it can have a real substantive impact.

1455 1456 1457 1458

Von Overbeck, para. 147, p. 403; Working Document No 44. Cheshire and North, 1040. Art. 14. See the discussion of “‘translating’ the trust into civil law systems” in relation to Art. 11,

above. 1459

A Dyer, “International Recognition of the Trust Concept”, (1996) 2 Trusts and Trustees 5, 8.

ARTICLE 16—INTERNATIONAL MANDATORY RULES OF THE FORUM The Convention does not prevent the application of those provisions of the law of the forum which must be applied even to international situations, irrespective of rules of conflict of laws. If another State has a sufficiently close connection with a case then, in exceptional circumstances, effect may also be given to rules of that State which have the same character as mentioned in the preceding paragraph. Any Contracting State may, by way of reservation, declare that it will not apply the second paragraph of this article.

1. SCOPE

This provision preserves the application of the forum’s international mandatory rules (sometimes called laws of immediate application (lois d’application immediate) or lois de police. These are those rules which the settlor may not derogate from and which are intended to be applied irrespective of the law governing the trust.1460 Given that Article 15 preserves the application of mandatory rules of the law designated by the forum’s conflict rules, it would have been extraordinary if the forum’s own mandatory rules of international application could not be allowed to override, or restrict the application of, the law applicable to the trust. In common law states, this will allow a state to apply its mandatory rules of trusts or of related areas of law and to superimpose these onto the applicable law. However, Article 16 extends beyond this and covers any international mandatory rule which conflicts with the trust. “The sense of these mechanisms is that they come into play regardless of the setting . . . in which they occur and thus regardless of the law which governs the . . . transaction.”1461 Presumably, it will cover, in particular, international mandatory rules of the forum relating to matters specified in Article 15, such as succession, property transfers, insolvency and the like. But Article 16 goes even further than this. Von Overbeck states that: “Among the laws which fall in this category, mention may be made of those which are intended to protect the cultural heritage of a country, public health, certain vital 1460 The wording used to describe these mandatory rules is much more emphatic than in Art. 15. Art. 16 speaks of rules which must be applied “even to international situations, irrespective of rules of conflict of laws.” See also von Overbeck, para. 149, pp. 404–5. 1461 E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 326.

Article 16(2)—Mandatory Rules of a State of Close Connection 381 economic interests, the protection of employees or of the weaker party to another contract. The French delegation asked that in the Report currency exchange regulations be mentioned in this connection”.1462

Perhaps most importantly of all, it must be recalled that Article 16 will also allow the forum to invoke its international mandatory rules of trusts, as well as those of related areas. In contrast, Article 15 is concerned solely with areas related to, but distinct from, the trust. Von Overbeck goes on to point out that states would inevitably have allowed their international mandatory rules to override the law applicable to the trust even without a special provision.1463 However, he suggests that this express provision will encourage states to apply mandatory rules, rather than take the step of ruling that the application of the law governing the trust infringes the forum’s public policy. Use of public policy necessarily involves some degree of negative comment on the content and scope of a foreign law, with the attendant negative consequences in terms of comity, and should be kept as narrow as possible. Nevertheless, it is evident that this provision has the potential greatly to undermine the work done elsewhere in the Convention and that, along with Articles 15 and 18, “such provisions may be either the Convention’s strength or downfall . . . In the final analysis much will depend on the will of the courts to implement the intentions of the Convention”.1464

An obstreperous attitude, which holds that any number of provisions of the law of property, insolvency, family law and succession are of international application, may dilute the Convention to the point of ineffectiveness.

2. ARTICLE 16(2) — MANDATORY RULES OF A STATE OF CLOSE CONNECTION ; EXCLUSION IN THE UNITED KINGDOM

Article 16(2) permits the application “in exceptional circumstances” of the international mandatory rules of a state of close connection with the case,1465 whose law is neither the law of the forum nor the applicable law. It means that “the same solidarity among States [can] . . . be shown as . . . in Article 13.” 1466 There, 1462

Von Overbeck, para. 149, pp. 403–4. See also E Gaillard and D Trautman, ibid., 326. Ibid., para. 150, p. 404. See also Re Lord Cable [1977] 1 WLR 7. 1464 A Paton and R Grosso, “The Hague Convention on the Law Applicable to Trusts and on their Recognition: Implementation in Italy”, (1994) 43 ICLQ 654, 661. 1465 Von Overbeck points out (at para. 158, pp. 406–7) out that the French phrase “avec l’objet du litige” is used rather than “case” and “does not necessarily correspond to a pending lawsuit.” Art. 7(1), Rome Convention uses the rather broader word “situation.” 1466 Ibid. para. 154, p. 405. E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 327 argue that application of the international mandatory rules of the forum or of a third state should be applied on the same criterion, namely that the state is the one most concerned with the issue in question. However, it goes without saying that the forum will have a particularly strong interest in apply its own mandatory rules in its home courts, so that discriminatory application of the mandatory rules of the forum from those of a third state is almost inevitable. 1463

382 International Mandatory Rules of the Forum it will be recalled, a state may refuse recognition to a trust which is objectively connected to another state which does not have the trust or the category of trust in question.1467 Under Article 16(2), a third state’s mandatory rules should only be applied if they are intended to apply to the trust in issue: “this point perhaps does not come out with sufficient clarity from the text”.1468 The phrase “effect may be given” emphasises that it is a matter of the judge’s discretion whether to apply such mandatory rules. Presumably, such discretion will be exercised by consideration of a number of factors, including: (i) the exact strength of connection to the third state. A choice of the law of state X to apply to a trust otherwise entirely connected to state Y would be an obvious case for application of the latter state’s mandatory rules; (ii) the extent to which the objects of the trust are to be effected in that third state, or the assets are primarily located in that state. Where performance of the trustee’s duties is likely to be rendered impossible or impracticable by a mandatory rule of a state exercising a degree of control over the trust and its assets, it may prove futile to act in a way which contravenes that state’s mandatory rules. This is especially likely if the trust property is land; (iii) whether the third state’s provisions are unambiguously intended to apply to such a situation. In this respect, the more express the legislation is as to its scope, the more likely that it might be applied; (iv) the extent to which the third state’s policies are shared by, or sympathised with, by the forum, or the degree to which it is important to maintain friendly relations with that foreign state.1469 In some states, Article 16(2) was felt to introduce an unacceptable level of uncertainty and derogation from the Convention’s aims.1470 Inevitably, problems would arise in determining when a state has a sufficient connection and as to whether the judge should exercise his discretion to apply its mandatory rules. For offshore states in particular, the unpredictability and unacceptability of a trust governed by local law being impugned by mandatory rules of a third state would have been intolerable. True, Article 16(2) does not require a court to invoke the third state’s mandatory rules and it might be thought that trust states would be unlikely to invoke the discretion. True also, it is only “in exceptional circumstances” that the discretion should be invoked. But the fear that a judge might choose to invoke Article 16(2) may be sufficient to deter a settlor from creating a trust which might need to be enforced in a contracting state. Article 16(3) offered contracting states the option to enter a reservation as to the application

1467

See “application of Article 13”, above. Von Overbeck Report, para. 151, p. 404. 1469 Compare the common law contract cases of Foster v. Driscoll [1929] 1 KB 470; Regazzoni v. KC Sethia (1944) Ltd [1956] 2 QB 490; [1958] AC 301. 1470 For an explanation of the evolution of Art. 16(2) and the various proposals considered, see von Overbeck, paras. 154–9, pp. 405–7. See also Working Documents Nos 10, 42, 44, 58 and 64. The United States of America, France, Finland and Switzerland all argued for the inclusion of a provision permitting the application of a third state’s mandatory rules. The proposal to permit a reservation to Art. 16(2) (but not Art. 16(1)) was an initiative of the German delegates. 1468

Whether a Rule has an International Mandatory Application 383 of Article 16(2).1471 The Recognition of Trusts Act 1987 simply omits Article 16(2),1472 so that only the international mandatory rules of the forum may be applied.1473 That said, Hayton points out1474 that an English court is very unlikely to require a trustee to act in a manner which is illegal in the place where the obligation is to be effected.1475 In other words, it is an English international mandatory rule that “Equity will do nothing in vain” and will not require performance where it is illegal.1476 It may well be, therefore, that an English court will indirectly continue to recognise the relevance of international mandatory rules of one particular third state, namely that of the place of performance.

3. DETERMINING WHETHER A RULE HAS AN INTERNATIONAL MANDATORY APPLICATION

Of course, the process of determining whether a rule is intended to have an international mandatory effect can be highly complex. In contract, Nygh argues that a mandatory rule is ultimately one which a given legal system evinces an intent to apply unilaterally.1477 Often, however, it is very difficult to determine the intended scope of legislation, when it appears that the legislators have given little thought to the matter. Since it is important to keep mandatory rules within strict boundaries, so as not to undermine the otherwise applicable choice of law 1471 The procedural mechanism for making a reservation is dealt with in Art. 26 (see below). The reservation is not reciprocal in nature, so that a contracting state which has not entered a reservation ought still, in an appropriate case, to apply the international mandatory rules of England or Scotland: von Overbeck, para. 159, p. 407. 1472 Nor does the provision apply in Alberta: s. 1(4) International Convention Implementation Act, 1990. 1473 The United Kingdom has done exactly the same in the contract sphere. Art. 7(1), Rome Convention allows a state to apply the international mandatory rules of “another country with which the situation has a close connection.” Art. 22(1)(a) allowed contracting states to enter a reservation on the application of this provision. The United Kingdom did so in s. 2(2), Contracts (Applicable Law) Act 1990, which states that Art. 7(1) “shall not have the force of law in the United Kingdom.” However, it should be noted that Australia has not entered this reservation, even though Australian courts appear to give very little effect to the mandatory rules of third states at common law: see P Nygh, Conflict of Laws in Australia, 6th edn., 307–8, 519–20. 1474 D Hayton, “The Hague Convention on Trusts: a Little is Better than Nothing but why so Little?” (1994) 3 J Int Corp P 23, 28. 1475 Such as where the exporting of assets is prohibited. “Since a Court of Equity will do nothing in vain and will not require a person to do an act that is illegal where it is to be done, so that sufficient protection exists, there seems no need for a common law country to adopt this paragraph [i.e. Art. 16(2)] of worryingly wide ambit” (D Hayton, in J Glasson (ed.) International Trust Laws, ch. C3, 15). 1476 He cites Re Lord Cable [1977] 1 WLR 7; Zivonestenska Bank v. Freeman [1950] AC 57; EuroDiam v. Bathurst [1987] 2 All ER 113. Compare Ralli Bros v. Cia Naviera Sota y Aznar [1920] 2 KB 287 (it being clear that an English court will not order performance of a contract which is illegal by the law of the place of performance (lex loci solutionis) when English law is the applicable law of the contract, but unclear whether it will do so if English law is not the applicable law.) 1477 P Nygh, Autonomy in International Contracts, 204.

384 International Mandatory Rules of the Forum rule, Nygh contends that three elements will normally be present in an international mandatory rule: (a) an express or clearly implied intention by the legislator to apply the legislation irrespective of the applicable law; (b) a reasonably close connection between the state and the transaction; and (c) a need for protection of the party seeking it. This third requirement can be critical, since it is one thing to determine a state’s general intention, but another to determine the limits thereof. Indeed, where a rule of e.g. forced heirship is concerned, different states will have quite differing views on whether there is a need to protect certain family members at all. Of course, in the case where the legislator’s intention is unambiguous, factor (a) will prove determinative. Such cases are, at least in English law, likely to be very rare. In England, certain administrative powers of the trustee have been regarded as applicable irrespective of the governing law of the trust.1478 In Re Wilks,1479 the court ruled that the powers conferred by the Administration of Estates Act 1925 on an English personal representative in respect of assets located in England may be exercised irrespective of the testator’s last domicile or, it seems, the law applicable to the trust.1480 In Re Kehr (dec’d),1481 it was held that where a grant of administration is made in England, the trustee’s statutory power to postpone sale under section 33(1), Administration of Estates Act 1925 applies irrespective of the law applicable to the trust. Such rules appear to be of an international mandatory nature. However, it is far from clear that they serve so pressing a public interest or protective function that they should be applied even where the governing law of the trust is not English and it is very doubtful whether such cases should be followed today.1482

4. INTERACTION WITH ARTICLE 15

The Convention does not expressly state what is to happen if the domestic mandatory rules of the law designated by Article 15 and the international mandatory rules of the forum themselves conflict. Since international mandatory rules are more powerful in nature than domestic ones, and since it is difficult to imagine a forum which would be willing to apply domestic mandatory rules of a foreign state which conflict with its own international mandatory rules, one can assume that Article 16 will prevail over Article 15. 1478

See the discussion of Art. 8(2)(a) and (d), above; and see section 13 of Part One of this book. [1935] Ch 645. 1480 This may not be a mandatory rule at all. Scott J in Chellaram v. Chellaram [1985] Ch 409, 430 explains it as an application of the choice of law rule that the administration of a deceased’s estate is governed by the law of the state from which he derives his authority. 1481 [1952] Ch 26; though see also the doubts expressed by Dankwerts J, 30; see also Re Ker’s Settlement Trusts [1963] Ch 553 (discussed by F Mann, “The Variation of Trusts Act, 1958, and the Conflict of Laws” (1964) 80 LQR 29). 1482 Of course, if the foreign law’s provisions on the administrative powers of trustees are repugnant to English standards of public policy, they will be disapplied to that extent. 1479

Do Mandatory Rules Undermine the Convention? 385

5. DO THE MANDATORY RULE PROVISIONS OF ARTICLES 15 AND 16 FATALLY UNDERMINE THE CONVENTION ?

Schoenblum strongly criticises Articles 15 and 16 as providing too easy an escape route for non-trust states from the consequences of trust recognition. “These articles reveal the ambivalence of the Convention regarding the burgeoning use of international trusts to defeat oppressive familial and unfounded creditors’ claims. The effects of these articles is to ensure further movement of trusts offshore, where unilateral trust legislation guarantees protection against such claims”.1483

The objection is a substantial one, especially to the extent that trust jurisdictions see themselves as competing in an international marketplace. Yet whilst the drafting of Article 15 and Article 16(2) is admittedly rather loose, one cannot expect the trust to prevail over all mandatory rules of succession, insolvency, property transfers etc. and at the same time expect the Convention to have any prospect of ratification in non-trust states. Offshore states which ratify the Hague Convention may have less to fear than it might seem. It was contended above that Article 15’s application is still at the judge’s discretion,1484 so that a judge in an offshore contracting state is only bound to apply the international mandatory rules of the forum (under Article 16(1))1485, which he would obviously wish to do, with or without the Hague Convention. Schoenblum goes on to contend that “If the Unites States ratifies the Convention, a disgruntled child may bring suit seeking to reach trust assets in the United States so as to satisfy a forced heirship claim”.1486

But it is not inevitable that a United States court would entertain such a claim.1487 Moreover, whilst it is true that offshore jurisdictions tend to be very much more protective of trust assets within their jurisdiction, this will not prevent non-trust states, in particular that of the testator’s domicile (in the case of a testamentary trust governed by the law of an offshore state), from subjecting the trust to the

1483 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5, 17. Contrast M Lupoi, Trusts: A Comparative Study, 363–4: “Article 15 is an extremely wise rule. . . . It makes recognition of trusts simpler, because it prevents all kinds of abuse right from the outset. . . .” 1484 See the discussion “must a judge apply the mandatory rules of the law designated by the conflict rules of the forum?” in relation to Art. 15, above. 1485 A contracting state may opt out of Art. 16(2), pursuant to Art. 16(3), as the United Kingdom has done. 1486 J Schoenblum, (1994) 3 J Int Corp P 5, 18. 1487 It might decide that invoking forced heirship rules infringes the forum’s own mandatory rules or public policy. Nor is it even clear that such a claim would fall within the Convention at all insofar as it relates to the “rocket-launching” aspect of the trust. In any event, if application of Art. 15 is indeed discretionary, a United States court would presumably simply decide not to exercise its discretion in respect of foreign mandatory rules of forced heirship.

386 International Mandatory Rules of the Forum claims of forced heirs in their courts,1488 regardless of whether the offshore jurisdiction whose law governs the trust has ratified the Convention.1489 The result is a clear conflict of laws between the offshore state and the non-trust state, which will render the place where proceedings are instituted all important. The Hague Convention tries to smooth the conflict between non-trust and trust states and compromise is an inevitability. In this author’s view, Articles 15 and 16(1) facilitate the recognition of trusts, not undermine it, because they remove a significant disincentive which non-trust states might otherwise have to ratifying the Convention.

1488 See D Hayton, “Whither Trusts in the Twenty-First Century: Part 2” [2000] Private Client Business 163 for a consideration of how common law courts might limit the application of offshore legislation. 1489 Which is open-ended and not reciprocal.

ARTICLE 17—EXCLUSION OF RENVOI In the Convention the word “law” means the rules of law in force in a State other than its rules of conflict of laws.

1. THE CASE FOR EXCLUSION EVALUATED

This provision excludes the application of the doctrine of renvoi.1490 In other words, if the law of Utopia is the governing law of a trust, the domestic trust law thereof will be applied without regard to Utopian choice of law rules. Suppose that a settlor chooses Utopian law to govern a trust. As a matter of Utopian private international law, the settlor may not choose the applicable law and a trust is governed by the law with which it has its closest connection. It may be that a Utopian judge would actually apply Ruritanian domestic law to the trust. Under the Convention, this fact is ignored and Utopian domestic law applied. Von Overbeck offers no explanation of why renvoi should be excluded, save that “this solution is that of all the modern conventions”.1491 Fashion is no substitute for principle. Not everyone is convinced that the need for renvoi has passed.1492 It goes without saying that such an approach has the merit of simplicity and avoids the difficulty of proving the choice of law rules of another state. It might convincingly be argued that the Convention’s stance is very much pro-settlor autonomy and the protection of his expectations. A settlor who provides that the law of Utopia should govern the trust almost certainly envisages that the domestic law of Utopia will be applied. To apply any other domestic law would defeat the point of allowing him to choose the domestic legal system which he wants to govern the trust. Hence, renvoi should be excluded in determining the law which governs a trust,1493 just as it is in contract. Against this, two main points may be made. First, the protection of expectations argument does not work where the applicable law is determined in the 1490 On which doctrine, see Dicey and Morris, ch. 4; Cheshire and North, ch. 5; see also section 7 of Part One of this book. 1491 Von Overbeck Report, 160, p. 407. In England, renvoi is excluded from contract by Art. 15, Rome Convention; in tort, it is excluded by s. 9(5), Private International Law (Miscellaneous Provisions) Act 1995, Part III. 1492 See A Briggs, “In Praise and Defence of Renvoi” (1998) 47 ICLQ 877; E Rimmell, “The Place of Renvoi in Transnational Litigation—a Pragmatic Approach to an Impractical Doctrine” (1998) 19 Hold LR 55. 1493 NB but not to the preliminary issue (outside the scope of the Convention) of the vesting of legal title in the trustee (on which, see section 7 of Part One of this book). In relation to “ordinary” property rules concerning inter vivos transfers, it is important to do as a judge in the courts of the situs would do. Indeed, a settlor seeking to vest Utopian property in a trustee, to be subjected to a trust governed by English law, will ignore the question what a judge in Utopia would do and which law he would actually apply at his peril, since the trust may subsequently need to be enforced in a Utopian court.

388 Exclusion of Renvoi absence of choice. There, the search is simply for a law of closest connection and there may be no expectation to protect. Suppose that the law of closest connection is Utopian. Can it really be thought sensible to apply Utopian law to the trust, even though a Utopian judge would himself not apply it, and would instead apply Ruritanian law? Why apply Utopian law if it is apparent that a Utopian judge would not even do so? If an English court applies a different law on the substance to that which a Utopian court would apply, the claimant has an incentive to forum shop and to sue in the state where the law applicable on the substance is more favourable to him. Yet von Overbeck is clear that “the taking into consideration of conflicts rules . . . [is] in any case excluded for the objective connection”.1494 In other words, when ascertaining the law of closest connection under Article 7, regard cannot be had to the private international law rules of the states whose (domestic) law might fall to be applied. In this writer’s view, simplicity and modernity can provide no convincing justification for such an approach. Secondly, the argument for renvoi is stronger in trusts than in contract, for the simple reason that the trust involves the assertion and administration of property rights. The rights, powers and duties of the trustee and beneficiary may need to be enforced in the state where the trust property is situated. In “general” property matters, the application of renvoi is well-established.1495 In relation to title to land, in particular, there is little point in deciding a dispute otherwise than in the manner that it would be decided in the courts of the situs. The difficulty for a beneficiary in asserting a beneficial interest in land situated in Utopia, which a Utopian judge, using his private international law rules, would decide was owned absolutely by the trustee or a third party, is all too-apparent.

2. EXPRESS CHOICE OF A STATE ’ S PRIVATE INTERNATIONAL LAW RULES

Nonetheless, for better or for worse, Article 17 is a clear, blanket exclusion of renvoi in relation to trusts matters under the Convention. But what if a settlor states in the trust instrument that “the trust shall be governed by Utopian law, including any rules of private international law that it might apply”? Though an unlikely scenario, one might think that the demands of protecting settlor autonomy and the parties’ expectations, which normally argue against application of renvoi, should in this case argue for application of the doctrine. A United States’ proposal would have allowed a settlor expressly to choose the private international law rules of a designated state.1496 However, von Overbeck simply explains that the proposal was rejected on the basis that “this did not seem 1494

Von Overbeck, para. 162, p. 407. See e.g. Re Ross [1930] 1 Ch 377; Re Duke of Wellington [1947] Ch 506; Winkworth v. Christie, Manson and Woods Ltd [1980] Ch 496, 514. See also sections 7 and 14 of Part One of this book. 1496 Working Document No 43. 1495

Renvoi and Related Areas of Law 389 necessary and it might complicate things”.1497 But it is far from obvious why a settlor should be able to choose a law to govern the trust of no objective connection1498 and/or pick and choose different laws to govern different aspects of a trust,1499 yet cannot state that renvoi shall be applied to the trust.

3. RENVOI AND RELATED AREAS OF LAW

Gaillard and Trautman argue that the literal wording of Article 17 speaks of the meaning of the word “law” under the Convention as a whole, and is consistent with the view that renvoi is excluded not just in relation to the choice of law rules for trusts, but wherever there is a reference to the word “law” in the Convention.1500 Accordingly, the application of the rules of law relating to property, insolvency, succession etc. under Article 15 would allow only for the domestic provisions of the law designated to be applied. That would be an extraordinary result. It would prevent an English court from continuing to apply renvoi in relation to questions of title of property.1501 It is difficult to imagine that such an outcome was intended by the drafters, or that it would be acceptable to the contracting states. Article 15 is intended to preserve the application of mandatory rules of the law designated by the forum’s choice of law rules; it is not intended to interfere in what those choice of law rules should be. It is suggested that Article 17 must be construed as referring only to the choice of law rules for trusts and does not affect a state’s right to apply the doctrine of renvoi to related areas of law such as property transfers.

1497

Von Overbeck, para. 162, p. 407. Art. 6. Art. 9. 1500 E Gaillard and D Trautman, (1987) 35 Am J Comp Law 307, 326. A Anton (with P Beaumont) Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 642 makes the same point. 1501 Or, at least, to “rocket” issues of title to property falling within the Convention’s ambit. 1498 1499

ARTICLE 18—PUBLIC POLICY The provisions of the Convention may be disregarded when their application would be manifestly incompatible with public policy (ordre public).

1. SCOPE

No private international law Convention is complete without its public policy derogation.1502 That said, it is inconceivable that states would not take it upon themselves to disapply provisions of the applicable law where they conflict with public policy, even were there is no express authorisation to do so. The provision encompasses the choice of law rules and the recognition of trusts. It may be applied to any1503 or all provisions of the Convention. Clearly, there are dangers with such a provision and much of the work of the Convention could be undone by its excessive use. It might, in particular, undermine asset protection trusts. Schoenblum concludes: “Thus, the recognition of trusts purportedly assured by the Convention seems a rather nebulous, if not empty gesture, in the very cases when trust recognition would be crucial—the use by corporations of the trust to protect foreign assets from hostile action at the domicile”.1504

Whilst it would be an overstatement to regard the recognition of trusts as fundamentally compromised by Article 18 (given that a residual public policy provision is inevitable), it is certainly true that the Convention is unlikely to be perceived by civil law states as an asset protection facilitator and that this might sound certain alarm bells in states where this might be an overriding objection, especially in the United States and in offshore jurisdictions.

1502 On the role of public policy in the conflict of laws, see Dicey and Morris, ch. 2; Cheshire and North, 123–32. Typical cases where public policy is invoked in contract law include where the applicable law is fundamentally contrary to basic principles of human liberty (e.g. Forbes v. Cochrane (1824) 2 B & C 448), English standards of justice (e.g. Royal Boskalis Westminster NV v. Mountain [1999] QB 674) or English morality (e.g. Lemenda Trading Co Ltd v. African Middle East Petroleum Ltd [1988] QB 448; though it may also be necessary in this “weaker” category that the place of performance take the same view as to the morality of the agreement). Public policy may also be invoked where the United Kingdom’s interests are adversely affected by the applicable law (e.g. Dynamit Actein-Gesellschaft v. Rio Tinto Co Ltd [1918] AC 260) or where good relations with a friendly foreign state may be adversely affected (e.g. Foster v. Driscoll [1929] 1 KB 479). 1503 Although Art. 18 strictly refers to disapplication of “the provisions of the Convention” and not to “any one or more of the Convention provisions”, it would be extraordinary if the public policy provision could only be applied to the Convention on an all or nothing basis. 1504 J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 246.

Scope 391 That said, it is likely that Article 18 will be interpreted restrictively,1505 in order not to undermine the other provisions of the Convention. The word “manifestly” indicates a strong threshold.1506 In particular, the fact that a technique necessitated by the trust is unfamiliar in a given state is not in itself a ground to invoke public policy.1507 Otherwise “Since the very concept of the trust could be and has been regarded as violative of public policy in civil law countries, it would seem that almost any provision of the Convention could be found to be against public policy”.1508

Application of public policy ought also to be resisted where the characteristic of the trust in question is one which is common to all trust states and is not an idiosyncrasy of the applicable law in question. Hence it would be inappropriate for a judge to invoke public policy so as to disapply the provision of the applicable law that e.g. the trust assets are not available to satisfy the interest of the trustee’s personal creditors. Hayton suggests1509 that likely examples of public policy in England would include valid non-charitable purpose trusts governed by a foreign law1510 and English property held on trust for English beneficiaries pursuant to a foreign law containing no rule against perpetuities.1511 Lewin on

1505 M Koppenol-Laforce, Het Haagse Trustverdrag, 270 remarks of Arts. 16 and 18: “From experiences with similar articles it follows that such articles are hardly ever applied. I see no reason why this should be different in trust-cases. Art. 15 is the real danger to recognition of the trust.” However, Art. 15 does not apply to the forum’s mandatory rules (unless the law of the forum is also the governing law of the relevant area of law); and neither Art. 15 nor 16 deal with public policy. The unique problems created by the recognition of the trust in states which do not know the institution in their domestic law provide a real prospect that Art. 18 may be used more widely than corresponding provisions of other Convention as a route of last resort to keep a trust out (even if the risk should not be overestimated). 1506 Although as J Schoenblum cogently points out “is not every violation of public policy ipso facto extreme?”: (1994) 3 J Int Corp P 5, 19, f/n 78. 1507 A fortiori, Art. 18 should not be invoked to undermine the trust’s recognition in its entirety on the basis that the trust is unknown in the state of recognition’s domestic legal system, since this would negate the point of the Convention. But cf. H Motulsky “De l’impossibilité de Creér un Trust AngloSaxon sous l’empire de la Loi Française” [1948] Revue Critique de Droit International Privé, 451. 1508 J Schoenblum, (1994) 3 J Int Corp P 5, 18. He goes on (at 19) to suggest that “every application of a foreign law or recognition of a legal entity unknown to a local law is technically violative of public policy.” It is suggested that this is incorrect: application of foreign law is the very essence of private international law and inevitably leads to claims being brought in respect of causes of action unknown in domestic law. Put differently, by having a system of private international law, states accept that, in principle, application of foreign law does not ipso facto impugn their public policy. 1509 D Hayton, in J Glasson (ed.) International Trust Laws, ch. C3, 15. 1510 But see the discussion of “STAR trusts” below and of “non-charitable purpose trusts” in relation to Art. 9, above. 1511 But cf. Lewin on Trusts, 17th edn., 289, who states that “. . . section 164 [Law of Property Act 1925] and the rule against perpetuities are not rules of public policy of the very strong kind that alone fall within that Article.” This assumes that the English rule against perpetuities is to be classified as a trusts matter and not a preliminary matter falling outside the Convention’s scope altogether. See the discussion of Art. 4, above and section 8 of Part One of this book; see also re Fitzgerald [1904] 1 Ch 573.

392 Public Policy Trusts gives the following examples:1512 discriminatory rules, rules violating human rights or oppressive foreign exchange control legislation.1513 It should be noted that the Convention contains the words “ordre public” in brackets at the end of the English version. These words are omitted from the Recognition of Trusts Act 1987. This may be thought to avoid unnecessary confusion. It may, however, also have a more subtle impact. Anton suggests that the doctrine of ordre public in continental legal systems may be wider than the doctrine of public policy.1514 If so, the omission might be a further indication that the Article should be construed narrowly. Article 18 says only that the applicable law may be disregarded. It imposes no obligation. However, it is difficult to imagine a state whose own public policy is manifestly infringed being sufficiently outward looking as to elect not to exercise its discretion. The only obvious case where a court might decline to exercise it would be where the policy is concerned with protection of individuals, and the individuals concerned choose to waive any rights that they may have under the law of the forum.1515

2. THE INTER - RELATIONSHIP OF ARTICLES 18 AND 15

It might be questioned whether an English court should be too ready to apply Article 18 to matters also affected by Article 15. For example, a testamentary trust of movable property governed by New York law might be said to infringe the forced heirship rules of France, the testator’s domicile. French mandatory rules of succession are preserved by Article 15(1)(c) and may be applied by an English court.1516 Could an English court nonetheless argue that application of the French forced heirship rules itself infringes English public policy?1517 In theory, the answer must be “yes”, since the public policy provision is a derogation from all of the Convention’s provisions and not just from its trusts choice of law rules, so that Article 18 trumps Article 15. However, insofar as French law has mandatory provisions requiring its application, an English court should not lightly find that application of these rules infringes English public policy

1512

It is notable how “extreme” are the violations of public policy which Lewin has in mind. Lewin on Trusts, 17th edn., 298. However, one example which they give, that of discriminatory rules on a settlor’s capacity to create a trust, is inappropriate, since capacity to create the trust is a preliminary matter falling outside the Convention and is thus, ex hypothesi, not governed by Art. 18. 1514 A Anton (with P Beaumont) Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 641. 1515 Or in deciding whether to apply the public policy of a third state (in so far as this is permitted by Art. 18): see “public policy of a state of close connection”, below. 1516 Re Annesley [1926] Ch 692. 1517 On the succession laws of England and France, see M Oldham, “Financial Obligations within the Family—Aspects of Intergenerational Maintenance and Succession in England and France” [2001] CLJ 128 (esp. 150–61). 1513

Public Policy of a State of Close Connection 393 merely because they differ from English law.1518 Such a finding is likely to make the trust “limp” between England and France.1519

3. PUBLIC POLICY OF A STATE OF CLOSE CONNECTION

Curiously, Article 18 does not state that only the forum’s public policy may be applied. In the light of the United Kingdom’s opt-out from Article 16(2), which would have allowed it to have applied international mandatory rules of a state of close connection, this is surprising. In England, it leaves open the question of whether the public policy of a friendly foreign state1520 or of the place where the trust is to be administered might apply.1521 Even if a third state’s public policy could, in principle, be applied, an English court will be unlikely to choose to do so.1522 It would indeed be very curious if a third state’s public policy were applied by an English court, when the international mandatory rules of such a state may not be.1523 That said, drawing the line between application of a forum’s mandatory rules and disapplication of the applicable law on grounds of public policy can frequently be highly complex, as can the process of determining whether it is English or foreign public policy that is being applied.1524 Public policy is an essentially negative process, leading to the disapplication of an otherwise

1518 However, claw-back rules of forced heirship would not be applied in an English court to validly created inter vivos trusts. This is because French law can only be applied to assets which an English court regards as forming part of the testator’s estate on death. Even if that argument were to fail, the English court would be likely to invoke public policy to prevent claw-back: D Hayton “Cross Border Estates: Part 3: the Hague Convention” [1994] Private Client Business 329, 334. 1519 It must be conceded, however, that in the example given, application of forced heirship rules in England would probably make the trust “limp” between England and France on the one hand, and New York on the other. 1520 See e.g. Foster v. Driscoll [1929] 1 KB 470; Regazzoni v. KC Sethia (1944) Ltd [1956] 2 QB 490; [1958] AC 301; Royal; Boskalis Westminster NV v. Mountain [1998] 2 WLR 538. See also Cheshire and North, 585–6. 1521 By analogy to Ralli Bros v. Cia Naviera Sota y Aznar [1920] 2 KB 287, where it was deemed to be contrary to English law (the proper law of the contract) to order performance of a contract which had subsequently been rendered illegal in the place of performance. Debate abounds about whether such a principle applied at common law to contracts governed by a foreign law: see Cheshire and North, 600–3. 1522 Art. 18 “does not prevent” the application of public policy; it does not require it. 1523 But see J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 238–9, who points out that Dicey and Morris changed their own position on this matter in successive editions, at one stage asserting that the strong public policy of the law with which the trust had its closest connection would also be applied, but subsequently referring only to English public policy. 1524 So much so that A Anton (with P Beaumont) Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 641 questions whether separate provision for application of the mandatory rules of the forum (in Art. 16(1)) and the public policy of the forum (Art. 18) was necessary, terming the former “something of a fifth wheel on the coach.”

394 Public Policy applicable law.1525 Mandatory rules, in contrast, perform a positive role, superimposing rules onto a trust. But consider the facts of Regazzoni v. KC Sethia (1944) Ltd.1526 The case involved a contract governed by English law which sought to evade an Indian law prohibiting the export of goods from India to South Africa. The House of Lords refused to enforce the contract. Nygh argues that the case is one where English public policy required the maintenance of good relations with another state.1527 However, there are several other possible interpretations. The decision could be construed as: (a) an application of English mandatory rules;1528 or (b) an application of Indian mandatory rules1529 or; (c) the operation of English public policy; or (d) the operation of Indian public policy. The distinction could be critical, particularly in a scheme where only the forum’s international mandatory rules or public policy are capable of application. If such cases can be regarded as an illustration of English public policy in operation, it is difficult to identify the limits thereof. Any recourse to the law of a foreign state other than the applicable law of the trust could potentially be an example of the forum’s public policy, because it is the forum which chooses to disapply the governing law. Rather, it is suggested that whether English, as opposed to foreign law is being applied depends in part on the degree to which the forum’s decision is based on its own state’s interests e.g. in protecting international relations with a particular country, as opposed to a situation where English law feels relatively neutral, but another state with an interest in the matter has especially strong views on the matter. Whether the matter concerns mandatory rules or public policy should depend upon whether a provision is being applied routinely due to its indispensable status by a particular law, or whether the applicable law’s impact is being limited, in the light of the law of another country, because of its unacceptable consequences on the facts of the instant case. The former looks like an application of mandatory rules; the latter, like use of public policy.1530

1525 For recent detailed discussion of public policy in the conflict of laws, see M RubinoSammartano and C Morse, Public Policy in Transnational Relationships (The Hague, Kluwer, 1992). See also, P Carter, “Rejection of Foreign Law: Some Private International Law Inhibitions” (1984) 55 BYIL 111; P Carter, “The Role of Public Policy in English Private International Law” (1993) 42 ICLQ 1; N Enonchong, “Public Policy in the Conflict of Laws: a Chinese Wall around Little England?” (1996) 45 ICLQ 633; J Harris and F Meisel, “Public Policy and the Enforcement of International Arbitration Awards: Controlling the Unruly Horse” [1998] LMCLQ 568; R Leslie, “The Relevance of Public Policy in Legal Issues Involving Other Countries and their Laws” (1995) 6 Jur Rev 477. 1526 [1958] AC 301. 1527 P Nygh, Autonomy in International Contracts, 234. 1528 And is consistent with the application of domestic or international mandatory rules of English law. 1529 These would be international mandatory rules, since Indian law was not applicable to the contract. 1530 Hence, Regazzoni looks more like an application of Indian mandatory rules applied due to their status in a “friendly foreign state” of export. See J Harris, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247, 260–4.

STAR Trusts 395

4. STAR TRUSTS

A subject of increasing importance is the enforcement in England of STAR trusts.1531 Originating from the Cayman Islands, the regime allows a settlor to create a trust for beneficiaries or for purposes, or for both. Such purposes may be charitable or non-charitable. The beneficiaries do not ordinarily have a right in the trust property itself unless and until it is transferred to them. The trusts are enforced by appointed enforcers. The beneficiary has no right to enforce the trust unless also appointed as an enforcer of the trust. Hayton suggests that the where no beneficiary is appointed as an enforcer, the absence of obligations owed directly by the trustee to the beneficiary means that English law would not regard STAR trusts as valid express trusts at all.1532 The scheme may be construed by an English court as creating a resulting trust for the settlor with a power to benefit certain objects.1533 But there is a real risk that a STAR trust of which no beneficiary is appointed as enforcer might prove unenforceable altogether in England on public policy grounds under Article 18.1534 Matthews also argues against recognition in England on the basis that the STAR trust may allow for non-charitable inalienable property dispositions.1535 However, in a heated exchange with Matthews, Duckworth argues (convincingly in this writer’s view) that STAR trusts should be recognised in England.1536 He points out that such trusts would appear to fall within the description contained in Article 2, Hague Convention, since they are a legal relationship created “when assets have been placed under the control of a trustee for 1531 See the Special Trusts (Alternative Regime) Law 1997. See A Duckworth, STAR Trusts; A Duckworth, “National Digest for the Cayman Islands” in J Glasson (ed.) International Trust Laws, ch. A7, pp. 28–36; G Thomas, “Purpose Trusts” ibid., ch. B4, 36–45. See also D Hayton, “STAR Trusts” (1998) 4 Amicus Curiae 13; D Hayton, “STAR Trusts”, (1998) 8 Offshore Taxation Review 43; P Matthews, “Shooting STAR: the New Special Trusts Regime from the Cayman Islands” (1997) 11 Trust Law International 67; A Duckworth, “STAR Wars: the Colony Strikes Back” (1998) 12 Trust Law International 16; P Matthews, “STAR: Big Bang or Red Dwarf?” (1998) 12 Trust Law International 98; A Duckworth “STAR Wars: Smiting the Bull” (1999) 13 Trust Law International 158. A Duckworth, “The New Frontier of Purpose Trusts”: http://www.il-trust-in-italia.it/ webtutti/@Congress%201999/Duckworth%20New%20frontier.htm. See also the discussion of “the administration and validity of non-charitable purpose trusts” in relation to Art. 9, above. 1532 Identified as an essential aspect of the trust by Millett LJ in Armitage v. Nurse [1998] Ch 241, 253. 1533 D Hayton, “STAR Trusts”, (1998) 8 Offshore Taxation Review 43, 45. But see Hayton’s far more non-charitable purpose trust-friendly discussion in, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96. See also P Matthews, “The New Trust: Obligations without Rights?” in A Oakley (ed.) Trends in Contemporary Trust Law, ch. 1. 1534 D Hayton, in J Glasson (ed.) International Trust Laws, ch. C3, 15. 1535 P Matthews, (1998) 12 Trust Law International 98, 99. Presumably on this view, STAR trusts for beneficiaries might be enforceable. See also P Matthews in A Oakley (ed.) Trends in Contemporary Trust Law, ch. 1. 1536 P Matthews, (1997) 11 Trust Law International 67; A Duckworth, (1998) 12 Trust Law International 16; P Matthews, (1998) 12 Trust Law International 98; A Duckworth, (1999) 13 Trust Law International 158.

396 Public Policy the benefit of a beneficiary or for a specified purpose”.1537 The assets form a separate fund from the trustee’s own estate, title to the assets stands in the name of the trustee and the trustee has powers and duties in respect of the trust assets “in respect of which he is accountable.” It matters not that he is accountable to the enforcer and not to the beneficiary. Duckworth argues that it is not inalienability which is offensive to English public policy, but infringement of the rule against perpetuities.1538 He rightly points out that the fact that a trust would not be valid if governed by English domestic law cannot itself be a reason to refuse its recognition for private international law purposes. Such an approach would render the choice of law process nugatory. The question then becomes whether the objections of English law to STAR trusts are sufficiently strong to merit application of English public policy, on the basis that the “irreducible core” of the trust is absent.1539 It is suggested they are not. Hayton has more recently suggested that the non-recognition of a noncharitable purpose trust with appointed enforcers is most unlikely in an English court.1540 Moreover, as Duckworth cogently argues, it would hardly do for an English court to reach for the public policy doctrine in the face of a trust with different characteristics to the English trust, when it simultaneously expects contracting non-trust states to recognise an institution not known in its legal system:. “. . . the primary purpose of the Convention was to require the recognition of trusts by non-trust states, states whose established principles of law have much more fundamental objections to English trusts than the English law has to purpose trusts. It is ironical that it should be an English lawyer who tries to evade the Convention by saying that it calls for recognition of an institution which is inconsistent with domestic legal principles”.1541

On balance, then, it is submitted that STAR trusts should be recognised in the English courts and Article 18 should not be invoked.1542

1537

A Duckworth, (1998) 12 Trust Law International 16, 22. A Duckworth, (1999) 13 Trust Law International 158, 162. 1539 See P Matthews, “The New Trust: Obligations without Rights?”, ch. 1 in A Oakley (ed.) Trends in Contemporary Trust Law, ch. 1; D Hayton, ibid., ch. 3; D Hayton, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96. 1540 D Hayton, (2001) 117 LQR 96, 100. 1541 A Duckworth, (1999) 13 Trust Law International 158, 162. 1542 That, in turn, might eventually lead to a more benevolent attitude in English courts to noncharitable purpose trusts in English domestic law: see “non-charitable purpose trusts” in the discussion of Art. 9, above. 1538

ARTICLE 19—EXCLUSION OF FISCAL MATTERS Nothing in the Convention shall prejudice the powers of States in fiscal matters.

1. A NEUTRAL APPROACH

The Convention’s neutrality on fiscal1543 matters has been criticised: “There were few if any tax advisors associated with the drafting of the Convention . . . This is a striking omission, since taxation is central to all trust and estate planning”.1544 Nobody could doubt the centrality of the issue of taxation. It is also imperative that states develop laws to prevent a settlor from being able to avoid an otherwise applicable tax through the simple expedient of choosing a different law to govern a trust.1545 However, the prospects of attempting to harmonise not merely the private international laws of trusts, but their taxation across the globe appear extremely remote and well beyond the scope of a private international law Convention. Without Article 19, the Convention’s “chances of ratification would be seriously compromised”.1546

2. THE UNITED KINGDOM

(A) Article 19 Not Enacted Article 19, although containing a principle important to trust and non-trust states, is not scheduled to the Recognition of Trusts Act 1987. It might be thought that it goes without saying that a Convention on the private international law of trusts will not affect the taxation of trusts in the United Kingdom. Nor was there any prospect of foreign taxation law being applied in the United Kingdom’s courts. The general rule of private international law which prevents enforcement of foreign revenue laws,1547 coupled, if necessary, with the invocation of public policy, would in any event prevent the application of foreign taxation rules. 1543 J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 247 points out that the word “fiscal” “could be construed as including within its scope not only taxation, but also exchange controls and even outright taking.” 1544 J Schoenblum, (1994) 3 J Int Corp P 5, 20. See also the discussion of the OECD Model Convention 1977 on the taxation of international trusts by I Koele, in F Sonneveldt and H Van Mens (eds.) The Trust—Bridge or Abyss between Common and Civil Law Jurisdictions, ch. 7. 1545 One which may have no objective connection with the trust at all. 1546 Von Overbeck, para. 165, p. 408. 1547 Government of India v. Taylor [1955] AC 491; Re Lord Cable [1977] 1 WLR 7; Williams and Humbert Ltd v. W and H Trade Marks (Jersey) Ltd [1986] AC 368.

398 Exclusion of Fiscal Matters

(B) Refusal to Enforce Foreign Revenue Laws1548 The leading case on enforcement of foreign revenue laws in England is Government of India v. Taylor.1549 A company registered in England which conducted its business in India was sold to the Indian government and the proceeds transferred to England. Upon the company’s subsequent liquidation, the Indian Commissioner of Income Tax alleged that substantial payment of capital gains tax was due in respect of the sale. The House of Lords refused to entertain the claim. It mattered not that the Indian taxation liability was neither punitive nor significantly different in nature to the equivalent English tax. The court will not allow the form of the action to blind it to its substance.1550 Hence, attempts indirectly to enforce a foreign revenue law will fail. In Rossano v. Manufacturers’ Life Assurance Co Ltd,1551 an Egyptian claimant sought to recover payments due under certain life assurance policies, two of which were allegedly payable in England. The defendants sought to resist the claim on the basis of garnishee orders already made in Egypt, which would render the defendants liable to pay taxation due from the claimant to the Egyptian government if they paid the sums due prior to the defendants discharging their tax liability. The defence was rejected, since to uphold it would be to allow rules of Egyptian taxation to determine the outcome of an English claim and would result in the indirect enforcement of a foreign revenue law.1552 It should be noted that there is no rule against the recognition only of a foreign revenue law. Cheshire and North give certain examples of where a foreign revenue law might be relevant in England:1553 where a fraudulent tax-evasion scheme has been established to avoid the revenue law of a friendly foreign state;1554 or where personal representatives have paid foreign taxes and seek an indemnity from assets of the estate situated in England.1555 Furthermore, the mere fact that an issue has tax implications should not take it outside the Convention, if the point at issue is clearly one of trusts law. For example, if an English court is asked to rule on whether certain formalities are 1548 Only brief comment is provided on this subject here. See further Dicey and Morris, ch. 5, (esp. 93–4); Cheshire and North, ch. 8 (esp. 108–10). 1549 [1955] AC 491. 1550 See generally, Dicey and Morris, 89–92; Cheshire and North, 108–110. 1551 [1953] 2 QB 352. 1552 For critical comment, see Cheshire and North, 109–10. Contrast Williams & Humbert Ltd v. W & H Trade Marks (Jersey) Ltd [1986] AC 368. See also A-G of New Zealand v. Ortiz [1984] AC 1. A foreign state is not precluded from seeking to obtain evidence in England in order to enforce foreign taxes: Re State of Norway’s Application (Nos 1 and 2) [1990] 1 AC 723. 1553 Cheshire and North, 110. 1554 Re Emery’s Investment Trusts [1959] Ch 410. 1555 Re Lord Cable [1977] 1 WLR 7. However, in so far as this requires an English court positively to indemnify a personal representative, it could be argued that this amounts in sum to indirect enforcement of the foreign revenue law.

Taxation and Civil Law States 399 required by the applicable law of the trust, the outcome of its decision may be determinative of whether the creation of a trust is subject to tax liability. However, the claim itself is a private law matter and not one to recover foreign taxes and, accordingly, within the Convention.1556

3. TAXATION AND CIVIL LAW STATES

(A) General Comment Of course, non-trust states ratifying the Convention will need to give careful consideration to the tax implications of the trust. Sonneveldt states that “Civil law jurisdictions are often at a loss when it comes to taxing trusts”.1557 This uncertainty leads Schoenblum to argue that “There are immense dangers in encouraging the availability of trusts in civil law jurisdictions, but not having the slightest notion of what it will cost”.1558 However, each civil law state is likely to evolve its own system of taxation and to insist upon the autonomy to do so and any attempted integrated approach may have jeopardised the prospects of ratification of the Convention.1559 Very brief consideration of the general approach to taxation in the two ratifying civil law states of Italy and the Netherlands will now be given.

1556 This is not free from doubt, however. In QRS 1 v. Fransden [1999] 1 WLR 2169, the Court of Appeal refused to entertain a claim brought by a liquidator against a Danish company at the behest of the Danish tax authorities. The purpose of the claim was to collect foreign taxes and amounted to indirect enforcement of a foreign revenue law. However, it has been pointed out that “. . . a company has a right to pursue those who have stripped it bare; and it is difficult to see why it matters very much or at all to know how any sums recovered will be applied by the company after any judgment in its favour has been executed:” A Briggs, “Decisions of British Courts during 1999 Involving Questions of Public or Private International Law: B) Private International Law” (1999) 70 BYIL 319, 341. Where a claim is genuinely brought to enforce a private law right, it is far from obvious that the purpose to which any sums recovered will be put involves the enforcement of a revenue matter, or, indeed, that it is any business of the English court. See P St J Smart, “The Rule against Foreign Revenue Laws” (2000) 116 LQR 360. See also Buchanan v. McVey [1955] AC 516. 1557 F Sonneveldt in F Sonneveldt and H van Mens (eds.) The Trust—Bridge or Abyss between Common and Civil Law Jurisdictions? ch. 1, 17. 1558 J Schoenblum, (1994) 3 J Int Corp P 5, 20. He argues that the Convention should have made an “integrated assault” on both tax matters and non-tax matters alike. 1559 For a discussion of taxation of trusts in a non-contracting state (Switzerland), see A Burnand, “Laws of Succession and Testamentary Dispositions—Conflicts of Laws and Recognition of Foreign Trust—a Swiss Overview” (1995) 1(10) Trusts and Trustees 15, 17–8; H van Mens, “The Trust and Swiss Tax Law” in F Sonneveldt and H van Mens (eds.) The Trust, Bridge or Abyss between Common and Civil Law Jurisdictions, ch. 4.

400 Exclusion of Fiscal Matters (B) The Netherlands1560 Koppenol-Laforce and Kottenhagen point out that, in the Netherlands, each trust may be considered by a tax authority committee to determine its “transparency” for tax purposes. Essentially, it appears that income tax will be due on the trust assets (and any trust income) if the settlor or beneficiary retains some control over the assets such that “it can be considered to be at his disposal as if it were his own wealth. . . .”1561 Accordingly, the trustee will not be liable to taxation, unless, of course, he is also a beneficiary.1562 Nor will the transfer to the trustee render him liable to gift tax. Inheritance tax will be payable on a testamentary trust created by a Dutch resident and a beneficiary acquiring a fixed interest under said trust will be liable thereto. In the case of a discretionary trust, it may be that the trustee will be considered to have inherited from the settlor and that inheritance tax should be paid by him and refunded by the beneficiary upon distribution. Even an inter vivos trust could incur inheritance liability if the settlor retained the power to revoke the trust prior to his death.1563

(C) Italy The position is also not clear-cut in Italy. Albisinni and Gambino suggest that “Applying the rationale that tax authorities may challenge the apparent possession of income by attributing it to the effective beneficiary rather than to the apparent possessor, to a trust, it is likely that tax authorities would attempt to attribute trust income to the beneficiary, and not the trustee or settlor, for taxation purposes”.1564 They go on to state that this principle is especially likely to 1560 “It is a pity that the Dutch legislator did not seize the opportunity to develop fiscal rules for the trust. On the other hand, the fiscal paragraph in the French Fiducie-proposal caused the proposal to be withdrawn altogether”: M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” 137,150. See also H Verhagen and N Faber, “a Trace of Chase Manhattan in the Netherlands?” (1998) 6 RLR 165, discussing the unreported Dutch decision in Ontvanger v. Hamm qq. 1561 M Koppenol-Laforce and R Kottenhagen, ibid., 151, considering a passage by H van Mens in “The Trust and Dutch Law: the Trust and the Netherlands’ Individual Income Tax, New Wealth Tax and Corporate Income Tax” in F Sonneveldt and H van Mens (eds.) The Trust, Bridge or Abyss between Common and Civil Law Jurisdictions, ch. 5:I, 54. See also J Schipper, “ Some Dutch Fiscal Aspects Regarding Trusts” (1997) 6 Trusts and Trustees, 21; H Verhagen, “Trusts in the Civil Law: Making Use of the Experience of ‘Mixed’ Jurisdictions”, (2000) 8 European Review of Private Law 477, 484; B Larking and J Dekker, “Use of Trust-Held Dutch Companies for International Tax Planning” in A Kaplan (ed.) Trusts in Prime Jurisdictions, ch. 14, 293 (and see the proposed changes to the income tax position in 2001, discussed at 295). 1562 Save in respect of any payment for his services. 1563 This paragraph briefly summarises some of the views expressed by B Leemreis in F Sonneveldt and H van Mens (eds.), The Trust, Bridge or Abyss between Common and Civil Law Jurisdictions, ch. 5: II. 1564 F Albisinni and R Gambino, “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 85.

Taxation and Civil Law States 401 apply to fixed trusts created overseas of assets located in Italy, where the settlor is resident in Italy and the income is paid regularly to the beneficiaries.1565 However, they consider that the “veil” is less likely to be lifted (and tax less likely to be attributed to the beneficiary) where the trustee has discretion whether to distribute the income to the beneficiaries and may instead reinvest it.1566

1565 1566

Ibid., 86. Ibid., 86.

ARTICLE 20—RIGHT TO EXTEND CONVENTION TO OTHER TYPES OF TRUST Any Contracting State may, at any time, declare that the provisions of the Convention will be extended to trusts declared by judicial decisions. This declaration shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands and will come into effect on the day when this notification is received. Article 31 is applicable to the withdrawal of this declaration in the same way as it applied to a denunciation of the Convention. This provision was discussed when we considered the scope of the Convention above.1567 It allows a contracting state to extend the scope of application of the Convention’s provisions. We saw above that the United Kingdom, in section 1(2), Recognition of Trusts Act 1987, took advantage of this provision, and indeed extended it1568 to any other trusts arising under the law of any part of the United Kingdom.1569 However, Article 20 is merely facilitative and creates no obligation on other contracting states to recognise these additional categories of trust. The provision itself is curious. It is easy to see why a state might wish to extend the Convention to other types of trust, such as resulting or constructive trusts.1570 It is much less obvious why it should want to extend it merely to the curious category of trusts declared by judicial decision.1571 The word “declared” could be construed so as to exclude those trusts which pre-exist and whose existence is merely reaffirmed by the court, whilst including those trusts imposed at the court’s discretion, such as remedial constructive trusts or trusts imposed e.g. as part of a matrimonial settlement. That said, the word “declares” might be construed so as also to include judicial affirmation of a pre-existing trust, such as an institutional constructive trust; just as a declaration of non-liability affirms a liability which already exists. However, the opaque drafting may, and has, led to anomalies in states which have taken advantage of Article 3 in their enacting provisions.1572 The United Kingdom refers in section 1(2), Recognition of 1567 See the discussion of “the statutory extension of the Convention rules in the United Kingdom” in relation to Art. 3. 1568 Even though Art. 20 does not expressly allow it to do so. 1569 The meaning and scope of s. 1(2) was considered above, in the discussion of Art. 3. 1570 To the extent that these are not already covered by the Convention. See the discussion of these trusts when posing the question, “which types of trust are within the Convention?” in relation to Art. 3, above. 1571 “. . . [T]he Convention makes no provision for constructive or resulting trusts, because, although it meant to make reference to them, it identifies a completely different category, which is that of trusts created pursuant to a judgment”: M Lupoi, Trusts: A Comparative Study, 238. 1572 See the discussion of s. 1(2), Recognition of Trusts Act 1987: “the statutory extension of the Convention rules in the United Kingdom” (when dealing with Art. 3, Hague Convention, above).

Right to Extend Convention to Other Types of Trust 403 Trusts Act 1987 to trusts “arising” (rather than “declared”) by virtue of a judicial decision. This does seem to exclude affirmations of a trust’s existence and to apply only where the court itself imposes the trust.1573 Koppenol-Laforce envisages that non-trust states will find the extension to trusts declared by judicial decision problematic and argues that it “should not be used by non-trust countries.”1574 However skilled they might be, or become, in dealing with the trust, the prospect of a non-trust state of competent jurisdiction being asked itself to determine the existence of a non-express trust and its applicable law and to give effect to that trust is one which few non-trust states are likely to relish.

1573 1574

See the discussion of “trusts ‘arising’ by judicial decision” in relation to Art. 3, above. M Koppenol-Laforce, Het Haagse Trustverdrag, 266.

ARTICLE 21—RIGHT TO RESTRICT RECOGNITION TO TRUSTS GOVERNED BY THE LAW OF A CONTRACTING STATE Any Contracting State may reserve the right to apply the provisions of Chapter III1575 only to trusts the validity of which is governed by the law of a Contracting State.

1. A COMPROMISE

This provision1576 (which is not enacted in the United Kingdom)1577 allows a state to restrict the otherwise open nature of the Convention and to limit the application of the Convention for recognition purposes to trusts governed by the law of a contracting state. Without it, a state can benefit from the recognition of its trusts in a contracting state without having to ratify the Convention itself.1578 It represents a compromise between those states which wanted an entirely open-ended Convention1579 and those, such as Italy, which wanted to restrict recognition to trusts governed by the law of a contracting state.1580 One consequence of this for states enacting the reservation is that it will permit them to exclude the recognition of trusts created in non-contracting offshore trust havens and subject to local laws which might appear unduly insular to the court of recognition.1581 The key factor under Article 21 is the law applicable to the validity of the trust. This must be taken to mean the law applicable to the essential validity of the trust.1582 It matters not that the law governing some other aspect, such as administration, is that of a non-contracting state. This is notwithstanding that the administration of a trust may be the most important issue confronting a state of recognition and that its courts may be asked to intervene in the administration of a trust (and, if they are asked to intervene, they will have to do so by applying the law of a non-contracting state).

1575

I.e. Arts. 11–14 (the provisions on recognition of trusts). For comment on its evolution, see the von Overbeck, paras. 170–2, pp. 409–10. It is omitted from the schedule to the Recognition of Trusts Act 1987. 1578 But enactment of Art. 21 does not render the Convention’s application reciprocal, since the reservation itself does not have reciprocal effect and the contracting state whose trust is to be recognised may not itself have enacted Art. 21 (von Overbeck Report, para. 170, p. 409). 1579 Actes et Documents, 303–5. 1580 Ibid., 304. See further M Lupoi, Trusts: A Comparative Study, 353. 1581 J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, 231. This may be a positive reason for offshore states to consider ratification of the Convention. 1582 Rather than its formal validity. 1576 1577

The Contracting Status of the State Whose Law is Applicable 405

2. TIME FOR ASCERTAINING THE CONTRACTING STATUS OF THE STATE WHOSE LAW IS APPLICABLE

Where the reservation is enacted, it seems that the relevant time for assessing the contracting status of the state whose law is putatively1583 applicable under the Convention is the time when recognition is sought.1584 This facilitates recognition where a state whose law is applicable was not party to the Convention at the time of the trust’s creation, but has subsequently become so.

3. CHANGE OF THE APPLICABLE LAW FROM THAT OF A NON - CONTRACTING STATE TO THAT OF A CONTRACTING STATE

A related question is: what should happen in a state which enters the reservation if a trust is initially governed by the law of a non-contracting state, but the governing law is subsequently changed to that of a contracting state?1585 Pragmatism dictates that the trust should still be recognised,1586 especially if one takes the view that the change of applicable law creates in substance a new trust to be recognised for the first time.1587 If this is correct, then it allows a trustee1588 of a trust initially governed by the law of a non-contracting state to change the governing law to that of a contracting state, where securing recognition of the trust in a particular state becomes paramount.1589 However, Hayton points out that there is no reason why a state should not provide that the relevant time be that of the creation of the trust, if the enacting state so desires. This would mean that a trust initially created under the law of a contracting state would not be vulnerable to a subsequent bona fide change of applicable law to that of a noncontracting state.1590

1583 Until a state has decided whether to recognise the trust, it does not know whether to apply the Convention’s choice of law rules to it and accordingly cannot definitively determine that it is governed by the law of a contracting state. It can only determine whether the law putatively applicable to the trust is that of a contracting state. 1584 Von Overbeck, para. 172, p. 410. He remarks that, “The time when the trust was created has no importance.” 1585 Pursuant to Art. 10. 1586 Recognition in this scenario is favoured by von Overbeck, para. 172, p. 410; and see D Hayton, Underhill and Hayton, Law of Trusts and Trustees, 15th edn., 952–3. 1587 See the discussion of “one trust or two?” in relation to Art. 10, above. 1588 Or whichever party or parties is/are authorised to change the applicable law. 1589 It also facilitates recognition where a state was not party to the Convention at the time of the trust’s creation, but has subsequently become so. 1590 D Hayton, (1987) 36 ICLQ 260, 281, f/n 51.

ARTICLE 22—TIME-FRAME OF THE CONVENTION The Convention applies to trusts regardless of the date on which they were created. However, a Contracting State may reserve the right not to apply the Convention to trusts created before the date on which, in relation to that State, the Convention enters into force. Article 22(1) applies the Convention’s rules to all trusts falling within its scope,1591 even if the trust was created prior to the entry into force of the Convention.1592 The lack of a date requirement is rather curious, given that it could mean the retrospective imposition of the Convention’s choice of law rules onto a pre-existing trust, with potentially serious consequences for the parties to the trust and for third parties. For example, a trustee might have assumed that a trust would not be recognised in the state of his habitual residence and have acted on the basis that he was absolutely entitled to the property transferred to him. The entry into force of the Convention in that state and the subsequent entitlement of the trust to recognition therein may radically alter the picture. Conversely, it is not inconceivable, although it is readily admitted that it is unlikely,1593 that a trust which was valid according to a trust state’s preConvention choice of law rules will be invalid by the Convention rules (e.g. if it is impugned by application of another state’s mandatory rules under Article 15 or Article 16(2)).1594 One might have expected at the very least a provision stating that the Convention should not adversely affect the trust’s validity1595 or any existing rights thereunder.1596 That said, “the general principle of Hague Conventions is that they are not intended to be retrospective unless expressly or necessarily so. It is thus implicit that the coming into force of the Convention is not to affect the law to be applied in respect of anything previously done or omitted”.1597 1591 But not to “any trust, wherever or whenever created” as O Elias, Judicial Remedies in the Conflict of Laws, 169 erroneously states (emphasis added). The Convention applies only to those trusts (i) demonstrating the characteristics listed in Art. 2; (ii) falling with the scope of Art. 3; and (iii) governed by the law of a trust state. 1592 See the decision of the Supreme Court of New South Wales in Saliba v. Falzon [1998] NSWSC 302. 1593 D Hayton, (1987) 36 ICLQ 260, 281 argues that “It seems hardly possible that a trust State might consider that the Convention’s very flexible rules for finding the applicable law . . . are different from the current rules.” 1594 In states which have enacted this latter provision. 1595 It ought to be possible for a previously invalid trust to be validated by the Convention’s rules, provided that the parties have not detrimentally relied on the trust’s invalidity (e.g. if the settlor, believing the property to be held on an automatic resulting trust, has now disposed of the trust assets). 1596 Although the rights of third party holders of the trust assets will be protected to some extent by Art. 11(3)(d), second sentence (see above). 1597 D Hayton, Underhill and Hayton, Law of Trusts and Trustees 15th edn., 953.

Time Frame of the Convention 407 The Convention may accordingly be interpreted so as to apply to trusts created pre-Convention, but only in respect of matters arising after the entry into force of the Convention.1598 Despite this, von Overbeck states that the drafting of Article 22(1) was quite uncontroversial, and that doubts centred rather on whether Article 22(2), which permits a state to enter a reservation to the effect that the Convention shall apply only to trusts created on or after the date on which the Convention entered into force in that state, was necessary or appropriate.1599 Hayton comments that: “the first sentence [of Article 22] has everything to commend it: the reservation allowed by the second sentence has little. There seems little point in recognising the effect of post-Convention trusts but leaving the effects of pre-Convention trusts to the unsatisfactory, uncertain position that currently prevails in non-trust states”.1600

Some delegates argued that retroactivity was only unacceptable if it altered substantive law.1601 However, it may be retorted that: (i) parties to the trust and strangers to the trust may detrimentally rely on choice of law rules in determining the validity of the trust at its inception and the rights and duties of the parties; and (ii) a change in the applicable law brought about by enactment of the Convention may retrospectively change the substantive law which will be applied to the trust. In the event, Article 22(2) was wisely retained and has been utilised by the United Kingdom in section 1(5), Recognition of Trusts Act 1987.1602 A question may arise as to the inter-relationship of Articles 21 and 22. Von Overbeck postulates a scenario where state A has entered the date reservation under Article 22 and state B has entered the reservation under Article 21 limiting its recognition obligations to trusts governed by the law of a contracting state. What if state B is asked to recognise a trust governed by the law of state A which was created prior to the commencement date of the Convention in state A? Is state A a “contracting state” for the purpose of Article 21? Von Overbeck suggests that the answer is “yes” and that state B should recognise the state A trust, in line with the “general spirit of the Convention”.1603 It is suggested that this interpretation is correct and also accords with the literal wording of Article 21. That Article seems only to require that the applicable law be that of a contracting state at the 1598 D Hayton, (1987) 36 ICLQ 260, 282 suggests that Art. 22(1) should simply have been qualified by a provision making clear that the Convention rules do not affect “the law to be applied in respect of anything done or omitted to be done before the coming into force of this Convention.” However, it will not always be possible to determine whether e.g. a continuing breach by a trustee occurred before or after the entry into force of the Convention. See “application of the Convention in the United Kingdom: the time factor”, above. 1599 Von Overbeck, para. 174, p. 410. 1600 D Hayton, (1987) 36 ICLQ 260, 281. 1601 This was apparently the view of the United Kingdom and United States delegates. 1602 On which see “application of the Convention in the United Kingdom: the time factor”, above. 1603 Von Overbeck, para. 175, pp. 410–1.

408 Time Frame of the Convention time when recognition is sought. A contracting state which has entered a reservation as to the time frame for which it will apply the Convention under Article 22(2) is nonetheless a contracting state.1604

1604 Of course, if state B has not made the reservation in Art. 21 (so that it is bound to recognise trusts governed by the law of a non-contracting state), it cannot refuse recognition to a state A trust on the basis that the Convention is not yet applicable in state A, since the Convention is not reciprocal and this would mean that “the trusts of Contracting State A would be less well treated than the trusts of non-Contracting States”: ibid., para. 175, pp. 410–1.

ARTICLE 23—STATES CONSISTING OF SEVERAL TERRITORIAL UNITS For the purpose of identifying the law applicable under the Convention, where a State comprises several territorial units each of which has its own rules of law in respect of trusts, any reference to the law of that State is to be construed as referring to the law in force in the territorial unit in question. This provision concerns states consisting of several units which each have a system of law, such as Canada. Each unit will be treated as a state for Convention purposes, so that a settlor may choose the law of such a unit (e.g. British Columbia) to govern the trust and the applicable law in the absence of choice under Article 7 will be the unit with which the trust is most closely connected.1605 If the settlor chooses e.g. Canadian law to govern a trust, it appears that the choice is simply treated as meaningless, so that the law of closest connection has to be sought.1606 Article 23 is not scheduled to the Recognition of Trusts Act 1987.1607 This is perhaps unfortunate, since an English court may still be faced with a choice of law clause specifying “Canadian” law, or with ascertaining the Canadian state with which a trust has its closest connection in the absence of choice. That said, it is almost certain that the English court will regard each territorial unit as a “state” for Convention purposes in any event,1608 since it can only sensibly apply the Convention by identifying the law of a place having one system of law, such as British Columbia.

1605 Compare Art. 19, Hague Convention on the Law Applicable to Agency 1978. See also von Overbeck, ibid., para. 178, p. 411. 1606 Ibid., para. 177, p. 411. The only exception may be where the dispute is between the application of e.g. Ontarian or New York law and where one can infer from the choice of “Canadian” law that Ontarian law was intended by the settlor to govern the trust; see further the discussion of “choice of the law of a state comprising several territorial units” in relation to Art. 6, above. 1607 Despite this, the Convention self-evidently applies to trusts governed by English or Scottish law. 1608 Even though Dicey and Morris note (at 27–8) that for private international law purposes the word “state” normally means “the whole of a territory subject to one sovereign power.” In contrast, the word “country” is normally understood in English private international law to refer to units having their own system of law, such as New York or Ontario (ibid., 26–7).

ARTICLE 24—APPLICATION OF CONVENTION BETWEEN TERRITORIAL UNITS OF A STATE A State within which different territorial units have their own rules of law in respect of trusts is not bound to apply the Convention to conflicts solely between the laws of such units.

1. THE PERMISSIBILITY AND DESIRABILITY OF EXCLUDING THE CONVENTION ’ S APPLICATION TO DISPUTES BETWEEN A STATE ’ S TERRITORIAL UNITS

This Article makes clear that the Convention’s provisions do not necessarily apply to disputes arising within a state comprising of different territorial units.1609 In Australia, the Law Reform Commission has taken the view that the Convention should be applied as between Australian states.1610 Moreover, Prior J said in In the Estate of Webb that an Australian court should be very slow to adopt a common law rule which departs from the Convention rules.1611 In Canada, many, but not all states have extended the Convention to conflicts between Canadian states.1612 The extension of the Convention is much to be preferred for several reasons. First, it provides states compromising of several territorial units with a single, harmonised set of choice of law rules for trusts falling within the Convention’s ambit. Secondly, it avoids complex questions which might arise as to whether a conflict is solely between the units. Imagine that a state does not apply the Convention to disputes solely between territorial units and that it has to determine the ambit of the Convention. Presumably, a trust with objective connections solely to Alberta and New Brunswick does not raise a purely internal conflict, if English law is chosen to govern the trust.1613 But what if e.g. one of the members of the qualifying class of objects under a discretionary trust resides in England, but the trust is otherwise objectively connected to Alberta and New Brunswick (and no applicable law is expressly or impliedly chosen)? It becomes 1609 In Australia, see s. 7(1), Trusts (Hague Convention) Act 1991. The states are permitted to adopt the Convention inter se, however: s. 7(2). 1610 Australia Law Reform Commission Report No 58, Choice of Law, para. 9.38. See also A Dyer, “International Recognition of the Trust Concept”, (1996) 2 Trusts and Trustees 5, 9. 1611 (1992) 57 SASR 193, 204. 1612 But not Alberta (s. 1(2), International Convention Implementation Act, SA 1990) or New Brunswick (s. 2, International Trusts Act, SNB 1988). Where the Hague Convention is inapplicable, the Conflict of Laws Rules for Trusts Act may be applicable, rather than the common law (on which see J-G Castel, Canadian Conflict of Laws, 4th edn., 549). 1613 There is no requirement that a trust be sufficiently “international” for a choice of law clause to be upheld (see the discussion of “‘international’ trusts” in relation to Art. 6, above). It would appear that a choice of a foreign law to govern a trust will also be sufficient to extend a conflict beyond the confines of the territorial units of Canada for Art. 24 purposes in the example given.

Intra-United Kingdom Conflicts 411 much more difficult to say whether the conflict is “solely” between the laws of the Canadian units. The issue becomes a question of degree, with consequent uncertainty as to the scope of application of the Convention. Thirdly, a trust which has objective connections solely with Alberta and New Brunswick at the time of its creation may come to have a connection with a “foreign” state, such as England, if e.g. trust assets are invested there. If the court determines the question whether the dispute is solely between Canadian units as of the time of the trust’s creation, it will fail to recognise what is now clearly an international trust and will fail to do justice to the spirit of the Hague Convention; but if it determines the matter as of the date of recognition, a settlor, who created a trust, the recognition of which he considered to be outside the Convention’s parameters, may find that it later qualifies for Convention recognition after all (or vice-versa, if all elements of what was an international trust come to be connected solely with Alberta and New Brunswick).

2. INTRA - UNITED KINGDOM CONFLICTS

Article 24 is not scheduled to the Recognition of Trusts Act 1987, so that whether the Convention rules should be applied to an Anglo-Scottish trust is at first sight unclear. However, it will be recalled that section 1(2), Recognition of Trusts Act 1987 applies to “any other trusts arising under the law of any part of the United Kingdom”.1614 This very broad wording indicates that the Act applies to disputes as to the application of English or Scottish law. Furthermore, section 1(4) of the Act indicates that the word “state”, as used in Article 17,1615 refers to “any country or territory (whether or not a party to the Convention and whether or not forming part of the United Kingdom) which has its own system of law.” Given that England and Scotland are states for these purposes one might expect more generally that conflicts between those states would be governed by the Act.1616 Indeed, any other solution, which required application of the common law to intra-United Kingdom trusts conflicts, would create a schism in the private international law of trusts in England for no obvious reason or benefit.

1614 Emphasis added. See the discussion of “trusts arising under the law of any part of the United Kingdom” in relation to Art. 3, above. 1615 Art. 17 excludes the doctrine of renvoi (see above). 1616 See also A Anton (with P Beaumont) Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., 630.

ARTICLE 25—RELATIONSHIP TO OTHER INTERNATIONAL CONVENTIONS The Convention shall not affect any other international instrument containing provisions on matters governed by this Convention to which a Contracting State is, or becomes, a Party. This provision seeks to avoid any conflict that might arise between the Hague Trusts Convention and other international instruments and subordinates the Hague Convention to such other instruments. It applies both to Conventions to which contracting states are already party and to those to which they might subsequently become party.1617 One Convention which might interact with this Convention is the Hague Convention on Succession to Deceased Persons’ Estates 1989,1618 were it to be signed and ratified in the United Kingdom. However, interaction with that Convention may be complicated by the fact that

1617 Compare Art. 21, Rome Convention; Art. 57, Brussels Convention; Case C–406/92 The Tatry [1994] ECR I–5439. Art. 71(1), Brussels Regulation (on which see f/n 5, above) provides that “this Regulation shall not affect any convention to which the Member States are parties and which in relation to particular matters, govern jurisdiction or the recognition or enforcement of judgments.” The Regulation does not provide (as does Art. 57(1), Brussels Convention) that states may enter into new conventions after the entry into force of the Regulation. This could mean that the European institutions will henceforth have an exclusive competence to act on behalf of their member states in negotiation of international treaties within the scope of the Brussels Regulation. This is especially significant in relation to the proposed Hague Convention on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial (“the Worldwide Judgments Convention”). However, the European institutions are not themselves a party to the Hague Conference, so that the individual member states of the EU (save Denmark, which is not bound by the Brussels Regulation) will need to agree a common position at the negotiation table. See further, P Beaumont, “Current Developments: EC Law” (1999) 48 ICLQ 225, 228. 1618 The Convention has not entered into force. It has been signed by Argentina, Luxembourg and Switzerland, but only ratified in the Netherlands. See D Hayton, “The Significance of the Hague Conventions on Trusts and Succession: a Common Law Perspective” in A Borras, A Bucher, T Struycken and M Verwilghen (eds.), E Pluribus Unum, 121; P North, “Reform, but not Revolution”, (1990–I) 220 Recueil des Cours 9, 278; see also section 18 of Part One of this book. One obvious point of conflict between the two Conventions would be that Art. 15(1)(c), Hague Trusts Convention allows for the application of the mandatory rules of the law governing succession only. The United Kingdom signed (on 2 October 1973), but has not ratified, the Hague Convention on the International Administration of the Estates of Deceased Persons 1973 (in force in the Czech Republic, Portugal and Slovakia). The United Kingdom has not signed or ratified the Hague Convention on the Law Applicable to Matrimonial Property Regimes 1978 (in force in France, Luxembourg and the Netherlands).The Hague Convention on the Law Applicable to Agency 1978 (in force in Argentina, France, the Netherlands and Portugal, but not signed or ratified in the United Kingdom) may interact with the Hague Trusts Convention, given the “shapeless” description of the trust in Art. 2 of the latter, which, in referring to the placing of “control” of assets in the hands of the trustee, may include agency. See “the characteristics considered: the ‘shapeless’ trust” in relation to Art. 2, above; M Lupoi, “The Shapeless Trust”, (1995) 1(3) Trusts and Trustees 15. See generally H Verhagen, Agency in Private International Law—the Hague Convention on the Law Applicable to Agency (Dordrecht/the Hague, Martinus Nijhoff/Asser Institute, 1995).

Relationship to other International Conventions 413 it too contains a give-way provision (in Article 23(1)),1619 so that the United Kingdom would need to determine whether to allow the Trusts Convention to prevail in case of conflict (on the basis that it entered into force first) or to develop a set of rules on the interaction of the two Conventions. Article 25 is not scheduled to the Recognition of Trusts Act 1987. Prima facie, this indicates that, insofar as a conflict arises with another international instrument to which the United Kingdom is party, the Hague Trusts Convention must still be applied.1620

1619 Although it does add the phrase “unless a contrary declaration is made by the States Parties to such instrument.” 1620 Whether the United Kingdom would still be of this mind were it to ratify the Hague Succession Convention is harder to say. In practice, such a conflict will probably be avoided either by classifying the issue as a matter other than trusts, or by treating it as a mandatory rule falling within Art. 15 or 16, Hague Trusts Convention.

ARTICLE 26—RESERVATIONS TO THE CONVENTION’S SCOPE CHAPTER V—FINAL CLAUSES Any State may, at the time of signature, ratification, acceptance, approval or accession, or at the time of making a declaration in terms of Article 29, make the reservations provided for in Articles 16, 21 and 22. No other reservation shall be permitted. Any Contracting State may at any time withdraw a reservation which it has made; the reservation shall cease to have effect on the first day of the third calendar month after notification of the withdrawal. 1. FINAL CLAUSES — CHAPTER V

These clauses (Articles 26–32) deal with procedural aspects of the Convention. None is scheduled to the Recognition of Trusts Act 1987. They are included here so that the procedures for acceding to, and modifying the provisions of, the Convention can be appreciated. Only brief comment is provided.

2. APPLICATION OF ARTICLE 26

A reservation must be made not later than the time when the enacting legislation is passed. Subsequent reservation is not permitted. In contrast, any reservation made may be withdrawn at any time. This may have the unfortunate effect of encouraging states to enter a reservation, on the basis that, if it is not made immediately, it cannot be made at all and that if it proves unnecessary, the reservation may be abandoned. The United Kingdom has reserved the application of Article 16(2), which would have allowed it to apply the international mandatory rules of a state of close connection,1621 and Article 22(1), by limiting the time frame in which the Convention operates in England in section 1(5), Recognition of Trusts Act 1987.1622 Article 26(2) makes it clear that the reservations detailed are exhaustive. However, we have seen that this has not stopped the United Kingdom from omitting to schedule certain other Articles of the Convention, which narrow its scope1623 or appear self-evident.1624 1621 See “Article 16(2)—mandatory rules of a law of close connection; exclusion in the United Kingdom”, above. 1622 See “application of the Convention in the United Kingdom: the time factor”, above. 1623 Such as Arts. 13 and 25. 1624 Such as Art. 19.

ARTICLE 27—ACCESSION OF MEMBER STATES The Convention shall be open for signature by the States which were Members of the Hague Conference on Private International Law at the time of its Fifteenth Session. It shall be ratified, accepted or approved and the instruments of ratification, acceptance or approval shall be deposited with the Ministry of Foreign Affairs of the Kingdom of the Netherlands. The Convention is stated to be open for signature and ratification only by states which were Conference members at the time of the Session where the Convention was concluded.1625 However, this should cause few practical problems, since Article 28 provides a mechanism by which non-member states may subsequently accede to the Convention

1625

For comment on the evolution of Art. 27, see von Overbeck, paras. 189–90, pp. 413–14.

ARTICLE 28—ACCESSION OF NON-MEMBER STATES Any other State may accede to the Convention after it has entered into force in accordance with Article 30, paragraph 1. The instrument of accession shall be deposited with the Ministry of Foreign Affairs of the Kingdom of the Netherlands. The accession shall have effect only as regards the relations between the acceding State and those Contracting States which have not raised an objection to its accession in the twelve months after the receipt of the notification referred to in Article 32. Such an objection may also be raised by Member States at the time when they ratify, accept or approve the Convention after an accession. Any such objection shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands. Non-member states were permitted to accede to the Convention, but only after the entry into force of the Convention and subject to any objections being raised by contracting states within one year.1626 If an objection is raised, the Convention will not enter into force as between the acceding state and the objecting state. It might be thought that this matters little, since the Convention is open-ended and does not work on a reciprocal basis. However, the importance of such an objection is that the objecting state will not treat the acceding state as a “contracting state” within the meaning of the Convention. Accordingly, if the objecting state has entered the reservation in Article 21, it will not apply the Convention to a trust governed by the law of the acceding “objectionable” state.1627 A member state which becomes party to the Convention may also raise an objection, even if the non-member state acceded to the Convention before it did. In this way, even after accession, non-member states are accorded a lower, more vulnerable status than member states of the Conference.

1626 It is not necessary for member states positively to indicate their acceptance: ibid., para. 192, p. 414; and see the Swiss proposal, Working Document No 56. 1627 Ibid., para. 194, p. 414.

ARTICLE 29—APPLICATION OF THE CONVENTION TO SELECTED TERRITORIAL UNITS OF A STATE If a State has two or more territorial units in which different systems of law are applicable, it may at the time of signature, ratification, acceptance, approval or accession declare that this Convention shall extend to all of its territorial units or only to one or more of them and may modify this declaration by submitting another declaration at any time. Any such declaration shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands and shall state expressly the territorial units to which the Convention applies. If a State makes no declaration under this Article, the Convention is to extend to all territorial units of that State. Where a state consists of more than one territorial unit, it may expressly provide that the Convention will apply to one or more of those units only. In the absence of an express provision, the Convention will be deemed to apply to all territorial units of that state. However, unlike the reservations specified in Article 26, the reservation in Article 29 may be made or altered at any time.1628 In Canada, the Convention applies only to certain states. The Convention was initially applied to Alberta, British Columbia, New Brunswick, Newfoundland and Prince Edward Island. Subsequently, it was extended to cover Manitoba and Saskatchewan.1629

1628 It is thought that Art. 29 may prove problematic where a state is responsible for the international relations of another state, but where the latter otherwise has its own system of law: see ibid., para. 196, pp. 414–15 and the concerns of the Dutch delegates in relation to the Dutch Antilles in Working Document No 57. 1629 But not to Ontario. See “current status of the Convention”, above. The United Kingdom initially ratified the Convention on behalf of Great Britain and Northern Ireland, Bermuda, British Antarctic Territory, the British Virgin Islands, the Falkland Islands, Gibraltar, the Isle of Man, Saint Helena and Dependencies, South Georgia and South Sandwich Islands, and the Sovereign base areas of Akrotiri and Dhekelia in Cyprus. The Convention was subsequently extended to Montserrat, Jersey, Guernsey and the Turks and Caicos Islands (see ibid., above).

ARTICLE 30—ENTRY INTO FORCE The Convention shall enter into force on the first day of the third calendar month after the deposit of the third instrument of ratification, acceptance or approval referred to in Article 27. Thereafter the Convention shall enter into force— a) for each State ratifying, accepting or approving it subsequently, on the first day of the third calendar month after the deposit of its instrument of ratification, acceptance or approval; b) for each acceding State, on the first day of the third calendar month after the expiry of the period referred to in Article 28; c) for a territorial unit to which the Convention has been extended in conformity with Article 29, on the first day of the third calendar month after the notification referred to in that Article. Little comment is required, save to note in the case of acceding non-member states that the three month period specified in Article 30(2)(b) is additional to the one year period for objections to be raised under Article 28. Accordingly, it will take fifteen months for the Convention to enter into force in non-member states.1630

1630

See von Overbeck, para. 197, p. 415.

ARTICLE 31—DENOUNCING THE CONVENTION Any Contracting State may denounce this Convention by a formal notification in writing addressed to the Ministry of Foreign Affairs of the Kingdom of the Netherlands, depository of the Convention. The denunciation takes effect on the first day of the month following the expiration of six months after the notification is received by the depository or on such later date as is specified in the notification. The Convention may be denounced at any time. It has no fixed minimum duration. The Convention will cease to have effect upon the expiry of the minimum six month period specified above. However, a state is free to specify a longer cessation period.

ARTICLE 32—NOTIFICATION PROCESS The Ministry of Foreign Affairs of the Kingdom of the Netherlands shall notify the States Members of the Conference, and the States which have acceded in accordance with Article 28, of the following— a) the signatures and ratifications, acceptances or approvals referred to in Article 27; b) the date on which the Convention enters into force in accordance with Article 30; c) the accessions and the objections raised to accessions referred to in Article 28; d) the extensions referred to in Article 29; e) the declarations referred to in Article 20; f) the reservations or withdrawals referred to in Article 26; g) the denunciations referred to in Article 31. No comment is required on these notification procedures, which are habitually used in Hague Conventions.

SIGNATURE CLAUSE The Convention closes with the words: In witness whereof the undersigned, being duly authorised thereto, have signed this Convention. Done at The Hague, on the first day of July, 1985, in English and French, both texts being equally authentic, in a single copy which shall be deposited in the archives of the Government of the Kingdom of the Netherlands, and of which a certified copy shall be sent, through diplomatic channels, to each of the States Members of the Hague Conference on Private International Law at the date of its Fifteenth Session. The English and French texts are of equal, official status.1631

1631 Of course, the trust is more familiar in the English speaking world and there may be subtle differences between the two texts (e.g. the French tendency to speak of the “trust” in the singular, where the English version tends to refer to “trusts” in the plural; or the French statement in Art. 2 that “les biens du trust constituent une masse distincte et ne font pas partie du patrimoine du trustee”, which Lupoi notes “tends to ‘entify’ the trust” see: M Lupoi, “The Civil Law Trust” in R Atherton (ed.) The International Academy of Estate and Trust Law: Selected Papers 1997–9, ch. 4, 35, 44).

III

Conclusion 1. MUCH ADO ABOUT LITTLE ?

The Convention has been lambasted by Schoenblum as “. . . an anachronism that constitutes little more than an abstract, academic exercise in conflicts of law theory”.1632 He criticises almost every aspect of its operation and concludes that the Convention “. . . is essentially useless in terms of fostering the widespread adoption of trusts for organizing and administering individual’s worldwide wealth”.1633 A number of specific objections can be traced throughout his work. (i) The choice of law rules of the Convention pale in comparison to the sophistication and pragmatism of offshore legislation. (ii) The choice of law rules are insufficiently protective of settlor autonomy, especially in protecting trusts from claims of forced heirship. (iii) The Convention fails to cover preliminary acts. A trust which would be valid by the law chosen by the settlor may fail due to the invalidity of a will according to a law which the settlor may not select. (iv) The choice of law rules within the Convention are themselves unclear, especially in relation to testamentary trusts in the absence of a choice of law. (v) The Convention does nothing to address the need for a substantive trust institution in non-trust jurisdictions. (vi) The Convention does not explain cogently how a common law trust might operate in practice in a civil law state. (vii) The Convention covers change of applicable law, but not the key question of trust migration. (viii) The Convention provides excessive derogations from the applicable law1634 in the form of Articles 15, 16 and 18, which will drive would-be settlors offshore to find more pro-settlor autonomy protective jurisdictions.

1632 J Schoenblum, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5. 1633 Ibid., 21. 1634 See also J Schoenblum, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, ch. 11, 246.

422 Conclusion (ix) (x) (xi)

Positive progress to the introduction of the trust into non-trust states is seriously undermined by the existence of Article 13. The Convention fails to consider the tax implications of trust recognition in non-trust states. The Convention values harmonisation at the expense of allowing each trust state to tailor its own choice of law rules to make itself competitive in the trusts marketplace.

This appears to be a formidable list of objections. When one adds to it the arguments of Lupoi as to the “shapeless trust”1635 and the consternation which this might create in onshore and offshore trusts states, it might make one wonder if the Convention is fatally flawed and incapable of appealing to the interests of non-trust and trust states, and of offshore and onshore states, alike. Indeed, Schoenblum concludes that: “It should be clear that the provisions of the Hague Convention do little good and can still do much that is bad”.1636 2. THE CRITICISMS EVALUATED

It would be foolish to contend that the Convention is unflawed. The “shapeless trust” described in Article 2 provides flexibility at the expense of some clarity of exposition and may fuel concern as to what exactly enacting the Convention will entail. Nor is it wholly clear which trusts fall within Article 3, or what should be classed as a “rocket-launcher” question within Article 4. The precise protection afforded to third parties to the trust is not as clear as it might be.1637 Other matters are not addressed at all, such as the law governing the validity of a choice of law clause or formal validity of the trust. Moreover, whilst the Convention strongly favours settlor autonomy for matters within its scope, the application of Article 7, which determines the applicable law in the absence of a choice, is not altogether clear and may vary between common law and civil law states. It may also lead to testamentary trusts being governed by a law other than that of the testator’s domicile. However, it is suggested that Schoenblum’s objections are overstated, and that many of his criticisms concern issues which could not possibly have been addressed by the Hague Conference on Private International Law and which, in any event, may have fatally impeded the prospects of ratification of the Convention. 1635 See the discussion of “the characteristics considered: the ‘shapeless’ trust” in relation to Art. 2, above; M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15. Lupoi states (in Trusts: A Comparative Study, 341) that the formulation of the shapeless trust is not a criticism and that his criticism is directed to the lack of rigorous comparative law methodology in the Dyer and van Loon Report and at the Conference. But, even if “trust” states might be unlikely to accept the full implications of the shapeless trust, the uncertainty to which it gives rise (both in terms of how other states will interpret Art. 2 and in terms of precisely what institutions might be covered by it) may deter onshore (and, especially) offshore states from ratifying the Convention. 1636 J Schoenblum, (1994) 3 J Int Corp P 5, 22. 1637 See, in particular, Arts. 11(3)(d), 15(1)(d) and (f), above.

The Criticisms Evaluated 423 The Hague Conference’s role is to develop harmonised private international law conventions. It is certainly not concerned with the harmonisation of domestic law, the provision of a substantive law of trusts or with tax liability.1638 Nor would one obviously expect a private international law convention to tackle the important question of trust migration, a matter which might more naturally be left to the law identified by the Convention’s choice of law rules to determine. More generally, Hayton explains that “in the time available, no agreement could be reached on common precise rules . . . and anyhow, precise rules different from those currently prevailing in the delegate’s jurisdiction could well prevent ratification of the Convention by such jurisdiction”.1639 Inevitably, this is a Convention of compromise; where trust and non-trust worlds meet, some progress is far better than none and may prove an important step towards introduction of the trust into the latter states. Moreover, the “shapeless trust” of Article 2 is by no means a wholly negative factor, as it extends the scope, and influence, of the Convention further than might have been anticipated and allows for a degree of reciprocal recognition of civil law “analogues” by common law states. As to Schoenblum’s criticisms of the exclusions and derogations from the Convention, it would be almost inconceivable for it to have reached agreement on harmonised choice of law rules on the preliminary acts necessary to create the trust, which may differ depending on whether the initial property transfer is inter vivos or testamentary, whether movable or immovable property is in issue and whether the issue relates to essential validity, form, or capacity. Nor could the trust states have expected the trusts choice of law rules not merely to be accepted by non-trust law states, but to be permitted to prevail over those states’ rules on matters such as forced heirship, protection of third party purchasers and the law of insolvency. As Hayton notes: “Civil law jurisdictions are going to be forced . . . to find some way of dealing with the powers and rights of trustees or beneficiaries within their borders . . . .The Convention is a readily available means to understand the trust concept and to give some effect to it without affording trusts preferential treatment as concerns existing laws on matters like insolvency, succession and the protection of transferees acting in good faith”.1640

Articles 16 and 18, which preserve the forum’s control through its own international mandatory rules and public policy, rather than undermining the trust, may well support it, since it is most unlikely that, e.g. an English court would allow an inter vivos trust to be subjected to a claw-back claim by forced heirs which exists under the law governing succession to the deceased’s estate.1641 Without Articles 15, 16 and 18, prospects for ratification would have been much 1638 See D Hayton, “The Hague Convention on Trusts: A Little is Better than Nothing but Why so Little?” (1994) 3 J Int Corp P 23. 1639 Ibid., 24. 1640 Ibid., 28. 1641 Although it may not need to resort to public policy here—an English court will apply the lex successionis only to assets which it regards as still forming part of the testator’s estate on death.

424 Conclusion reduced. It is true that the application of these Articles in non-trust states is difficult to predict. However, in only rare cases will proceedings concerning the internal mechanics of the trust be brought in civil law states, which might in any event be inclined to stay their proceedings in favour of a trust jurisdiction, where their law so permits.1642 Where a judgment has been delivered in a trust state, its recognition may be required by a non-trust state according to its rules on recognition of foreign judgments.1643 Article 11 specifies the key consequences of recognition of the trust in a civil law state. It would, however, have been impossible to have spelt out the detailed consequences of recognition of the trust, since these must ultimately be determined by the law applicable to the trust and what it requires. Nor should the mitigating effect of Article 13 be overstated. Schoenblum comments that “An attorney who fails to advise the client of the risks involved and the unpredictability as to enforcement of a choice of law clause could be liable for malpractice”.1644 This, however, is precisely the point of the Article, which seeks to allay fears in non-trust states1645 that ratification of the Convention will lead to the introduction of the trust into their domestic legal systems through the back-door. Schoenblum himself states that “international trusts with worldwide assets beg for a uniformity of approach”.1646 That must be so, since even an apparently domestic trust may ultimately require recognition in a non-trust state, if e.g. assets are subsequently acquired or moved overseas. Yet Schoenblum then goes on to compare the Convention’s rules to offshore legislation, arguing that those states’ “unilateral efforts in response to market forces have reduced the debate over the Hague Convention to a side-show”.1647 It may be true that the autonomy protecting paradises to be found offshore might appear more attractive to some settlors than the circumscribed autonomy of the Convention and that individual states which view themselves primarily as competing in a global trusts market might feel more inclined to develop their own enticing solutions than to ratify the Convention. But Schoenblum’s call for a harmonised approach sits uncomfortably with his later remark that “ratification of the Convention represents . . . an undue preoccupation with multilateral harmony”.1648 Much depends on the attitude with which trust states approach the Convention. If they view the international trusts world as one where development of the most autonomy-protecting trust environment is crucial, then ratification appears undesirable. More cynically, such a state might take the view 1642 See D Hayton, (1994) 3 J Int Corp P 23. But a court with Brussels Convention Regulation jurisdiction would not be able to stay its proceedings in favour of England. 1643 Especially if the Brussels Convention’s Regulation’s recognition rules are applicable. 1644 J Schoenblum, (1994) 3 J Int Corp P 5, 7. 1645 On Lupoi’s view, the “shapeless” trust of the Convention means that few states will be nontrust states. 1646 J Schoenblum, (1994) 3 J Int Corp P 5, 13. 1647 Ibid., 21. 1648 Ibid., 22.

Looking Back and Looking Forward 425 that because the Convention is not reciprocal, its trusts can gain the benefit of recognition in contracting civil law states without taking on the perceived burden of the Convention’s choice of law rules itself. But progress on recognition of the trust in civil law states requires an attitude of flexibility and compromise and a demonstration of faith in the Convention. Compromise must be seen as better than failure, at least as a first step. If one disapproves of the idea of multilateral agreement in this area altogether, the prospects for recognition of trusts in non-trust states are poor; if one wants a precise, wholly autonomy-endorsing Convention straight away, this is something which will not happen and any attempt to formulate one would almost inevitably prevent its ratification in a great many states. In short, the Convention is an enormously significant step along the road to more widespread recognition of the trust in non-trust states. Its choice of law rules reflect the diversity of state interests which the Convention serves, as do the derogations therein. But it can only thrive in an atmosphere of co-operation, harmonisation and compromise. It is unlikely ever to gain a foothold in many offshore states or in other trust states principally concerned with developing clearer-cut, autonomy embracing rules which almost entirely “insulate” the trust from the claims of third parties.

3. LOOKING BACK AND LOOKING FORWARD

On balance, it would be fair to say that the Convention has enjoyed limited success, without having the impact in civil law states which it was hoped it would have.1648a Unfamiliarity with the trust in domestic law has proved a deterrent to its introduction into private international law.1649 This is despite the fact that the Convention contains many reservations and safeguards to protect a state from the requirement and consequences of recognition and from the use of the trust by a settlor to avoid local rules relating to such matters as forced heirship. The sporadic ratification of the Convention in trust states has hardly assisted the calls for ratification of the Convention in non-trust states.1650 Those non-trust states have little to fear from the Convention, given that, at a bare 1648a Although it seems that the Hague Convention will be implemented in Luxembourg and that its prospects of implementation in Switzerland are good: see further D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. 1649 For civil law states which do ratify the Convention, the unequal treatment of the international trust and the nearest domestic analogue can lead to strong calls for reform of that state’s domestic law. For example, M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” 137, 151 argue that the legislator in the Netherlands must decide either “that the trustee cannot have his domicile in the Netherlands, or . . . will have to provide for a separate fund within the Dutch fiducia cum amico.” 1650 T Trumpy, “Non-Corporate Entities in Private International Law: the Trust” describes the Hague Convention as “a flawed . . . effort (which probably does not merit ratification)”, [1989] Hague Yearbook of International Law 211.

426 Conclusion minimum,1651 it necessitates little more than recognising the legal personality of the trustee to sue and be sued in his official capacity. But that minimal requirement might make a civilian state legitimately pause to consider whether the ratification game is really worth the candle.1652 On the other hand, a more inclusive ratification, which recognises the separate nature of the private and fiduciary patrimony of the trustee, might prove a step too far, unless and until such a distinction is introduced into that state’s domestic legal system.1653 For some, the prospects of a happy marriage between the trust and civil law rules of property, succession etc. is a long way off: “En vérité le résultat est un progès comparé à un refus total de reconnaître une institution si essentielle à la vie juridique d’une grande partie de la planète”.1654 Although it is in the essence of private international law that a state, when called upon to apply a foreign law, will be required to recognise the existence of legal rights, obligations and constructs which do not exist in domestic law, the practicalities of attempting to enforce a trust in a state which lacks certain key features which a common lawyer would associate with that construct1655 will inevitably impede ratification. That said, the challenge facing the Convention’s drafters in formulating a Convention which would appeal to common and civil lawyers, onshore and offshore states alike was monumental. The Convention manages to produce rules which should hold little fear for common lawyers, and frequently makes little change to existing choice of law rules. It manages also to produce a scheme which offers guidance, flexibility and compromise for civil lawyers.1656 As the

1651 If maximum use of the derogations, especially those in Art. 15, is made, even the requirement that the fund be treated as separate from the trustee’s personal wealth could, on one view, be overridden. 1652 M Koppenol-Laforce, Het Haagse Trustverdrag, 277 argues that it is: “The Trust Convention is . . . a first step towards the bringing together of civil law and common law. It is to be wished that in the coming years non-trust countries will be prepared really to open up their borders to the trust.” 1653 Of course, such a distinction may already exist in would-be ratifying states which do not have the trust. 1654 G Droz, “Regards sur le Droit International Privé”, (1991–IV) 229 Recueil des Cours 9, 249. 1655 E.g. the ability to recover the trust property or its value from a third party, even where it has been mixed or substituted (which is in principle a feature of the Convention under Art. 11(3)(d), first sentence). 1656 The approach of the Convention’s drafters is described by L Maerten as “audacieuse et pragmatique”: “Le Regime International du Trust après la Convention de La Haye du 1er Juillet 1985” [1988] La Semaine Juridique 3319. E Gaillard argues that “L’urgence me paraît incontestablement dans la ratification de la Convention de la Haye tant par les États qui ignorent l’institution (pour résoudre les difficultés et les inéquités qui pesistent en la matière) que par ceux qui la connaissent (pour permettre la reconnaissance internationale de leurs trusts): “Les Enseignements de la Convention de la Haye du 1er Juillet 1985 relative à la Loi Applicable au Trust et à sa Reconnaissance”, [1990] Revue Juridique et Politique Indépendance et Cooperation 304, 314. D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation—Kollision und Vereinheitlichung—Conflicts and Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème Anniversaire, 591, 609 describes the Convention as having “a considerable and exciting potential.”

Creation and Transfer 427 influence of the trust in domestic legal systems spreads,1657 there is reason to suppose that the Convention may prove a more widespread success.1658

1657

Projects such as the European Principles of Trust Laws will aid this development. For a sympathetic attitude towards possible ratification in Switzerland, see C Reymond, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, 9e Journée des Professeurs Suisses de Droit International Privé, at http://www-isdc.ch/e/Reymond97.asp; A von Overbeck “Trusts: the Hague Convention and the Law of the Netherlands, at http://wwwisdc.unil.ch/f/vO97.asp. A report commissioned by the Swiss government has recently recommended implementation of the Convention; see L Thévenoz, Trusts in Switzerland. Ratification of the Hague Convention on Trusts and Codification of Fiduciary Transfers; see also D Hayton, “Some Major Developments in Trust Law” [2001] Private Client Business 361, 369–70. In Luxembourg, a Bill was introduced on 14 December 2000 to implement the Hague Convention and was endorsed by the Luxembourg Chamber of Commerce on 14 March 2001 (see http://www. cc.lu/avis2001/2414wje.htm). 1658

Conclusion There is little need, or point, in summarising the conclusions of the previous parts here. Only a few very general comments will be made. We have seen throughout this book the problems created by the trust, whose characteristics (and characterisation) may differ from trust state to trust state and which creates considerable complications when, as is ever more frequent, it comes into contact with the law of a non-trust state. Those complexities are substantially magnified by the close relationship of the trust to other areas of law, such as the transfer and acquisition of property, the law of succession and the protection of creditors. The rules on creating the trust in the first place may be extremely varied and complex. The Hague Trusts Convention juggles the need for settlor autonomy in trust states with the concerns of other areas of the law, within the limits of a Convention expressly designed not to introduce the trust into the domestic law of non-trust states. This is no easy task, but it may be an essential step towards the introduction of a domestic trust in civil law states, as the fertile development of the trust in Italy, in situations almost entirely objectively connected to that state, has proved. A Convention which seeks to meet the needs of various trust states (both offshore and onshore) and of non-trust states is unlikely completely to suit the needs of any of them. However, states which adopt very individualised approaches to the private international law of trusts run the risk that trusts created in their states will not be recognised as having the same effects in another jurisdiction with which the trust may come into contact, or that it might even not be recognised at all. What is needed from each state is a shared appreciation of the benefits of trust recognition on a global scale and, with that in mind, a willingness to compromise, in order ultimately to go forward. To states so inclined, the Hague Trusts Convention should be strongly commended.

Table The Current Status of the Convention1 State

Signature

Ratification/date of extension

Entry into force

Australia

17 October 1991

17 October 1991

1 January 1992

Canada (Alberta, British Columbia, New Brunswick, Newfoundland, Prince Edward Island) extended to Manitoba extended to Saskatchewan

11 October 1988

20 October 1992

1 January 1993

14 April 1994 8 June 1994

1 July 1994 1 September 1994

30 March 1990 (extension by the United Kingdom). Now applied to the Hong Kong Special Administrative Region from 1 July 1997

1 January 1992

Hong Kong

Italy

1 July 1985

21 February 1990

1 January 1992

Malta (accession)

7 December 1994

31 December 1995

1 March 1996

The Netherlands

1 July 1985

28 November 1995

1 February 1996

17 November 1989

1 January 1992

United Kingdom 10 January 1986 (on behalf of Great Britain and Northern Island, Bermuda, British Antarctic Territory, British Virgin Islands, Falkland Islands, Gibraltar, Isle of Man, Saint Helena and Dependencies,

1 A full and regularly updated status report on the Convention is provided by the Hague Conference on Private International Law at http://www.hcch.net/e/status/stat30e.html

Table 433 State

Signature

Ratification/date of extension

Entry into force

extended to Montserrat

10 January 1992

1 January 1992

extended to Jersey

20 December 1991

1 March 1992

extended to Guernsey and to Turks and Caicos Islands

28 April 1993

1 July 1993

South Goergie and South Sandwich Islands, Sovereign base areas of Akrotiri and Dhekelia in Cyprus

Cyprus

11 March 1998

France

26 November 1991

Luxembourg

1 July 1985

United States of America

13 June 1998

Appendix 1 Hague Conference on Private International Law Convention on the Law Applicable to Trusts and on their Recognition (Concluded 1 July 1985) The States signatory to the present Convention, Considering that the trust, as developed in courts of equity in common law jurisdictions and adopted with some modifications in other jurisdictions, is a unique legal institution, Desiring to establish common provisions on the law applicable to trusts and to deal with the most important issues concerning the recognition of trusts, Have resolved to conclude a Convention to this effect, and have agreed upon the following provisions—

CHAPTER I—SCOPE Article 1 This Convention specifies the law applicable to trusts and governs their recognition. Article 2 For the purposes of this Convention, the term “trust” refers to the legal relationships created—inter vivos or on death—by a person, the settlor, when assets have been placed under the control of a trustee for the benefit of a beneficiary or for a specified purpose. A trust has the following characteristics— a) the assets constitute a separate fund and are not a part of the trustee’s own estate; b) title to the trust assets stands in the name of the trustee or in the name of another person on behalf of the trustee;

436 Appendices c) the trustee has the power and the duty, in respect of which he is accountable, to manage, employ or dispose of the assets in accordance with the terms of the trust and the special duties imposed upon him by law. The reservation by the settlor of certain rights and powers, and the fact that the trustee may himself have rights as a beneficiary, are not necessarily inconsistent with the existence of a trust. Article 3 The Convention applies only to trusts created voluntarily and evidenced in writing. Article 4 The Convention does not apply to preliminary issues relating to the validity of wills or of other acts by virtue of which assets are transferred to the trustee. Article 5 The Convention does not apply to the extent that the law specified by Chapter II does not provide for trusts or the category of trusts involved.

CHAPTER II—APPLICABLE LAW Article 6 A trust shall be governed by the law chosen by the settlor. The choice must be express or be implied in the terms of the instrument creating or the writing evidencing the trust, interpreted, if necessary, in the light of the circumstances of the case. Where the law chosen under the previous paragraph does not provide for trusts or the category of trust involved, the choice shall not be effective and the law specified in Article 7 shall apply. Article 7 Where no applicable law has been chosen, a trust shall be governed by the law with which it is most closely connected. In ascertaining the law with which a trust is most closely connected reference shall be made in particular to— a) the place of administration of the trust designated by the settlor; b) the situs of the assets of the trust; c) the place of residence or business of the trustee; d) the objects of the trust and the places where they are to be fulfilled.

1. Text of the Hague Trusts Convention 437 Article 8 The law specified by Article 6 or 7 shall govern the validity of the trust, its construction, its effects, and the administration of the trust. In particular that law shall govern— a) the appointment, resignation and removal of trustees, the capacity to act as a trustee, and the Devolution of the office of trustee; b) the rights and duties of trustees among themselves; c) the right of trustees to delegate in whole or in part the discharge of their duties or the exercise of their powers; d) the power of trustees to administer or to dispose of trust assets, to create security interests in the trust assets, or to acquire new assets; e) the powers of investment of trustees; f) restrictions upon the duration of the trust, and upon the power to accumulate the income of the trust; g) the relationships between the trustees and the beneficiaries including the personal liability of the trustees to the beneficiaries; h) the variation or termination of the trust; i) the distribution of the trust assets; j) the duty of trustees to account for their administration. Article 9 In applying this Chapter a severable aspect of the trust, particularly matters of administration, may be governed by a different law. Article 10 The law applicable to the validity of the trust shall determine whether that law or the law governing a severable aspect of the trust may be replaced by another law.

CHAPTER III—RECOGNITION Article 11 A trust created in accordance with the law specified by the preceding Chapter shall be recognized as a trust. Such recognition shall imply, as a minimum, that the trust property constitutes a separate fund, that the trustee may sue and be sued in his capacity as trustee, and that he may appear or act in this capacity before a notary or any person acting in an official capacity. In so far as the law applicable to the trust requires or provides, such recognition shall imply, in particular—

438 Appendices a) that personal creditors of the trustee shall have no recourse against the trust assets; b) that the trust assets shall not form part of the trustee’s estate upon his insolvency or bankruptcy; c) that the trust assets shall not form part of the matrimonial property of the trustee or his spouse nor part of the trustee’s estate upon his death; d) that the trust assets may be recovered when the trustee, in breach of trust, has mingled trust assets with his own property or has alienated trust assets. However, the rights and obligations of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum. Article 12 Where the trustee desires to register assets, movable or immovable, or documents of title to them, he shall be entitled, in so far as this is not prohibited by or inconsistent with the law of the State where registration is sought, to do so in his capacity as trustee or in such other way that the existence of the trust is disclosed. Article 13 No State shall be bound to recognize a trust the significant elements of which, except for the choice of the applicable law, the place of administration and the habitual residence of the trustee, are more closely connected with States which do not have the institution of the trust or the category of trust involved. Article 14 The Convention shall not prevent the application of rules of law more favourable to the recognition of trusts.

CHAPTER IV—GENERAL CLAUSES Article 15 The Convention does not prevent the application of provisions of the law designated by the conflicts rules of the forum, in so far as those provisions cannot be derogated from by voluntary act, relating in particular to the following matters— a) the protection of minors and incapable parties; b) the personal and proprietary effects of marriage; c) succession rights, testate and intestate, especially the indefeasible shares of spouses and relatives; d) the transfer of title to property and security interests in property; e) the protection of creditors in matters of insolvency;

1. Text of the Hague Trusts Convention 439 f) the protection, in other respects, of third parties acting in good faith. If recognition of a trust is prevented by application of the preceding paragraph, the court shall try to give effect to the objects of the trust by other means.

Article 16 The Convention does not prevent the application of those provisions of the law of the forum which must be applied even to international situations, irrespective of rules of conflict of laws. If another State has a sufficiently close connection with a case then, in exceptional circumstances, effect may also be given to rules of that State which have the same character as mentioned in the preceding paragraph. Any Contracting State may, by way of reservation, declare that it will not apply the second paragraph of this Article.

Article 17 In the Convention the word “law” means the rules of law in force in a State other than its rules of conflict of laws.

Article 18 The provisions of the Convention may be disregarded when their application would be manifestly incompatible with public policy (ordre public).

Article 19 Nothing in the Convention shall prejudice the powers of States in fiscal matters.

Article 20 Any Contracting State may, at any time, declare that the provisions of the Convention will be extended to trusts declared by judicial decisions. This declaration shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands and will come into effect on the day when this notification is received. Article 31 is applicable to the withdrawal of this declaration in the same way as it applies to a denunciation of the Convention. Article 21 Any Contracting State may reserve the right to apply the provisions of Chapter III only to trusts the validity of which is governed by the law of a Contracting State.

440 Appendices Article 22 The Convention applies to trusts regardless of the date on which they were created. However, a Contracting State may reserve the right not to apply the Convention to trusts created before the date on which, in relation to that State, the Convention enters into force. Article 23 For the purpose of identifying the law applicable under the Convention, where a State comprises several territorial units each of which has its own rules of law in respect of trusts, any reference to the law of that State is to be construed as referring to the law in force in the territorial unit in question. Article 24 A State within which different territorial units have their own rules of law in respect of trusts is not bound to apply the Convention to conflicts solely between the laws of such units. Article 25 The Convention shall not affect any other international instrument containing provisions on matters governed by this Convention to which a Contracting State is, or becomes, a party.

CHAPTER V—FINAL CLAUSES Article 26 Any State may, at the time of signature, ratification, acceptance, approval or accession, or at the time of making a declaration in terms of Article 29, make the reservations provided for in Articles 16, 21 and 22. No other reservation shall be permitted. Any Contracting State may at any time withdraw a reservation which it has made; the reservation shall cease to have effect on the first day of the third calendar month after notification of the withdrawal. Article 27 The Convention shall be open for signature by the States which were Members of the Hague Conference on Private International Law at the time of its Fifteenth Session. It shall be ratified, accepted or approved and the instruments of ratification, acceptance or approval shall be deposited with the Ministry of Foreign Affairs of the Kingdom of the Netherlands.

1. Text of the Hague Trusts Convention 441 Article 28 Any other State may accede to the Convention after it has entered into force in accordance with Article 30, paragraph 1. The instrument of accession shall be deposited with the Ministry of Foreign Affairs of the Kingdom of the Netherlands. The accession shall have effect only as regards the relations between the acceding State and those Contracting States which have not raised an objection to its accession in the twelve months after the receipt of the notification referred to in Article 32. Such an objection may also be raised by Member States at the time when they ratify, accept or approve the Convention after an accession. Any such objection shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands. Article 29 If a State has two or more territorial units in which different systems of law are applicable, it may at the time of signature, ratification, acceptance, approval or accession declare that this Convention shall extend to all of its territorial units or only to one or more of them and may modify this declaration by submitting another declaration at any time. Any such declaration shall be notified to the Ministry of Foreign Affairs of the Kingdom of the Netherlands and shall state expressly the territorial units to which the Convention applies. If a State makes no declaration under this Article, the Convention is to extend to all territorial units of that State. Article 30 The Convention shall enter into force on the first day of the third calendar month after the deposit of the third instrument of ratification, acceptance or approval referred to in Article 27. Thereafter the Convention shall enter into force— a) for each State ratifying, accepting or approving it subsequently, on the first day of the third calendar month after the deposit of its instrument of ratification, acceptance or approval; b) for each acceding State, on the first day of the third calendar month after the expiry of the period referred to in Article 28; c) for a territorial unit to which the Convention has been extended in conformity with Article 29, on the first day of the third calendar month after the notification referred to in that Article. Article 31 Any Contracting State may denounce this Convention by a formal notification in writing addressed to the Ministry of Foreign Affairs of the Kingdom of the Netherlands, depositary of the Convention.

442 Appendices The denunciation takes effect on the first day of the month following the expiration of six months after the notification is received by the depositary or on such later date as is specified in the notification. Article 32 The Ministry of Foreign Affairs of the Kingdom of the Netherlands shall notify the States Members of the Conference, and the States which have acceded in accordance with Article 28, of the following— a) the signatures and ratifications, acceptances or approvals referred to in Article 27; b) the date on which the Convention enters into force in accordance with Article 30; c) the accessions and the objections raised to accessions referred to in Article 28; d) the extensions referred to in Article 29; e) the declarations referred to in Article 20; f) the reservations or withdrawals referred to in Article 26; g) the denunciations referred to in Article 31. In witness whereof the undersigned, being duly authorized thereto, have signed this Convention. Done at The Hague, on the first day of July, 1985, in English and French, both texts being equally authentic, in a single copy which shall be deposited in the archives of the Government of the Kingdom of the Netherlands, and of which a certified copy shall be sent, through diplomatic channels, to each of the States Members of the Hague Conference on Private International Law at the date of its Fifteenth Session.

Appendix 2 Recognition of Trusts Act 1987 (1987 c 14) An Act to enable the United Kingdom to ratify the Convention on the law applicable to trusts and on their recognition which was signed on behalf of the United Kingdom on 10th January 1986 [9 April 1987] Northern Ireland. This Act applies; see s 2(1) post.

1 Applicable law and recognition of trusts (1) The provisions of the Convention set out in the Schedule to this Act shall have the force of law in the United Kingdom. (2) Those provisions shall, so far as applicable, have effect not only in relation to the trusts described in Articles 2 and 3 of the Convention but also in relation to any other trusts of property arising under the law of any part of the United Kingdom or by virtue of a judicial decision whether in the United Kingdom or elsewhere (3) In accordance with Articles 15 and 16 such provisions of the law as are there mentioned shall, to the extent there specified, apply to the exclusion of the other provisions of the Convention. (4) In Article 17 the reference to a State includes a reference to any country or territory (whether or not a party to the Convention and whether or not forming part of the United Kingdom) which has its own system of law. (5) Article 22 shall not be construed as affecting the law to be applied in relation to anything done or omitted before the coming into force of this Act. 2 Extent (1) This Act extends to Northern Ireland. (2) Her Majesty may by Order in Council direct that this Act shall also from part of the law of the Isle of Man, any of the Channel Islands or any colony. (3) An Order in Council under subsection (2) above may modify this Act in its application to any of the territories there mentioned and may contain such supplementary provisions as Her Majesty considers appropriate. (4) An Order in Council under subsection (2) above shall be subject to annulment in pursuance of a resolution of either House of Parliament.

444 Appendices 3 Short title, commencement and application to the Crown (1) This Act may be cited as the Recognition of Trusts Act 1987. (2) This Act shall come into force on such date as the Lord Chancellor and the Lord Advocate may appoint by an order made by statutory instrument. (3) This Act binds the Crown.

SCHEDULE Section 1 CONVENTION ON THE LAW APPLICABLE TO TRUSTS AND ON THEIR RECOGNITION

CHAPTER I—SCOPE Article 1 This Convention specifies the law applicable to trusts and governs their recognition. Article 2 For the purposes of this Convention, the term “trust” refers to the legal relationship created—inter vivos or on death—by a person, the settlor, when assets have been placed under the control of a trustee for the benefit of a beneficiary or for a specified purpose. A trust has the following characteristics (a) the assets constitute a separate fund and are not a part of the trustee’s own estate; (b) title to the trust assets stands in the name of the trustee or in the name of another person on behalf of the trustee; (c) the trustee has the power and the duty, in respect of which he is accountable, to manage, employ or dispose of the assets in accordance with the terms of the trust and the special duties imposed upon him by law. The reservation by the settlor of certain rights and powers, and the tract that the trustee may himself have rights as a beneficiary, are not necessarily inconsistent with the existence of a trust. Article 3 The Convention applies only to trusts created voluntarily and evidenced in writing. Article 4 The Convention does not apply to preliminary issues relating to the validity of wills or of other acts by virtue of which assets are transferred to the trustee. Article 5 The Convention does not apply to the extent that the law specified by Chapter II does not provide for trusts or the category of trusts involved.

2. The Recognition of Trusts Act 1987 445 CHAPTER II—APPLICABLE LAW Article 6 A trust shall be governed by the law chosen by the settlor. The choice must be express or be implied in the terms of the instrument creating or the writing evidencing the trust, interpreted if necessary, in the light of the circumstances of the case. Where the law chosen under the previous paragraph does not provide for trusts or the category of trust involved, the choice shall not be effective and the law specified in Article 7 shall apply. Article 7 Where no applicable law has been chosen, a trust shall be governed by the law with which it is most closely connected. In ascertaining the law with which a trust is most closely connected reference shall be made in particular to— (a) the place of administration of the trust designated by the settlor; (b) the situs of the assets of the trust; (c) the place of residence or business of the trustee; (d) the objects of the trust and the places where they are to be fulfilled. Article 8 The law specified by Article 6 or 7 shall govern the validity of the trust, its construction, its effects and the administration of the trust. In particular that law shall govern— (a) the appointment, resignation and removal of trustees, the capacity to act as a trustee, and the devolution of the office of trustee; (b) the rights and duties of trustees among themselves; (c) the right of trustees to delegate in whole or in part the discharge of their duties or the exercise of their powers; (d) the power of trustees to administer or to dispose of trust assets, to create security interests in the trust assets, or to acquire new assets; (e) the powers of investment of trustees; (f) restrictions upon the duration of the trust, and upon the power to accumulate the income of the trust; (g) the relationships between the trustees and the beneficiaries including the personal liability of the trustees to the beneficiaries; (h) the variation of termination of the trust; (i) the distribution of the trust assets; (j) the duty of trustees to account for their administration. Article 9 In applying this Chapter a severable aspect of the trust, particularly matters of administration, may be governed by a different law. Article 10 The law applicable to the validity of the trust shall determine whether that law or the law governing a severable aspect of the trust may be replaced by another law.

446 Appendices CHAPTER III—RECOGNITION Article 11 A trust created in accordance with the law specified by the preceding Chapter shall be recognised as a trust. Such recognition shall imply, as a minimum, that the trust property constitutes a separate fund, that the trustee may sue and be sued in his capacity as trustee, and that he may appear or act in this capacity before a notary or any person acting in an official capacity. In so far as the law applicable to the trust requires or provides, such recognition shall imply in particular— (a) that personal creditors of the trustee shall have no recourse against the trust assets; (b) that the trust assets shall not form part of the trustee’s estate upon his insolvency or bankruptcy; (c) that the trust assets shall not form part of the matrimonial property of the trustee or his spouse nor part of the trustee’s estate upon his death; (d) that the trust assets may be recovered when the trustee, in breach of trust, has mingled trust assets with his own property or has alienated trust assets. However, the rights and obligations of any third party holder of the assets shall remain subject to the law determined by the choice of law rules of the forum. Article 12 Where the trustee desires to register assets, movable or immovable, or documents of title to them, he shall be entitled, in so far as this is not prohibited by or inconsistent with the law of the State where registration is sought, to do so in his capacity as trustee or in such other way that the existence of the trust is disclosed. Article 14 The Convention shall not prevent the application of rules of law more favourable to the recognition of trusts.

CHAPTER IV—GENERAL CLAUSES Article 15 The Convention does not prevent the application of provisions of the law designated by the conflicts rules of the forum, in so far as those provisions cannot be derogated from by voluntary act, relating in particular to the following matters— (a) the protection of minors and incapable parties; (b) the personal and proprietary effects of marriage; (c) succession rights, testate and intestate, especially the indefeasible shares of spouses and relatives; (d) the transfer of title to property and security interests in property; (e) the protection of creditors in matters of insolvency;

2. The Recognition of Trusts Act 1987 447 (f) the protection, in other respects, of third parties acting in good faith. If recognition of a trust is prevented by application of the preceding paragraph, the court shall try to give effect to the objects of the trust by other means. Article 16 The Convention does not prevent the application of those provisions of the law of the forum which must be applied even to international situations, irrespective of rules of conflict of laws. Article 17 In the Convention the word “law” means the rules of law in force in a State other than its rules of conflict of laws. Article 18 The provisions of the Convention may be disregarded when their application would be manifestly incompatible with public policy. Article 22 The Convention applies to trusts regardless of the date on which they were created.

Appendix 3 Explanatory Report by Professor Alfred E. von Overbeck TRANSLATION OF THE PERMANENT BUREAU Introduction THE DEVELOPMENT OF THE WORK DONE BY THE HAGUE CONFERENCE ON PRIVATE INTERNATIONAL LAW 1 The Hague Conference on Private International Law decided to include trusts in its agenda at its Fourteenth Session in 1980.1 2 There was excellent preparation for the work by way of a very thorough report by Messrs A. Dyer and H. van Loon, secretaries at the Conference’s Permanent Bureau.2 This report describes in detail the functions of trusts in those countries which have them and the private international law problems that they raise. It is indispensable to read that report in order to comprehend the Conference’s work, and reference is here made to it once and for all. 3 A Special Commission was convened which held three sessions, from 21 to 30 June 1982, from 28 February to 11 March 1983 and from 24 to 28 October 1984. It chose as its Chairman Mr C.D. van Boeschoten (Netherlands), Mr P.H. Pfund (United States) was named Vice-Chairman of the first session, and Mr D.T. Trautman (United States) as Vice-Chairman of the second and third sessions. Mr A.E. von Overbeck (Switzerland) served as Reporter starting with the second session. Experts participated from the following countries: Argentina, Australia, Austria, Canada, Denmark, France, the Federal Republic of Germany, Greece, Ireland, Israel, Italy, Japan, Luxemburg, Netherlands, Portugal, Spain, Sweden, Switzerland, United Kingdom, United States and Venezuela. Also participating were observers from the International Institute for the Unification of Private Law, the Bank for International Settlements, the Commonwealth Secretariat and the International Union of Latin Notaries. 1 Final Act of the Session, Part E, No 3. See Acts and Documents of the Fourteenth Session, 1980, Book I. Miscellaneous matters, pp. 64, 167, 189, 230, 241. 2 Report on trusts and analogous institutions, Preliminary Document No 1, May 1982 (referred to as “Dyer/Van Loon Report”).

450 Appendices 4 An ad hoc Drafting Committee, presided over by Mr A.E. Anton (United Kingdom) and comprising also Mr J.-P. Beraudo (France) and Mr B. Wiesbauer (Austria), met between the first and the second sessions, from 2 to 5 November 1982. It prepared “draft articles on the law applicable and the recognition of trusts”.3 This same Drafting Committee, joined by Mr D.T. Trautman (United States) and Mr A.E. von Overbeck, served during the second and third sessions. 5 The Special Commission adopted a “preliminary draft Convention on the law applicable to trusts and on their recognition” on 28 October 1983. The preliminary draft and the report were submitted to the governments in March 1984.4 The Governments of Argentina, the Netherlands and the Federal Republic of Germany as well as the Bank for International Settlements and the Commonwealth Secretariat offered written comments on the preliminary draft.5 6 The Fifteenth Session of the Conference was held at The Hague from 8 to 20 October 1984. It chose as its President Mr J.C. Schultsz (Netherlands). General affairs of the Conference were entrusted to the First Commission; the preparation of a Convention on trusts, which was the only subject on the agenda for definitive treatment, was entrusted to Commission II. That Commission chose as its Chairman Mr C.D. van Boeschoten (Netherlands) and as its Vice-Chairman Mr D.T. Trautman (United States). The Reporter’s task was entrusted to Mr A.E. von Overbeck (Switzerland). The number of States represented at the Fifteenth Session was markedly higher than the number of countries which had sent experts to the Special Commission meetings. The following countries were represented by delegates: Argentina, Australia, Austria, Belgium, Canada, Cyprus, Czechoslovakia, Denmark, Egypt, Finland, France, Federal Republic of Germany, Greece, Ireland, Israel, Italy, Japan, Luxemburg, Norway, Netherlands, Poland, Portugal, Spain, Sweden, Switzerland, Suriname, Turkey, United Kingdom, United States, Uruguay, Venezuela, Yugoslavia. Panama sent an observer. The observers for the Bank for International Settlements, the Commonwealth Secretariat, the International Union of Latin Notaries and the International Office for Land Registration and Surveys brought a very valuable contribution to the meetings founded on their practical experience. 7 A Drafting Committee was established. Since Mr A.E. Anton was no longer a member of the United Kingdom delegation, this Committee was chaired by the Reporter, Mr A.E. von Overbeck, and included Messrs J.-P. Beraudo (France), D.J. Hayton (United Kingdom), R. Loewe (Austria) and D.T. Trautman (United States). 8 The Commission also designated a Sub-Committee on general and final clauses, charged in particular with drafting the federal clauses but also dealing with the other final clauses. Mr J. Ru° žička (Czechoslovakia), who had chaired an analogous committee during the preparation of the draft Convention on the law applicable to contracts for 3 Preliminary Document No 7 of November 1982 (hereinafter referred to as “Draft articles of November 1982”). 4 Trusts—Applicable law and recognition, preliminary draft Convention adopted by the Special Commission and Report by Alfred E. von Overbeck, Preliminary Document No 9 of March 1984 for the attention of the Fifteenth Session (hereinafter referred to as “Report of the Special Commission”). 5 Preliminary Document No 10.

3. The Explanatory Report by Professor Alfred E. von Overbeck 451 the international sale of goods,6 chaired this Sub-Committee which included Mr A. Boggiano (Argentina), Miss M. Langlois (Canada), Mr P.H. Pfund (United States) and Mr P. Volken (Switzerland). 9 The Second Commission held fourteen sittings and the Drafting Committee and the Sub-Committee on general provisions and final clauses met, as needed, on numerous occasions between the sittings. Stress should be placed here on the important part that the members of the Permanent Bureau, in particular Mr C.A. Dyer and Mr J.H.A. van Loon, as well as the ad hoc recording secretaries, never failing in their devotion, Messrs E. Gaillard, E. Kerckhove, Mrs S. Lahne, Misses H. Muir-Watt and K.S. Williams, as well as all the secretariat of the Conference, brought to the work. 10 The draft Convention in its entirety was adopted unanimously during the plenary session of 19 October 1984 and the Final Act, containing the draft Convention, was signed on 20 October 1984.

PURPOSE OF THIS REPORT 11 This Report is intended in the first place to explain the Convention to persons who did not participate in its preparation, but it also endeavours to recount its genesis. In the part devoted to the commentary on the Convention the reader will find in general, at the beginning, a succinct account of the most important points of each article or paragraph, and then more detailed explanations based on the Fifteenth Session’s discussions and where necessary on those of the Special Commission. Reference is made to the Report of the Special Commission for additional details on its work.

THE CONVENTION’S CHARACTERISTICS 12 This Convention is intended to deal with an institution, the trust, which is known in certain Member States of the Conference, most often States of common law, but which is unknown in the majority of the civil law States of the Members of the Conference. In this it differs essentially from the other Hague Conventions which deal on the level of conflict of laws, of conflict of jurisdictions or of recognition and enforcement of judgments, with institutions such as adoption, divorce, sales contracts or maintenance obligations governed to be sure by divergent rules of private international law in different States, but known everywhere. If certain of these Conventions sought to reconcile the countries having the nationality principle and the countries having the principle of domicile, this Convention is more particularly intended to build bridges between countries of common law and countries of civil law. 6 Law applicable to contracts for the international sale of goods, draft Convention adopted by the Special Commission and Report by Arthur Taylor von Mehren, Preliminary Document No 4 of August 1984 for the attention of the Diplomatic Conference of October 1985 (hereinafter referred to as “Draft Sales Convention”).

452 Appendices 13 The particular character of the Convention also justifies requesting, in respect of it, a certain amount of indulgence on the part of those persons for whom law should be comprised only of absolutely positive and strict rules. It was impossible to deal with all the problems of detail which might arise and the Convention deliberately leaves a place for interpretation and adaptation. It will be noted for example that if the application of the Convention to trusts properly speaking is not in doubt, the question of whether analogous institutions existing in certain civil law countries also meet the criteria of the Convention will be more difficult to resolve (infra, No 26, concerning article 1). We note, along with the Chairman of the Commission, that it will be necessary to distinguish those institutions which are structurally analogous to the trust, and which fall under the Convention, from those which are only functionally analogous and which are not covered. 14 The Member States of the Conference can be divided into two groups, those which have trusts or analogous institutions and those which do not have them. The interest on the part of one group and the other for ratification of the Convention is, consequently, not identical. For the States which have trusts, the principal interest is obviously to have the trusts created under their laws recognized in the countries which do not have this institution. However, it appeared that, even in the relations among them, the common law States would find the conflict rules of the Convention useful, given that their national systems of private international law differ in this field. The interests of States which do not have trusts appear less clearly. These may on the contrary fear that the principles of their judicial systems might be shaken by the inclusion of a somewhat troubling foreign institution. In this connection it should be emphasized that there was never any question of introducing the trust into the civil law countries, but simply of furnishing to their judges the instruments which are appropriate to grasp this legal device. And this is precisely where the interest of the Convention lies for those States which do not have trusts. Since the institution is not provided for in their substantive laws, they of course also do not have rules of private international law to govern it, and they are therefore reduced to seeking laboriously to introduce the elements of the trust into their own concepts. To the contrary, the Convention puts at their disposal conflictof-laws rules for trusts; then it indicates what the recognition of a trust should consist of, but also the limits of this recognition. Several delegations from civil law countries pointed out that their jurisdictions encountered problems concerning trusts with a certain amount of frequency, particularly in connection with the estates of deceased persons. These questions may, in particular, arise in countries which like Switzerland allow foreigners domiciled there to submit the succession to their estates to their national laws, but also they may arise in countries where the objective connecting factor for testate and intestate successions is the national law. 15 From these remarks and from other sources of information it could be determined that the Convention had considerable practical interest. This is so especially where trusts are involved as to which it is not obvious at first glance that they are subject to a determined law, but for which several laws may be in line for consideration.

3. The Explanatory Report by Professor Alfred E. von Overbeck 453 The usefulness of the Convention in this field is also demonstrated by the exceptional interest with which it was greeted by the practitioners who belonged to the International Union of Latin Notaries. This organization sent as observers German, Spanish, French and Dutch notaries. 16 On the whole the delegates showed that they were very open to the arguments which were advanced on one side or the other, even where the points of view were very divergent. Solutions limiting the reach of the Convention were often rejected; thus the delegations did not want to limit the scope of the Convention by excluding certain categories of trusts, such as business trusts (infra, No 27), but, to the contrary, they encompassed within the Convention institutions analogous to the trust (infra, No 26). 17 At certain moments the fear of the delegations of countries which did not have the trust of seeing this institution utilized for fraudulent purposes did none the less manifest itself, and this brought on the fear of other delegations that no trust might ever be recognized by reason of the number of escape clauses. The delegations of the States which have trusts did all that they could to reassure their colleagues and to demonstrate that the trust is essentially a useful device; they showed that they were ready to accept the necessary precautions against those abuses which might arise. 18 In the end the delegations of the countries which do not have trusts seemed to consider the escape clauses inserted into the Convention to be sufficient. Those clauses on the other hand did not appear excessive to the delegations of the common law countries, which to the contrary indicated at the end of the session their satisfaction as to the results attained. An English delegate had at the beginning of the session designated the discussions as a “unique and historic opportunity” to bring together the countries of civil law and common law. A Canadian delegate expressed at the end of the conference with enthusiasm his conviction that full advantage had been taken of this opportunity.

THE STRUCTURE OF THE CONVENTION 19 The Convention is divided into five chapters. The first chapter on the scope of application identifies the institutions which are covered, delimits them in relation to other institutions and provides finally that the Convention does not apply to the extent that its conflict-of-laws rules lead to a law which does not provide for trusts. 20 The second chapter on the applicable law provides primarily a subjective connection to the intent of the settlor and subsidiarily an objective connection to the law with which the trust is most closely connected. This chapter deals with the scope of application of the law governing the trust, the possibility of subjecting certain of the trust elements to a different law and changes of the law applicable to the trust. 21 The third chapter on the recognition of trusts sets out what the recognition of a trust will consist of at a minimum and specifies the form in which the trust may appear in public registers. It contains provisions permitting the non-recognition of certain trusts which may appear to be improper and, on the other hand, preserves those rules of law which are more favourable to the recognition of trusts than are those of the Convention. The

454 Appendices application of Chapter III may be limited by the use of a reservation to trusts the validity of which is governed by the law of a Contracting State. 22 Chapter IV on general clauses contains rules of several types. Articles 15 and 16 allow the application of the mandatory rules of laws which are applicable to matters other than trusts under the conflict rules of the forum and under laws of immediate application. Articles 17, 18 and 19 exclude renvoi, provide for the public policy exception and exclude all effects of the Convention in fiscal matters. Articles 20, 21 and 22 respectively permit the Convention’s provisions to be extended to trusts declared by judicial decisions or to the contrary, permit the application of Chapter III to be limited to trusts connected with Contracting States and to those created after the entry into force of the Convention. Articles 23 and 24 concern States comprised of several territorial units while article 25 gives priority to other existing or future conventions dealing with the same matters. 23 Chapter V, devoted to the final clauses, contains the customary provisions of the Hague Conventions on signature, ratification, accession, entry into force and denunciation, as well as those on the implementation of the Convention in certain territorial units of composite States.

Commentary on the Convention PREAMBLE AND FIRST CHAPTER—SCOPE Article 1 24 Article 1 indicates on the one hand the institution with which the Convention deals: the trust; in this respect reference should also be made to the preamble. On the other hand, article 1 indicates the objectives of the Convention: to determine the law applicable to trusts and to govern their recognition. In contrast, a provision on the recognition of judicial or administrative decisions concerning trusts, which appeared as article 13 of the preliminary draft, was unanimously eliminated on a proposal offered by the United States delegation (Working Document No 1). Since article 1 makes no allusion to Contracting States the Convention is of general application (applicable erga omnes; infra, Nos 32–35). The institutions covered 25 The Conference’s work was aligned first of all on trusts of common law countries. However, trusts declared by judicial decisions are, in the absence of a contrary declaration by a State, excluded (infra, Nos 49–51, dealing with article 3; article 20, infra, Nos 166–9). The Special Commission debated whether the Convention’s rules should apply to institutions of the civil law countries, whether these be patterned on the trusts of common law countries or otherwise bearing the essential traits of trusts. It finally decided to limit the preliminary draft to “Anglo-American” trusts, but without so specifying in the text. The

3. The Explanatory Report by Professor Alfred E. von Overbeck 455 restriction was to appear in a preamble to be drafted later (Report of the Special Commission, Nos 20–21). 26 The Fifteenth Session, to the contrary, decided to include institutions other than the trust of common law legal systems properly speaking on the condition that these institutions met the criteria of article 2. The preamble recites that the trust “. . . adopted with some modifications in other jurisdictions, is a unique legal institution”. This phrase perhaps does not indicate with all of the clarity which one might wish that analogous institutions meeting the criteria of article 2 are included, even if they do not constitute a reception of the trust of the common law systems. Such is however the conclusion which emerges from the entirety of the Fifteenth Session’s discussions. Indeed, from the first approach to this subject and then when the problem was taken up again, voices which were favourable to the inclusion of analogous institutions were making themselves heard. Thus the Egyptian Delegate and the Polish Delegate indicated that their countries have the equivalent of the charitable trust; the Delegates of Japan and of Luxemburg wanted their countries’ institutions to be encompassed. The Delegate of Venezuela pointed out that his country had introduced the trust in 1956 but that the institution was rarely utilized. A number of delegates from other countries were also favourable to the inclusion of analogous institutions, an inclusion which finally was accepted without a formal vote. 27 The Bank for International Settlements at Basel, in its memorandum containing comments on the preliminary draft (Prel. Doc. No 10), and the Observer for that institution at the Fifteenth Session proposed the exclusion of business trusts and of trusts intended to create security interests from the scope of the Convention (Working Document No 19; cf. Dyer/Van Loon Report, Prel. Doc. No 1, p. 102 et seq.). The Observer in particular placed the conference on guard against the dangers which the recognition of trusts could involve in respect of the protection of the rights of third persons and the maintenance of the systems for security interests in the civil law countries. These dangers would be posed in particular when the trusts came from common law countries which had very lax regulation of trusts. The governmental delegates were very sensitive to the dangers pointed out by the Observer of the BIS. However they thought that they should ward the dangers off by means of various escape clauses provided in the Convention and not by the exclusion of certain categories of trusts, particularly business trusts which would very much diminish the coverage of the Convention. Also the concept of the ‘business trust’ seemed too uncertain to be utilized in the Convention. The law applicable to trusts 28 The first objective of the Convention—to determine the law applicable to trusts—is easily understood: what is involved is the creation of conflict-of-laws rules for trusts. Such rules will be particularly useful for the civil law States which, since they do not provide for trusts in their internal law, have no conflicts rules for the purpose of dealing with them. These countries are therefore reduced to constructions which are sometimes strained in order to try to capture this institution, which is unknown in their conflict-oflaws systems, as a French delegate very well explained (cf. Report of the Special Commission, No 22).

456 Appendices It is also in this respect that the Convention seems to have an interest in relations between the common law States; indeed, controversies exist over the connecting factor for trusts which the Convention might usefully eliminate. Recognition of trusts 29 The aim of the second objective of the Convention—to govern the recognition of trusts—is more difficult to grasp. It might indeed be argued that, from the very fact that a trust is valid under the law designated by the conflicts rules of the Convention, it should be recognized and deploy its effects in the other Contracting States. Doubts about the utility of this chapter devoted to recognition were expressed again at the Fifteenth Session. 30 With this same view the Government and the delegation of the Netherlands proposed not to speak of recognition but rather of the effects of trusts (Prel. Doc. No 10; Working Document No 14). In support of this proposal they invoked the argument mentioned in the preceding paragraph as well as the fear that the concept of recognition might imply that the trust would deploy its effects by means of a sort of writ of exequatur. Objection was made to this proposal on the grounds that the text already mentioned the effects of a trust in another sense in article 8 and that the term “recognition” was not in current usage reserved for judicial decisions. Reference was also made to the fact that a trust, without being a legal entity, offered certain analogies with such entities which justified the use of the term “recognition” (Report of the Special Commission, No 23, last paragraph). The proposal was rejected by a strong majority. The conference also decided to delete the words “effects of . . .” which appeared within square brackets in the title to Chapter III of the preliminary draft. 31 The conference took into account the fact that for trusts the situation is posed in a way which is different from that for other institutions, such as marriage or contracts of sale for example, which are known in all countries. On the contrary, the trust is an institution which in principle is unknown to the judges of civil law countries. They might stand perplexed before the pure and simple affirmation that a trust, from the mere fact that it is in conformity with the law designated by the Convention, should deploy effects in their countries. If the Convention is wanted to be efficacious, it is necessary that it indicate, at least on the principal points, what recognition will consist of and what the effects will be that the trust will deploy. At the same time the provisions on recognition offer guarantees for countries which do not have trusts since they limit the effects which must be recognized (see in particular article 15). The application of the Convention erga omnes 32 Article 1 of the preliminary draft had two variants. Under both variants the provisions on the applicable law were to be of general application and cover all trusts which might be encountered in the Contracting States. In contrast, where recognition was concerned, variant A limited this to relations between Contracting States, which meant under article 11, first paragraph, trusts created in accordance with the law of a Contracting State. 33 The problem which the choice between the two variants raised was usually referred to as one of “reciprocity”. In reality this was not the most appropriate term. Indeed, it

3. The Explanatory Report by Professor Alfred E. von Overbeck 457 designates in general the fact that a Contracting State recognizes for example a judgment coming from another Contracting State, on the condition that the latter State do the same. In this case, to the contrary, the States which do not have trusts will recognize trusts established under the laws of other Contracting States. But the reciprocity on the part of these latter States is not a real counterpart since precisely the former States do not have the institution. The question was rather whether the States which do not have trusts wanted in a general way to declare themselves ready to recognize this institution, or to recognize it only when it is connected to another Contracting State. In favour of the second solution the argument could be advanced on one hand that it was preferable for the civil law countries to recognize only trusts deriving from a limited number of known legal systems and not from just any country in the world. On the other hand the fact that their trusts would be recognized only if they ratified the Convention could stimulate the interest of the common law States in becoming Parties to the Convention. The Special Commission preferred on an indicative vote the restrictive variant and decided to remit the decision to the plenary conference (Report of the Special Commission, Nos 25–29). 34 At the Fifteenth Session the question of the Convention’s application erga omnes was left open at the time of first discussion of article 1, the idea being that it would be preferable to decide when the Convention’s contours had been sketched with more precision. This discussion took place at the end of the first reading on the text of variant A as it appeared in the preliminary draft and of variant B slightly modified as to its drafting (Working Document No 55). An evolution towards the application of the Convention erga omnes then appeared. It was in particular pointed out that this system was in conformity with the objectives of the Convention, and of private international law in general, to facilitate the legal relationships of individuals on the international level. The delegations of the common law countries were all favourable to the system of universality, but this point of view was also shared by a number of delegates from civil law countries. It appeared also at this point in time that the common law States wanted to establish a Convention even for the relations among themselves, which would take the importance away from the argument that the restrictive system would facilitate the accession of these States. On the other hand the diverse guarantees in favour of States which did not have trusts now appearing in the text made it less necessary to limit recognition to those trusts which are connected to Contracting States. The technical difficulties of a Convention which was partially of general application (for the applicable law) and partially of a limited scope of application (for recognition) were also pointed out. How would one deal with a trust to which several laws might be applicable under article 9 when one of these laws might not be the law of a Contracting State? Finally it was pointed out that the criterion of a trust subject to the law of a Contracting State was not always relevant since the settlor could very well have designated that law, even though the trust in fact had closer ties with the law of an non-Contracting State. 35 None the less a certain number of delegations from civil law countries still wanted to restrict recognition to trusts created under the law of a Contracting State. In order not to

458 Appendices prevent these States from becoming Parties to the Convention, the Secretary General suggested a reservation. The Fifteenth Session accepted the principle of the application erga omnes by a majority of three-fifths. But unanimously it accepted the possibility of a reservation against this system. Following these decisions article 1 was adopted in its present text and the possibility of a reservation was inserted in article 21 (infra, Nos 170–172). Article 2 36 This provision may appear as a definition of a trust. In reality, article 2 simply tries to indicate the characteristics which an institution must show—whether this is a trust from a common law country or an analogous institution from another country—in order to fall under the Convention’s coverage. 37 Several delegates fought against the idea of defining trusts in the Convention. Given the numerous types of trusts and the divergent definitions offered by the authors in common law countries, this would have been moreover an impossible task. 38 The text was taken in its essentials from article 4 of the preliminary draft. The discussions at the Fifteenth Session bore above all on the most appropriate drafting of the provision. The text of article 4 of the preliminary draft was criticized for having a first paragraph which looked like a definition, while the second and third paragraphs seemed to contain substantive rules. This is the reason why the second paragraph of article 2 now appears in the form of an enumeration of the characteristics of trusts. First paragraph 39 This provision sets out the elements which are considered in English and American legal writings as being essential in order that there be a trust: a transfer of assets, a settlor, a trustee and beneficiaries. It was with a thought above all to charitable trusts that the statement was added that the assets could be devoted to a specified purpose. 40 The term “legal relationships” gave rise to some difficulties. The Special Commission had employed the expression “fiduciary relationship”. This formulation was inspired by the fiduciary duties that English and American law impose on a trustee. Certain experts from civil law countries however had objections to the use of the term “fiduciaire”, which had no meaning in French law. They preferred to speak of “legal relationships”. There was moreover criticism of the use of the plural in “legal relationships”, since this might make one think that a trust was a triangular operation analogous for example to agency relationships. In fact, the essence of a trust lies in the fiduciary relationship between the trustee and the beneficiaries. But it was feared that the use of the singular might lead to misunderstanding of the complexity of the relationships that arise from a trust, and so the plural was finally preferred. 41 The words “a person, the settlor” also apply to limited companies and corporations. In contrast, a trust created by a court does not fall within the scope of the Convention (infra, Nos 49–51, dealing with article 3). It goes without saying that the trustee may also be a limited company or corporation. The same goes for the beneficiary, particularly where a charitable trust or a trust for the purpose of creating security interests is involved.

3. The Explanatory Report by Professor Alfred E. von Overbeck 459 42 It was not absolutely certain under the preliminary draft that the Convention would apply to testamentary trusts. Thus the specification “inter vivos or on death” was inserted during the discussions at the Fifteenth Session. 43 The drafting of the first paragraph was also improved by the replacement of the words “have the effect of placing assets under the control. . .” by “when assets have been placed under the control. . .”. This language expressed the fact that the transfer of the assets is a prior condition to the creation of a trust (see also infra, No 54, dealing with article 4). Second paragraph 44 Sub-paragraph a has to do with the assets of the trust. The conference wanted to emphasize that the trust assets constitute a fund separate from the trustee’s own estate. This important element will appear again in article 11, second paragraph, in the context of the recognition of trusts. 45 It may be that a third person will hold the title to certain assets belonging to the trust, but on behalf of the trustee. Sub-paragraph b picks up the idea expressed by the second sentence of the second paragraph of article 4 of the preliminary draft in a more detailed formulation which had already been proposed at the Special Commission (Report of the Special Commission, No 44). 46 Sub-paragraph c describes the fiduciary duties of the trustee. It expresses the trustee’s power and duty to manage the assets and to dispose of them in accordance with the settlor’s intent and the rules of law and equity. The statement that the trustee is accountable expresses the idea that the beneficiaries have rights against the trustee which they may assert in a court of equity and also suggests the supervisory powers of those courts. Third paragraph 47 This paragraph makes it clear that, contrary to the “traditional” image of trusts, the roles of the different persons involved may be mingled. Indeed, the settlor of the trust does not necessarily disappear completely; he may still retain for himself certain decisions. In his turn, the trustee may have certain interests which make him also a beneficiary. It seemed desirable to draw attention to these possibilities although they seem to be rarely utilized in practice. Article 3 48 This provision delimits in two respects the circle of the trusts covered by the Convention; such trusts must be created voluntarily and evidenced in writing. The text was taken just as it was from the preliminary draft. The limitation to voluntary trusts 49 Only trusts created voluntarily by the settlor fall under the coverage of the Convention. In particular, the Convention is not applicable to trusts created by operation

460 Appendices of law or by judicial decision. However, article 20 permits Contracting States to extend the Convention’s application to these latter trusts (infra, Nos 166–169). The exclusion of judicial trusts extends to constructive trusts imposed by the courts and to trusts that the courts create by virtue of an express provision of law. In contrast it seems that a trust voluntarily established by a divorced spouse in order to carry out the obligation which has been imposed on that spouse to transfer certain assets to the other spouse and to their children would fall under the coverage of the Convention. A trust may also be considered as being voluntarily created, for example, if it is established for the purpose of fulfilling a maintenance obligation and is then approved by a court. 50 The Fifteenth Session also decided that mention should be made in the report of the fact that certain elements of a voluntary trust may derive from a judicial decision. Thus it may happen that a court has to designate the person who will serve as trustee because the settlor has failed to do so. Finally the variation of a trust by judicial decision should not take away from it its voluntary character. 51 During the discussion on judicial trusts, the problem of the resulting trust was raised. This may be involved, for example, in cases where the purposes of an express trust have been fulfilled, but the trustee continues to hold certain assets under a resulting trust. The opinion was expressed at the Fifteenth Session that such resulting trusts could fall under the coverage of the Convention (cf. Dyer/Van Loon Report, No 110). This interpretation also corresponds to that of the Special Commission which had specifically omitted the words “constituted expressly” which would have excluded resulting trusts (Report of the Special Commission, Nos 34 and 35). Evidence in writing 52 Article 3 states in addition, drawing its inspiration from article 5, No 6, of the Brussels Convention,7 “that the trust must be evidenced in writing”. In fact, a trust may be established orally or by simple delivery of the goods to the trustee, although the execution of an instrument in writing is more customary. From this it seems reasonable to require at least evidence in writing. This may even consist of a writing in which the trustee recites the intentions expressed orally by the settlor (on the question of the form of the trust itself, see infra, No 55, dealing with article 4, and Nos 82–83, dealing with article 8). Article 4 53 Article 4 deals with the delimitation—which is particularly delicate—of the Convention’s scope of application in respect of certain institutions to which resort must be had in order for a trust to spring into life and which give rise to legal operations which are distinct from the trust and preliminary to its inception. The image employed was that of the launcher and the rocket; it is always necessary to have a “launcher’” for example 7 Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters of 27 September 1968, amended by the Accession Conventions of 9 October 1978 and of 25 October 1982, Official Journal of the European Communities, No C 97/2 dated 11 April 1983 (hereinafter referred to as “Brussels Convention”).

3. The Explanatory Report by Professor Alfred E. von Overbeck 461 a will, a gift or another act with legal effects, which then launches the “rocket”, the trust. The preliminary act with legal effects, the “launcher”, does not fall under the Convention’s coverage. 54 A transfer of assets to the trustee is a sine qua non condition for the creation of the trust. But the law designated by the Convention applies only to the establishment of the trust itself, and not to the validity of the act by which the transfer of assets is carried out. This act is entirely governed by the law to which the conflicts rules of the forum submit it. It may be moreover that different laws will be applicable for the substance and for the form of this act, or yet for the capacity of the person who has effected it. If it turns out that under the applicable law the transfer is not valid, one may consider at the start that the trust has not come into existence since an essential element is lacking. It may be that the law applicable to a trust recognized as such will encroach on the area of another law designated by the forum’s conflict rules. Under article 15, first paragraph, it is then that other law which will prevail, but only—and this is its difference from article 4—as concerns the mandatory rules of that other law (infra, Nos 136–146). By virtue of article 15, the law applicable to the trust must yield to the provisions for forced heirship of the law governing the succession to property upon death as designated by the forum’s conflicts rules but not to the rules of that law designating in the absence of testamentary dispositions the heirs who do not have forced shares. To the contrary, a will by which assets are placed in trust completely escapes the reach of the Convention under article 4. 55 Article 4 is intended to exclude from the Convention’s scope of application both the substantive validity and the formal validity of the transfers which are preliminary to the creation of the trust. As concerns form, it may be difficult in certain cases to distinguish the form of validity of the transfer from that of the trust itself. However, according to the explanation given by the delegates of common law countries, what counts is that the instrument of transfer, the “launcher”, be valid as to its form since the formal validity of the trust itself is a rather theoretical question (see infra, Nos 82–83, dealing with article 8, and the Report of the Special Commission, No 70). 56 As concerns the question of the substantive validity, it was pointed out that certain provisions such as the rule against perpetuities might, depending on the case, concern either the will, in other words the preliminary instrument of transfer, or the trust itself. 57 The words “assets or transfers to the trustee” are completely clear when the settlor and the trustee are distinct persons. In contrast, one may doubt whether they cover the case of the declaration of trust in which these two persons are mingled: the owner of assets declares that henceforth he will hold these assets as a trustee. The Commission unanimously accepted that the acts by which this change in the capacity in which the assets were held was effectuated must also be envisaged by article 4 and therefore excluded from the Convention’s scope. A proposal was offered to express this by the words “acts by which assets are placed under the control of a trustee for trust purposes” (Working Document No 2). The Fifteenth Session, after having accepted this proposal, returned however in the end to the terms used in article 2 of the preliminary draft. It was thought on the one hand that this expressed rather clearly the idea that article 4 also applied to the declaration of trust, and on the other hand it was not desired to take up here the allusion to control which appears in a different perspective in article 2, first paragraph. A proposal to make

462 Appendices express allusion to the declaration of trust (Working Document No 60) was offered on the second reading, but did not receive the absolute majority required in order that the discussion be reopened. 58 Finally, the word “preliminary” expresses the idea that the instrument of transfer such as the will pre-exists the trust; a proposal to delete this word was rejected by a large majority. 59 Concerning article 4, the question was also discussed as to whether capacity was one of the preliminary questions excluded from the Convention’s scope. Here were opposed the points of view of the civil law States which deal with the connecting factor for capacity separately, and that of the common law countries which tend to subject capacity to the law governing the validity of the act which is in question. The conference received a proposal to exclude expressly ‘the capacity of the settlor, the trustee or the beneficiary’ (Working Document No 30 see also the Report of the Special Commission, No 71). This formulation gave rise in particular to the objection that the beneficiaries do not always need to have capacity. In the end the Commission did not expressly exclude general capacity, although a consensus emerged that this was not to be governed by the Convention, while the specific capacity to become a trustee was to be so governed (see infra, No 85, dealing with article 8, second paragraph, sub-paragraph a). 60 In the end article 4 did not undergo important modifications by comparison with article 2 of the preliminary draft. The only changes were, in the French text, the replacement of the word “acte” by “acte juridique” and, in the English text, the replacement of the word “instruments” by “acts’”this latter term seeming to be somewhat broader. Article 5 61 This article provides that the Convention is not applicable if its conflicts rules designate the law of a State which does not have trusts. The same result occurs if the applicable law does not provide for the category of trusts that the settlor tried to create. In fact, the legal systems which provide for this institution in principle differ substantially as to types of trusts which are permitted. Article 5 should not come into play very often. For one thing, it is not very probable that the settlor will choose a law which does not provide for the trust which is envisaged. And even if this occurs, the trust will still fall under the Convention’s coverage if the objectively applicable law provides for such a trust (article 6, second paragraph, and 7, first paragraph, infra, Nos 70 and 73). One would think that, given the broad power to assess the situation that article 7 confers on a court, the judge will have a tendency to conclude that a trust is most closely connected with a State which has this institution. 62 What will happen if the law governing a severable aspect of the trust, in particular administration, does not provide for trusts or the category of trusts involved? The Special Commission left this question open (Report of the Special Commission, No 50). The Fifteenth Session had before it an Australian proposal to apply the Convention none the less in this case (Working Document No 7). There was no decision on this proposal.

3. The Explanatory Report by Professor Alfred E. von Overbeck 463 In contrast, another Australian proposal to consider as being void the choice of a law which does not provide for trusts was discussed in detail and accepted in principle. This is recounted in connection with article 9 on dépaçage (infra, Nos 95–96). The result is that the choice of a law which does not provide for trusts or the category of trusts involved, to govern a severable aspect of the trust, is void and that such aspect should be subjected either to the law which is objectively applicable under article 7 or to the law governing the validity of the trust. In the end it seems very improbable that a law which does not provide for trusts or the category of trusts involved will have to be applied to an aspect of a trust the validity of which is governed by the Convention. If this situation were to arise, the aspect in question would escape from the Convention’s coverage.

CHAPTER II—APPLICABLE LAW Article 6 63 This provision adopts for trusts the principle of party autonomy. The first paragraph repeats the substance of the first two paragraphs of article 6 of the preliminary draft, while the second paragraph, which takes away all effects from the choice of a law which does not provide for trusts or the category of trusts involved, was added by the Fifteenth Session. To the contrary, the third paragraph of article 6 of the preliminary draft has disappeared, the question of dépaçage now being entirely governed by article 9. First paragraph 64 The first sentence gives the settlor the possibility of choosing the applicable law. The second sentence governs the manners in which the choice may be made, which can be express or implied. The implied choice must derive from the terms of the instrument creating or the writing evidencing, the trust (article 3), interpreted if necessary in the light of the circumstances of the case. This choice goes to the substantive law, not to the conflict rules (article 17). The circle of laws which can be chosen is unlimited. 65 The experts of the Special Commission and the delegations at the Fifteenth Session were unanimous in accepting the possibility of a choice of the applicable law. In contrast, the requirement of an objective connection between the trust and the chosen law, which was set aside by a weak majority at the Special Commission, was taken up again at the Fifteenth Session in the same terms by the Greek delegation (Working Document No 32; Report of the Special Commission, No 54): This choice may be disregarded where there is no [substantial] connection with the chosen law. This idea met with a certain amount of sympathy, but the text was criticized for not furnishing the judge with sufficiently precise criteria. The view prevailed that it was preferable to curb improper choices in the provision which was going to become article 13, or else by means of a reservation in accordance with a proposal which the conference then

464 Appendices had before it (Working Document No 28). So the proposal quoted above was rejected by a clear majority. 66 The Fifteenth Session also did not adopt an intermediate proposal suggested in the course of the discussion under which there would be complete freedom of choice of the applicable law, but only on the condition that the trust be international in character. 67 The second sentence of article 6, first paragraph, gave rise to rather long discussions, both of substance and of form as concerns choices which are not express. First of all, a proposal dealing with the substance offered to take into account in the search for the settlor’s intent the settlor’s interest or the settlor’s purpose (Working Document No 11, subsequently made more precise). This proposal was criticized for introducing a purely hypothetical intention or even an element of the objective connecting factor. It was also pointed out that the interest of one party to a lawsuit might be that the trust be valid and that the other party’s interests might be that it not be valid and thus that preference could not be given systematically to the validity of the trust. Finally it was observed that these terms referred to the intent of the settlor to create a valid trust, which intent was self-evident. These observations led to the withdrawal of the proposal, but this was done with the idea that the settlor’s intent to create a trust should be understood in the expression “circumstances of the case”. 68 The conference discussed the form of the entirety of the second sentence on the basis of a proposal which followed closely the Rome Convention of 19 June 1980 on the Law Applicable to Contractual Obligations8 (Working Document No 20): A trust shall be governed by the law chosen by the settlor. The choice must be expressed or demonstrated with reasonable certainty by the terms of the instrument creating or the writing evidencing the trust, interpreted if necessary in the light of the circumstances of the case. The positive form of this provision was generally preferred to the negative form of the preliminary draft, but on the other hand it was pointed out that it was not absolutely necessary to conform, where trusts were involved, to the Rome Convention’s formulations and that if harmonization was being sought it was better to draw closer to the texts of the other Hague Conventions. In particular reference was made to the preliminary draft Convention on sales of goods, which contains the following formulation (article 7, first paragraph): A contract of sale is governed by the law chosen by the parties. The parties’ agreement on this choice must be express or be demonstrated, in the light of the situation as a whole, by the terms of the contract and the circumstances of the case. Such a choice may be limited to a part of the contract. A proposal to take this approach was adopted in principle (Working Document No 20), adapted by the Drafting Committee, it became the present second sentence of article 6, first paragraph.

8 Convention of 19 June 1980 on the Law Applicable to Contractual Obligations, Official Journal of the European Communities L 266/1, of 9 October 1980 (hereinafter referred to as “Rome Convention”).

3. The Explanatory Report by Professor Alfred E. von Overbeck 465 69 The meaning of the second sentence is that the judge should start out by taking as a basis the instrument creating or the writing evidencing the trust. If that does not allow him to answer the question, he may interpret these writings in the light of the circumstances of the case. The words “if necessary” which were already in the preliminary draft, were maintained by a large majority at the Fifteenth Session; they are intended to express the idea that the circumstances should only be consulted if the writings do not allow for a clear answer. Inversely, the formulation adopted signifies that an implied intention may not be extracted only from the circumstances of the case while completely ignoring the writings. (See also Report of the Special Commission, Nos 58 and 59). Second paragraph 70 The second paragraph deals with the cases where the settlor has chosen a law which does not provide for trusts, or for the type of trust that he intends to create. Such a choice is void and the trust’s connection will be found objectively under article 7. The choice of a law which does not provide for trusts for a severable element of the trust is also void (supra, No 62, dealing with article 5, and infra, No 73, dealing with article 7, Nos 95–96, dealing with article 9). Article 7 71 Article 7 provides the objective connecting factor of the trust in two cases, first of all when no applicable law has been designated in accordance with article 6, first paragraph, and secondly, pursuant to article 6, second paragraph, when the law chosen does not provide for trusts or the category of trusts involved. 72 The principle of the linkage is expressed in the first paragraph: the trust is governed by the law with which it is most closely connected. In order to assist in grasping this concept of the closest connection, the second paragraph of article 7 offers as examples the four elements set out in sub-paragraphs a, b, c and d. These criteria are all in principle on the same footing; however, the conference has given them their places by order of importance so that it might be said that there is among them a certain implicit hierarchy (infra, No 77). 73 First of all it should be emphasized that, in contrast to article 7 of the preliminary draft, article 7 of the Convention applies not only when there has been no choice, but also where the choice of the applicable law is ineffective because the law chosen does not provide for trusts or the category of trusts involved. Such a choice cannot therefore take the case out from under the Convention’s application (supra, No 61, dealing with article 5). This solution results from the adoption of an Israeli proposal (Working Document No 39). The Fifteenth Session was of the opinion that it was not necessary in the first paragraph of article 7 to refer expressly to article 6. 74 The first paragraph was the subject of a proposal by the delegation of the United States, which would have led in principle to abandonment of the idea of an applicable law. This proposal, which ultimately was withdrawn is discussed in connection with article 9 (infra, No 94). 75 For the second paragraph, the text of the preliminary draft was reproduced almost without change. Only the order of the criteria was somewhat modified, the former

466 Appendices sub-paragraph a becoming sub-paragraph d. The formulation “Reference may be made. . .” was replaced by “Reference shall be made. . .” following an observation by the Austrian Delegate, to the effect that the first formulation would encounter in his country constitutional difficulties. On a proposal of the Irish delegation (Working Document No 41), the conference examined again the possibility of including certain other criteria, but finally it decided not to extend the list which had already been set out in the preliminary draft. 76 The system of the preliminary draft, as concerns the second paragraph, was however brought back fundamentally into issue by a Spanish proposal which provided a system of fixed connecting factors, subsidiary one to the other (Working Document No 29). The law of the place of administration of the trust was to apply in the first place. For the case in which that law did not provide for trusts or the category of trusts involved, there was to be applicable subsidiarily the law of the situs of the trust assets and further, subsidiarily to that law, the law of the habitual residence or of the place of business of the trustee. The Fifteenth Session considered this text only under the aspect of its fixed connecting factors, without taking into account the solution in favour of trusts which this proposed cascade system brought along with it, in that it gave three possibilities for subjecting the trust to a law which provided for this institution. The proposal found a certain echo among the observers and among certain delegates from the civil law countries who regretted the lack of precision and the too great flexibility of the preliminary draft’s approach. However, the delegations of the common law countries were unanimous in thinking that a fixed hierarchy of factors was not acceptable. Certain delegates from civil law countries then favoured an intermediary system involving recourse to the concept of the closest connection, but matching that with a hierarchy of criteria. 77 In the end the idea prevailed that the preliminary draft constituted a reasonable compromise between the civil law conceptions and the common law conceptions. The delegations of the civil law countries also agreed that the second paragraph, by the order in which the criteria were set forth comprised an implicit hierarchy which would satisfy the needs of practice. But the text will also allow the judges of the common law countries to take into account, as they have the habit of doing, all factors at the same time. With this in view the conference placed the reference to the objects of the trust and the places where they were to be fulfilled in the last place, although it appeared in the first place in the preliminary draft. 78 The Fifteenth Session received an Irish proposal to delete from sub-paragraph a the qualification “designated by the settlor”. Under this proposal the place of administration would be an important element, whether or not it was designated by the settlor. The reply was however given to this argument that, given the implicit hierarchy of the criteria of article 7, it was important that sub-paragraph a be limited to cases where the settlor had designated the place of administration. This element might especially come into consideration when the designation by the settlor of the place of administration of the trust was not sufficient to establish, under article 6, an implied choice of the applicable law.

3. The Explanatory Report by Professor Alfred E. von Overbeck 467 Finally, it was thought that the place of residence or business of the trustee, which is provided for in sub-paragraph c, amounts in fact to designating the actual place of administration. Thus this proposal was rejected by a strong majority. 79 The other sub-paragraphs of the second paragraph were taken over from the preliminary draft with minor drafting changes. 80 As was the case with article 7 of the preliminary draft, article 7 of the Convention does not indicate the time as of which the diverse criteria should be evaluated. Several delegations had however wanted this to be specified, but during the discussion it turned out that fixing the time would involve difficulties and that all of the criteria could not be evaluated as of the same time. The conference decided not to add anything to article 7, but deferred the question of the effects of changes in the connecting factors to the discussion of articles 9 and 10 of the preliminary draft. It will be seen below that in the end article 10 dealt with the question of the replacement of one law by another, but not with the time as of which the criteria of article 7, second paragraph, should be evaluated (infra, Nos 97–103). Article 8 81 Article 8 contains a detailed but non-exhaustive enumeration of the questions which are subject to the law governing the trust under articles 6 and 7. It follows in this the example of several recent Conventions. In this Convention such an enumeration may contribute to showing what the problems are which arise in connection with trusts and thus complement article 2. The law specified by article 6 or 7 will not always be a single law since questions concerning a severable aspect of the trust may be subjected to a different law (article 9; see infra, No 88). 82 Article 8 deals with substantive issues without expressly excluding issues of form. It will be recalled that the Convention applies only to trusts evidenced in writing (article 3). Must a trust also meet the formal requirements of the law which is applicable to it under articles 6 and 7? The Special Commission discussed this question in considerable depth, it decided in the end that formal validity should not be governed by the Convention and it deleted a supplementary paragraph to article 8 which included this issue (Report of the Special Commission, Nos 69–71). 83 At the Fifteenth Session the opinion was expressed that in addition to the requirement of evidence in writing supplementary formal requirements of the law governing the trust would have to be observed in the application of article 8. This issue, raised in connection with the discussion of the current article 4, was deferred but was not taken up again thereafter. It appeared in the course of the discussion on article 4 that problems of form concerned rather the preliminary instrument by which the goods were transferred to the trust, which instrument does not fall under the Convention’s coverage, than the trust itself (supra, No 55, dealing with article 4). 84 The list of article 8, second paragraph, already appeared in its essentials in article 6 of the draft articles of November 1982.

468 Appendices The Fifteenth Session did not bring many modifications to the text. Substantive issues were taken up only in connection with sub-paragraphs a, c and g; for the rest the changes were essentially of a drafting nature. Thus the reference to the power to accumulate the income of the trust was moved from sub-paragraph e to sub-paragraph f. 85 In sub-paragraph a the words “the capacity to act as a trustee” were included following a Canadian proposal (Working Document No 25). The proposal gave rise to controversy between delegates from civil law countries and delegates from common law countries. The former thought for example that the age required to be a trustee was a question of general capacity for which a connecting factor would have to be found separately and not a question of a person’s fitness to become a trustee. To the contrary, it was pointed out that the law applicable to the trust may require that the trustee’s age be higher than the general age of majority, that the trustee not have declared bankruptcy, or even may provide that limited companies or corporations may be trustees only if they are so authorized by law. In the end the proposal was accepted by a strong majority. Thus the Fifteenth Session reversed a contrary decision of the Special Commission (Report of the Special Commission, No 71; see also on the question of general capacity, supra, No 59, dealing with article 4). The Fifteenth Session experienced some difficulty in rendering in French the idea of “the capacity to act as a trustee” without engendering confusion with the concept of general capacity. 86 Sub-paragraph c gave rise on second reading to a discussion because of the fact that the text of the preliminary draft and that which emerged from the first reading (Working Document No 58) said in French “déléguer . . . ses pouvoirs” and in English “delegate . . . the performance of their duties” (Working Document No 61). This led to the present text which indicates clearly that either duties or powers may be delegated. 87 In sub-paragraph g finally it was made clear that only the liability of the trustee to the beneficiaries was meant. 88 By virtue of article 9 some of the issues set out in article 8, second paragraph, may be subjected to a different law, and this is particularly true for matters of administration. It then becomes necessary to know which law will govern the drawing of the dividing line between the laws governing validity, on the one hand, and administration on the other. Article 9 of the preliminary draft referred for this matter to the law governing the validity of the trust as a whole. The Fifteenth Session, to the contrary, decided by a large majority to eliminate all rules dealing with the characterization of issues as belonging to validity or administration. It would seem that in English law this question is governed by the law applicable to the validity of the trust, while in American law the contrary solution might prevail. The first solution appears preferable within the framework of the Convention by analogy to the role attributed by article 10 to the law governing validity as concerns the possibility of replacing the applicable law by another law. 89 The preliminary draft subjected to the law governing the validity of the trust, in particular, the issue of whether the place of administration may be changed (article 9). This

3. The Explanatory Report by Professor Alfred E. von Overbeck 469 had to do therefore with the physical removal of the place of administration as opposed to a change in the law applicable to administration, which was governed by article 10 of the preliminary draft. Since the rule of article 10 on the change of the applicable law had become obsolete, following amendments to the drain, the sole question which remained to be resolved was whether something should be said expressly, and then very probably in article 8, second paragraph, as to whether the possibility of moving the place of administration was subject to the law governing the trust’s validity. Unanimously the conference finally decided not to include any provision on this point. 90 The Report of the Special Commission mentioned the possibility of establishing a form, delivered by the authority of the country where the trust is established and specifying the trustee’s powers (Report of the Special Commission, No 13). The question was raised by the French delegation during the discussion of article 8 at the Fifteenth Session, but without any specific proposal being made. An observer for the International Union of Latin Notaries also showed interest in this idea. The Delegate of the United Kingdom was opposed to this idea. He pointed out that the Convention of 2 October 1973 concerning the International Administration of the Estates of Deceased Persons, which provided for such a certificate, had caused the greatest difficulties for the Law Commission of England and Wales. That Commission had had to suspend its work after having drawn up six drafts of a report. Thus, for lack of a precise proposal, the suggestion was rejected. For much the same reason a suggestion by the Greek delegation, which would have provided a system for communication of information about institutions analogous to the trust, was not pursued. Article 9 91 Article 9 accepts dépeçage, which is the practice of subjecting certain elements of the trust to different laws. The provision sets out the most frequent example, which is that it may be desirable to apply to the administration of the trust a law other than that which governs its validity; the law of the place where the administration of the trust is located will particularly come to mind. Also it is possible to envision as severable aspects of a trust issues concerning property which is situated in different countries or beneficiaries domiciled in different countries. Dépeçage may just as well be applied in the context of the subjective connection of article 6 as in the context of the objective connection of article 7. However, where the settlor has chosen a law to govern the entirety of his trust, it should not be possible in application of article 7 to extract certain aspects in order to apply a different objective connection to them. 92 In a general way it can be said that dépeçage was desired by the delegations of the common law countries and in particular by the United States delegation, while the delegates of the civil law countries viewed it with distrust. The preliminary draft had dealt with this question at article 6, third paragraph, for the subjective connection and at article 7, third paragraph, for the objective connection, specifying that in this latter case dépeçage should only occur exceptionally.

470 Appendices 93 The Fifteenth Session received a Canadian proposal to delete the third paragraph of article 7 and to restrict the choice of a different law to questions of administration at article 6 third paragraph (Working Documents Nos 36–38). This very restrictive proposal as regards dépeçage encountered in particular the opposition of the United States Delegate who considered dépeçage going beyond questions of administration as being essential. Following that, the Canadian delegation withdrew its proposal. That proposal was none the less put to a vote at the request of the German Delegate who thought, to the contrary, that unlimited dépeçage would make the Convention unacceptable. The Canadian proposal was rejected by a weak majority. 94 To the contrary, an American proposal, which should be mentioned here, would have in the context of the objective connection of article 7 completely dispensed with any law being considered applicable to the entirety of the trust (except for severable aspects) in order to subject each aspect to the law with which it is the most closely connected. This proposal read as follows (Working Document No 12): To the extent that the applicable law has not been chosen in accordance with article 6, the applicable law shall be the law of that State with which the particular aspect of the trust in question is most closely connected. Acceptance of this proposal would have brought about the elimination of article 7, third paragraph, of the preliminary draft. Several delegations of civil law countries remarked that this text was based on a fundamentally different conception, since the jurists of common law countries like to reason issue by issue, while continental jurists wish to deal with a law which is applicable in principle while accepting that certain aspects may be governed by another law. The proposal carried none the less, after a first discussion, by a weak majority. At the next sitting the Delegate of the United States himself proposed however to set aside this proposal, following numerous unfavourable comments that he had received concerning it. He thought that the concerns of the United States on this point could be quieted in another way, in particular if the words “law governing the validity of the trust as a whole”, which appeared in various articles of the preliminary draft, could be eliminated. Unanimously, except for two abstentions, the conference decided to annul the vote of the previous session. The American Delegate also renounced a change in article 7, third paragraph, which he had envisaged. 95 A particular problem of dépeçage which the Special Commission left without a solution (Report of the Special Commission, No 50) was the subject of an Australian proposal; the problem could arise if the settlor chose, to govern a severable aspect of the trust, a law which did not provide for trusts or which did not provide for the category of trusts involved. According to this proposal the choice would be ineffective in such a case. Its text was as follows (Working Document No 6): If a settlor chooses a law to govern a severable aspect of a trust other than validity, and the chosen law contains no provisions relating to trusts or the category of trust involved, the choice shall not be effective. The Australian delegation itself was in doubt as to whether there was still a need for this proposal since the conference also had received an Israeli proposal, by virtue of which, generally in cases where a law was chosen which did not provide for trusts or the category

3. The Explanatory Report by Professor Alfred E. von Overbeck 471 of trusts involved, this choice would be ineffective and the objective connection of article 7 would take its place (Working Document No 39). The Australian proposal was none the less discussed, in particular because certain delegations preferred it to a general rescue operation by means of the objective connecting factor for trusts which have been subjected to a law which does not provide for them. This proposal was accepted by a large majority. However, it was left for the Drafting Committee to decide whether, in case the choice was ineffective under the Australian proposal, the applicable law would be that which governs the validity of the trust or the aspect in question would be linked up under article 7. 96 At the same sitting the conference adopted the Israeli proposal mentioned above. The Drafting Committee incorporated this proposal in articles 6, second paragraph, and 7, first paragraph (Working Document No 58). It thought that the problem, which was the subject of the Australian proposal, was governed by those provisions which, under article 9, also determine the law applicable to aspects of the trust which are severable and connected separately. Under the system of the Convention as a whole it seems that the response to be given to the question left open after the adoption of the Australian proposal—to which question the Convention does not expressly give an answer—is to be found by going in the direction of making an objective connection under article 7 for the severable aspect. It might be then that this aspect would be governed by a law which does not provide for trusts and therefore would escape from the Convention’s coverage. Also the inverse point of view should be recognized, which is that the submission of the severable issue to the law governing the validity of the trust would offer an advantage. The choice of a law which does not provide for trusts could be considered as an absurdity which should be completely ignored, not only as concerns the choice itself but also as concerns the principle of dépeçage. Article 10 97 Since trusts last a long time, the question may arise as to whether during the existence of the trust the law which was initially applicable may or must be replaced by another law. Article 10 remits this question to the very law which governs the validity of the trust when it is created, to decide whether a change of the applicable law is possible and under what conditions. It does not deal with the possible changes that the legislator or case law might bring in the contents of the law which was initially applicable. 98 The delegates of the Fifteenth Session were looking above all at the case of a change of the applicable law by the trustee. It appeared that in certain common law systems the trustee has this possibility only if it has been expressly conferred upon him by the instrument creating the trust; in other countries, to the contrary, the law itself says whether the trustee has such powers. Article 10 also deals with a change of applicable law which may result from a change in the factors which are to be taken into consideration for the objective connection under article 7 (in French “conflit mobile”).

472 Appendices 99 The question of the change of the applicable law can also be posed for one of the aspects of the trust which is severable and can be subjected under article 9 to a different law. It is still the law applicable to the validity of the trust—and not the law which governs the aspect in question—which will say whether, and under what conditions, such a replacement of the applicable law is possible. 100 The present article 10 is the result of rather long discussions within the Special Commission and then at the Fifteenth Session. These discussions were complicated by the fact that the change of the applicable law was discussed at the same time as two other points dealt with in articles 9 and 10 of the preliminary draft these being the change of the place of administration of the trust and the distinction between questions relating to the administration of the trust and questions relating to its validity (supra, Nos 88–89, dealing with article 8). 101 The Special Commission had not been able to examine these problems in depth, and articles 9 and 10 of the preliminary draft left standing a number of issues (Report of the Special Commission, Nos 72–79). The Reporter proposed a different formulation (ibid., No 77) and these articles were criticized in the comments of the Governments. Following these criticisms a number of working documents were submitted to the Fifteenth Session (Australia, Working Document No 9; United States, Working Document No 13; Netherlands, Working Document No 26; Canada, Working Documents Nos 36 and 38; Reporter, Working Document No 37). The discussions of the Fifteenth Session came to grips first of all with a Canadian proposal, dealing not with old articles 9 and 10, but with article 8. Under this proposal a supplementary sub-paragraph would have been inserted in the second paragraph of article 8, reading as follows (Working Documents Nos 36, 38): (i) [the validity of] the power to change the law governing the validity of the trust, or the power to change the law governing the administration of the trust; This proposal indicated clearly that the trustee’s power to change the law applicable to the trust or to its administration is governed by the law which is applicable under article 6 or 7. It was adopted as to its principle by a large majority, with the question of the article at which it would be included being left open. 102 Following these discussions the Drafting Committee submitted a draft of article 7bis, the text of which was as follows (Working Document No 46, slightly modified): The law applicable to the validity of a trust under the preceding articles determines whether or not it is possible to change the law applicable to the validity of the trust or its administration. Subsequently the conference discussed this text at the same time as it discussed the diverse proposals submitted concerning articles 9 and 10 of the preliminary draft. This text as to its principle attracted the conference’s acceptance and led in particular to the withdrawal of the Canadian proposal dealing with article 10 (Working Documents Nos 36, 38). This provision was also accepted by the Netherlands delegation which had proposed the pure and simple deletion of articles 9 and 10 of the preliminary draft (Working Document No 26).

3. The Explanatory Report by Professor Alfred E. von Overbeck 473 The provision was not put in issue again as to its principle, but a satisfactory drafting was found only after it was twice sent back to the Drafting Committee or to its Chairman. In particular the words “replaced by another law” were chosen in order to show clearly that this did not refer to a change in the text of the law (Working Documents Nos 58, 63 and 65). 103 According to a Canadian proposal, a change in the applicable law by virtue of a power conferred on the trustee would have had to be made in writing or, at least, evidenced in writing (Working Documents Nos 36, 38, proposal for article 9). The English text of this proposal included also the requirement that the power to change the applicable law be contained in the instrument creating the trust. This requirement was withdrawn following comments by the Australian and United States delegations But finally the proposal in its entirety was tacitly abandoned.

CHAPTER III—RECOGNITION Article 11 104 This provision sets out, in its first paragraph. the principle of the recognition of the trust created in accordance with the law specified by Chapter II and, more particularly, by articles 6 and 7. The second paragraph requires the recognition of essential elements for all trusts, while the third paragraph enumerates the supplementary aspects of recognition which may flow from the law applicable to the trust. 105 The application of this article might lead to inadmissible encroachments on the forum’s substantive law or on the law applicable under its normal conflict rules; thus article 15 preserves the mandatory rules of these laws. In the preliminary draft the third paragraph contains an express reference to article 19, which corresponds to the present article 15. This reference was eliminated in the draft Convention, specifically in order not to give rise to the argument a contrario that article 15 did not apply to the second paragraph. The clear intention of the conference was on the contrary that the effects of article 11 in its entirety be limited by article 15 (Report of the Special Commission, No 93). The same applies for article 16 which deals with laws of immediate application. Finally article 13 contains another limitation on the recognition of trusts which might be improper, and the rights of third party holders of the trust assets are preserved by the second sentence of subparagraph d of the third paragraph of article 11. First paragraph 106 The condition required in order for a foreign trust to be recognized as such is that it have been created by the law specified by Chapter II. This requirement did not absolutely go without saying: it would have been conceivable to recognize—by analogy to the recognition of foreign judgments, without regard to whether the law applied by the foreign court was the same law as that which would have been applicable in the courts of the recognizing country—certain effects of trusts created under other laws. But in the system of the preliminary draft, such a solution would have brought on a number of difficulties.

474 Appendices In contrast there was no need, as had been provided provisionally in the preliminary draft, that the law under which the trust was created be the law of a Contracting State (supra, Nos 32–35). 107 During the second reading at the Fifteenth Session the question was raised as to whether the expression “trusts created in accordance with. . .” was adequate and whether it would not be better to say “trust subject to”. The text, which had already appeared in the preliminary draft, was however retained, with the idea being that what was involved here was the law which was applicable under Chapter II at the time when the trust came into being, and not another law which might possibly be applicable by reason of a change allowed by article 10. Second paragraph 108 This paragraph determines first of all that the assets of the trust are separate from those of the trustee. This is an essential element of a trust, without which its recognition would have no meaning. The formulation was taken with a minimum change in the English text from the preliminary draft. It will be noted that the idea is expressed in somewhat different fashions in French and in English, but this seemed best to express the conference’s intentions. 109 In the course of the Fifteenth Session the Greek delegation submitted a proposal which would have eliminated the reference to a separate fund in the second paragraph (Working Document No 48). This delegation thought that there was a contradiction between having a substantive rule in the second paragraph and a reference to a separation between the trust assets and the trustee’s estate under the law applicable to the trust which derived from subparagraph b of the third paragraph. This proposal was rejected by a strong majority. 110 The second point which was judged to be essential in the recognition of a trust was the capacity of the trustee to sue and be sued as such. Specific language was added to show that this principle applied not only in lawsuits but also for appearances before an authority or a notary or any person acting in an official capacity. Third paragraph 111 This provision sets out four consequences of a trust which must be recognized if the law applicable to the trust so dictates. 112 Sub-paragraphs a, b and c are easily understandable and did not give rise to much discussion. To the text drafted by the Special Commission the Fifteenth Session added, on proposal of the French delegation (Working Document No 15), a reference to the matrimonial property of the trustee and his spouse, in addition to that to the trustee’s estate upon his death in sub-paragraph c. 113 Sub-paragraph d of the third paragraph deals with tracing, which is a right that the legal systems which have trusts grant to the trust beneficiary when the trustee has violated his obligations by mingling trust assets with his own property or by alienating such assets (Report of the Special Commission, No 46). The second sentence specifies—although this might go without saying—that the relations of the trustee and the beneficiary with any third party, who holds trust assets, are

3. The Explanatory Report by Professor Alfred E. von Overbeck 475 not governed by the law applicable to the trust but rather by the law designated by the forum’s choice-of-law rules. This provision concerns for example the relations between a trustee and a bank in a country which does not have trusts; such a bank will therefore not incur the liability—which is, to tell the truth, rather limited—that the law applicable to the trust may impose on banks in case of a trustee’s breach of his obligations. 114 Sub-paragraph d gave rise to rather long discussions during the Fifteenth Session. The Netherlands Government in its comments had proposed its deletion (Working Document No 10). In the course of the discussions this proposal was however withdrawn. 115 The second sentence, which did not appear in the preliminary draft, had its origin in a document submitted by the Observer for the Bank for International Settlements (Working Document No 17), the reasons for which were set out in the memorandum submitted by that international organization (Preliminary Document No 10, No 6). It read as follows: The responsibility of the holder of trust assets however remains governed by the law applicable at the place of his habitual residence. (translation by the Permanent Bureau) The proposal was above all intended to ensure that the liabilities of bankers would escape the coverage of the law applicable to the trust. It gave rise to various objections. First of all, it was asked whether the problem could not be dealt with in article 19 of the preliminary draft (which was to become article 15 of the Convention). However, to this suggestion the objection was raised that article 19 preserved only the mandatory rules of the law designated by the forum’s conflicts rules, while the intent of the proposal of the Bank for International Settlements was to take out completely from under the coverage of the law applicable to the trust, even where ordinary legal provisions were concerned, relations between trustee and banker. Moreover, the reference to the law which was applicable at the place of habitual residence gave rise to criticism. 116 At the following sitting the United Kingdom delegation proposed to replace subparagraph d in its entirety by the following text (Working Document No 47): d that the interest of the beneficiary under the trust shall be given effect to, to the extent that this does not conflict with the law governing the acquisition of the trust asset in respect of which the beneficiary seeks to assert his interest. This proposal was intended on one hand to make the right to tracing depend on the law applicable to the trust and, on the other hand, to subordinate this right to the law under which the assets had been acquired. It gave rise to various objections. First of all, it was thought that reference should not be made here to questions concerning the acquisition of assets, which were covered by article 19 of the preliminary draft (article 15 of the Convention). Then, the proposal was judged to be too abstract, and the thought was that it would cause sub-paragraph d to lose much of its substance. In the end the provision was sent to the Drafting Committee which was charged with taking the various proposals into account. 117 Although the proposal of the Bank for International Settlements referred to the law of the habitual residence, and that of the United Kingdom to the law applicable to the

476 Appendices acquisition of the assets in question several delegations would have preferred to make reference to the law designated by the forum’s conflicts rules. This is the solution which was proposed by the Drafting Committee (Working Document No 58) and finally adopted by the conference. The English text was still slightly modified in order to avoid any possible confusion between the third parties referred to here and the persons in the names of whom title to the trust assets may stand on behalf of the trustee under sub-paragraph b of the second paragraph of article 2. Article 12 118 Article 12 deals with a problem which could raise difficulties in the practice of States which do not have trusts: this problem is that of reflecting as faithfully as possible in the public registers of such States the peculiarities of trusts. The context shows that this provision applies only to trusts which have been recognized under article 11, in other words those which have been created under the law provided for in Chapter II. 119 A trustee who wishes to register a movable or an immovable asset in a register may ask to appear in this register as a trustee; in cases where this is not possible the registration may take place in such a manner that the existence of the trust relationship appears in another way. All of this is possible only to the extent that the law of the State where registration is to take place so permits. This restriction might possibly have been included in article 15; however, it was preferred to insert it directly in this special provision. In addition, the escape clauses of articles 13, 15 and 16 may also limit the scope of article 12. 120 It is of course intended that the trustee will have to produce, in order to obtain registration, the necessary supporting documents, and that he will have to fulfil the formalities required by the law of the country where registration is to take place. It is also the law of this country which will decide who may require that the existence of a trust be disclosed. 121 The reach of article 12 is not limited to public registers; this provision also deals with private registers having a public function such as a shareholders’ register. The Fifteenth Session reproduced the text of the preliminary draft with minor drafting changes. It is necessary therefore to consult the Report of the Special Commission in order to know the origin of this provision (Report of the Special Commission, Nos 96–104). Article 13 122 Article 13 allows refusal of recognition to a trust which has been placed under a law which provides for trusts, through the designation of the applicable law (article 6, first paragraph), the fixing of the place of administration (article 7, second paragraph, subparagraph a) or the choice of a trustee who has his habitual residence in such a country (article 7, second paragraph, sub-paragraph c), while the other objective elements of the

3. The Explanatory Report by Professor Alfred E. von Overbeck 477 situation connect the trust more closely to one or several States which do not have trusts. These “significant” elements may in particular be the habitual residences or the nationalities of the persons involved or the location of property. 123 The option offered by article 13 is open to the judges of all the Contracting States, but it is clear that it is in fact an escape clause in favour of States which do not have trusts. The clause will be used above all by judges who think that the situation has been improperly removed from under the application of their own laws. But it might also be utilized by the judge of one State which does not have trusts as a matter of solidarity with another State, which also does not have them and to which the situation is objectively connected. 124 It will also be noted that this provision allows a judge of a State which does not have trusts to refuse recognition to the trust because he thinks that the situation involved is internal to his State. In contrast this possibility does not exist for those States which have trusts, but those States do not seem to feel the need for it. Article 13 does not say the time at which the conditions that it sets out must be fulfilled. Reasoning from the purpose of the provision one might think that the time in question should be that of recognition, rather than the time of creation of the trust. 125 The article of the preliminary draft which corresponded to article 13 of the Convention was article 14 which consisted, as a result of rather long discussions of two variants, designated A and B. Variant B allowed only the non-recognition of trusts, all of the objective elements of which were localized in the State which was asked to give recognition. Variant A, on the contrary, corresponded in its essentials to the present article 13. The only differences concerned the addition of the trustee’s habitual residence as a third element by means of which the settlor may seek to place the trust in a country which has provision for trusts, and the elimination of the reference to the time of the trust’s creation (Report of the Special Commission, Nos 115–124; see also Nos 137–146). 126 The reference to the time of creation was eliminated without discussion. Reasoning from the purpose of this provision it seems moreover that the time of recognition and not that of the creation of the trust ought to be determinative. The present text leaves a certain amount of freedom to the judge. 127 The solution of article 13, although in the end it is very near to that of variant A of article 14 of the preliminary draft, was adopted only after rather long discussions. A proposal by the Netherlands (Prel. Doc. No 10; Working Document No 28) and an Australian proposal relating to variant B of article 14 of the preliminary draft (Working Document No 4) lost their pertinency in the aftermath of other proposals and were not discussed. 128 The discussion, on the contrary, took its departure from a proposal by the delegations of Argentina, Egypt, France, Greece and Spain, which read as follows (Working Document No 33): The States Parties shall not recognize trusts involving property which, at the time of the trust’s creation, is primarily located in States which do not provide for trusts. The same rule applies where, at the time of creation of the trust, the settlor and the beneficiaries have the nationality of a State which does not provide for trusts and reside habitually in States which do not have this institution. (Translation by the Permanent Bureau).

478 Appendices This proposal departed from the preliminary draft in two respects which were discussed in parallel but which should be distinguished here. On one hand, it obligated the Contracting States to refuse recognition to certain trusts and, on the other hand, it replaced the reference to the significant elements of the trust with precise factors, such as the situs of property, and the nationalities and habitual residences of the interested persons. 129 As to its first aspect, the authors of the proposal pointed out that their countries, which had civil law legal traditions, were inclined to recognize trusts, but that as a counterpart they would like for the common law States to take their interests into account by refusing to recognize trusts which dealt in essence with situations connected to States which did not have trusts. This was desired particularly in order to protect these States from fraudulent trusts. It appeared in the end that the authors of the proposal feared that, despite article 14 of the preliminary draft, the Convention might lead to introduction of trusts as an institution in the civil law countries. Several delegates from common law countries sharply opposed this idea. They could not agree for the Convention to obligate common law States to refuse recognition to trusts created under their own law or under the laws of other States having trusts, where under their ordinary rules of law they would have recognized them. However, these delegates declared that they were ready to protect the civil law States against fraudulent trusts in an appropriate fashion, for example through the use of mandatory rules. The proposal of the five delegations in contrast would, it seemed to them, lead to the non-recognition of numerous trusts which were entirely proper. Several delegates of countries which do not have trusts moreover, thought that this proposal could bring the whole Convention into question or, in any case, that it went too far. To require in certain cases the non-recognition of trusts of common law countries in common law jurisdictions would have profoundly altered the spirit of the preliminary draft, which was intended to be favourable to the recognition of trusts. This would have made pointless the present article 14 (article 15 of the preliminary draft) which allows the recognition of trusts, even when the Convention does not require it. Finally it was also pointed out that the judge who found himself under an obligation not to recognize the trust might be perplexed. Should he then, for example, rule that the trustee was the owner in his own right? These arguments in the end convinced the authors of the proposal mentioned above and, upon resumption of the discussion, they consented to make it optional by replacing the words “shall not recognize” by “shall not be bound to recognize”. 130 The second aspect, the factors set out in the proposal of the five delegations as a basis for non-recognition, was also the subject of discussion and criticism. A trust would not have to be recognized either where the assets were primarily located in States which did not have trusts (first paragraph, supra, No 128), or where the settlor and the beneficiary had the nationality of a State which did not have trusts and at the same time had their habitual residences in such a State (second paragraph, supra, No 128). First of all an Italian counter-proposal, which was not discussed in detail, set out as the condition for non-recognition that three factors, the location of the assets, the nationality of the settlor and the beneficiaries, and finally the habitual residence of these persons,

3. The Explanatory Report by Professor Alfred E. von Overbeck 479 designate the same State which does not have the institution of the trust (Working Document No 34). Then, from a technical point of view, there was criticism of the reference to the assets being “primarily located. . .”. By speaking of assets being “primarily” located in a State the drafters had wanted to avoid obligating a country to recognize a trust merely by reason of the fact that a bank account was opened in a country which had this institution. But this expression was criticized for its lack of precision. The objection was also made that for movable assets and, in particular, for receivables, commercial paper and investment securities, the situs was very difficult to determine under the current conditions of business life. 131 The examples which were advanced against the proposal of the five delegations showed that its acceptance would have brought on many difficulties. Thus the English delegation gave the example of a Frenchman, owner of a prosperous French limited company, who emigrated to England where he married an English woman and had children. The proposal would obligate an English judge to refuse to give effect to a testamentary trust in which this Frenchman had placed the shares of the French company for the benefit of his wife and children. To this example the French delegation replied that it scarcely seemed conceivable that a lawyer in a common law country might advise a Frenchman domiciled in Great Britain to create a trust covering assets in France. In reality it would seem that a trust covering shares in the French company—shares which were physically located in England—might certainly be justified; in contrast, an English trust covering assets which were actually located in France, and in particular immovables, would certainly not be appropriate. The English delegation then gave another example in which, contrary to the second paragraph of the proposal, there might be justification for persons of French nationality having their habitual residence in France to make a trust; the case mentioned was that in which these persons owned a house for use on weekends in the south of England, used by all the family in turn. 132 The discussion was resumed the following day and the Chairman submitted to the conference a new proposal, which was discussed at the same time as that of the five delegations (supra, No 128). Only the first variant of this proposal, which contrasted with the original proposal of the five delegations, not only by the abandonment of the fixed criteria for non-recognition but also by the fact that non-recognition was optional (supra, No l29), was discussed (Working Document No 50, variant A): No Contracting State shall be bound to recognize a foreign trust all of the relevant elements of which, except for the choice of the applicable law, the place of administration or the habitual residence of the trustees, are most closely connected with the law of the State which does not have the institution of the trust or that category of trust. The English text specified that the other elements referred to were the “relevant elements” while the French text did not so specify. This formulation was criticized as to certain details, in particular because of its reference to a “foreign” trust, which was a new concept that was difficult to identify, and this reference was abandoned without an express decision. As to the rest, the proposal attracted the support of a number of delegations.

480 Appendices 133 The conference voted then on the proposal of the five delegations (amended to make non-recognition optional) and rejected it by a weak majority. To the contrary, the text proposed by the Chairman was accepted without opposition, although with several abstentions. During the second reading, article 13, refined by the Drafting Committee (Working Document No 48), no longer contained the specification “all the significant elements. . .”. The reinsertion of this word was requested by certain delegates; others, to the contrary, were strongly opposed to this, in the thought that then even the creation of a bank account in a State which had trusts would make article 13 inoperative. Other delegations were of the opinion that this addition would not change the meaning of the text. In the end the text proposed—which is that of the Convention—was adopted by a very strong majority. Article 14 134 This provision reflects very well the objectives of the Convention: its purpose is to facilitate the recognition of trusts, following the example of what is usually done in the conventions which deal only with the recognition and enforcement of judgments. It was not intended to prevent States from recognizing trusts, even in cases which are not covered by the Convention. As has happened for other Hague Conventions, one may imagine that in practice the Convention’s solutions will be extended to trusts which are not formally covered by it. 135 Article 14 reproduces the substance of article 15 of the preliminary draft in a simplified drafting, borrowed from article 13 of the Convention of 14 March 1978 on Celebration and Recognition of the Validity of Marriages. This article did not give rise to discussion at the Fifteenth Session, but it would have become obsolete if the proposal for article 13 which provided for the obligatory refusal of recognition in certain cases had been accepted (supra, Nos 128–129).

CHAPTER IV—GENERAL CLAUSES Article 15 136 The first paragraph of this provision preserves the mandatory rules of the law designated by the conflicts rules of the forum for matters other than trusts. The second paragraph invites the judge to mitigate, where necessary, the effects of the first paragraph. First paragraph 137 Given the multiple functions which trusts can fulfil, it may well be that the provisions of a trust, or certain provisions of the law which governs it, will be incompatible with the law applicable to another matter under the forum’s conflicts rules. The conference thought that the law applicable to these other institutions should prevail over the law of the trust, but only as to its mandatory rules, in other words those rules which cannot be derogated from by voluntary act. Mention may be made in this connection of the

3. The Explanatory Report by Professor Alfred E. von Overbeck 481 rules providing indefeasible shares for the heirs, which are mandatory for a testator. The heirs may, naturally, renounce the invocation of these rights and it would perhaps have been more exact to say in accordance with an Austrian proposal: “derogated in advance. . .” (translation of the Permanent Bureau) (Working Document No 49). 138 It will be recalled that article 3 removes entirely from the Convention’s coverage the acts done prior to the creation of a trust, which are necessary for its creation. To the contrary, article 15 looks to those hypothetical situations in which an existing trust may deploy effects which are incompatible with the mandatory rules of the forum or of the law of a third country for a field other than trusts. The intent was to preserve above all the forum’s substantive law in cases where its conflicts rules designated its own law, but this provision is also intended to preserve a foreign substantive law which is designated by the private international law of the forum. The last sentence of sub-paragraph d of the third paragraph of article 11 also preserves a law which is designated by the forum’s conflicts rules, but unlike article 15 that provision of article 11 looks to the entirety of the law thus determined and not only to its mandatory rules (supra, Nos 113–117). 139 It should be emphasized that the enumeration of the first paragraph of article 15 is by way of example. Mandatory rules in matters which are not listed may therefore also override the trust’s rules. Not without reason, it was said that a hostile judge might always find in article 15 a means of frustrating the trust. 140 The first paragraph of article 15 reproduces the essence of article 19 of the preliminary draft, with the second paragraph having been incorporated at the beginning of the first paragraph (Working Document No 54). The rather complicated origin of article 19 of the preliminary draft, which was tied to that of article 20 (present article 16) and of article 14 (present article 13; see Report of the Special Commission, Nos 135–150) should be recalled. The Fifteenth Session began by discussing articles 15 and 16 (articles 19 and 20 of the preliminary draft) jointly. Following the example of the Rome Convention (article 5, second paragraph, and article 6, first paragraph, on one hand, and article 7 on the other), the preliminary draft utilized the term “mandatory rules” in two different senses. In article 19 these words designated simply mandatory rules as opposed to dispositive rules, while in article 20 they referred to laws of immediate application, sometimes known as “lois de police”. These terms were in the end eliminated, both from article 15 and from article 16. 141 The principle already expressed in article 19 of the preliminary draft, which is respect for the mandatory rules of the law designated by the forum’s conflicts rules, was brought into question only by a proposal of the United States delegation offered as a combination of articles 19 and 20, but in reality placing the whole question on the level of the laws of immediate application (Working Document No 42). This proposal was not discussed on the merits and was not put to a vote; it was certainly contrary to the intentions of the States which do not have trusts, since they wanted to have respect given in a general manner to the mandatory rules of the law designated by the forum’s conflicts rules, and not simply to laws of immediate application. 142 At the end of the first discussion of these two articles, the principle of article 15 was adopted unanimously. The list set out in article 15 still however gave rise to a number of proposals.

482 Appendices First of all the order followed in the preliminary draft was modified in such a way as to put in the first place the two cases concerning individuals. Sub-paragraphs a, b and c reproduce without change the texts of sub-paragraphs d, c and a of the first paragraph of article 19 of the preliminary draft. 143 Sub-paragraph d (sub-paragraph b of article 19 of the preliminary draft) gave rise to more extended discussions. The issues with which it deals had already been taken up when the suggestion of the Bank for International Settlements to exclude business trusts and trusts intended to create security interests from the scope of the Convention had been discussed (supra, No 27). The decision had then been remitted to the time of the discussion of articles 11 and 15 (article 19 of the preliminary drain). Subsequently the Observer for the Bank for International Settlements proposed the following formulation (Working Document No 18): “rights in rem, especially in respect of third parties” (translation of the Permanent Bureau). Before the time when the discussion was held he had, however, reached agreement with the Permanent Bureau on the following proposal (Working Document No 53): The transfer of title to property, especially in relation to third parties, and security interests. For its part the German delegation proposed to speak in a broader fashion simply of “property” (Working Document No 44). The German Delegate hoped in this way to avoid having a separate sub-paragraph dealing with the protection of third parties. During the discussion a reference to rights in rem or to property in general met with opposition, for it seemed to go too far. It was thought that, property rights being an essential element of every trust in the common law countries, such a reference would contradict the provisions on recognition. Thus this idea was not adopted. In contrast, the conference decided, without opposition but with several abstentions, to make reference to security interests, specifying that those referred to were security interests in property. In addition it was decided to add that prior creditors were meant to be protected. In the end the Drafting Committee decided not to include this specification and its decision was accepted tacitly. Finally, once sub-paragraph f, dealing with the protection of third parties in general, had been included, the reference to third parties could be eliminated from sub-paragraph d. 144 Sub-paragraphs e and f had their origin in sub-paragraph e of the first paragraph of article 19 of the preliminary draft and in a text submitted by the German delegation (Working Document No 44). This latter text proposed to set forth in separate subparagraphs on one hand “the protection of third parties” and on the other “insolvency”, even though it was recognized that the reference to insolvency in a general manner went very far and was difficult to accept for the common law countries. Following various objections the conference decided by a rather weak majority to include a special sub-paragraph providing for the protection of third parties in general. Without opposition, it decided to limit this sub-paragraph to third parties acting in good faith. The idea of a general reference to insolvency was not pursued. 145 The Drafting Committee set down as follows sub-paragraphs e end f (Working Document No 58):

3. The Explanatory Report by Professor Alfred E. von Overbeck 483 e the protection of third parties in insolvency; f the protection of third parties in good faith in other respects. It was then made clear, in respect of sub-paragraph e, that in case of insolvency good faith was not required. The reference there to third parties was replaced by a reference to creditors; several other drafting modifications were then made in the text. 146 The German delegation proposed to add yet another sub-paragraph dealing with the law of societies and legal persons (Working Document No 44). It turned out to be rather difficult to determine the exact reach of this proposal. The German delegation pointed out that under German law an individual merchant always has personal liability, which can be enforced against all of his property, unless he sets up a limited liability company. A trustee finding himself in financial difficulties should not be able to escape this rule. In the course of the discussion, the delegates of the common law countries pointed out that the trustee was always entirely liable for his acts, and that this could be enforced against all of his property, but that his liability for acts done in his personal capacity could not be enforced against the trust assets. The German proposal was in the end rejected with many abstentions. Second paragraph 147 This provision is intended to mitigate the overly rigorous consequences which the first paragraph may have. It indicates to the judge that if the trust cannot be recognized, he should try by other means—thought will turn in particular to adaptation—to give effect in so far as possible to the settlor’s intentions. This is an appeal to goodwill rather than a rule of strict law, but in this Convention, which cannot deal with all of the details of the subject-matter, the rule seems to have a place. 148 The provision had been proposed for possible inclusion by the German delegation as a third paragraph of a proposal replacing articles 19 and 20 of the preliminary draft, with the following text (Working Document No 44; see also infra, No 155, dealing with article 16) [Insofar as the effects of the trust cannot be achieved by result of the application of the preceding paragraphs the court shall try, and have discretion, to achieve an equivalent result by other technical means of the applicable law.] During the discussion the Austrian delegation raised objections to the words “and have discretion”, which would give the judge a discretionary power which was unacceptable under the Austrian constitution. It was also determined that it would be pointless to refer to the applicable law and that the term “technical means” created difficulties. The proposal thus amended was adopted by a strong majority and forwarded to the Drafting Committee. On the second reading, the present text was adopted without discussion. Article 16 149 Article 16 deals with the provisions known among legal authors under the name of “laws of immediate application” and designated in the Rome Convention as “mandatory

484 Appendices rules”. These are rules which are intended to apply to certain legal situations and which take preference over the law which would be applicable according to the normal play of rules of conflicts of laws. They are sometimes referred to as mandatory rules, but these are in fact special mandatory rules and this idea is expressed in article 16 as follows: “provisions . . . which must be applied even in international situations. irrespective of rules of conflict of laws”. Among the laws which fall in this category, mention may be made of those which are intended to protect the cultural heritage of a country, public health, certain vital economic interests, the protection of employees or of the weaker party to another contract. The French delegation asked that in the Report currency exchange regulations be mentioned in this connection. 150 The first paragraph of article 16 provides that the forum’s laws of immediate application will override the other provisions of the Convention. Without doubt, the national courts will apply the same solution, even without a special rule, but this rule may contribute to clarifying matters and may also avoid that, in line with certain opinions of the legal writers, public policy be invoked in these cases. 151 The second and third paragraphs have to do with the laws of immediate application of third States to which effect may be given exceptionally if the lawsuit has a sufficiently close connection with such a State. The third State will then be neither the State of the forum nor the State the law of which is designated by the Convention’s conflicts rules. It should be emphasized that the laws of immediate application of the third State will be taken into consideration only if they are intended to apply to the trust in question; this point perhaps does not come out with sufficient clarity from the text. Finally, the question of whether effect will be given to the laws of immediate application of a third State will always depend on the judge’s wisdom. The formulation “effect may. . . be given” makes it clear that the court is never obligated to do so. The second paragraph will be applied only in those States which have not excluded its application by means of the reservation allowed in the third paragraph. 152 The solutions of article 16 are the result of rather long discussions. Reminder should be made here again of the genesis of articles 13, 15 and 16 (articles 14, 19 and 20 of the preliminary draft) within the Special Commission (Report of the Special Commission, Nos 135–153). This type of provision originated with article 7 of the Rome Convention. Although the theory of laws of immediate application had become familiar in the European States, and particularly in those of the Common Market, it caused difficulties for the delegations of countries which were further away from these developments, as was recalled during the Fifteenth Session by a delegate of the United States. It will be observed however that article 16 may also serve to cause certain rules, considered as being fundamental in a country which has trusts (for example the rule against perpetuities), to prevail as against a trust which is subject to the law of another country. 153 The respect for the forum’s laws of immediate application was scarcely ever put in doubt. One might moreover argue that, since article 15 preserves the mandatory rules of the laws designated by the forum’s conflicts rules, it calls a fortiori for respect to be given to the rules of immediate application of the forum itself.

3. The Explanatory Report by Professor Alfred E. von Overbeck 485 154 To the contrary, the inclusion of a rule on the laws of immediate application of third States was very much contested. The discussions of the Fifteenth Session reflected here the controversies which exist on this subject, in particular in the countries of the Common Market with regard to article 7 of the Rome Convention. As is known, that article may be set aside by a reservation (sub-paragraph a of the first paragraph of article 22) which several States seem to intend to employ. The elimination of article 20 of the preliminary draft, dealing with the laws of immediate application of third States, had been proposed in the comments of the German Government (Prel. Doc. No 10) and then, above all, by a Working Document and by the remarks of the Austrian Delegate (Working Document No 10), who argued in particular that the only precedent in favour of a rule of this type was the Rome Convention. In the work on revision of the 1955 Convention on Sales of Goods the question was very controversial, and the preliminary draft does not contain a provision on this point. In that Delegate’s opinion such a provision has to do with public international law; its application may distort the results to which the conflicts rules of the forum would lead, and certain persons might employ it in order to avoid their obligations. The partisans of this solution, to the contrary, remarked that there was in any case yet another precedent in article 16 of the Convention of 14 March 1978 on the Law Applicable to Agency; it is true that this Convention has not yet entered into force. The argument was also advanced that, by giving effect to the laws of immediate application of a third State, the same solidarity among States could be shown as had been involved during the discussion of article 13 (supra, No 129). This rule was also seen as a means to frustrate certain attempts to escape from mandatory rules by means of a trust. 155 During the first discussion of articles 19 and 20 of the preliminary draft, the conference received two proposals, which were however of rather different scope, to merge these two provisions. A proposal of the United States of America (Working Document No 42) placed the entirety of these questions in the framework of the laws of immediate application of the forum or of a third State. For this reason it was unacceptable to the adversaries of this theory. We have seen in connection with article 15 that this proposal was not discussed in more detail (supra, No 141). A proposal of the German delegation, to the contrary, reproduced in their essentials the provisions of the first paragraph of article 15 as a category of mandatory laws, but provided as another category the laws of immediate application of the forum, to the exclusion of those of a third State. The text of this proposal was as follows (Working Document No 44): This Convention does not prevent the application of a mandatory rule: 1) of the law of the forum, if and so far as that rule applies irrespective of the law specified by choice-of-law rules, or 2) of the law designated by the choice-of-law rules of the forum outside the applicability of the Convention governing in particular— (the proposals contained in sub-paragraphs a to g) The conference adopted the principle of this proposal. Indeed, unanimously it accepted the system of article 15 (article 19 of the preliminary draft) and respect for the forum’s

486 Appendices laws of immediate application. In contrast, it rejected by a clear majority any provision dealing with the laws of immediate application of third States. 156 Following the vote, the Drafting Committee presented a draft of article 16, corresponding to the current first paragraph and therefore dealing only with the forum’s laws of immediate application (Working Document No 58). 157 The delegations of the United States of America, of France, of Finland and of Switzerland proposed jointly to add the current second paragraph of article 16. Since the inclusion of a rule on the laws of immediate application of third States had been rejected in the first vote, reopening of the discussion required a decision by an absolute majority. The Austrian delegation opposed the motion for reopening of the discussion; the German delegation indicated that in case of acceptance of this proposal, a reservation would be necessary. The reopening was then decided on by the conference with a single opposing vote and several abstentions. The French and Swiss delegations defended the joint proposal by recalling the arguments in favour of a reference to foreign laws of immediate application, and by pointing out the imbalance which could result from mentioning only the forum’s laws of immediate application. They also contended that the new proposal was clearer than the text of the preliminary draft and also constituted an advance in comparison with the formulation of the Rome Convention. To the contrary, the English and Australian delegations, in the light of practical examples which were submitted, thought that their legal systems would allow foreign laws of immediate application to be taken into account, without any special provision for this purpose. The principle of the proposal by the four delegations was then adopted by a three-fifths’ majority. 158 The conference then examined three proposals, which would permit States not to apply this provision. The Austrian delegation revived an amendment to article 20 of the preliminary draft, which it had made as a subsidiary proposal (Working Document No 10): The Convention does not prevent mandatory rules from being applied, exceptionally, where the law of the forum so permits . . . (Translation by the Permanent Bureau) This delegation contemplated, moreover, the transformation of the proposed rule into a simple option for the States. Finally, the German delegation revived its proposal for a reservation against the second paragraph, following the example of article 22 of the Rome Convention which also allows a reservation dealing only with foreign laws of immediate application. The Reporter would have preferred for the reservation to cover the entirety of the provision, in order to avoid the imbalance resulting from having a rule only on the forum’s laws of immediate application. The Austrian proposal to make the provision’s application depend on the law of the forum was criticized as being harmful to foreseeability; it was rejected by a clear majority with many abstentions. The idea of an option was not put to the vote.

3. The Explanatory Report by Professor Alfred E. von Overbeck 487 The conference adopted, with only one opposing vote, the principle of a reservation against only the second paragraph of article 16. The making of the reservation is governed by article 26. At the time of the second reading, the texts of the second and third paragraphs were adopted as they appear in the Convention (Working Document No 64). It was then explained that it was intentional that the French and English texts had somewhat different formulations. It was also pointed out that, although the French phrase “avec l’objet du litige” could appear narrower than the English phrase “with the case” the French phrase did not necessarily refer to a pending lawsuit. We should point out that the Rome Convention speaks in this connection of “situation”. 159 The conference did not discuss the question of the reciprocity of the reservation provided for in the third paragraph. Could Contracting State B refuse to apply the laws of immediate application of Contracting State A, which had made the reservation? Such a solution would have the strange result that State B would apply the laws of immediate application of a third non-Contracting State but not those of Contracting State A. Thus it has to be admitted that this reservation has no reciprocal effect.9 Article 17 160 This provision indicates that the law designated by the Convention is always the substantive law, to the exclusion of rules of conflict of laws. This solution is that of all the modern conventions. Until the recent past, the same idea was expressed in the Hague Conventions by the employment of the term “internal law”. 161 This principle appears in article 16 of the preliminary draft. It gave rise to some discussions within the Special Commission (Report of the Special Commission, Nos 128–131). 162 The Fifteenth Session received a proposal of the United States which read as follows (Working Document No 43): In this Convention. the word “law” means the law in force in a State, other than its rules of private international law, unless the settlor has specifically indicated otherwise. During the discussion it was argued that the taking into consideration of conflicts rules was in any case excluded for the objective connection, and that, for the subjective connection, this did not seem necessary and it might complicate things. It was pointed out that such a solution had never been adopted in the recent Hague Conventions. The proposal was rejected by a strong majority and the conference adopted a text proposed by the Sub-Committee on general and final clauses, modelled on the preliminary draft Convention on sales of goods (Working Document No 45, article 16). 163 In plenary session the words “private international law” were replaced by the words “conflict of laws”. 9 On reservations and their reciprocities, see “Note on reservations and options in the Hague Conventions” in Acts and Documents of the Thirteenth Session, 1976, Book 1, Miscellaneous matters, p. 102.

488 Appendices Article 18 164 This provision contains the public policy clause which is customary in the Hague Conventions. The formulation of article 18 of the preliminary draft raised at the Fifteenth Session several comments which were more of a drafting nature. It was in particular argued that there was no need to say “. . . may be refused only”. Indeed, there are other articles and in particular article 5 which set aside the application of the Convention. On the other hand, it was pointed out that the English formulation of the preliminary draft, which spoke of “the application of the Convention”, was preferable to the French formulation, which spoke about the application “of one of the laws designated. . .”, since this Convention deals not only with conflict of laws but also with recognition. The text of article 18 of the Convention took these remarks into account. Article 19 165 It was necessary to state in the Convention that tax law would not be affected; indeed, if the Convention appeared to allow, through means of trusts, the evasion of certain taxes, its chances for ratification would be seriously compromised. Article 17 of the preliminary draft was adopted without discussion by the Fifteenth Session. Article 20 166 By virtue of this article, a State may declare that it will apply the Convention’s provisions also to trusts declared by judicial decisions, which otherwise would not be covered (supra, Nos 49–51 dealing with article 3). Such a declaration gives rise to no obligation for the other Contracting States to recognize judicial trusts of the State which has made the declaration. The second and third paragraphs deal with the procedures in connection with such a declaration. 167 The Special Commission had decided to exclude trusts created by judicial authorization from the Convention’s scope (Report of the Special Commission, Nos 33–34). This decision was brought back into issue at the Fifteenth Session, and a proposal to include trusts created by judicial decision (Working Document No 16) gave rise to a rather extended discussion. On this proposal, both the delegates from civil law countries and those from common law countries were divided. In favour of the inclusion of these trusts it was argued that they offered in general more guarantees than trusts created by individuals and that, in addition, they were subject to the lex magistratus, which excluded the possibility of a fraudulent choice of a foreign law. Further arguments were based on the Brussels Convention on jurisdiction and the recognition of judgments. Against this solution it was argued first of all that the Convention was essentially centred on the protection of the intentions of individuals who want to create trusts, and that on

3. The Explanatory Report by Professor Alfred E. von Overbeck 489 the other hand to include judicial trusts would amount in some way to recognizing foreign decisions by way of a detour, which did not present the ordinary guarantees. It was also pointed out that in the common law countries judicial trusts were most often a method of providing a legal remedy. (See Dyer/Van Loon Report, No 107). Now, it was unthinkable to require recognition in trust form of something which would not be recognized in the form of an ordinary judgment for the payment of a sum of money. At the end of this discussion the general inclusion of judicial trusts was rejected by a clear majority. 168 The Commission then examined a more restrictive proposal, under which the Convention would apply “to trusts of a continuing administrative nature which are created by a judicial instrument” (Working Document No 35). This proposal was intended to exclude constructive trusts, which involve simply a transfer of assets, but to cover lasting trusts created for example in order that a testamentary executor might be trustee of property left to the deceased’s children. However, this proposal also encountered the objection that it would involve the recognition of foreign judgments, and it was set aside. 169 The solution of article 20 was then proposed by the French and English delegations (Working Document No 51). The authors of this proposal recalled that the members of the European Community were already bound to recognize judicial trusts by virtue of the Brussels Convention. They wanted a Contracting State to be able to declare that it would recognize either the judicial trusts of all other States or only those coming from certain States, for example States of the Common Market. The proposal was adopted in principle after a brief discussion, the final text, which was designed to meet various questions which had been posed, was voted in without discussion (drafting amendments were made in plenary session). Article 21 170 This article allows the Contracting States to apply the provisions of Chapter III on recognition only to trusts the validity of which is governed by the law of a Contracting State. This provision was included following discussions on the application of the Convention erga omnes (supra, Nos 32–35). The reservation eliminating trusts governed by the law of a non-Contracting State is not to have reciprocal effect. The procedure for the reservation is governed by article 26. 171 Following decisions taken on the universality of the Convention and the insertion of a reservation, the Drafting Committee proposed the current article 21 in the following text (Working Document No 58): A Contracting State may reserve the right to apply the provisions of Chapter III only to trusts governed by the law of a Contracting State. During the second reading the German delegation brought up, however, the following proposal (Working Document No 62):

490 Appendices A Contracting State may reserve the right to apply the provisions of Chapter II only to trusts the significant elements of which are connected with a Contracting State and of Chapter III only to trusts governed by the law of a Contracting State. It can be seen that this formulation would also allow, under different criteria, the making of a reservation concerning the provisions on the applicable law. This proposal was rejected after a brief discussion by a clear majority, with several abstentions. 172 During the same discussion, it was decided without opposition to specify that the determinative criterion under article 21 is the law governing the validity of the trust. If this law is the law of a Contracting State, the trust must be recognized even if certain of its aspects, for example its administration, are governed by the law of a non-Contracting State. Although this was not discussed at the Fifteenth Session, it should be accepted that it will suffice, in order that Chapter III be applicable in a State which has made the reservation, that the validity of the trust be governed by the law of a Contracting State at the time when recognition is requested. The time when the trust was created has no importance. It may be that, at that time, the State the law of which governs the validity was not yet a Contracting State, or even that the law governing the validity when the trust was created has been replaced by another law, pursuant to article 10. We should recall, however, that a trust has to be recognized only if it has been created under a law designated by articles 6 and 7 (article 11, first paragraph). A trustee who has the power to change the applicable law might then replace the law of a non-Contracting State, which was initially applicable, by the law of a Contracting State, in order to assure the recognition of the trust in another Contracting State. Article 22 173 The first paragraph of this article sets forth the principle that the Convention applies to all trusts, regardless of the date on which they were created. The second paragraph, however, allows a Contracting State to make a reservation under the terms of which it will not apply the Convention to trusts created before the Convention entered into force for it. Article 26 governs the procedures for this reservation. 174 This article was proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 24). During the discussion the principle of the first paragraph was not contested; in contrast, the necessity for and the justification of the reservation were placed in doubt. Against it, the objection was raised that it was not very appropriate, given the open character of the Convention, and the fact that a similar reservation in the Convention of 5 October 1961 on the Conflicts of Laws Relating to the Form of Testamentary Dispositions had not been employed. Certain delegations, however, expressed fear about the retroactivity which the first paragraph seemed to involve. Thus the Australian Delegate thought that his country would perhaps have need for the reservation. To the contrary, the Delegates of the United States and the United Kingdom pointed out that retroactivity was dangerous only if it changed

3. The Explanatory Report by Professor Alfred E. von Overbeck 491 the substantive law, which was not the case for this provision. It was also feared that the reservation would sow confusion in connection with trusts which have ties with several countries. In the end, the proposal in its entirety was adopted without opposition. An exchange of views then took place on the possibility of making a partial reservation. The conclusion drawn from this was that such would be possible without changing the drafting. 175 At the second reading, a question was raised as to how this reservation would combine with that concerning the limitation of Chapter III to trusts created under the law of a Contracting State (article 21). Supposing that State A makes the reservation as to time of article 22, and State B makes the reservation of article 21 concerning Chapter III, it is quite clear that State A will recognize the trusts of State B only if they have been created after the entry into force of the Convention for State A. Should it be concluded from this that State B is not bound to recognize trusts conforming with the law of State A, which were created before the entry into force of the Convention for that State? The question was not resolved but under the general spirit of the Convention and the discussion it seems that the response should be in the negative. Reciprocity certainly would not come into play if State B had not made the reservation of article 21, for without that the trusts of Contracting State A would be less well treated than the trusts of non-Contracting States. Article 23 176 This provision has to do with the States which are comprised of several territorial units, each having its own legislation. Such a unit will be treated for the purposes of the Convention as if it were a State; the choice of the law of one of these component units by the settlor will be effective and, in the absence of choice, the criteria of the second paragraph of article 7 will be applied to determine with which unit the trust has the closest connection. What will happen if the settlor chooses the law of the non-unified State as such, for example American law? According to a declaration by the delegation of the United States, which was not contradicted, such a choice would have no effect and the applicable law would have to be determined under article 7. 177 The text of the provision was proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 21). This text was inspired by an Australian proposal (Working Document No 3) and the preliminary draft Convention on sales of goods. The discussion showed that two solutions were possible for the problem posed by the composite States; either as the proposal did, apply directly the Convention’s conflicts rules to the territorial unit, or apply the inter-provincial rules of the State concerned to the extent that some exist. This latter solution appears in a number of Hague Conventions, for example in article 16 of the Convention of 2 October 1973 on the Law Applicable to Maintenance Obligations. In the context of the Convention under discussion, the first solution was however unanimously preferred.

492 Appendices 178 The question was posed as to whether the solution adopted should be expressed by means of the proposed article or, to the contrary, by means of a provision inspired by article 19 of the Convention of 14 March 1978 on the Law Applicable to Agency, which reads as follows Where a State comprises several territorial units each of which has its own rules of law in respect of agency, each territorial unit shall be considered as a State for the purposes of identifying the law applicable under this Convention. In the end the formulation of the current article 23 was preferred to that of article 19 of the Agency Convention by all the delegations, except for one opposing and five abstentions. Article 24 179 Article 24 allows the States with composite systems not to apply the Convention to internal conflicts. This provision, proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 22), was inspired by the preliminary draft Convention on sales of goods. It was adopted without discussion, despite certain doubts as to its utility. Article 25 180 Article 25, which is intended to avoid conflicts of conventions, was also proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 23) and adopted without discussion. This rule, which is also found in a number of other Conventions, was borrowed from article 22 of the Convention of 14 March 1978 on the Law Applicable to Agency.

CHAPTER V—FINAL CLAUSES 181 The final clauses follow the customary lay-out of the Hague Conventions. Several innovations, which were also under discussion in connection with the revision of the Convention of 15 June 1955 on the Law Applicable to International Sales of Goods, were in the end rejected. It was thought that it was not a good idea to mix tested solutions with new formulations, which had not yet been examined by the Governments. 182 In particular, the Fifteenth Session gave up the idea, after rather long discussions, of including a clause dealing with revision of the Convention. The Sub-Committee on general and final clauses had proposed such a clause (Working Document No 45, article 31). That clause had provided that when the Convention had been in force for four years, a quarter of the Contracting States, with a minimum of five, might request the convocation of a conference for the Convention’s revision. The result of the revision was to take the form of a new international instrument and accession to that instrument would involve ipso jure the denunciation of the old Convention.

3. The Explanatory Report by Professor Alfred E. von Overbeck 493 This proposal was inspired by the preliminary draft Convention on sales. In particular, the automatic replacement of the old Convention by the new was intended to ward off the difficulties that had been encountered by reason of the fact that the 1955 Convention on Sales could be denounced only at five-year intervals. 183 During the first discussion the usefulness of clauses for revision was placed in doubt by the Austrian Delegate, but the proposal met with the agreement of all the other delegates except for some abstentions. The proposal then appeared in article 31 of the revised text, submitted for second reading (Working Document No 64), with some drafting changes. 184 On second reading, the criticism bore in particular on the last paragraph, which was drafted as follows (Working Document No 64): The consent to be bound by the new revising instrument shall ipso jure involve the immediate denunciation of this Convention, notwithstanding the provisions of its article 32 [article on denunciation, current article 31] if and when the new revising instrument shall have come into force in respect of the State concerned. The Austrian Delegate and the Reporter asked what the formulation “consent to be bound” meant; did this refer to signature or ratification? It was also asked whether the reference to the article on denunciation meant only that the formalities did not have to be observed, or also that the six-months’ period of delay would not apply. The usefulness of this clause was placed in doubt, in particular because, unlike the Sales Convention of 1955 the Convention on trusts can be denounced at any time with a delay of six months before the denunciation takes effect. There was also criticism of the solution on its merits, since the maintenance of the old Convention for relations with the States which do not wish to ratify the new one may be justifiable, given that certain provisions of the Convention refer to Contracting States. The proposal to delete this paragraph was however rejected by a weak majority and the provision was sent back to the Drafting Committee. 185 At a subsequent sitting the Swiss delegation asked formally for the deletion of the fourth paragraph. The Fifteenth Session decided by a large majority to open the discussion again. It voted then by a strong majority to delete the fourth paragraph. 186 The plenary session therefore found before it an article 31 which consisted only of the first three paragraphs (plenary session, Working Document No 2, article 31). Already, at the time of the second reading, the comment had been made that, without the last paragraph, the revision clause had no longer very much meaning. In the end the conference decided in plenary session to delete completely the article on revision of the Convention. Article 26 187 Article 26 governs the procedure for the reservations provided in articles 16, 21 and 22. It will be noted that, in application of the first paragraph, a reservation must be made at the latest at the time of the act which gives effect to the Convention in a State or in a territorial unit. In contrast, a reservation may be withdrawn at any time.

494 Appendices The second paragraph contains the important information that the possible reservations are set out exhaustively in the Convention. 188 This provision was proposed by the Sub-Committee on general and final clauses, following the models contained in other Hague Conventions (Working Document No 45, article 25) and it was adopted without discussion. Article 27 189 Article 27 sets forth the traditional solution of the Hague Conventions, which can be signed and ratified only by States which were Members of the Conference at the time of the Session at which they were drawn up. 190 The Fifteenth Session had received in this connection an alternative proposal from the Sub-Committee on general and final clauses (Working Document No 45, article 26, variants I and II). A first variant provided that the Convention would be open for signature by all States; the States moreover would also have been able to accede. without signing. This formulation was borrowed from the preliminary draft Convention on Sales, which will go to a special session of the Conference to which all States of the world, and not merely the Members of the Conference, are invited. The second variant, to the contrary, corresponded to the traditional system of the Hague Conventions. During the discussion a great majority of the delegations expressed themselves in favour of the second variant; so this variant was accepted with a single opposing vote and several abstentions. Article 28 191 This article provides the possibility for non-Member States to accede to the Convention after its entry into force. Under the third paragraph, such accessions will have no effect as regards relations between the acceding States and those Contracting States which raise an objection within one year. The Conference’s Member States may also raise such an objection if they become Parties to the Convention after the accession. 192 This provision was proposed by the Sub-Committee on general and final clauses for the situation which would arise from the acceptance of variant II of article 27 which, indeed occurred (Working Document No 45 article 27bis). Following the discussion, the specification that the accession can only take place after entry into force of the Convention, which had been omitted from certain recent Hague Conventions, was added following the example of some Conventions which were a little older such as, for example, the Convention of 2 October 1973 on the Law Applicable to Maintenance Obligations, article 21. 193 A proposal by the Swiss delegation to provide, in the third paragraph of article 28, for the more restrictive system of “positive acceptance”, in place and instead of that of “preventive opposition”, was withdrawn following the conference’s decision to adopt a convention which would be applicable erga omnes (Working Document No 56).

3. The Explanatory Report by Professor Alfred E. von Overbeck 495 194 It may be asked what significance the third paragraph of article 28 still has in a Convention which provides that its rules on applicable law and recognition are applicable to all trusts, whatever their source (supra, Nos 32–35). The reply may be made that the clause will have effects for those States which make the reservation of article 21 dealing with Chapter III. Any State, against the accession of which they will have raised an objection, will be for them a non-Contracting State, and they will not be bound to recognize trusts created under its law. The clause will also have effects for a State which will have made a limited use of article 20, by declaring that it will recognize trusts declared by judicial decisions in Contracting States. Article 29 195 This article provides the possibility for territorially non-unified States to apply the Convention only in certain territorial units, and subsequently to change the situation. The text proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 27) met in itself no objection. 196 It still contained the passage “. . . different systems of law are applicable in relation to matters dealt with in this Convention . . .”. These words were deleted in order to try to resolve a difficulty raised by the Netherlands delegation. That delegation pointed out that article 29 would not cover the case of its relations with the Netherlands Antilles, for which it handled international relations, but which, for the rest, had their independent legislative policy. This delegation desired the insertion of the following provision which appeared regularly in the Hague Conventions, and in particular at article 33 of the Convention of 25 October 1980 on International Access to Justice (Working Document No 57): Any State may, at the time of signature, ratification, acceptance, approval or accession, declare that the Convention shall extend to all the territories for the international relations of which it is responsible, or to one or more of them. Such a declaration shall take effect at the time the Convention enters into force for that State. This proposal was fought by reference to the solutions adopted in the preliminarly draft Convention on sales of goods. It was also said that this text was contrary to the decolonization policy of the United Nations, though it was not explained exactly what this contradiction consisted of. Given the concrete situation existing in the Netherlands, article 29 did not appear to be able to apply and the proposed clause seemed necessary. It was none the less eliminated by a clear majority. It is doubtful that the deletion of the passage quoted above will suffice to resolve the difficulty. Article 30 197 Article 30 provides for the Convention’s entry into force after the deposit of the third instrument of ratification, acceptance or approval. It distinguishes thereafter for the entry into force between States ratifying, accepting or approving—in other words, the Member States of the Conference—the acceding States and extensions to territorial units. It should be noted that in sub-paragraph b the three-months’ time period commences to run only after the expiry of the one-year time period referred to in the third

496 Appendices paragraph of article 29. An accession will therefore take effect only 15 months after the deposit of the instrument. 198 This provision is based on an article proposed by the Sub-Committee on general and final clauses (Working Document No 45, article 28), the drafting of which was made more precise in the meeting and by the Drafting Committee. Article 31 199 Article 31 concerns the denunciation of the Convention; it will be noted that unlike older Conventions the present Convention does not have a fixed duration, but may be denounced at any time. If a State purely and simply denounces it, the Convention will cease to have effect at the end of a period of three months. But the denouncing State may also set a later date; in contrast, it cannot make the effects of the denunciation depend upon a future event. 200 This clause was taken, with drafting changes, from article 30 of the proposal of the Sub-Committee on general and final clauses (Working Document No 45, article 30). Article 32 and signature clause 201 These provisions also correspond to the usual final clauses of the Hague Conference. It will be noted that the English and French texts are the official texts. Bettmeralp, January 1985

Bibliography F Albisinni, “National Digest for Italy”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A51 ——and R Gambino “The Italian Civil Law System and the Hague Convention on Trusts” (1993) 2 J Int Corp P 73, 76–9 ——and V Sinisi “Trusts from the Italian Perspective” in Trusts for Europe (Legal Studies and Services Ltd, 1988), 53 American Restatement (2d) on the Conflict of Laws, (St Paul, Minn, American Law Institute, 1971) Annual Report of the Charity Commissioners, 1992 A Anton (with P Beaumont), Private International Law: A Treatise from the Standpoint of Scots Law, 2nd edn., (Edinburgh, W Green, 1990) J de Arespacochaga El Trust: la Fiducia y Figuras Afines, (Madrid, Marcial Pons, 2000) E Arosemena, “Concept and Regulation of the Protector in Offshore Jurisdictions”, [2000] Private Client Business 287 T Arrigo (ed.), I Trusts nell’ordinamento Italiano, (Milan, Giuffrè, 1999) R Ashton, An Analysis of the Guernsey Law of Trusts, (London, Key Haven, 1998) R Atherton (ed.) International Encyclopaedia of Estate and Trust Law: Selected Papers 1997–1999, (The Hague, Kluwer, 2000) Australia Law Reform Commission Report No 58, Choice of Law K Baker and E Devenport, “National Digest for Jersey” in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A15 L Barnard, “Choice of Law in Equitable Wrongs: a Comparative Analysis” [1992] CLJ 474 H Batiffol & P Lagarde, “Droit International Privé” [1983] LGDJ, vol 2, no 575 B Beat Aussonderung von Treugut im Schweizerischen Partikularkonkurs, (Zurich, Schulthess, 1997) A Beck, “Floating Choice of Law Clauses” [1987] LMCLQ 523 D Benest, “The Trustee’s Dilemma: to Disclose or not to Disclose?” (2000) 6 Trusts and Trustees 6 ——, “Re Rabaiotti Revisited: are Letters of Wishes Disclosable?” (2001) 7 Trusts and Trustees 16 I Beneventi (ed.) I Trusts in Italia Oggi, (Milan, Giuffrè, 1996) J-P Béraudo, Les Trusts Anglo-Saxons et le Droit Français, (Paris, LGDJ, 1992) S Berti, “Trusts and the Lugano Convention—does it Matter? and what about the Hague Trust Convention” in N Vogt (ed.) Disputes Involving Trusts, (Saffron Walden, Trusts & Trustees, Helbing & Lichtenhahn, 1999), 9 W Binchy, Irish Conflict of Laws, (Ireland, Butterworths, 1988) J Bird, “Choice of Law” in F Rose (ed.) Restitution and the Conflict of Laws, (Oxford, Mansfield Press, 1995), chapter 3 ——, “Choice of Law Rule of Priority Disputes in Relation to Shares” [1996] LMCLQ 57

498 Bibliography P Birks, “Proprietary Rights as Remedies” in P Birks (ed.) “The Frontiers of Liability, (Oxford, Oxford University Press, 1994), volume 2, chapter 16 V Bolgár, “Why no Trusts in the Civil Law?” (1953) 2 Am J Comp Law 208 A Briggs, “The Validity of ‘Floating’ Choice of Law and Jurisdiction Clauses” [1986] LMCLQ 508 ——, “The Formation of International Contracts” [1990] LMCLQ 192 ——, “The International Dimension to Claims for Contribution” [1995] LMCLQ 437 ——, “From Complexity to Anticlimax: Restitution and Choice of Law” (1996) 4 RLR 88 ——, “In Praise and Defence of Renvoi” (1998) 47 ICLQ 877 ——, “Decisions of British Courts during 1999 Involving Questions of Public or Private International Law: B) Private International Law” (1999) 70 BYIL 319 A Briggs and P Rees, Civil Jurisdiction and Judgments, 2nd edn., (LLP, London, 1997) G Broggini “Folgen der Ratifikation des Trusts—Übereinkommens in Italien”, 9e Journée des Professeurs Suisses de Droit International Privé, Lausanne, 7 March 1997, available at http://www-isdc.ch/e/Broggini97.asp A Bureau, Le Contrat de Fiducie: Études de Droit Comparé Allemagne, France, Luxembourg: http://juripole.univ-nancy2.fr/memoires/compare/Antoine_Bureau/ E Burn, Cheshire and Burn’s Real Property, 16th edn., (London, Butterworths, 2000) A Burnand, “Laws of Succession and Testamentary Dispositions—Conflicts of Laws and Recognition of Foreign Trust—a Swiss Overview” (1995) 1(10) Trusts and Trustees 15 A Burrows, “Proprietary Restitution: Unmasking Unjust Enrichment” (2001) 117 LQR 412 E Calo, Dal Probate al Family Trust, (Milan, Giuffrè, 1996) D Campbell (ed.) “Offshore Trusts” [1995] Comparative Law Yearbook of International Business: Special Issue, (the Hague, Kluwer, 1996) J Campbell, “National Digest for Bermuda” in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter B5 P Carter, “Rejection of Foreign Law: Some Private International Law Inhibitions” (1984) 55 BYIL 111 ——, “The Foreign Limitation Periods Act 1984” (1985) 101 LQR 68 ——, The Role of Public Policy in English Private International Law (1993) 42 ICLQ 1 J-G Castel, Canadian Conflict of Laws, 4th edn., (Toronto and Vancouver, Butterworths, 1997) D Cavers “Trusts Inter Vivos and the Conflict of Laws”, (1930) 44 Harv LR 161 C Clarkson and J Hill, Jaffey on the Conflict of Laws, 2nd edn., (London, Butterworths, 1997) J Collier, [1989] All ER Rev, 61 L Collins (ed.), (with A Briggs, J Hill, J McClean and C Morse) Dicey and Morris, The Conflict of Laws, 13th edn., (London, Sweet and Maxwell, 2000); 1st Supplement to the 13th edn., (London, Sweet and Maxwell, 2001) A Conder “The Office of the Protector” (1995) 4 Trusts and Trustees 12 E Corso, Trustee e Gestione dei Beni in Trust, (Milan, Giuffrè, 2000) E Crawford, International Private Law in Scotland, (Edinburgh, W Green, 1998) P Croucher, “Trusts of Moveables in Private International Law” (1940) 4 MLR 111 C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs Fiducie in a Business Context, (Brussels, Bruylant, 1999) P Czermak Der Express Trust im Internationalen Privatrecht, (Frankfurt am Main, Verlag Peter Land, 1986)

Bibliography 499 C Davies, “Jersey: Disclosure of Trusts Documents” (2000) 6 Trusts and Trustees 33 V Delany, “Charitable Trusts and the Conflict of Laws”, (1961) 10 ICLQ 385 C Deneuville, “French Attitude Towards Trusts, in R Atherton (ed.) International Encyclopaedia of Estate and Trust Law: Selected Papers 1997–1999, (The Hague, Kluwer, 2000), chapter 2 J Denker, “The British Isles: England” in D Hayton (ed.), European Succession Laws, (Bristol, Jordans, 1998), chapter 4 C de Wulf, The Trust and Corresponding Institutions in the Civil Law, (Brussels, Emile Bruylant, 1965) W Diamond, D Diamond and V Kaplan, International Trust Laws and Analysis (Boston, Warren, Gorham and Lamont, loose-leaf) ——, ——and J Sullivan, International Trust and Laws Analysis, (The Hague, Kluwer, 2001—(loose-leaf)) G Droz, “La Conférence de la Haye et le Droit International Privé Notarial: Deux Récentes Conventions en Matière de Trusts et de Successions” [1989] Journal des Notaires et des Avocats 507 ——, “Regards sur le Droit International Privé”, (1991–IV) 229 Recueil des Cours 9 J Dubard, “‘La Fiducie’: Trusts in France” (1992) 3 International Company and Commercial Law Review 324 A Duckworth, “The Offshore Trust in Transition” in D Kozusko and J Schoenblum (eds.) International Estate Planning: Principles and Strategies”, (Chicago, American Bar Association, 1991), 129 ——, “Attacking International Trusts—the Problem of ‘Blowing Hot and Cold’ ” [1996] Private Client Business 97 ——, STAR Trusts, (Saffron Walden, Gostick Hall, 1998) ——, “STAR Wars: the Colony Strikes Back” (1998) 12 Trust Law International 16 ——, “STAR Wars: Smiting the Bull” (1999) 13 Trust Law International 158 ——, “An Offshore View of Forced Heirship—Global Conflict and its Planning Implications: Part 1” [1995] Private Client Business 270 ——, “An Offshore View of Forced Heirship—Global Conflict and its Planning Implications: Part 2” [1995] Private Client Business 334 ——, “An Offshore View of Forced Heirship—Global Conflict and its Planning Implications: Part 3” [1995] Private Client Business 408 ——, “Trust Law in the New Millennium: Part 1: Retrospective” (2000) 7(1) Trusts and Trustees 12 ——, “Trust Law in the New Millennium: Part 2: Prospective” (2000/1) 7(2) Trusts and Trustees 11 ——, “Trust Law in the New Millennium: Part 3: Fundamentals” (2000/1) 7(3) Trusts and Trustees 9 ——, “The New Frontier of Purpose Trusts”: http://www.il-trust-in-italia.it/webtutti/ @Congress%201999/Duckworth%20New%20frontier.htm ——, “National Digest for the Cayman Islands” in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A7 ——, “Forced Heirship and the Trust”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter B1 A Dyer, “International Recognition of the Trust Concept” (1996) 2 Trusts and Trustees 5 ——, “International Recognition and Adaptation of Trusts: the Influence of the Hague Convention” (1999) 32 Vanderbilt Journal of Transnational Law 989

500 Bibliography A Dyer and H van Loon, Report on Trusts and Analogous Institutions, Preliminary Document No 1, May 1982, Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable Law and Recognition, (The Hague, Permanent Bureau of the Hague Conference/SDU Publishers) H Dyson, “The Proposed New Law of Trusts in France” [1992] Conv 407 O Elias, Judicial Remedies in the Conflict of Laws, (Oxford, Hart Publishing, 2001) N Enonchong, “Public Policy in the Conflict of Laws: a Chinese Wall around Little England?” (1996) 45 ICLQ 633 D Evans, “Administering Foreign Settlements” (1986) 102 LQR 28 R Fentiman, Foreign Law in English Courts, (Oxford, Clarendon Press, 1998) H Fischer, “The Law Governing Capacity with Regard to Bills of Exchange” (1951) 14 MLR 144 P Fletcher, “Grupo Torras in the Bahamas Court of Appeal” (1998) 4(8) Trusts and Trustees 28 D Forbes-Jaeger and E Stormann, “The Trust in Switzerland” (1996) 4 J Int Corp P 75 ——, “The Trust in Switzerland—Revisited” (2000) 8 J Int Corp P 141 C Forsyth, “Characterisation Revisited: an Essay in the Theory and Practice of the English Conflict of Laws” (1998) 114 LQR 141 E Gaillard, “Les Enseignements de la Convention de la Haye du 1er Juillet 1985 relative à la Loi Applicable au Trust et à sa Reconnaissance”, [1990] Revue Juridique et Politique Indépendance et Cooperation 304 ——and D Trautman, “Trusts in Non-Trust Countries: Conflict of Laws and the Hague Convention on Trusts” (1987) 35 Am J Comp Law 307 A Gambaro, Il Trust nell Prassi Bancaria e Finanziariai, (Milan, CEDAM, 1998) M Giuliano and P Lagarde, “Council Report on the Convention on the Law Applicable to Contractual Obligations” OJ 1980 C282/4 J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf) H Goldstein, Trusts of Movables in the Conflict of Laws: a Comparative Study of the Laws of the United States and Germany with Reference to the Laws of Belgium, France, Italy and the Netherlands (dissertation, Köln, 1966) J Goldsworth, “Validity of Change of Law from the Bahamas to Cayman Islands” (1997) 3 Trusts and Trustees 7 ——, “A Better Definition of Trusts” (2000) 6 Trusts and Trustees 34 C Gonzales Beilfuss El Trust. Law Institución Angloamericana y el Derecho Internacional Privado Espagñol, (Barcelona, Bosch, 1997) Sir W Goodhart, “Trust Law for the Twenty-First Century” in A Oakley (ed.) Trends in Contemporary Trust Law, (Oxford, Clarendon Press, 1996), chapter 11 WGoodman,“VulnerabilityofTruststoAttack—RetentionofPowersandInterestsbySettlor” inJGlasson(ed.),InternationalTrustLaws,(Bristol,Jordans,loose-leaf),chapterB8 A Goodwin, “Links to China: the new PRC Trust Law” (2001) 29 Trusts and Estates Law Journal 19 M Graziadei, “Trusts in Italian Law”, in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs Fiducie in a Business Context, (Brussels, Bruylant, 1999), chapter E4, G Gretton, “Trusts without Equity” (2000) 49 ICLQ 599 M Grimaldi and F Barrière, “La Fiducie en Droit Français” in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs Fiducie in a Business Context, (Brussels, Bruylant, 1999), chapter 3

Bibliography 501 J Grodecki, “Conflict of Laws in Time” (1959) 35 BYIL 58 F Guillaume, “ Incompatibilité du Trust avec le Droit Suisse? Un Mythe s’effrite” [2000] Revue Suisse de Droit International et de Droit Européen R Ham, “Protectors”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter B3 P Hargreaves, “Problem Children: a Jersey Perspective on Assets Held by and for Minors” [2000] Private Client Business 222 J Harris, “Rights in Rem and the Brussels Convention”, (1997) 22 EL Rev 179 ——, “Choice of Law in Tort—Blending in with the Landscape of the Conflict of Laws?” (1998) 61 MLR 33 ——, “Justiciability, Choice of Law and the Brussels Convention”, [1999] LMCLQ 360 ——, “Forum Shopping in International Libel”, (2000) 116 LQR 562 ——, “Contractual Freedom in the Conflict of Laws” (2000) 20 OJLS 247 ——, “Ordering the Sale of Land Situated Overseas” [2001] LMCLQ 205 ——, “Transnational Trusts Litigation—Jurisdiction and the Enforcement of Foreign Judgments”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter C1 ——, “Launching the Rocket: Capacity and the Creation of Inter Vivos Transnational Trusts”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter C2 —— and F Meisel, “Public Policy and the Enforcement of International Arbitration Awards: Controlling the Unruly Horse” [1998] LMCLQ 568 D Hayton, “Trusts in Private International Law” (1985) 135 NLJ 18 ——, “The Hague Convention on the Law Applicable to Trusts and on their Recognition”, (1987) 36 ICLQ 260 ——, “Developing the Law of Trusts for the Twenty-First Century” (1990) 106 LQR 87 ——, “Trusts and Forced Heirship Problems”, (1993) 2 J Int Corp P 3 (also published in (1993–4) 4 KCLJ 12) ——, “The Hague Convention on Trusts: a Little is Better than Nothing but why so Little?” (1994) 3 J Int Corp P 23 ——, “Cross Border Estates: Part 1: an Introduction” [1994] Private Client Business 168 ——, “Cross Border Estates: Part 2: the Problems of Diversity” [1994] Private Client Business 250 ——, “Cross Border Estates: Part 3: the Hague Conventions” [1994] Private Client Business 329 ——, Underhill and Hayton, The Law Relating to Trusts and Trustees, 15th edn., (London, Butterworths, 1995), chapter 23 ——, Hayton and Marshall, Commentary and Cases on the Law of Trusts and Equitable Remedies, 11th edn., (London, Sweet and Maxwell, 2001) ——, “The Netherlands Implementation of the Hague Trusts Convention”, (1996) 5 J Int Corp P 127 ——, “The Significance of the Hague Conventions on Trusts and Succession: a Common Law Perspective” in A Borras, A Bucher, T Struycken and M Verwilghen (eds.), E Pluribus Unum (The Hague, Martinus Nijhoff, 1996) 121 ——, “STAR Trusts”, (1998) 8 Offshore Taxation Review 43 ——, “STAR Trusts” (1998) 4 Amicus Curiae 13 ——, “The Problems of Diversity”, in D Hayton (ed.), European Succession Laws, (Bristol, Jordans, 1998), chapter 1

502 Bibliography D Hayton, (ed.) Principles of European Trust Law, (The Hague, Kluwer, 1999) —— (ed.) Modern International Developments in Trust Law (The Hague, Kluwer, 1999) ——, “The Development of the Trust Concept in Civil Law Jurisdictions” (2000) 8 J Int Corp P 159 ——, “The Developing European Dimension of Trust Law (2000) 11 KCLJ 48 ——, “Whither Trusts in the Twenty-First Century? Part 1” [2000] Private Client Business 94 ——, “Whither Trusts in the Twenty-First Century?: Part 2” [2000] Private Client Business 163 ——, “Whither Trusts in the Twenty-First Century?: Part 3” [2000] Private Client Business 244 ——, “Developing the Obligation Characteristic of the Trust” (2001) 117 LQR 96 ——, “Some Major Developments in Trust Law” [2001] Private Client Business 361 ——, “The ‘Trusts Interni’ seen from England”: http://www.il-trust-in-italia.it/webtutti/ @Congress%201999/Hayton.htm ——, “International Recognition of Trusts” in J Glasson (ed.) International Trust Laws (Bristol, Jordans, loose-leaf), chapter C3 ——and S Kortmann (ed.) Vertrouwd met de Trust; Trust and Trust-Like Arrangements (Deventer, Tjeenk Willink, 1996) R Hendrickson and N Silverman, Changing the Situs of a Trust (New York, Law Journal Seminars Press, loose-leaf) J Herbots, and D Philippe, Le Trust et la Fiducie—Implications (Brussels, Bruylant, 1997) L Hessey, “When in Rome . . .” (2000) 82 In-House Lawyer 31 Il trust in Italia: Congresso Nazionale, Rome, 19–20 November 1999, available at: http://www.il-trust-in-italia.it/webtutti/@Congress%201999/Congresso%201999% 20tabella.htm A Jaffey, “Essential Validity of Contracts in the English Conflict of Laws” (1974) 23 ICLQ 1 ——, “Offer and Acceptance and Related Questions in the English Conflict of Laws” (1975) 24 ICLQ 603 C Jauffret-Spinosi “La Convention de la Haye Relative à la Loi Applicable au Trust et à sa Reconnaissance (1er Juillet 1985)” (1987) 114 Journal du Droit International 23 F Juenger, Choice of Law and Multistate Justice, (The Hague, Nijhoff, 1993) A Kaplan (ed.) Trusts in Prime Jurisdictions (The Hague, Kluwer, 2000) G Keeton, “Trusts in the Conflict of Laws” (1951) 4 CLP 107 M Keller and K Siehr, Allgemeine Lehren des Internationalen Privatrechts, (Zurich, Schulthess, 1986) P Kinsch “Trusts in a Civil Law Environment: Jersey Trustees may Sue in Luxembourg” (1996) 5 J Int Corp P 120 F Klein “A Propos de la Convention de La Haye du 1er Juillet 1985 Relative à la Loi Applicable au Trust et à sa Reconnaissance” in F Sturm (ed.) Mélanges Paul Piotet (Berne, 1990) I Koele, “France: a Law Proposal Introducing ‘la Fiducie’” in F Sonneveldt and H van Mens (eds.) The Trust—Bridge or Abyss between Common and Civil Law Jurisdictions?, (Deventer, Kluwer, 1992), chapter 6 M Koppenol-Laforce, Het Haagse Trustverdrag, (Deventer, Kluwer, 1997) ——, “The Trust, the Hague Trusts Convention and Civil Law Countries: a Mission Impossible?” (1998) 3 Notarius International 27

Bibliography 503 M Koppenol-Laforce and R Kottenhagen “The Institution of the Trust and Dutch Law” in Netherlands Reports to the XVth International Congress of Comparative Law, Bristol 1998 (Antwerpen/ Groningen Intersentia Rechtswetenschappen), section II A 3, p. 137, (also published in C Cumyn (ed.) La Fiducie face au Trust dans les Rapports des Affaires: Trust vs Fiducie in a Business Context, (Brussels, Bruylant, 1999), chapter 5) H Kötz, “Die 15 Haager Konferenz und das Kollisionsrecht der Trust” (1986) 50 Rabels Zeitschrift 562 ——, “The Hague Convention on the Law Applicable to Trusts and their Recognition”, in D Hayton (ed.) Modern International Developments in Trust Law (The Hague, Kluwer, 1999), chapter 3, (also published in A Kaplan (ed.) Trusts in Prime Jurisdictions (The Hague, Kluwer, 2000), chapter 1) B Labesse, “Why Litigate in Jersey?” (2001) 87 In House Lawyer 74 R Lafer and A Siegel, “Trusts of Movables in the Conflict of Laws” (1961) NYULR 713 P Lagarde “Le Nouveau Droit International Privé des Contrats” [1991] Revue Critique de Droit International Privé 288 P Lalive, The Transfer of Chattels in the Conflict of Laws: a Comparative Study, (Oxford, Clarendon Press, 1955) W Land, Trusts in the Conflict of Laws; Validity, Construction, Administration and Taxation of Trusts: what Law Governs (New York, Baker, Voorhis & Co, 1940) J Langbein, “The Contractarian Basis of the Law of Trusts” (1995) 105 Yale LJ 625 B Larking and J Dekker, “Use of Trust-Held Dutch Companies for International Tax Planning” in A Kaplan (ed.) Trusts in Prime Jurisdictions, (The Hague, Kluwer, 2000), chapter 14 V Latham, “The Creation and Administration of a Trust in the Conflict of Laws” (1953) 6 CLP 176 Law Commission Report No 193 on Choice of Law in Tort and Delict (London, HMSO, 1990) Law Commission Report No 251 on the Rules against Perpetuities and Excessive Accumulations (London, HMSO, 1998) R Lawrence, “The Role of the Protector—an Insulator for Corporate Beneficiaries?” (1993) 2 J Int Corp P 88 N Lawton, “The Cayman Islands: Sham Trusts” (1999) 17 Amicus Curiae 31 R Leslie, “The Relevance of Public Policy in Legal Issues Involving Other Countries and their Laws” (1995) 6 Jur Rev 477 D Libling, “Formation of International Contracts” (1979) 42 MLR 169 K Lipstein, “Trusts” in K Lipstein (ed.) International Encyclopaedia of Comparative Law, (Dordrecht, Mohr (Paul Siebeck), Tübingen and Martinus Nijhoff, 1994), chapter 23 Y Loussouarn, “La Convention de la Haye d’octobre 1985 sur la Loi applicable aux Contrats de Vente Internationale de Marchandises” [1986] Revue Critique de Droit International Privé 271 A Lowenfeld, “Tempora Mutantur—Wills and Trusts in the Conflicts Restatement”, (1972) 72 Col LR 382 M Lupoi, Appunti sulla Real Property e sui Trust nel Diritto Inglese (Milan, Giuffrè, 1971) ——, Introduzione ai Trusts: Diritto Inglese, Convenzione dell’Aja, Diritto Italiano (Milan, Giuffrè, 1994)

504 Bibliography M Lupoi, “The Shapeless Trust” (1995) 1(3) Trusts and Trustees 15 ——, “Trusts in the Civil Law—an Introduction” (1996) 2 Trusts and Trustees 21 ——, Trust Laws of the World (Rome, ETI, 1996), ——, “I Trusts Interni”: http://www.il-trust-in-italia.it/webtutti/@Congress%201999/ Lupoi.htm ——, “Effects of the Hague Convention in a Civil Law Country” in The Reform of Property Law (Aldershot, Ashgate, 1997), chapter 12, p 222 ——, “Effects of the Hague Convention in a Civil Law Country” (1998) 4 Trusts and Trustees 15 ——, “The Domestic Trust Theory is Upheld in Italy”, (1998) 4 Trusts and Trustees 24 ——, “Trusts and Civilian Categories” in D Hayton, S Kortmann and K Verhagen (eds.), Itinera Fiduciae, Trust & Treuhand in Historical Perspective, (Berlin, Duncker & Humblot, 1998) ——, “The Civil Law Trust” (1999) 32 Vanderbilt Journal of Transnational Law 967 ——, “The Civil Law Trust” in R Atherton (ed.) the International Academy of Estate and Trust Law: Selected Papers 1997–1999 (The Hague, Kluwer, 2000), chapter 4 ——, Trusts: A Comparative Study (Milan, Giuffrè, 1997), translated by S Dix, (Cambridge, Cambridge University Press, 2001) ——(and others), Fiducia, Trust, Mandato ed Agency (Milan, Giuffrè, 1991) S Mackintosh, “An Onshore View of Forced Heirship—Domestic Conflict and Its Planning Implications” [1996] Private Client Business 161 and 197 L Maerten, “Le Régime International du Trust après la Convention de La Haye du 1er Juillet 1985” [1988] La Semaine Juridique 3319 F Mann, “The Time Element in the Conflict of Laws” (1954) 32 BYIL 217 ——, “The Variation of Trusts Act, 1958, and the Conflict of Laws” (1964) 80 LQR 29 ——, “Subsequent Conduct as Aid to Construction”, (1973) 89 LQR 464 J Martin, Hanbury and Martin, Modern Equity, 16th edn., (London, Sweet and Maxwell, 2001) F Mascré, “National Digest for France” in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A10 P Matthews, “Constructive Trusteeship: Proprietary Claims, Personal Claims and the Hague Convention” (1995) 1 Trusts and Trustees 7 ——, “Protectors: Two Cases: Twenty Questions” (1995) 9 Trust Law International 108 ——, “The New Trust: Obligations without Rights?” in A Oakley (ed.) Trends in Contemporary Trust Law, (Oxford, Clarendon Press, 1996) ——, Trusts: Migration and Change of Proper Law, (London, Key Haven, 1997) ——, “Italian Trust for Debenture Holders” (1997) 11 Trust Law International 20 P Matthews, “Shooting STAR: the New Special Trusts Regime from the Cayman Islands” (1997) 11 Trust Law International 67 ——, “STAR: Big Bang or Red Dwarf?” (1998) 12 Trust Law International 98 ——, “Trusts & Equity: How Dumb is the Blind Trust?” (1998) 3 Amicus Curiae 22 ——, “Un Trust per l’Italiano” (1998) 12 Trust Law International 104 ——, “New Draft Italian Trust Law” (2000) 14 Trust Law International 33 ——and T Sowden, The Jersey Law of Trusts, 3rd edn., (London, Key Haven, 1993) P Mayer, Droit International Privé, 4th edn., (Paris, Montchrestien, 1991), T Mayer, Die Organisierte Vermogenseinheit Gemass Art 150 des Bundesgesetzes über das Internationale Privatrecht: unter besonderer Berucksichtigung des Trust (Basel, Helbing & Lichtenhahn, 1998)

Bibliography 505 P Maynard, “The Bahamas: the New Trustee Act” (1999) 16 Amicus Curiae 29 B McCutcheon and P Soares (eds.), Euro-Trusts: the New European Dimension for Trusts (Southampton, Legal Studies Publishing Ltd, 1993) J McGhee, Snell’s Equity, 30th edn., (London, Sweet & Maxwell, 2000) E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, (Oxford Clarendon Press, 1992) P McNair, “Anderson in the Court of Appeals—Contempt, Impossibility and Foreign Trusts” (1999) 5(9) Trusts and Trustees 27 S Meikeljohn, “The British Isles: Jersey” in D Hayton (ed.), European Succession Laws (Bristol, Jordans, 1998), chapter 4 O Merren in D Campbell (ed.) “Offshore Trusts” [1995] Comparative Law Yearbook of International Business: Special Issue, (The Hague, Kluwer, 1996), 78 P Messina, “Trusts: the Italian Approach” in N Vogt (ed.) Disputes Involving Trusts, (Trusts & Trustees; Saffron Walden, Helbing & Lichtenhahn, 1999), 295 G Miller, International Aspects of Succession, (Aldershot, Ashgate, 2000) J Morris, “The Time Factor in the Conflict of Laws” (1966) 15 ICLQ 422 ——and W Leach, “The Rule Against Perpetuities”, 2nd edn., (London, Stevens, 1962), ——and P North, Cases and Materials on Private International Law, (London, Butterworths, 1984) M Moshinsky, “The Assignment of Debts in the Conflict of Laws” (1992) 108 LQR 591 H Motulsky “De l’impossibilité de Creér un Trust Anglo-Saxon sous l’empire de la Loi Française” [1948] Revue Critique de Droit International Privé, 451 J Mowbray, L Tucker, N Le Poidevin and E Simpson, Lewin on Trusts, 17th edn., (London, Sweet & Maxwell, 2000), chapter 11 K Neuhoff and U Pavel (eds.) Trusts and Foundations in Europe: a Comparative Survey (London, Bedford Square Press, 1971) P North, “Reform, but not Revolution”, (1990–I) 220 Recueil des Cours 9 ——, Essays in Private International Law, (Oxford, Clarendon Press, 1993) ——, Private International Law Problems in Common Law Jurisdictions, (the Hague Martinus Nijhoff, 1993) ——and J Fawcett, Cheshire and North’s Private International Law, 13th edn., (London, Butterworths, 1999) Proceedings of the Fourteenth Session of the Hague Conference on Private International Law 1980, Book I—Miscellaneous Matter (The Hague, Permanent Bureau of the Hague Conference/SDU Publishers) Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition (The Hague, Permanent Bureau of the Hague Conference/SDU Publishers) P Nygh, Conflict of Laws in Australia, 6th edn., (Sydney, Butterworths, 1995) ——, Autonomy in International Contracts, (Oxford, Clarendon Press, 1999) ——and M Pryles “The Hague Convention on the Law Applicable to Trusts and on their Recognition” in Thirteenth International Trade Law Conference 1–3 November 1986, (Canberra, Australian Government Publishing Service, 1987) 199 A Oakley, Parker and Mellows, The Modern Law of Trusts, 7th edn., (London, Sweet & Maxwell, 1998), (esp. chapter 25) M Oldham, “Financial Obligations within the Family—Aspects of Intergenerational Maintenance and Succession in England and France” [2001] CLJ 128

506 Bibliography M O’Sullivan, “The Hague Convention on Trusts—Further Considerations” (1993) 2 J Int Corp P 65 G Palermo, “It Trust di Diritto Interno” http://www.il-trust-in-italia.it/webtutti/ @Congress%201999/Palermo.htm E Paltzer, “Trusts in Switzerland: some Recent Decisions of the Swiss Courts”, (2000) 8 J Int Corp P 137 G Panagopoulos, “Cross Border Tracing: Conflict of Laws Issues” (1998) 6 RLR 73 ——, Restitution in Private International Law, (Oxford, Hart Publishing, 2000) A Paton and R Grosso, “The Hague Convention on the Law Applicable to Trusts and on their Recognition: Implementation in Italy”, (1994) 43 ICLQ 654 R Pearce and J Stevens, the Law of Trusts and Equitable Obligations, 2nd edn., (London, Butterworths, 1998) P Piccoli “La Convenzione dell’ Aja sulla Legge Applicabile ai Trust” in Fiducia, Trust, Mandato ed Agency, (Milan, Giuffrè, 1991) D Pierce, “Post-Formation Choice of Law in Contract” (1987) 50 MLR 176 S Rae, “Trusts: the Scottish Experience” in N Vogt (ed.) Disputes Involving Trusts, (Trusts & Trustees; Saffron Walden, Helbing & Lichtenhahn, 1999) W Reese “Dépeçage: a Common Phenomenon in Choice of Law” (1973) 73 Col L Rev 58 M Revillard, “France” in D Hayton (ed.), European Succession Laws, (Bristol, Jordans, 1998), chapter 7 C Reymond, “Réflexions de Droit Comparé sur la Convention de la Haye sur le Trust” (1991) LXVIII Revue de Droit International et de Droit Comparé 7 ——, “La Suisse et la Convention de la Haye sur la Reconnaissance du Trust”, 9e Journée des Professeurs Suisses de Droit International Privé, available at http://wwwisdc.ch/e/Reymond97.asp M Rheinstein and M Glendon, International Encyclopaedia of Comparative Law Vol IV, Persons and Family (Dordrecht, (Mohr (Paul Siebeck), Tübingen and Martinus Nijhoff, 1980) , chapter 4 E Rimmell, “The Place of Renvoi in Transnational Litigation—a Pragmatic Approach to an Impractical Doctrine” (1998) 19 Hold LR 55 C Robertson, “International Succession Law” (1995) 2 Trusts and Trustees 20 P Rogerson, “The Situs of Debts in the Conflict of Laws—Illogical, Artificial and Misleading” [1990] CLJ 441 ——, “Choice of Law—Notice of Assignment” [2001] CLJ 14 F Rose (ed.) Restitution and the Conflict of Laws, (Oxford, Mansfield Press, 1995) M Rosenberg, P Hay and R Weintraub, Cases and Materials: Conflict of Laws, 10th edn., (New York, Westbury, 1996) M Rubino-Sammartano and C Morse, Public Policy in Transnational Relationships, (The Hague, Kluwer, 1992) P St J Smart, “The Rule against Foreign Revenue Laws” (2000) 116 LQR 360 A Saker and J McKeown, “Reservation of Powers to Settlors” [2001] PCB 154 V Salvatore, Il Trust: Profili di Diritto Internazionale e Comparato (Padova, CEDAM, 1996) (esp. part 2, chapters 4–6) J Schipper, “ Some Dutch Fiscal Aspects Regarding Trusts” (1997) 6 Trusts and Trustees 21 J Schoenblum, “Choice of Law and Succession to Wealth: a Critical Analysis of the Ramifications of the Hague Convention on Succession to Decedent’s Estates (1991) 32 Vanderbilt Journal of International Law 84

Bibliography 507 ——, “The Hague Convention on Trusts: Much Ado about Very Little” (1994) 3 J Int Corp P 5 ——, “Adaptation of Asset Protection Trusts” in E McKendrick (ed.) Commercial Aspects of Trusts and Fiduciary Obligations, (Oxford, Clarendon Press, 1992), chapter 11 E Scoles, P Hay, P Borchers and S Symeonides, The Conflict of Laws, 3rd edn., (St Paul, West Group, 2000) A Scott and W Fratcher, the Law of Trusts, 4th edn., (Boston, Little, 1989) J Smalley, “National Digest for the Isle of Man”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A14 J Smith, “Capacity in the Conflict of Laws—a Comparative Study” (1952) 1 ICLQ 446 L Smith, The Law of Tracing, (Oxford, Oxford University Press, 1997) ——, “Unjust Enrichment, Property and the Structure of Trusts” (2000) 116 LQR 412 P Soares, Non-Resident Trusts, 5th edn., (London, Sweet and Maxwell, 1998) F Sonneveldt, De Anglo-Amerikaanse Trust in het Nederlandse Civiele en Fiscale Recht (Deventer, FED, 1990) F Sonneveldt and H van Mens, The Trust: Bridge or Abyss Between Common and Civil Law Jurisdictions?, (Deventer, Kluwer, 1992) E Spiro, “The Incidence of Time in the Conflict of Laws” (1969) 19 ICLQ 357 J Stevens, “Restitution or Property? Priority and Title to Shares in the Conflict of Laws” (1996) 59 MLR 741 R Stevens, “The Law Applicable to Priority in Shares” (1996) 112 LQR 198 —— in F Rose (ed.) Restitution and the Conflict of Laws, (Oxford, Mansfield Press, 1995), chapter 5 ——, “Resulting Trusts in the Conflict of Laws” in P Birks and F Rose (eds.) Restitution and Equity, Volume 1: Resulting Trusts and Equitable Compensation, (London, Mansfield Press, 2000), 147 P Stibbard, “Swiss Courts Declare Guernsey Trust a Sham” [2001] Private Client Business 111 P Stone, “Time Limitation in the English Conflict of Laws”, [1985] LMCLQ 497 P Supino and A Limburg, “A Swiss Perspective on Trusts” in A Kaplan (ed.) Trusts in Prime Jurisdictions, (The Hague, Kluwer, 2000), chapter 20 W Swabenland, “The Conflict of Laws in Administration of Express Trusts of Personal Property”, (1936) 45 Yale LJ 438 W Swadling, “A Claim in Restitution?” [1996] LMCLQ 63 E Sykes and M Pryles, Australian Private International Law, 3rd edn., (Sydney, Law Book Co, 1991) K Takahashi, Claims for Contribution and Reimbursement in an International Context: Conflict of Laws Dimensions to Third Party Procedure, (Oxford, Clarendon Press, 2000) M Taylor, [2001] European Legal Business 57 L Thévenoz, Trusts in Switzerland. Ratifiction of the Hague Convention on Trusts and Codification of Fiduciary Transfers (Zurich, Schulthness, 2001) G Thomas, “Purpose Trusts”, in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter B4 H Thompson, “The Bahamas: Amendments to the Trusts (Choice of Governing Law) Act 1989” (1997) 3 Trusts and Trustees 4 A Thomson, “A Different Approach to Choice of Law in Contract” (1980) 43 MLR 650

508 Bibliography A Travers and J Appleyard, “Trusts in the Cayman Islands”, in A Kaplan (ed.) Trusts in Prime Jurisdictions, (The Hague, Kluwer, 2000), chapter 11 T Trumpy, “Non-Corporate Entities in Private International Law: the Trust” [1989] Hague Yearbook of International Law, 211 C van Boeschoten, Het Haagse Trustverdrag in Nederlands Perspectief Preadviezen uitgebracht voor de Vereniging voor Burgerlijk Recht (Lelystad, Koninklijke Vermande, 1994) H van Mens, “The Trust and Swiss Tax Law” in F Sonneveldt and H van Mens (eds.) The Trust, Bridge or Abyss between Common and Civil Law Jurisdictions?, (Deventer, Kluwer, 1992), chapter 4 —— “The Trust and Dutch Law: The Trust and the Netherlands’ Individual Income Tax, New Wealth Tax and Corporate Income Tax” in The Trust, Bridge or Abyss between Common and Civil Law Jurisdictions?, (Deventer, Kluwer, 1992), chapter 5 R Venables, Non-Resident Trusts, 8th edn., (London, Key Haven, 2001) H Verhagen, “Trusts in the Civil Law: Making Use of the Experience of ‘Mixed’ Jurisdictions”, (2000) 8 European Review of Private Law 477 H Verhagen and N Faber, “A Trace of Chase Manhattan in the Netherlands?” (1998) 6 RLR 165 G Virgo, “Interest, Constructive Trusts and the Conflict of Laws” (2000) 8 RLR 122 A Von Overbeck, Explanatory Report on the Convention on the Law Applicable to Trusts and on their Recognition, Proceedings of the Fifteenth Session of the Hague Conference on Private International Law 1984, Book II—Trusts—Applicable law and Recognition 370 (also reproduced in (1986) XXV International Legal Materials 593) A von Overbeck, “Law Applicable to, and Recognition of Trusts in Switzerland: the Possible Future under the Hague Convention” (1996) 4 Trusts and Trustees 6 ——, “Trusts: the Hague Convention and the Law of the Netherlands”, 9e Journée des Professeurs Suisses de Droit International Privé, available at http://wwwisdc.ch/e/vO97.asp ——, “La Ratification de la Convention de la Haye sur le Trust par les Pays Bas: un Exemple pour la Suisse?” in Collisio Legum, Mélanges offerts à G Broggini, (Milan, Giuffrè, 1997) ——, “National Digest for Switzerland” in J Glasson (ed.) International Trust Laws, (Bristol, Jordans, loose-leaf), chapter A52 F von Savigny, A Treatise on the Conflict of Laws, 2nd edn., (W Guthrie translation, 1880) R Vos, “Migration of Trusts and Change of Proper Law”, (2001) 7(5) Trusts and Trustees 24 A Wallace, “Choice of Law for Trusts in Australia and the United Kingdom” (1987) 36 ICLQ 454 D Waters “Unification or Harmonization? Experience with the Trust Concept” in W Stoffel and P Volken (eds.) Conflits et Harmonisation—Kollision und Vereinheitlichung—Conflicts and Harmonization: Mélanges en l’honneur d’Alfred von Overbeck à l’occasion de son 65ème anniversaire (Fribourg, Éditions Universitaires, 1990) 591 ——, Explanatory Report on the convention on the Law Applicable to Succession to the Estates of Deceased Persons by Professor D Waters, Proceedings of the Sixteenth Session of the Hague conference on Private International Law 1990, Book II— Succession to Estates—Applicable Law, 526

Bibliography 509 ——, “The Institution of the Trust in Civil and Common Law” (1995) 252 Recueil des Cours 117, 427–435 ——, “The Protector: New Wine in Old Bottles?” in A Oakley (ed.) Trends in Contemporary Trust Law (Oxford, Clarendon Press, 1996), chapter 4 ——, “The Concept Called ‘the Trust’ ” (1999) 53 Bulletin for International Fiscal Documentation 118 ——, The Trust in Civil Law Jurisdictions—the Dutch Experience (1999) 7 J Int Corp P 131 P Webb, “Trusts, Conflicts and New Zealand” [1975] NZLJ 128 M Whincop, M Keyes, R Posner Policy and Pragmatism in the Conflict of Laws (Aldershot, Dartmouth, 2001) R White, “Equitable Obligations in Private International Law: the Choice of Law” (1986) 11 Sydney LR 92 K Williams, “Trusts and the Hague Convention” (1986) 7 Company Lawyer 95 W Wilson (ed.), Trusts and Trust-Like Devices (United Kingdom Comparative Law Series, Vol. 5) (London, Chameleon Press, 1981) G Wittuhn, Das Internationale Privatrecht des Trusts (Frankfurt, Lang, 1987) T Yeo, “Choice of Law for Fiduciary Duties” (1999) 115 LQR 571 ——, “Constructive Trustees and the Brussels Convention” (2001) 117 LQR 560 G Zaphiriou, The Transfer of Chattels in Private International Law: A Comparative Study (London, Athlone Press/University of London, 1956)

Index Absence of choice, applicable law (Article 7), 215–32 administration, place of, 218–20 application, 215–18, 227–8 beneficiaries, place of residence, 224–5 chosen law, objective connection, 179 claw-back, 58, 59 closest connection, law of: Brussels Convention/ Regulation and, 232 common law, 215 invalidity of certain terms, 229–31 time for assessment, 228–9 validity, 227 dépeçage, 162 forced heirship claims, 57 non-trust state, objective factors pointing to, 226–7 objects of trust, 222–4 origin, 215–18 place of fulfilment, objects of trust, 223–4 settlor, place of residence, 224–5 situs of trust assets, 220–1 splitting the applicable law, 289 testamentary trusts, 225–6 trustee, place of residence or business, 221–2 trusts, intention to create, 26 see also Splitting applicable law: absence of choice, in Account, meaning, 271–2 Accountability issues, trustees, 109–10 Accumulations, 33–4 Administration: applicable law, scope of, 234 trustees, 241-2 place of, 218–20 determining and changing, 308–10 trustees: accountability for, 271–2 scope of applicable law, 241–2 validity, distinguishing from, 283–9 Administration of estates: mandatory rules, 373 testamentary trusts: applicable law, 45 Commonwealth, 45 delimiting, 45–6 grant of probate, English law, 43–4 jurisdiction of courts, 44–5 Northern Ireland, 45 Scotland, 45

Agency law: characteristics of trusts, 106 recognition of trusts, 334 Albisinni, F., 339, 400 America, see United States of America American Restatement (2d) on the Conflict of Laws, 204, 169, 286, 287, 290, 292 Ante-nuptial contracts, 63–4 absence of, 62–3 Applicable law: absence of choice, 215–32 administration, place of, 218–20 Article 7, application, 215–18, 227–8 beneficiaries, place of residence, 224–5 Brussels Convention/Regulation, 232 closest connection, law of, 227, 228–31 invalidity of certain terms, 229–31 non-trust state, objective factors pointing to, 226–7 objects of trust, 222–4 origin and application, 215–18 place of fulfilment, objects of trust, 223–4 settlor, place of residence, 224–5 situs of trust assets, 220–1 testamentary trusts, 225–6 trustee, place of residence/business, 221–2 validity, 227 changing, 297–310 administration, place of, 308–10 change of law, 299 charitable purpose trusts, 304 closest connection law, new, 298–9 content of applicable law, 306–8 forum, law of, 301 motivation for and permissibility, 298 numbers of trusts, 304–6 “one trust” approach, 304–5 other laws, 301–3 retrospective change, 301 role of law initially governing trust, 299–301 severable elements of trust, 303–4 validity, 302 variation of trusts, 299 scope of: administration matters, 46, 234 choice of law clause, consent to, 189–90 construction, 233–4 distribution of trust assets, 271 “effects” of trust, 234

512 Index Applicable law (cont.): scope of (cont.): essential validity, 276–8 existence of terms, 233 formal validity, 233, 272–6 offshore jurisdictions, 73 trustees, 235–56, 271–2 validity of terms, 233 variation of trusts, 256–71 splitting: absence of choice, in, 289 administration, 283–9, 292–3 charitable purpose trusts, 291–4 choice of law, non-trust states, 289–90 cy-près doctrine, settling 293–4 matters uncontrolled by terms of trust, 290–1 non-charitable purpose trusts, 295–6 right to split, 281–2 severable parts of trust, 282–3, 303–4 validity, 283–9, 291–2 Article 1 (functions of Convention): choice of law, 100–1 claw-back, 57–8 recognition of trusts, 101–2 Article 2 (characteristics of trusts): 103–22 characterisation, 116–20 express trusts, 104 general characteristics, 105–6 Rome Convention, 120–2 “shapeless” trust, 111–16, 352, 422, 423 specific characteristics, 107–10 Article 3 (types of trust governed by Convention): 123–50 constructive, 128–33 UK, extension of Convention rules, 140–1 formalities, 123–4 Hayton’s views, 133–4 judicially created trusts, exclusion, 134–9 continuing obligation, trusts imposing, 138–9 foreign judgments, recognition “problem”, 135–7 generally, 134–5 “remedial” judicial trusts, 137 resulting trusts, 125–7 UK, extension of Convention rules, 140–1 statutory trusts, 133 statutory extension, UK: trusts governed by law of any part of UK, 141–4 choice of law rule, resulting and constructive trusts, 140–1 relevant provisions, 139–40 trusts “arising” by judicial decision, 145–9 voluntary, 124–5 Article 4 (preliminary matters, exclusion): 151–7

constitution of trusts, 7 generally, 3–78 perpetuity rules, 33 “rocket”/“rocket-launcher”, distinguishing, 153–7 validity, 5 Article 5 (non-trust jurisdictions): 158–65 application, 158–61 categories of trust, 163–5 dépeçage, 161–3 Article 6 (choice of law, express or implied): autonomy principle, 166–9 consent, choice of law clause, 188–90 contracts and trusts, 186–8 implied choice, 191–8 “circumstances of the case”, 196–8 common law authorities, 198–208 law of closest connection, 194–5 nature of, 191–4 objective proper law and, 198–208 “reasonable certainty” test, 194 settlor’s purpose, 198 “international” trusts, 182–4 mandatory rules and, 190–1 multiple settlements, 186 objective connection to chosen law, need for, 179–82 objects of trusts and, 223 recognised legal systems, 185–6 several territorial units, state comprising, 184–5 “specific relationship” test, 187 “substantial relation” test, 181 trusts of land, 169–78 see also Chosen law Article 7 (absence of choice, applicable law): 215–32 administration, place of, 218–20 application, 215–18, 227–8 beneficiaries, place of residence, 224–5 chosen law, objective connection, 179 claw-back, 58, 59 closest connection, law of: Brussels Convention/Regulation and, 232 common law, 215 invalidity of certain terms, 229–31 time for assessment, 228–9 validity and, 227 dépeçage, 162 forced heirship claims, 57 non-trust state, objective factors pointing to, 226–7 origin, 215–18 settlor, place of residence, 224–5 situs of trust assets, 220–1 splitting the law, 289 testamentary trusts, 225–6 trustee, place of residence or business, 221–2

Index 513 trusts, intention to create, 26 Article 8 (scope of applicable law): administration matters, 46, 234 choice of law clause, consent to, 189–190 construction, 233–4 “effects” of trust, 234 essential validity, 276–8 existence of terms, 233 formal validity, 233, 272–6 offshore jurisdictions, 73 termination of trusts, 270–1 trust assets, distribution, 271 trustees: appointment, 235–7 beneficiaries, relationship with, 245–56 capacity, 6, 21, 237–9 delegation, 240, 245 duration of trust and, 244–5 duty to account for administration, 271–2 income of trust, accumulation, 244–5 investment powers, 243–4 removal, 235–7 rights and duties among themselves, 239–40 third party liability, 254–6 trust assets, administration of, 241–2 trust property, powers relating to, 242–3 vicarious liability, 240, 245 validity of terms, 233 variation of trusts: and administration, 270–1 discretion, exercise of, 267 divorce settlements, 257–61 domestic mandatory rules, 264–7 English law, application by foreign court, 269 foreign trusts, English statutes, 264–9 forum, rules of, 267–8 public policy, English, 268–9 statutory provisions, 261–4 and termination, 270–1 Article 9 (splitting the applicable law): absence of choice, in, 289 charitable purpose trusts: administration, 292–3 cy-près doctrine, settling, 125, 293–4 validity, 291–2 choice of law, non-trust state, 289–90 “horizontal” splits, 283 matters uncontrolled by terms of trust, 290–1 non-charitable purpose trusts, 295–6 right to split, 281–2 severable parts of trust, 282–3, 303–4 validity, distinguishing from administration, 283–9 “vertical” splits, 283

Article 10 (changing applicable law): administration, place of, 308–10 charitable purpose trusts, 304 closest connection, new law of, 298–9 content of law, 306–8 forum, law of, 301 Italian courts, separate fund, 315–17 motivation for and permissibility, 297–8 number of trusts, 304–6 “one trust” approach, 304–5 other laws, 301–3 retrospective change, 301 role of law initially governing trust, 299–301 severable elements of trust, 303–4 trust assets and personal assets: Italy, interpretation in, 318–19 Netherlands, 319–21 separate status of, 317–21 validity, 302 variation of trusts, 299 Article 11 (recognition of trusts): characteristics of trusts, 107 charitable purpose trusts and, 294 criticisms of Convention and, 424 mandatory rules, 315 need for, 311–13 and non-trust states, 159 settlor, claims against, 314–15 specific requirements of recognition, 314–15 “translating” trust into civil law systems: agency and mandate, 334 contractual approaches, 331–2 problems, 335–6 revocation powers, 334–5 scope of discussion, 330 testamentary trusts, 334–5 trust as company, 332–3 types of trust, 330–1 trust assets: following and tracing, 321–9 status, 314–15 uniform law, 313, 314 Article 12 (registration of trust): 337–40 chose in action, title to, 32 English reactions, 338–9 facility or obligation, 340 Italy, 339–40 prohibition of registration, 337–8 registration permissible, 337 Article 13 (right to refuse recognition, connection to non-trust states): chosen law, objective connection, 179 general application, 343–6 Italian courts, application in, 350–3 non-recognition, consequences, 349–50 partial recognition, 349 purpose of, 341–2

514 Index Article 13 (cont.): “significant elements”, 346–8 time factor, 348–9 United Kingdom, 342–3 Article 14 (freedom to adopt more liberal recognition rules), 354 Article 15 (mandatory rules): application, 363 characterisation, 357–60 creditors, protection in insolvency matters, 376–7 domestic, 264–7, 356–7 forum, rules of, 361–2 function, 355 incapable parties, protection of, 363–4 international, 356–7 marriage, effects of, 264, 364–5 matters uncontrolled by terms of trust, 291 minors, protection of, 363–4 mutual exclusivity, 357–60 nature, 355–6 offshore jurisdictions: Bermuda, 70 Isle of Man, 75–6 otherwise giving effect to objects of trust, 378–9 public policy (Article 18), inter-relationship of, 392–3 “rocket-launching” and, 360–1 succession rights, testate and intestate: administration of estate, 373 civilian perspective, 370–1 claw-back claims, 368–9 forced heirship, 365–7 inter-vivos trusts, 368–9 offshore dimension, 371–3 third party protection, 377–8 transfer of title, 373–6 variation of trusts, 264–5, 266, 267 see also Mandatory rules Article 16 (international mandatory rules of forum): 380–6 Article 15, interaction with, 384 determination of international application, 383–4 scope, 380–1 UK, exclusion in, 381–3 undermining of Convention, 385–6 variation of trusts and, 265, 267–8 Article 17 (renvoi): case for exclusion evaluated, 387–8 choice of law, recognised legal systems, 186 express choice of state’s private international rules, 388–9 related areas of law and, 389 see also Renvoi doctrine Article 18 (public policy): close connection, states of, 393–4

mandatory rules (Article 15), interrelationship, 392–3 scope, 390–2 STAR trusts, 395–6 and variation of trusts, 268–9 Article 19 (fiscal matters, exclusion): civil law states: general, 399 Italy, 400–1 Netherlands, 400 neutral approach, 397 United Kingdom, 397–9 refusal to enforce foreign revenue laws, 398–9 Article 20 (extension of Convention to other types of trust), 402–3 Article 21 (recognition, right to restrict to trusts governed by law of contracting state): ascertainment of contracting status, 405 change in a state’s contracting status, 405 compromise, 404 Article 22 (time factor): 406–8 Hague Convention, application in UK, 94–7 Article 23 (states consisting of several territorial units), 307, 409 Article 24 (application between territorial units of a State): intra-United Kingdom conflicts, 411 permissibility and desirability, 410–11 Article 25 (relationship to other international Conventions), 412–13 Article 26 (reservations to scope of Convention), 414 Article 27 (accession of Member States), 415 Article 28 (accession of Non-Member States), 416 Article 29 (application of Convention to selected territorial units of a state), 417 Article 30 (entry into force of Convention), 418 Article 31 (denouncing Convention), 419 Article 32 (notification process), 420 Assets of trust, see Trust assets Assignment/assignability of debts, 31, 32 Attorney, power of, 251 Australia: application of Convention between territorial units, 410 choice of law, 169, 195, 250–1 chosen by parties, 13–14 express or implied, 180, 200, 208, 211 determining place of administration, 310 dépeçage, 161 tracing trust assets, 323 validity, 227, 285 vesting property in the trustee inter vivos, 24 Autonomy principle, choice of law, 166–9

Index 515 Bahamas: changing of applicable law, 300 trusts of land, 175 Bare trusts, 109 Barnard, L., 147, 250 Beneficiaries: administration of trust and rights of, 285–6 capacity of, 21–3 defined, 245–6 “identification and extent” of rights, 258 residence, place of, 224–5 settlors as, 110 termination of trusts, seeking, 82 transfer of equitable title to, 24 trustees as, 110 trustees, relationship with, 245–56 fiduciary duties, 246–53 remedies for liabilities, 253–4 trusts, not express in nature, 96 Béraudo, J-P., 130–1 Bermuda, creation of transnational trusts, 69–72 Bona vacantia rule, intestate succession, 51 Brussels Convention/Regulation: characterisation, 116 closest connection, 232 judgments, recognition, 135–6 mutual exclusivity, 358 objects of trust, place of fulfilment, 223 recognition and enforcement of trust judgments, 85 Business, place of, trustee, 221–2 Canada: application of Convention between territorial units, 410–11 choice of law rule, 142, 184–5 fiducie concept, 107–8 states consisting of several territorial units, 409 testamentary trusts, absence of choice, 225–6 Capacity of settlor: beneficiaries, of, 21–3 characterisation, 8–9 choice of law rule, 10–20 contracts, 11–15 lex domicilii, 11 property, transfer of, 15–16 trusts, 16–20 declarations of trust, 6, 19 defined, 7 and essential validity, 7–10 exclusion from Hague Convention, 5–6 inter vivos trusts, 11, 13, 52 lex domicilii (law of settlor’s domicile), 9, 11, 20

lex loci contractus (place of conclusion of contract), 11, 13 status and, 7 testamentary powers of appointment, 60 testamentary trusts, 48, 50, 52 testate succession: immovable property, 50 movable property, 48 trustees, of, 6, 21, 237–9 trusts, 16–20 corporate trustees, 19–20 creation of structure, 18–19 declarations by trustee, 6, 19 property transfer, 17–18 see also Settlors Categories of trust: mandatory rules, 357 non-trust jurisdictions, 163–5, 345–6 Cayman Islands: changes of applicable law, 300, 304 creation of transnational trusts, 66–9 STAR trusts, 295, 395–6 Trusts (Foreign Element) Law 1987 (TFEL), 66, 175 trusts of land, 175 Characterisation process: 116–20 capacity, 8–9 defined, 116 essential validity, 9–10 mandatory rules, 357–60 non-discretionary approach, 120 Characteristics of trusts (Article 2): characterisation, 116–20 express trusts, 104 general, 105–6 Rome Convention, 120–2 “shapeless” trust, 111–16, 352, 422, 423 specific, 107–10 Charitable purpose trusts: administration, 292–3 applicable law, changing, 304 cy-près doctrine, 125, 293–4 and non-charitable purpose, 109–10, 163–4, 295–6 objects of trust and, 223 validity, 291–2 Chattels, inter vivos trusts of, 30 Choice of law: capacity of settlor, 10–20 contracts, 11–15 lex domicilii, 11 transfer of property, 15–16 constructive trusts, 140–1 express or implied, 166–214 autonomy principle, 166–9 common law authorities, 198–208 consent to choice of law clause, 188–90 contracts and trusts, 186–8

516 Index Choice of law (cont.): express or implied (cont.): implied choice, 191–8 “international” trusts, 182–4 mandatory rules and, 190–1 multiple settlements, 186 objective connection to chosen law, 179–82 objective proper law and, 198–208 objects of trust and, 223 recognised legal systems, 185–6 several territorial units, state comprising, 184–5 “specific relationship” test, 187 “substantial relation” test, 181 trusts of land, 169–78 “floating”, 298 formal validity, 233, 272–6 functions of Convention (Article 1), 100–1 inter vivos trusts, 33 matrimonial property regimes, 62–4 ante-nuptial contracts existing, 63–4 ante-nuptial contracts not existing, 62–3 immovable property, 63 movable property, 62–3 succession, law applicable to, 64 non-trust states, law governing part of trust, 289–90 resulting trusts, 140–1 Chose in action, title to, 32 “Circumstances of case”, implied choice, 196–8 Civil law states: taxation and: general, 399 Italy, 400–1 Netherlands, 400 “translating” of trust into: agency and mandate, 334 contracts, 331–2 problems of translation, 335–6 revocation powers, 334–5 scope of discussion, 330 testamentary trusts, 334–5 trust as company, 332–3 types of trust, 330–1 see also Non-trust jurisdictions Civil Procedure Rules (CPR), 82 Civilopian law (law of a mythical civilian state): characterisation process, 118, 119, 120 constructive trusts, 131 continuing obligation, trusts imposing, 138–9 discretion and, 119 foreign judgments, recognition “problem”, 136 governing part of trust, 289, 290 Hayton’s views on, 134

implied choice of law, 192 non-trust states, refusal of recognition of trusts connected to (Article 13), 340, 341, 345, 347 recognition of trusts, 101, 102 resulting trusts, 126 trusts of land, 170–1 Claw-back claims: Hague Convention on Law Applicable to Estates of Deceased Persons (1989), 57–8 mandatory rules, 368–9 Closest connection, law of: absence of choice, 215 Brussels Convention/Regulation and, 232 cases, 14 changing, 298–9 common law doctrine, 215 implied choice and, 194–5 invalidity of certain terms, 229–31 non-trust jurisdictions, 160–1 time for assessment, 228–9 validity and, 227 Closest connection, state of: mandatory rules, 381–3 exclusion in UK, 383 public policy, 393–4 Commonwealth, grants of probate, 45 Constitution of trusts: capacity of settlor, 7 inter vivos, 23 Rome Convention, exclusion from, 31 Constructive trusts: 128–33 choice of law, 140–1 institutional, 146, 147 judicial decision, trusts “arising” by, 145–7 “statutory jurisdiction trusts”, 145 third parties and, 326–7 transfers of assets, 138 Consultative Group on Trusts, formation of, 183 Contracts: choice of law, capacity of settlor, 11–15 trusts and, 186–8 Contrat sans loi, 185 Corporate trustees, capacity of settlor, 19–20 Corpus alienum, 81 CPR (Civil Procedure Rules), 82 Creation of trusts: accumulations, 33–4 capacity of settlor, see Capacity of settlor donatio mortis causa, 42 equitable interests: situs and dealings, 34–9 transfers, 39–40 forced heirship, 54–6

Index 517 Hague Convention on Law Applicable to Estates of Deceased Persons (1989), potential impact, 56–9 incompletely constituted, 41–2 intention, certainty of, 25–6, 124 inter vivos, 5, 23–33 matrimonial property regimes, 61–4 offshore jurisdictions, 65–77 perpetuities, rule against, 33 rocket launch imagery, 3–5 structure of, 18–19 sub-trusts, 41 testamentary, see Testamentary trusts and transfer, 4–5 Creditors, protection, insolvency matters, 376–7 Criticisms of Hague Trusts Convention: 421–2 evaluation of, 422–4 Cy-près doctrine, charitable purpose trusts, 125, 293–4 Debts, assignment of, 31, 32 Declarations of trust, trustees: assets, control of, 106 capacity of settlor, 6, 19 inter vivos trusts, 25 Definitions and interpretations: “account”, 271–2 beneficiaries, 245–6 capacity, 7 essential validity, 7 inter vivos transfers, 23–4 lex situs, 28–9 mandatory rules, 355–6 “non-trust” states, 341 recognition of trusts, 101 rocket-launch imagery, 3–5 “trusts”, 105 Denmark, specific characteristics of trusts, 107 Denouncing of Convention (Article 31), 419 Dépeçage: absence of choice, 216 administration of trust assets, 241 choice of law, recognised legal systems, 185 content of law, changes to, 307 non-trust jurisdictions, 161–3 scope of applicable law, 233 splitting of trust, 281 Dicey and Morris, Rule in: administration and, 241, 283–4 ante-nuptial contracts, 63 beneficial interests, enforcement of trusts, 35, 37 choice of law rule, capacity, 12, 14 constructive trusts, 147 divorce settlements: applicable law of trust, role, 259

jurisdiction, exercise by English judge, 259–60 substance and procedure, distinguished, 258 fiduciary duties of trustees, 250 marriage, personal and proprietary effects of, 364–5 tangible movable property, 29 validity and, 283–4 Divorce settlements: effect of divorce on trusts, 264 exercise of discretion, English judge, 259–61 general, 257 substance or procedure, English law, 257–8 Domicile, law of: matrimonial capacity, 12–13 testate succession, 47, 48 Donatio mortis causa, 42 Double recovery, 64 Double renvoi, 27–8 Drafting Committee, Hague Convention, 86 Droz, G., 370 Duckworth, A., 57–8, 304, 367 Duration of trust, restrictions upon, 244–5 Dutch Civil Code, transfers of property, 319–21 Dyer, A., 81, 86, 112, 218, 227, 275, 284, 287, 308, 309, 330–2, 337, 368 Enforcement rights, beneficial interests, 35–9 English law: accumulations, 33 characterisation process, 117–18 charitable purpose trusts, 292 contracts governed by, capacity, 10 forced heirship claims, 58 freedom of choice, 15 grant of probate, 43–4 incompletely constituted trusts, 43 institutional constructive trusts, 146 invalidity of trust terms, 230–1 mandatory rules, domestic and international, 356–7 public policy, 391 registration of trusts, 338–9 renvoi doctrine, 27 termination of trusts, seeking, 82 variation of trusts: application of law, foreign courts, 269 discretion, exercise of, 267 forum, rules of, 267–8 jurisdiction, 256, 259–61 mandatory rules, 264–5 public policy, 268–9 substance or procedure, 257–8 Variation of Trusts Act (1958), 261–4

518 Index Equitable interests, situs of: determination, 34 enforcement of trusts, beneficial interests and, 35–9 legal interests and, 38 lex situs, 39, 40 “trustee residence” rule, 38–9 trusts, 34–5 unadministered estates, 34–5 Equitable interests, subsisting, transfers of, 39–40 Equitable title, transfer of, 24–5 Essential validity: ante-nuptial contracts, 63 bootstraps approach, 276–7 and capacity, 7–10 characterisation, 9–10 existence of trust, 278–9 formation of trusts, 276–8 inter vivos trusts, 33 lex situs, 47, 60 meaning, 7 terms of trust, 280 testamentary powers of appointment, 59–60 testamentary trusts: executor trustees, 53–4 general rule, 52–3 immovable property, 49 movables, 47 testate succession: immovable property, 49 movable property, 47 Estates, administration of, see Administration of estates Evolution of Hague Trusts Convention, 86–7 Exécuteur testamentaire , trustee as, 334–5 Executor trustees, 53–4 Existence of trust, essential validity, 278–9 Express trusts: categories of trust, 163 and constructive trusts, 129, 130 intentions of parties, 134 and resulting trusts, 126 and voluntary trusts, 125 see also Article 6 (choice of law, express or implied) “Family home” constructive trusts, 132 Fiduciaries, unauthorised profits made by, 132, 137 Fiduciary duties of trustees, 246–53 accounting for administration, 271–2 breaches, 247 Fiducie concept, Canada, 107–8 Fiscal matters, exclusion (Article 19): civil law states, taxation: general, 399 Italy, 400–1

Netherlands, 400 neutral approach, 397 UK, 397–9 refusal to enforce foreign revenue laws, 398–9 “Flee” clauses, 308 Forced heirship: applicable law, 365–7 creation of trusts, 54–6 Hague Convention on Law Applicable to Estates of Deceased Persons (1989), potential impact, 57 inter vivos trusts, 55, 58 “rocket-launcher” classification, 155 testate succession, 47 types of rules, 367 von Overbeck Report, 265 Foreign trusts, English statutes, 264–9 application of English law, 269 discretion, exercise of, 267 domestic mandatory rules, 264–7 forum, rules of, 268–9 public policy, English, 268–9 Formal validity: applicable law, scope of, 233 choice of law rule, 233, 272–6 testamentary powers of appointment, 61 testate succession: immovable property, 50 movable property, 48 Forum, law of: applicable law, changing, 301 Article 16 (international mandatory rules), 380–6 Article 15, interaction with, 384 determination of international application, 383–4 scope, 380–1 UK, exclusion of Article 16(2) in, 381–3 undermining of Convention, 385–6 conflicts rules designated by, 321–3 fiduciary duties of trustees, 248 mandatory rules designated by, 361–2 offshore jurisdictions, 71–2 standard of proof, 189 trust assets, following and tracing, 321–3 variation of trusts, 267–8 France: claw back claims and inter vivos trusts, 368 contract approach to “translating” trust, 331 express trusts, 168 testamentary trusts and powers of revocation, 335 trust as company, 332 trust assets and claims against settlor, 314

Index 519 Functions of Convention (Article 1): choice of law, 100–1 and claw-back, 57–8 recognition of trusts, 101–2 Gaillard, E.: on mandatory rules, 362 on recognition: right to refuse, 344–5 uniform law, 314 on renvoi, 389 on “rocket-launching” issues, 155 on “translation” of trust into civil law systems, 335 Gambino, R., 400 Germany: protection of creditors, 377 shapeless trust, 111 specific characteristics of trusts, 107, 108 specificity principle (Bestimmtheitsgrundsatz), 377 succession rights, 365 third party assertion of title, 375 Giuliano, M., 193, 196 Grant of probate, English law, 43–4 Guernsey, 71, 72, 73, 113 forced heirship, 55 Hague Conference on Private International Law: genesis of Convention, 86 judicially created trusts, exclusion, 134 ratification status of Convention and, 90 registration of trusts, 339 role, 423 Hague Convention on the Conflict of Laws Relating to the Form of Testamentary Dispositions (1961), 48 Hague Convention on the Law Applicable to Contracts for the International Sale of Goods Act (1955), 182–3 Hague Convention on Law Applicable to Estates of Deceased Persons (1989), potential impact: claw-back, 57–8 forced heirship, 57 general, 56–57 Hague Trusts and Succession Conventions, 57, 58, 59 interaction with Hague Trusts Convention, 412 Hague Trusts Convention: accession of Member States (Article 27), 415 accession of Non-Member States (Article 28), 416 application between territorial units (Article 24), 410–11

application to selected territorial units (Article 29), 417 capacity of settlor, exclusion from, 5–6 challenges of, 425–6 criticisms, 421–2 evaluation, 422–5 current status, 88–9 denouncing (Article 31), 419 draft, 86–7 entry into force (Article 30), 418 evolution, 86–7 Final Act, signature of, 87 final clauses, 414 non-trust states, implications of ratification, 90–2 open-ended nature of, 93–4 Preamble, 99, 112, 119 ratification, 88–9 implications, non-trust states, 90–2 trust and non-trust states, 425–6 relationship to other international conventions (Article 25), 412–13 reservations to scope (Article 26), 414 Second Commission, 86 time-frame (Article 22), 406–8 application in UK, 94–7 UK, application in, 94–7 von Overbeck report, status, 89 see also individual Articles Half-secret trusts, 131 Harmonisation of law: characterisation, 116 choice of law rules, 100 trust and non-trust states, 84 Hayton, D: on agency and mandate, 106 on civil law systems, “translating” trust into, 330–1 on claw-back, 57, 58 on constructive trusts, 129–32 on evaluation of criticisms of Convention, 423 on international mandatory rules of forum (Article 16), 265 on mandatory rules, 362 on non-charitable purpose trusts, 295 on non-trust states, 226 on perpetuities, rule against, 154 on public policy, 391 on recognition, restriction of, 405 on specific trust characteristics, 107, 109 on trust assets, tracing, 329 on trustees, control of assets, 106 on variation of trusts, international mandatory rules of forum, 265 views of, considered, 133–4

520 Index Immovable property, 26–8 general rule, 26 intestate succession, 51 matrimonial property regimes, 63 renvoi, 26–8 testate succession: capacity of testator, 50 construction, 49 essential validity, 49 formal validity, 50 revocation, 50 trusts of land, 178 Implied choice of law: “circumstances of the case”, 196–8 common law authorities, 198–208 cases evaluated, 209 conclusions, 209–11 inferred choice, 212–13 inter vivos trusts, 201–8, 210, 211 relevance, 198 testamentary trusts, 199–201 valid trust, implied intention to create, 208 law of closest connection and, 194–5 nature, 191–4 “reasonable certainty” test, 194 settlor’s purpose, 198 time for ascertaining, 213–14 see also Choice of law In personam orders/proceedings, 173, 246 In rem, interests, 173 Incapable parties, protection of, 363–4 Income of trust, accumulation, 244–5 Incompletely constituted trusts, 41–2 Insolvency matters, creditor protection, 376–7 Intangible movable property: capacity of settlor to transfer, 16 inter vivos trusts, 30–3 assignability of debts, 31 the assignment of the debt, 32 assignment of debts, issues raised, 31 chose in action, title to, 32 situs, 16, 30–1 Intention of settlor: creation of trusts, 25–6, 124 express trusts, 134 formalities, 123–4 implied, 208, 209 Inter vivos trusts: administration, validity distinguished, 283–4 beneficiaries: capacity of, 23 transfer of equitable title, 24 capacity, 11, 13, 52 chattels, of, 30 choice of law rule, 33 claw-back claims and, 368–9 closest connection, 228 constitution, 23

declarations of trust, trustees, 25 definitions, 23–4 essential validity, 33 forced heirship claims, 55, 58 formal validity, 273 generally, 5 immovable property, 26–8 general rule, 26 renvoi, 26–8 implied choice, common law authorities, 201–8, 210, 211 incompletely constituted, 42 intention to create, 25–6 lex situs rule, 47 movable property, intangible, 30–3 assignability of debts, 31 the assignment of the debt, 32 assignment of debts, issues raised, 31 chose in action, title to, 32 situs, 30–1 movable property, tangible: lex situs, 16, 28–9 lex situs rule, exceptions, 29 renvoi, 29–30 offshore jurisdictions, 72–3 trustees: new, vesting of property in, 237 transfer of legal title, 24 validity, administration distinguished, 283 “International” trusts, choice of law, 182–4 International trusts convention, desirability, 83–6 Intestate succession: immovables, 51 mandatory rules: administration of estates, 373 civilian perspective, 370–1 claw-back claims, 368–9 forced heirship, 365–7 inter vivos trusts, 368–9 offshore dimension, 371–3 movables, 50–1 renvoi, 51–2 Invalidity, certain terms, of, 229–31 Investment powers of trustees, 243–4 Isle of Man: creation of transnational trusts, 75–7 mandatory provisions, 371–2 Isle of Man’s Trust Act 1995 (IMTA), offshore jurisdictions, 75, 76 Italy: recognition of trusts: refusal of, 350–3 separate fund, 315–17 status of assets, 318–19 registration of trusts, 339–40 rules of succession, 370 taxation and, 400–1

Index 521 trust interno, 184 weaknesses of Convention and, 91–2 Japan: forced heirship, 365 protection of trust assets from creditors, 377 shapeless trust, 111 specific characteristics of trust, 108 wrongful disposition of trust property, 375 Jauffret-Spinosi, C., 84 Jersey: creation of transnational trusts, 72–4 mandatory rules, 372–3 Judicial decision, trusts “arising” by: complexities raised, 145 constructive trusts, 145–7 judgments entitled to recognition, 148–9 presumed resulting trusts, 127, 148–50 Judicially created trusts, exclusions, 134–9 continuing obligation, trusts imposing, 138–9 foreign judgments, recognition “problem”, 135–7 general, 134–5 “remedial” trusts, 137 Koppenol-Laforce, M.: on extension of Convention to other types of trust (Article 20), 403 on mandatory rules, 362 on non-recognition, consequences, 349 on non-trust states, implications of ratification, 91 on taxation, civil law states, 400 on trust assets, tracing, 327, 328–9 Kottenhagen, R., 91, 327, 400 Lagarde, P., 193, 196 Land: capacity of settlor to transfer, 15–16 trusts of: leasehold interests, 174 lex situs, need for, 169–71, 176 movable and immovable property, 178 no law chosen to govern, 175–8 offshore, 175 sale, for, 172–4 statutory provisions, 171–2 Latin America, fideicomiso structure, 107 Law of closest connection, see closest connection Law Society of England and Wales, registration of trusts, 339 Leach, W., 33 Leasehold interests, trusts of land, 174 Legal title, transfer of: 24–5 chose in action, title to, 32 essential validity, testamentary trusts, 52–3

Lewin on Trusts: capacity: beneficiaries, of, 21 declarations of trust, 19 marriage settlements, 12–13 constructive trusts, 145 executor trustees, 53–4 forced heirship, 366 formalities, 123 movable property, renvoi, 29–30 public policy, 391–2 “rocket-launcher” classification, 155 situs of beneficial interests, 34 time factor, 94 Lex causae (governing law), 118 Lex domicilii(law of settlor’s domicile): capacity, 9, 11, 20 offshore jurisdictions, 72 Lex fori (law of the forum): characteristics of trusts, 118 inferred choice, 212 invalid trust terms, 230 trustees: appointment and removal, 235 liability and remedies, 253 Lex loci contractus (place of conclusion of contract), capacity, 11, 13 Lex mercatoria, 185, 186 Lex situs (law of place where property situated): autonomy principle, 166, 168 capacity: beneficiaries, 23 corporate trustees, 19 declarations of trust, 19 intangible movables, 16 land, conveyance of, 15–16 transfer of property, 18 choice of law rule, 274 chose in action, title to, 32 declarations of trust, 25 capacity, 6, 19 defined, 28–9 equitable interests, situs, 39, 40 essential validity, 47, 60 exceptions to rule, 29 forced heirship, 366 grant of probate, English jurisdiction, 44 immovable property, 49, 178 general rule, 26 inter vivos property transfers, 23–4, 47 leasehold interests, 174 legal title, initial vesting of, 237 limits of, definition, 242 matrimonial property regimes, 63 matters uncontrolled by terms of trust, 291 offshore jurisdictions: Bermuda, 72

522 Index Lex situs (cont.): offshore jurisdictions (cont.): Cayman Islands, 67, 68, 69 Isle of Man, 77 “rocket-launching” issues, 152 tangible movable property, 16, 28–9, 30 termination of trusts, 270 testamentary trusts: capacity to create, 52 essential validity, 53 testate succession, 49, 50 third parties, liability to trustees, 254 trust assets, tracing, 325, 329 trusts of land, 169–71, 176 trusts for sale, 173, 174 see also Situs Lex successionis (law applicable to succession): absence of choice, testamentary trusts, 226 creation of trusts, 152, 153 English courts, jurisdiction, 44 essential validity, 53, 60 forced heirship claims, 55, 57 incompletely constituted trusts, 42, 43 intestate succession, 51 mandatory rules, 155 matrimonial property regimes, 64 testate succession, 47 Locus standi, non-charitable purpose trusts, 296 Lugano Convention, recognition and enforcement of trust judgments, 85 Lupoi, M.: on constructive trusts, 128 on general trust characteristics, 105 on “international” trusts, 183, 184 on private international law instruments, 84 on ratification, implications for non-trust states, 91 on registration of trusts, 338 on rocket-launching imagery, 4 on “shapeless” trusts, 111–15, 352, 422, 423 on trust assets, tracing, 326 Luxembourg: agency and mandate, 334 capacity of trustees to sue, 328 process of ratification of the Hague Trusts Convention, 56, 89, 104, 318, 334, 425, 427 Mandatory rules: application, 363 Article 15, see Article 15 (mandatory rules) characterisation, 357–60 and choice of law process, 190–1 civil law of succession, testamentary trusts, 83 domestic: and international, 356–7

variation of trusts, 264–7 forum, rules of, 361–2 function, 355 international, 356–7 lex successionis, 155 marriage, personal and proprietary effects of, 264, 364–5 mutual exclusivity, 357–60 nature, 355–6 offshore jurisdictions: Bermuda, 70 Isle of Man, 75 public policy and, 392–3 recognition of trusts and, 315 “rocket-launching” and, 360–1 and testamentary trusts, 83 variation of trusts, 264–5 see also Article 16 (international mandatory rules of forum) Marriage, personal and proprietary effects of: mandatory rules, 264, 364–5 Matrimonial capacity settlements, 12–13 Matrimonial Causes Act (1973), 257–61, 264–7 Matrimonial property regimes, 61–4 ante-nuptial contracts existing, 63–4 ante-nuptial contracts not existing, 62–3 choice of law rules, 62–4 general, 61–2 immovable property, 63 lex situs, 63 movable property, 62–3 succession, law applicable to, 64 see also Divorce settlements Matthews, P.: on capacity, 10 on changes of law, 304, 306 on Jersey law, 73, 74 on residence, place of, 224 on splitting of law, 286 Member States, accession of (Article 27), 415 Minors, protection of, 363–4 Mixed funds, tracing of trust assets, 323–4 Morris, J., 33 Movable property: capacity of settlor: choice of law, 16 testate succession, 48 choice of law, capacity of settlor, 16 inter vivos trusts: lex situs, 28–9 lex situs rule, exceptions, 29 renvoi, 29–30 intestate succession, 50–1 matrimonial property regimes, 62–3 tangible, inter vivos trusts, 28–30 testate succession, 47–50 capacity of testator, 48 construction of will, 47–8

Index 523 essential validity, 47 formal validity, 48 revocation, 48–9 trusts of land, 178 Multiple settlements, chosen law, 186 Mutual exclusivity, mandatory rules, 357–60 Negotia, legal, 4 Netherlands: recognition of trusts: Convention, weaknesses of, 91 status of trust assets, 319–21 tracing trust assets, 327–9 registration of trusts, 338 specific characteristics of trusts, 107 taxation and, 400 Non-Member States, accession of (Article 28), 416 Non-trust jurisdictions: Article 5: application, 158–61 categories of trust, 163–5 dépeçage, 161–3 choice of law, governing part of trust, 289–90 difficulties, 81 harmonisation of law, 84 objective factors pointing to, absence of choice, 226–7 ratification, implications of, 90–2 recognition difficulties, 81 trust world, frequency of contact, 83 see also Article 13 (right to refuse recognition, connection to non-trust states); Civil law states North, P., 301 Northern Ireland: grants of probate, 45 variation of trusts, statutory provisions, 262 Notification process (Article 32), 420 Nygh, P.: on changes of law, 301–2 content, 306 on contracts, 13 on trusts governed by Convention, 130 on validity, 227 Objects of trust, 222–4, 348 giving effect to, 378–9 Offshore trusts, 65–77 Bermuda, 69–72 Cayman Islands, see Cayman Islands choice of law (Article 6), 175 insulation, 78 Isle of Man, see Isle of Man Jersey, 72–4 lex situs, 67, 68, 69 “rocket-launching”, 66, 71, 72, 77, 78

Panagopoulos, G., 250–1, 324 Perpetuities, rule against: applicable law, scope of, 244, 245 creation of trusts, 33 mandatory rules (Article 15), 361 “rocket-launching” issues, 154 Preliminary matters, exclusion (Article 4), 151–7 constitution of trusts, 7 perpetuities, 33 “rocket”/“rocket-launcher”, distinguishing, 153–7 validity, 5 Principles of European Trust Laws, 108–9 Private international law: common law principles, 95 constructive trusts, 146–7 express choice of rules, 388–9 Hague Convention, weaknesses of, 90–1 harmonisation in, 116 international trusts convention, desirability, 84 mutual exclusivity, 359–60 renvoi doctrine, 27, 28 UK, application of Convention in, 144 see also Hague Conference on Private International Law Probate, grant of, 43–4 Procedure, substance distinguished, 257–8 Property transfers: Dutch Civil Code, 319–21 equitable interests, 39–40 legal title, transfer of, 24, 373–6 security interests, 373–6 see also Immovable property; Matrimonial property regimes; Movable property; Transfers Provisions of Convention, see individual Articles Pryles, M., 9, 10, 13–14, 16–17 Public policy (Article 18): close connection, states of, 393–4 mandatory rules (Article 15), interrelationship with, 392–3 scope, 390–2 STAR trusts, 395–6 and variation of trusts, 268–9 Ratification of Convention, 88–9 implications, non-trust states, 90–2 trust and non-trust states, 425–6 Recognition of trusts: characteristics of trusts (Article 2), 107 charitable purposes and, 294 civil law states, 93 common law states, 93 criticisms of Convention and, 424 definitions, 101

524 Index Recognition of trusts (cont.): foreign judgments, recognition “problem”, 135–7 functions of Convention, 101–2 Italian courts, separate fund, 315–17 judgments entitled to recognition, 148–9 liberal rules, freedom to adopt (Article 14), 354 mandatory rules and, 315 need for Convention Article, 311–13 and non-trust states, 159 partial, 349 refusal of, connection to non-trust states: general application of Article 13, 343–6 Italy, 350–3 non-recognition consequences, 349–50 partial recognition, 349 purpose of Article 13, 341–2 “significant elements”, 346–8 time factor, 348–9 United Kingdom, 342–3 right to restrict to trusts governed by law of contracting state (Article 21), 404–5 specific requirements, 314–15 “translating” into civil law systems: agency and mandate, 334 contractual approaches, 331–2 problems, 335–6 scope of discussion, 330 trust as company, 332–3 types of trust, 330–1 trust assets: following and tracing, 321–9 Italy, interpretation in, 318–19 Netherlands, 319–21 separate status of, 317–21 settlor, claims against, 314–15 uniform law, 313, 314 Registration of trusts: chose in action, title to, 32 English reaction to proposals, 338–9 facility or obligation, 340 Italy, 339–40 prohibition of registration, 337–8 registration permissible, 337 Renvoi doctrine: Article 17: case for exclusion evaluated, 387–8 choice of law, recognised legal systems, 186 express choice, state’s private international rules, 388–9 related areas of law and, 389 double, 27–8 exclusion of in trusts, 190 immovable property, 26–8 inter vivos trusts: immovable property, 26–8

movable property, 29–30 movable property, 29–30 offshore jurisdictions, Cayman Islands, 69 private international law, 27, 28 single, 27, 28 situs, law of, 26, 27 testamentary trusts, succession, 51–2 Residence, place of: beneficiaries, 224–5 settlor, 224–5 trustee, 221–2 Resulting trusts, 125–7 automatic, 130 choice of law, 140–1 presumed resulting, 127 judicial decision, 148 Revocation: immovable property, 50 movable property, 49 powers of, recognition of trusts, 334–5 subsequent marriage: immovable property, 50 movable property, 49 testamentary powers of appointment, 61 subsequent will or codicil: immovable property, 50 movable property, 49 testamentary powers of appointment, 61 “Rocket-launching”: administration of trust assets, 242 capacity of settlor, 238 description, 3–5 equitable interests, transfers of, 39 existence of trust, 278 formal validity, 273 generally, 3–78 incompletely constituted trusts, 42 mandatory rules, 360–1, 374 matrimonial property regimes, 62 offshore jurisdictions, 66, 71, 72, 77, 78 perpetuities, 33, 244 preliminary matters, exclusion of (Article 4), 151–7 property transfers, 18 recognition issues, 350 “rocket”/“rocket-launcher”, distinguishing, 4, 153–7 trust assets, tracing, 325 trustee, vesting of property inter vivos, 24, 25 Rome Convention on the Law Applicable to Contractual Obligations: ante-nuptial contracts, exclusion from, 63 choice of law: consent to, 188 formal validity, 275 objective connection to, 179, 180 “circumstances of the case”, 196

Index 525 comparison of Article 13, Hague Convention, with Article 3(3), Rome Convention, 345 constitution of trusts, exclusion from, 31 contracts, capacity, 11 corporations, capacity to contract, 19 equitable interests, transfers of, 40 existence of trust, 279 implied choice, nature, 191, 192, 193 intangibles, assignment of, 16 lex mercatoria, 185 splitting of trust, 282 trusts and, 120–2 Schoenblum, J.: on changes of law, 302 criticisms of Convention by, 421–4 on mandatory rules, undermining of Convention, 385 on non-trust states, refusal of recognition to objectively connected trusts (Article 13), 341 on objects of trust, absence of choice, 222 on place of administration, determining and changing, 309 on preliminary matters, exclusion, 152, 153 on public policy, scope of Article 18, 390 on translation difficulties, 335 on trust assets, tracing, 329 on trustees, removal of, 237 Scotland: choice of law rule, 142, 143 grants of probate, 45 intra-UK conflicts, 411 variation of trusts, statutory provisions, 262 Settlors: autonomy principle, 166–9 as beneficiaries, 110 choice of law, 56 claims against, 314–15 intentions of, see Intention of settlors lex mercatoria, choice of, 186 as “persons”, 105–6 purpose of, 198 residence, place of, 224–5 Severable parts of trust: changing applicable law, 303–4 splitting applicable law, 282–3, 303–4 “Shapeless” trusts: characteristics of trusts (Article 2), 111–16 criticisms of Convention, 422, 423 non-trust states and, 352 Signature clause, 420 Situs: equitable interests: determination of, 34 enforcement of trusts, beneficial interests and, 35–9

trusts, 34–5 unadministered estates, 34–5 intangible movable property, 16, 30–1 non-trust states, recognition issues, 347 renvoi doctrine, 26, 27 transfer of property, 17 trust assets: absence of choice, 220–1 implied choice of law, 200, 205 see also Lex situs South Africa: land, capacity to transfer, 16 Sowden, T., 10, 73, 74, 224 Spanish law, renvoi doctrine, 27 Special Commission meetings, 86 Splitting applicable law: absence of choice, in, 289 administration: charitable purpose trusts, 292–3 validity, distinguishing from, 283–9 charitable purpose trusts: administration, 292–3 cy-près doctrine, settling, 125, 293–4 non-charitable purpose and, 295–6 validity, 291–2 choice of law, non-trust states, 289–90 matters uncontrolled by terms of trust, 290–1 right to split, 281–2 severable parts of trust, 282–3, 303–4 validity: administration, distinguishing from, 283–9 charitable purpose trusts, 291–2 STAR trusts: non-charitable purpose, 295 public policy, 395–6 Status: and capacity, 7 Hague Convention, 88–90 and recognition of trusts, 405 trust assets, 317–21 Statutory trusts, 133 Stevens, R., 126 Sub-trusts, 41 Subsequent marriage, revocation: immovable property, 50 movable property, 49 testamentary powers of appointment, 61 testamentary trusts, succession, 49, 50 Subsequent will or codicil, revocation: immovable property, 50 movable property, 49 testamentary powers of appointment, 61 testamentary trusts, succession, 49, 50 Substance, procedure distinguished, 257–8 “Substantial relation” test, choice of law, 181 Succession law, matrimonial property regimes, 64

526 Index Succession rights, mandatory rules: administration of estates, 373 civilian perspective, 370–1 claw-back claims, 368–9 forced heirship: law applicable, 365–7 types of rules, 367 inter vivos trusts, 368–9 offshore dimension, 371–3 Succession, testamentary trusts, 46–51 intestate, 50–1 renvoi, 51 testate, 47–50 Switzerland: possible ratification of Hague Trusts Convention, 56, 100, 318, 334, 427 recognition of trusts, 331–2 trust as company, 100, 333 Sykes, E., 9, 10, 13–14, 16–17 Tangible movable property: capacity of settlor to transfer, 16 inter vivos trusts: lex situs, 16, 28–9 lex situs rule, exceptions, 29 renvoi, 29–30 Termination of trusts, 270–1 Terms of trust, essential validity, 280 Territorial units: application of Convention between units of a state, 410–11 application of Convention to selected units of a state, 417 choice of law of state comprising several, 184–5 states consisting of several, 307, 409 Testamentary powers of appointment: capacity of settlor, 60 construction, 60 essential validity, 59–60 formal validity, 61 general, 59 revocation, 61 Testamentary trusts: absence of choice, applicable law, 225–6 administration of estates: applicable law, 45 Commonwealth, 45 delimiting, 45–6 grant of probate, English law, 43–4 jurisdiction of courts, 44–5 Northern Ireland, 45 Scotland, 45 capacity of testator to create, 52 civil law of succession, mandatory rules, 83 closest connection law, time for assessment, 228–9

essential validity: executor trustees, 53–4 general rule, 52–3 formal validity, 273 implied choice, common law authorities, 199–201 intention, certainty of, 26 recognition of trusts and, 334–5 renvoi doctrine, movables, 29–30 succession, law applicable to, 46–51 intestate succession, 50–1 mandatory rules and, 83 renvoi, 51 testate succession, 47–50 and time factor, 95–6 Testate succession, 47–50 administration of estates, 373 claw-back claims, 368–9 forced heirship: applicable law, 365–6 types of rules, 367 immovable property: capacity of testator, 50 construction, 49 essential validity, 49 formal validity, 50 revocation, 50 inter vivos trusts, 368–9 mandatory rules: administration of estates, 373 civilian perspective, 370–1 claw-back claims, 368–9 forced heirship, 365–7 inter vivos trusts, 368–9 offshore dimension, 371–3 movable property: capacity of testator, 48 construction of will, 47–8 essential validity, 47 formal validity, 48 revocation, 48–9 Third parties: good faith, purchasers acting in, 324–5, 377–8 liability to trustees, personal and proprietary claims, 254–6 liability of trustees to, 254 trust assets, following and tracing: conflicts rules, law of forum, 321–3 constructive trusts, 326–7 good faith, purchasers acting in, 324–5 mixed funds, 323–4 Time-frame of Convention (Article 22), 406–8 application in UK, 94–7 Tort, “general” law of, 249 Transfers: capacity of settlor: intangible movables, 16

Index 527 land, 15–16 tangible movables, 16 trusts, 17–18 constructive trusts, 138 and creation, 4–5 equitable interests, subsisting, 39–40 mandatory rules, 373–6 property, capacity to transfer, 15–16 Trautman, D., work with E. Gaillard, see Gaillard, E. Trust assets: administration, 241–2 and constructive trusts, 129 control by trustee, 106 distribution, 271 following and tracing: constructive trusts, 326–7 dominance of law governing trust, 321 Netherlands, application in, 327–9 proprietary effects of Convention, 327 third parties, 321–5, 326–7 UK, consequences in, 326 powers of trustees to deal with and acquire, 242–3 and settlor, claims against, 314–15 situs: absence of choice, 220–1 implied choice of law, 200, 205 “Trustee residence” rule, equitable interests, situs of, 38–9 Trustees: accountability issues, 109–10 administration of trust assets, 241–2 appointment, 235–7 beneficiaries, as, 110 beneficiaries, relationship with, 245–56 fiduciary duties owed to, 246–53 liabilities, remedies for, 253–4 trusts, not express in nature, 96 capacity, 6, 21, 237–9 corporate, 19–20 declarations of trust: assets, control of, 106 capacity of settlor, 6, 19 inter vivos, 25 delegation, 240, 245 duties: accounting for administration, 270–1 among themselves, 239–40 fiduciary, see Fiduciary duties of trustees as exécuteur testamentaire, 334–5 executor, 53–4 income of trust, power to accumulate, 244–5 investment powers, 243–4 liability: personal, 245 remedies and, 253–4 third parties, to, 254

vicarious, 240, 245 personal liability, 245 place of residence or business, 221–2 powers: administration of assets, 241–2 income accumulation, 244–5 investment, 243–4 trust property, acquiring and dealing with, 242–3 property of trust, power to deal with and acquire, 242–3 removal, 235–7 rights and duties among themselves, 239–40 third parties, liability to: lack of provisions, 254 personal and proprietary claims, 254–6 transfer of legal title to, 24 vesting of property in inter vivos, see Inter vivos trusts vicarious liability, 240, 245 Trusts: bare, 109 capacity of settlor, 16–20 corporate trustees, 19–20 creation of trust structure, 18–19 declarations of trust, 6, 19 property transfer, 17–18 categories of: mandatory rules, 357 non-trust jurisdictions, 163–5, 345–6 constitution: capacity of settlor, 7 inter vivos trusts, 23 Rome Convention, exclusion from, 31 duration, powers of trustees, 244–5 “effects” of, 234 enforcement rights, beneficial interests, 35–9 half-secret, 131 income, accumulation of, 244–5 incompletely constituted, 41–2 inter vivos, see Inter vivos trusts “international”, 182–4 meaning, 105 objects of, 222–4, 348 recognition, 101–2 sale, for, 172–4 severable parts, 282–3, 303–4 sub-trusts, 41 termination, 270 testamentary, see Testamentary trusts unadministered estates and, 34–5 variation, 256–71 and administration, 270–71 application of English law, foreign courts, 269 change of law and, 299 discretion, exercise of, 267 divorce settlements, 257–61

528 Index Trusts (cont.): variation (cont.): domestic mandatory rules, 264–7 foreign trusts, English statutes, 264–9 forum, rules of, 267–8 international mandatory rules of forum (Article 16), 265, 267–8 mandatory rules (Article 15), 257–61, 264 Matrimonial Causes Act (1973), 257–61, 264–7 public policy, English, 268–9 and termination, 270–1 Variation of Trusts Act (1958), 261–4 Trusts (Foreign Element) Law 1987 (TFEL): Cayman Islands, 66, 175 Isle of Man, approach compared, 75, 76 Trusts of land: leasehold interests, 174 lex situs, need for, 169–71 movable and immovable property, 178 no law chosen to govern, 175–8 offshore, 175 sale, for, 172–4 statutory provisions, 171–2 Types of trusts covered (Article 3), 123–50 constructive, 128–33 formalities, 123–4 Hayton’s views, 133–4 judicially created trusts, exclusion, 134–9 continuing obligation, trusts imposing, 138–9 foreign judgments, recognition “problem”, 135–7 general, 134–5 “remedial” judicial trusts, 137 resulting, see Resulting trusts statutory, 133 United Kingdom, statutory extension of Convention rules: any part of UK, 141–4 constructive trusts, choice of law rule, 140–1 judicial decision, trusts “arising” by, 145–9 relevant provisions, 139–40 resulting trusts, choice of law rule, 140–1 voluntary, 124–5 Unadministered estates, beneficial interests, 34–5 Uniform law, recognition of trusts, 313, 314 United Kingdom: fiscal matters, exclusion, 397–9 refusal to enforce foreign revenue law, 398–9 Hague Convention: application in, 94–7

Recognition of Trusts Act (1987), 89 intra-UK conflicts, application of Article 24, 411 judicial decision, trusts “arising” by: complexities raised, 145 constructive trusts, 145–7 judgments entitled to recognition, 148–9 presumed resulting trusts, 148–50 mandatory rules of forum, exclusion in, 381–3 non-trust states, refusal of recognition to connected trusts (Article 13), 342–3 statutory extension of Convention rules: any part of UK, 141–4 choice of laws, 140–1 constructive trusts, choice of law rules, 140–1 judicial decision, trusts “arising” by, 145–9 relevant provisions, 139–40 resulting trusts, choice of law rules, 140–1 time factors for application of Convention, 94–7 trust assets, following and tracing, 326 trusts arising under law of any part: domestic law, 141–4 imposed or affirmed in UK, 144 private international law, 144 see also English law; Northern Ireland; Scotland United States of America: American Restatement (2d) on the Conflict of Laws, 204, 169, 286, 287, 290, 292 capacity of beneficiary, 22 change in place of administration, 309 changing the applicable law: one trust or two?, 304 charitable trusts, 292 choice of law in contract, 344 forced heirship, 55 implied choice, 198 inter vivos trusts, 204 lex situs, 169, 220 matrimonial property regimes, 64 testamentary powers of appointment, 60 trustee’s power to delegate, 240 trusts for sale, 173 recreating a trust, 304 validity and closest connection, 227 Validity: administration, distinguishing from, 283–9 applicable law, changing, 302 closest connection, law of, 227 see also Essential validity; Formal validity; Invalidity Van Loon, H., 81, 86, 112, 218, 227, 275, 284, 287, 308, 309, 330–2, 337, 368

Index 529 Variation of trusts, 256–71 and administration, 270–1 change of law and, 299 divorce settlements: discretion, exercise by English judge, 259–61 effect of divorce on trusts, 264 general, 257 substance or procedure, English law, 257–8 foreign trusts, English statutes: application of English law, foreign courts, 269 discretion, exercise of, 267 forum, rules of, 267–8 public policy, English, 268–9 international mandatory rules of forum (Article 16), 265, 267–8 mandatory rules (Article 15) and, 264–5, 266, 267 Matrimonial Causes Act (1973), 257–61, 264–7 and termination, 270–1 Variation of Trusts Act (1958), 261–4 Vicarious liability, trustees, 240, 245 Voluntary trusts, 124–5 Von Overbeck Report: absence of choice, 217, 218 administration, place of, 218 capacity of settlor, 5 characteristics of trusts, 104 “circumstances of the case”, 197 closest connection, law of, 160–1 constructive trusts, 129 continuing obligation, trusts imposing, 138 declarations of trust, 6

dépeçage, 161, 162 divorce, effects on trust, 264 essential validity, 276 forced heirship rules, 265 foreign judgments, recognition “problem”, 135 formal validity, 272 implied choice of law, 192, 198 mandatory rules: application of Article 15 , 363 international (Article 16), 380, 381 nature, 356 non-trust states: application of Article 5, 158 refusal of recognition to objectively connected trusts (Article 13), 346 “persons”, meaning, 105 preliminary matters excluded (Article 4), 151–2 ratification, non-trust states, 91 recognition principle, 312, 314, 348 “remedial” judicial trusts, 137 renvoi, exclusion of, 387 resulting trusts, 126 rocket-launching imagery, 3, 151–2 splitting of trusts, 290 status of, 89 time-frame of Convention, 407 trust assets, tracing, 322–3 Waters, D., 114–15, 251, 326–7 Wills: capacity to make, 48 construction, 47–8, 49, 60 and constructive trusts, 128 subsequent, revocation, 49, 50