Property Law: Comparative, Empirical, and Economic Analyses 1009236598, 9781009236591

The first book of its kind, Property Law: Comparative, Empirical, and Economic Analyses, uses a unique hand-coded data s

278 18 33MB

English Pages 452 [453] Year 2023

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Property Law: Comparative, Empirical, and Economic Analyses
 1009236598, 9781009236591

Citation preview

PROPERTY LAW

The first book of its kind, Property Law: Comparative, Empirical, and Economic Analyses uses a unique hand-coded data set on nearly 300 dimensions on the substance of property law in 156 jurisdictions to describe the convergence and divergence of key property doctrines around the world. This book quantitatively analyzes property institutions and uses machine-learning methods to categorize jurisdictions into ten legal families, challenging the existing paradigms in economics and law. Using other cross-country data, the author empirically tests theories about property law and comparative law. Using economic efficiency as both a positive and a normative criterion, each chapter evaluates which jurisdictions have the most efficient property doctrines, concluding that the common law is not more efficient than the civil law. Unlike prior studies on empirical comparative law, this book provides detailed citations to laws in each jurisdiction. Data and documentation are publicly available on the author’s website. Yun-chien Chang  is the Jack G. Clarke Professor in East Asian Law, Cornell Law School. Professor Chang is Associate Reporter for the Restatement of the Law Fourth, Property and President of the Asian Law and Economics Association (2023–2024). He has published more than 10 books, including Private Property and Takings Compensation: Theoretical Framework and Empirical Analysis (2013), and three edited volumes with Cambridge University Press.

Published online by Cambridge University Press

Published online by Cambridge University Press

PROPE RT Y L AW Comparative, Empirical, and Economic Analyses

YUN-CHIEN CHANG Cornell Law School

Published online by Cambridge University Press

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009236591 DOI: 10.1017/9781009236553 © Academia Sinica 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library. A Cataloging-in-Publication data record for this book is available from the Library of Congress ISBN 978-1-009-23659-1 Hardback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published online by Cambridge University Press

Dedicated to my wonderful colleagues at Academia Sinica, Taiwan – particularly Professor Tzu-Yi Lin, whose decision in 2013 as Director of Institutum Iurisprudentiae to fund my ambitious project, despite negative reviews, makes this book possible.

Published online by Cambridge University Press

Published online by Cambridge University Press

CONTENTS

List of Figures  page xvi List of Tables  xviii Citation Format  xix Acknowledgments  xx List of Abbreviations  xxiii Introduction  1



part i Foundation  21 1 Property Law around the World: An Empirical Overview  23

2 Economic Framework  46

3 Limited Number of Limited Property Rights: Less Is More   62



4 Transfer of Ownership: Transaction Cost versus Information Cost  101



Part II Immovable Property  123



5 Acquisitive Prescription: Hardly Justified in Modern, Developed Countries  125



6 Building Encroachment: In Search of an Efficiency Justification  160



7 Co-Ownership Partition: Proposing a New Auction-Based Design  179



8 Managing Co-ownership: Tragedy of the Common-Ownership?   205



9 Access to Landlocked Land: Hybrid Entitlement Protection  230 vii

Published online by Cambridge University Press

viii



contents

Part III Movable Property  257



10 Good-Faith Purchase: Proposing Fractional Ownership and Internal Auction   259



11 Finders, Keepers: A Minority Rule   289



12 Specificatio: Do What the Romans Did   309



13 Accessio and Confusio: No Sign of Convergence   335

Conclusion  352 Data Appendix  361 Method Appendix  362 References  380 Index  415

Published online by Cambridge University Press

DETAILED CONTENTS

List of Figures  page xvi List of Tables  xviii Citation Format  xix Acknowledgments  xx List of Abbreviations  xxiii Introduction  1





I Three Approaches  2 A Comparative Analysis  2 B Empirical Analysis  3 C Economic Analysis  4 II Engagement with Existing Theories   5 A Convergence and Divergence of Property Law   6 B Law versus Meta-Law   11 C Are Judge-Made Laws Generally More Efficient?   12 D Legal Origins and Legal Families   12 III Terminologies  13 IV Overview of the Book   15

Part I Foundation  21 1 Property Law around the World: An Empirical Overview  23 I Legal Families  23 A Findings and Implications   24 1 Legal Family Tree   24 2 Distinctive Features  30 B Technical Details  36 II Correlation of Property Law   39 III Conclusion  45

2 Economic Framework  46

I Ex Ante versus Ex Post Viewpoints  46 II Efficiency versus Welfare   47 A Institution Costs  48

ix

Published online by Cambridge University Press

x

detailed contents



B Allocative Benefits  49 C Other Welfare Concerns   50 D Assessing Efficiency by Cost–Benefit Analysis   51 III Property Rules and Liability Rules   52 A Rule 1 Is More Efficient than Rule 2 When Transaction Costs Are Low   53 B The Problem of Asymmetric “Cathedral”   54 C A Framework for Preferring Rule 2 over Rule 1   56 1 Transaction or Information Costs Are High   56 2 Transferring Entitlements Likely Increase Allocative Benefits  57 3 Ex Ante Investment Not Important   58 4 Courts Make Fewer Errors   58 5 When Rule 2 Is the Only Choice   59 D Radical Market: Limited Applicability   60





3 Limited Number of Limited Property Rights: Less Is More  62 I Comparative Overview  63 A The Numerus Clausus Principle   63 B Limited Property Forms   66 1 Use Rights on Immovables   67 2 Security Rights on Immovables   74 3 Security Rights on Movables   82 C Future Interests  86 II Economic Analysis  87 A Why Close the Number of Property Forms?   87 1 Optimal Standardization  87 2 Externalization  90 3 Highly Standardized  92 B Limited Property Rights as Lego Bricks and Modules   93 C Future Interests Not Necessarily Create High Information Cost   96 D Recursiveness  97 E Registry Capabilities  98 III Conclusion  99

4 Transfer of Ownership: Transaction Cost versus Information Cost  101 I Comparative Overview  101 A Ownership Transfer Rules for Immovables   101 B Public Faith Principle   105 C Ownership Transfer Rules for Personal Properties   109 II Economic Analysis  113 A What Type of Registration Is More Efficient   113 B Absolutism Not Always Realistic    115

Published online by Cambridge University Press

detailed contents





C Constructive Notice Outside of Registration   116 D Causa versus Non-Causa   119 E Default Rules and Menus in Movable Transfer Designs  119 III Conclusion  121

Part II Immovable Property  123 5 Acquisitive Prescription: Hardly Justified in Modern, Developed Countries  125 I Comparative Overview  127 A Registration-Based Acquisitive Prescription   129 B Possession-Based Acquisitive Prescription   132 II Modern Justifications for Acquisitive Prescription   138 A Registration-Based, with Good Faith and Apparent Title   138 B Registration-Based, with Bad Faith and Apparent Title   143 C Possession-Based, with Good Faith and Apparent Title   145 D Adverse Possession: Possession-Based, without Apparent Title   146 1 In Recording Systems   147 2 In Registration-of-Right Systems   152 3 Dysfunctional Registries  155 E Registration-Based, without Apparent Title   157 III Conclusion  157

6 Building Encroachment: In Search of an Efficiency Justification  160

I Three Roads Diverge in the Doctrinal Woods   161 A Building Encroachment Doctrine   161 1 Claims to Sell Land Ownership   163 2 Claims to Sell Easement or Use Right   166 3 Discretionary Power  166 4 Protest: How Prompt Is Fast Enough?   167 B Acquisitive Prescription Doctrine   168 C Accession Doctrine  169 II Economic Analysis  169 A Safety Valve: Good Faith Will Lose Its Import   170 B A Two-Tiered Structure   173 1 Ex Post Inefficiency  173 2 Ex Post Efficiency  174 3 Safe Harbor and Sure Shipwreck   175 C Put-Option Rule Efficient?   175 D Institution Cost  177 III Conclusion  178

Published online by Cambridge University Press

xi

xii





detailed contents

7 Co-Ownership Partition: Proposing a New Auction-Based Design  179 I Comparative Overview  180 A Prefer in Kind to Public Auction   181 B Prefer in Kind to Internal or Public Auction   182 1 Internal Auction as Default   182 2 Public Auction as Default   183 3 No Default  183 C Prefer in Kind, with Call or Put Options   183 1 Public or Internal Auctions, plus Put Option   183 2 Public Auction, plus Put Option, without Internal Auction   184 3 Public Auction, with Call Option   184 4 Partition of Share, with Put Option   184 D Prefer Public Auction to in Kind   185 E Trustees Decide  185 F Judicial Discretion  185 II The Fragmentary Land Problem   186 III A Proposal for a More Efficient Partition Approach   188 A Beyond a Binary Ex Post Analysis  188 1 Subjective Valuations and Economies of Scale   188 2 Partition’s Shadow  189 3 Intermediate Partition Approaches   192 B Concrete Proposal  195 1 Self-Assessment-Based Partition Rules   195 2 Three Steps  199 C Why the CF Approach Is More Efficient   201 IV Conclusion  204

8 Managing Co-ownership: Tragedy of the Common-Ownership?   205 I Comparative Overview  205 A Covenant Not to Partition   206 B Administration  211 1 Always Less than Unanimity   212 2 Sometimes Less than Unanimity   213 3 Always Unanimity  213 4 A Fiduciary Decides   214 C Sale  214 II Tragedy of the Common-Ownership?   215 A Underuse or Overuse?   216 1 Common-Law Jurisdictions  216 2 Civil-Code Jurisdictions  218 3 The Calculus of Consent   219 B Underinvestment  221

Published online by Cambridge University Press

detailed contents



xiii

C Not Necessarily Tragic   222 D Partial Solutions to Underuse and Underinvestment   224 III Conclusion  229

9 Access to Landlocked Land: Hybrid Entitlement Protection  230



I Comparative Overview  231 A Easement of Necessity   232 B Statutory Easements  234 II A Bargaining Power Theory   237 A Overcoming Asymmetrical Bargaining Power and Information   240 B Bilateral Monopoly Theory Reconsidered   242 III The Compensation Element   243 IV The Necessity Requirement   244 A MR = MC Leads to Too Much Passage   245 B “Too Little” Is More Efficient than “Too Much”   246 C Necessity as MR > MC   247 V Passage Location: The Least Damage and Other Rules   248 A Four Prototypes: “Least Damage” Probably Most Efficient   248 B Length and Width   250 VI Ex Ante Viewpoint and Easements of Necessity   251 A Restriction on the Choice of Servient Land   252 B The Gratuity Rule   254 VII Conclusion  256



Part III Movable Property  257 10 Good-Faith Purchase: Proposing Fractional Ownership and Internal Auction   259 I Comparative Overview  261 A The Nemo Dat Doctrine   262 B Good-Faith Purchase Doctrine Not Found   263 C Good or Bad Faith Not Distinguished   263 D Categorical Approach: Stolen Goods Always Revert   264 1 Non-Stolen Goods Not Always Protected   264 2 Purchasers Always Keep Non-Stolen Goods   265 3 Purchasers May Be Reimbursed for Non-Stolen Goods   266 E Binary Approach: Stolen or Not Matters   266 1 Good-Faith Enough to Keep Non-Stolen Goods   266 2 Good-Faith Not Enough to Keep Non-Stolen Goods with Certainty   267 F Unitary Approach: Stolen or Not Does Not Matter   268 1 Strongest Good-Faith Purchase Protection   268

Published online by Cambridge University Press

xiv

detailed contents



2 Reimbursement Rule  268 3 Market Overt and/or Merchant Dealer Rules   268 4 Unique Chinese Rule   269 II A Revised Optimization Theory   269 A Maximize Efficiency in Property Law   271 B Original Owners’ Prevention Costs   273 C Original Owners’ Search Costs   276 D Purchasers’ Verification Costs   277 E Double Negligence  279 F Double Non-Negligence  279 III Fractional Ownership and Internal Auction   281 A Myerson and Satterthwaite Impossibility Theorem   282 B Equal-Share Co-Ownership in Bidding Games Ensures Allocative Efficiency   283 C Determining and Assigning Fractional Ownership   284 D Incorporation with the Doctrine   285 IV Conclusion  287







11 Finders, Keepers: A Minority Rule   289 I Comparative Overview  289 A Finders of Lost Things   290 1 Finders Never Acquire Ownership   291 2 Finders May Acquire Ownership   292 3 Finders Have Property Rights against all the World but Owner  293 4 No Tailor-Made Rule   293 B Reward Fraction and Waiting Period   294 C Finders of Treasure Trove   294 1 Finders Never Acquire Ownership   296 2 Finders May Acquire Ownership   298 3 Finders and Locus Owners Share Ownership   298 4 Locus Owners Acquire Ownership   299 II Economic Analysis  299 A Rewards Increase Incentives to Turn in Found Movables   299 B Schedule Approach to Reward Design   301 C Cultural Variation as Explanation for Variation?   303 D Should Treasure Trove Law Be Different?   306 III Conclusion  308

12 Specificatio: Do What the Romans Did   309 I Comparative Overview  310 A Restoring to Status Quo Ante   311 B Co-Ownership versus Sole Ownership   312 C Two Tests  312

Published online by Cambridge University Press

detailed contents



D Good versus Bad Faith   315 E Summary  318 II Ownership Transfer Rule   319 A High Institution Cost?   322 1 Hold-Out Problem?  322 2 Verification Cost and Benefit   322 3 Good Faith versus Bad Faith   323 B Allocative Benefits?  326 C Other Justifications?  328 III The Compensation Rule   330 IV Conclusion  333

13 Accessio and Confusio: No Sign of Convergence   335 I Comparative Overview  336 A Accessio between Two Movables   337 1 Rule: Sole Ownership   337 2 Rule: Co-Ownership  341 3 Standard  343 B Confusio (Mixture)  343 C Accessio: Seeds or Plants to Land   344 D Accessio: Buildings to Land   346 II Economic Analysis  347 A Economic Formula for Separability   347 B Bad-Faith Party Get Nothing   348 C The Efficient Portugal Fractional Ownership and Internal Auction Design   349 D Simple (First-Order) Rule   350 III Conclusion  351

Conclusion  352

I Theoretical Implications  352 A Converged or Diverged?   352 1 Structural Aspects  352 2 Interconnected and Divergent   353 3 Isolated and Convergent   354 B The Use of Meta-Law   355 C Judge-Made Laws Are Not More Efficient   355 D The Changing Familial Picture   356 II More Topics  357 III Extension  359

Data Appendix  361 Method Appendix  362 References  380 Index  415

Published online by Cambridge University Press

xv

FIGURES

Legal family tree   page 27 Legal families in world map   30 Key factors driving the grouping   35 Distribution of correlation coefficients   40 Correlation coefficients with German property law   42 Correlation coefficients with French property law   42 Correlation coefficients with American property law (proxied by New York law)   42 1.8 Correlation coefficients with Chinese property law   43 1.9 Correlation coefficients with Brazilian property law   43 1.10 Correlation coefficients with South African property law   43 1.11 Correlation coefficients with Israeli property law   44 1.12 Correlation coefficients with Ukrainian property law   44 1.13 Correlation coefficients with Dutch property law   44 3.1 The numerus clausus principle   66 3.2 Structure of use rights for immovables   68 3.3 Security rights on immovables: typology   75 3.4 Security rights on movables: typology   76 3.5 Distribution of limited property forms   92 3.6 Most frequently used limited property forms   93 3.7 Registry capability, workload, and the numerus clausus principle   99 4.1 Transfer ownership of real property    102 4.2 Absolutism in real estate registration   107 4.3 How to transfer ownership of personal property   110 4.4 Absolutism and World Bank’s doing business (DB) data    117 5.1 Adoption of registration- and possession-based acquisitive prescription   133 5.2 Types of possession-based acquisitive prescription   136 5.3 Typology of possession-based acquisitive prescription   137 6.1 Illustration of the issue   161 6.2 Typology of building encroachment doctrine    163 7.1 Typology of partition approaches   181 7.2 Homeacre  192 1.1 1.2 1.3 1.4 1.5 1.6 1.7

xvi

Published online by Cambridge University Press

list of figures 8.1 Judicial power to postpone partition and terminate non-partition covenant  207 8.2 Administration of co-owned land   212 8.3 Voting rule versus individualism   221 8.4 Condominium, cooperative, and commonhold   228 9.1 Typology of easement of necessity   232 9.2 Typology of statutory easement   235 9.3 Marginal costs and marginal benefits of passage   238 9.4 Hypothetical scenario for easements of necessity   253 10.1 Taxonomy of good-faith purchase doctrines   262 10.2 Stolen goods and market overt rule   276 11.1 Typology of the finder law   290 11.2 Reward fraction and waiting period   295 11.3 Typology of the treasure trove law   296 11.4 Reward % by Hofstede data   304 11.5 Reward % by Schwartz data   304 11.6 Expected payoff by Hofstede data   305 11.7 Expected payoff by Schwartz data   305 12.1 Use of the two tests   314 12.2 Structural relationship in Types 0–6   316 12.3 Different ways to treat bad-faith party   318 12.4 Overall typology: co-ownership part   319 12.5 Overall typology: sole ownership part   320 13.1 Typology of accessio   338 C.1 The second general concurrent ownership form   359

Published online by Cambridge University Press

xvii

TABLES

0.1 Convergence–divergence hypothesis  page 11 1.1 Classification of jurisdictions into legal families   31 1.2 Stability values for 10 groups   38 1.3 Stability values for 20 groups   39 2.1 The conventional typology of Rules 1–6   52 3.1 Changes in marginal (social institution) cost   89 3.2 Externalizing costs and benefits under the numerus apertus principle   91 4.1 Constructive delivery in movable sales   112 5.1 Registration-based acquisitive prescription: two dimensions   130 5.2 Typology of possession-based acquisitive prescription   134 5.3 Transparency of land ownership information   154 5.4 Functioning of registries   156 6.1 Typology of building encroachment doctrine   163 6.2 Conditions for neighbors’ losing claims for removal   164 6.3 Courts that have discretionary power to preserve construction   167 7.1 Valuations of Homeacre’s areas   193 7.2 Payoffs for Ann, Burt, and Casper   194 8.1 Covenants for not partitioning   207 8.2 Typology of running covenants between co-owners   208 9.1 Location for statutory easement   235 9.2 Choosers’ utility function under four prototypical methods of determining passage location   249 10.1 Good-faith purchase rules in the English group   265 10.2 Summary of design   270 11.1 Fraction type  295 12.1 Two tests  313 13.1 Comparison of terms under the accession principle   336

xviii

Published online by Cambridge University Press

CITATION FORMAT

Prior comparative data set often lacks citations to the sources for coding decisions. To provide citations and to save space, this book uses § to represent article (or section) numbers in statutes. Country names in English are used. Statutes other than civil codes are cited with abbreviations (see Abbreviations) in parenthesis in the text; for example, Land Code in Sweden is expressed as “Sweden (LC).” Civil codes will not be noted; for example, France §2276 means Article 2276 of the French Civil Code. When articles/sections are renumbered for each part or chapter, the part/ chapter is expressed in Roman numerals; for example, Part 3 Section 101 in the Dutch Civil Code is expressed as “the Netherlands §III:101.” The format used in the book also has the advantage of avoiding the use of the adjective form of a country’s name. This makes it easier for indexing and digital searching. Each chapter will label the prototypes in the format of Type A1a. Type A1 is a subtype of Type A, and Type A1a is a subtype of Type A1, and so on. Type A1a in Chapter 3 can be cited as Type 3A1a. For citations to academic literature, this book uses the author-date system.

xix

Published online by Cambridge University Press

ACKNOWLEDGMENTS

In collecting the raw materials (civil codes and other statutes) for coding property law around the world, I was helped by the librarians at the University of Chicago Law School, Cornell Law School (particularly Alison Shea), University of Illinois College of Law, University of Iowa College of Law, and NYU School of Law. Professor XU Guodong deserves a special thanks. Although we have never met in person, he allowed me to xerox his civil code collections. In coding the property law, I was privileged to work with RAs from Taiwan: Jung-Han Chang, Kai-ping Chang, Hana Chang, Chih-jui Chen, Kuan-ting Chen, Jung Chen, Yea-Shou Chen, Yi-sin Chen, Yu-chih Chen, Yi-sin Chen, Man-Ting Chien, Ming-hsi Chu, Tzu-Yuan Chu, Yi chen Chu, Ming-hsi Chu, Po-jen Huang, Melanie Lee, Yenhsuen Li, Tin-jun Liu, Yi-Chun Ou, Yu-June Tseng, Ming Xuan Xu, and HsienYung Yi. In addition, I benefited from the inputs from RAs from other parts of the world: Winnie Awino (Uganda), Harika Bakaraju (India), Gahli Berger (Israel), Paloma Carreno (Colombia), Danlin Chang (China), Gina Chavarry (Peru), ​Iris Druyan (Israel), Huseyin Guzeler (Turkey), Christina Lee (Hong Kong), Ingrid Lee (Hong Kong), Calvin Lim (Singapore), Hannah Musgrave (New Zealand), Maria Oluyeju (South Africa), Walter Romero (Ecuador), Anne-Line Schwint (France), and Daniella Weinrauch (Israel). Most of them were LLM students at the University of Chicago Law School, Cornell Law School, or NYU School of Law. A few were law students in Israel, Hong Kong, and Singapore. In the final writing stage, I received invaluable assistance from Christine Yuan (Cornell JD 21’), who was first my summer intern (sponsored by the Ministry of Science and Technology, Taiwan) in the summer of 2014, and then my RA in 2021 when I visited Cornell. She is instrumental in doing key research for this book. Also, Kuan-ting Chen, Yi-Chun Ou, Jhih-Wen Yang, Ming-hsi Chu, and Daniel Chang double-checked citations in the book and provided helpful comments. xx

Published online by Cambridge University Press

acknowledgments

xxi

I presented chapters of this book at 2021 AsLEA, 2021 AEDE, 2021 PSEAP, 2021 & 2022 ALPS, 2021 EALE, 2022 AsLEA online webinar series, 2022 Asia-Pacific Private Law Conference, 2021 Ius Commune Conference, NYU CLI Coffee talk, and lectures and workshops at Yale Law School, Cornell Law School, Duke Law School, Humboldt University, and Bucerius Law School. Larissa Katz organized a half-day workshop (virtually from Toronto) on the book manuscript on February 23, 2022. CHEN Ruoying and HOU Meng organized a half-day workshop (virtually from Beijing) on the book manuscript on June 10, 2022. For helpful comments on book manuscripts and answering my blackletter law questions, I thank Bram Akkermans, Benito Arrunada, Kemal Baslar, Jarosław Bełdowski, Gary Bell, Cynthia Bowman, Vanessa CasadoPérez, Aurore Chaigneau, Felix Chang, CHANG Pengao, Jianlin Chen, Weitseng Chen, Chung-Wu Chen, CHEN Ruoying, Dawn Chutkow, Heather Conway, Pierre Crocq, DAI Xin, Gijs van Dijck, Carrie Ding, Sjef van Erp, Chris Essert, Lee Anne Fennell, Tom Gallanis, Nuno Garoupa, Rafael Ibarra Garza, Marco Giraudo, David Goddard, Carlos Ignacio Gomez Ligüerre, Fernando Gómez, Valerie Hans, Lusina Ho, Adam Hofri-Winogradow, HOU Meng, Ulf Jensen, Yotam Kaplan, Larissa Katz, Dan Klerman, Lars Klöhn, Mitja Kovac, Malcolm Lavoie, Joyman Lee, Amnon Lehavi, Monika Leszczynska, Odette Lienau, Ching-Yang Lin, Yael Lifshitz, Ejan Mackaay, MATSUOKA Hisakazu, Giuseppe DariMattiacci, Tom Merrill, Geof Miller, Jessie Owley, Eduardo Peñalver, Víðir Smári Petersen, Jeff Rachlinski, Aziz Rana, Brian Richardson, Alessandra Rossi, Vincent Sagaert, Arthur Salomons, Ronit Levine-Schnur, Alvin See, Daniela Sele, Roy Shapira, Emily Sherwin, Jan Smits, Henry Smith, Jed Stiglitz, Katarzyna Metelska-Szaniawska, Nelson Tebbe, Chantal Thomas, Io Cheng Tong, Cynthia Alejandra Sánchez Torres, Hugo Tremblay, Eva Vermeulen, Lars van Vliet, Konstanze von Schuetz, WANG Yang, Rachael Walsh, Eleanor Wilking, Una Woods, Tzung-Mou Wu, Ying Chieh Wu, XIONG Bingwan, YOSHIDA Katsumi, and ZHU Hu. Professor Rafael Ibarra Garza and Cynthia Alejandra Sánchez Torres, Departamento Académico de la Escuela de Derecho, Universidad de Monterrey, helped me understand the adverse possession and good-faith purchase law in each Mexican state. Part of the book is drawn on my prior works. Introduction is partly adapted from Chang and Smith (2012). Chapter 1 uses the methods developed in Chang et al. (2021) but uses the recently enlarged comparative property law data set. Chapter 2 draws on Chang (2015c). Part I.A.

Published online by Cambridge University Press

xxii

acknowledgments

of Chapter 3 updates the analysis in Chang and Smith (2019). Chapter 4 is a substantial revision of Chang (2018b). Chapter 5 corrects errors in, and is adapted from, Chang (2022a). The theory in Chapter 7 is adapted from Chang and Fennell (2014b). The theory in Chapter 8 is adapted from Chang (2012d). Chapter 9 updates Chang (2016b). Part I of Chapter 10 is based on Chang (2020a). Chapter 12 updates Chang (2015a). I thank my co-authors for allowing me to include our joint works in this book and the journals that have published my works for permission to adapt my prior works.

Published online by Cambridge University Press

ABBREVIATIONS

ADA AOORR APA ARA BG CA

CA CA CC CCLA CDR CF CGSA CL CPC CPO CRE DL DRA EA ELC EV FA FCA FPA FPL GLC ICA IRA ITTA

Assignment of Debts Act (Israel) Act on Ownership and Other Real Rights (Croatia and Macedonia) Acquisition of Property Act (Botswana) The Access to Road Act (Uganda) Baurechtsgesetz (Austria) Co-ownership Act (Finland, Norway, and Sweden) [Lag (1904:48 s.1) om samäganderätt in Swedish; Lov om sameige in Norwegian] [180/1985 in Finland] Conveyancing Act of 1919 (New South Wales, Australia) Company Act of 2013 (India) Commercial Code (California) Contract and Commercial Law Act (New Zealand) Code des Droits Reels (Tunisia) Code de la Famille (Senegal) Civic Government (Scotland) Act 1982 (Scotland) Companies Law (Israel) Civil Procedure Code (California; Macau; Poland) Conveyancing and Property Ordinance (Cap. 219) (Hong Kong) Code of Real Estate (Finland) Decreto Legislativo Nº 1400 (Peru) Deeds Registries Act (Zimbabwe) Easement Act of 1882 (India and Pakistan) Estate and Land Code [Ordonnance N°00-027/P-Rm du 22 Mars 2000 Portant Code Domanial et Foncier] (Mali) Erbbaurechtsverordnung (Germany) Factors Act of 1994 (Singapore) Floating Charge Act of 1958 (Japan) Factory Pledge Act (South Korea) Framework Pledge Law (Bosnia and Herzegovina) General Law of Contracts and Other Obligations (DR Congo) Indian Contract Act (India) Indian Registration Act, 1908 (India and Pakistan) Indian Treasure Trove Act, 1878 (India and Pakistan)

xxiii

Published online by Cambridge University Press

xxiv LA LA LAA LBOPR LC LCLRA LCT LDRA Ley 1676 LF LGM LH LL LL LLLS LO Loi 1973

Loi 2013 LoP LoPA LoPA LP LPA LPA LRA LRA LRA LRERS LRL LRO LRRP LTA LTA MCL MIPA MPL NLC

list of abbreviations Land Act (Bhutan, Maldives, Norway, Tanzania, and Zambia) Limitation Act (Malaysia, Ontario of Canada, India (1963), Pakistan, Singapore, Uganda) Limitation of Actions Act (Victoria Province of Australia) Law on Basis of Ownership and Proprietary Relations (Serbia) Land Code (Brunei, Kazakhstan (2003), and Sweden) Land and Conveyancing Law Reform Act 2009 (Ireland) Liberian Code, Title 29 (Property Law) Land and Deeds Registration Act (Zambia) Ley 1676 de 2013 (Colombia) Loi Foncière (Democratic Republic of Congo) Ley de Garantías Mobiliarias N° 9246 (Costa Rica) Loi Hypothecaire of 1851 (Belgium) Land Law, 5729-1969 (Israel) Land Law 1980 (Royal Decree No. 5/80) (Oman) Limitation Law of Lagos State (Nigeria) Limitation Ordinance (Hong Kong) Loi N° 73-021 du 20 Juillet 1973 Portant Regime General des Biens, Regime Foncier et Immobilier et Regime Des Suretes, Telle Que Modifiee et Completee par la Loi N° 80-008 du 18 Juillet 1980 (Democratic Republic of Congo) Loi n° 2013-01 portant code foncier et domanial (Benin) Law on Pledge (Moldova) Law on Property Act (Slovenia) Law of Property Act 1925 (England) Law on Pledge (Kyrgyzstan) Lost Property Act (Denmark and Norway) Law of Property Act 1925 (England) Land Registration Act 2012 (Scotland) Land Registration Act (Norway) Land Registration Act, Chapter 334 (Tanzania) Law No. 14 of 1964 on the Real Estate Registration System (Qatar) Land Reform Law Cap. 297 (Sri Lanka) Land Registration Ordinance (Hong Kong) Law on Registrable Real Property (Morocco) [la législation applicable aux immeubles immatriculés (B.O. 7 juin 1915)] Land Transfer Act 1952 (New Zealand) Land Titles Act (Singapore; Ontario) Movable Collateral Law [Ley de Garantías Mobiliarias] (Honduras) Moveable and Immovable Property Act (Bhutan) Movable Property Law (Israel) National Land Code (Malaysia)

Published online by Cambridge University Press

list of abbreviations OA OLRS

PA PL PL PL PLA PLA PO PPIP PPL PPSTA PRC PRL PsL RA RERL RLA RLPL RPA RPAPL RPL RTA SD SGA SGSSA SGO SL SL SR TCA TLPA TPA TTA UCC UPHPA

xxv

Ownership Act (Bulgaria) Order relating to Land Registration System (Madagascar) [ORDONNANCE N°60-146 DU 3 OCTOBRE 1960 relative au régime foncier de l’immatriculation (J.O.R.M. n° 129 du 22.10.60, p.2205)] Prescription Act (Zimbabwe) Partition Law (No. 21 of 1977) (Sri Lanka) Property Law of 1990 (Laos) Prescription Law of 1958 (Israel) Property Law Act (Australia 1974; New Zealand 2007) Prescription and Limitation Act 1973 (Scotland) Partition Ordinance (Hong Kong) The Punjab Partition of Immovable Property 2012 (Pakistan) Personal Property Law (New York) Personal Property Secured Transactions Act (Taiwan) Public Resource Code (California) Property Registration Act (Bahrain) Pledges Law 5727-1967 (Israel) Rentcharges Act (United Kingdom) Real Estate Registration Law (Dominican Republic) Registered Land Act (Malawi) Restoration of Lost Property Law 5733-1973 [Hok Hashavat Ave’da] (Israel) Real Property Act (New South Wales Province of Australia) Real Property Actions and Proceedings Law (New York) Real Property Law (New York) Registration of Titles Act (Uganda) Sultani Decree 1980 (Oman) Sale of Goods Act (all common-law jurisdictions) Sale of Goods and Supply of Services Act (Uganda) Sale of Goods Ordinance (Hong Kong) Sale Law 1968 (Israel) Statute of Limitations (Ireland, New Zealand) Sachenrecht (Liechtenstein) Title Conditions Act 2003 (Scotland) The Law of Property Act (Estonia) Transfer of Property Act of 1882 (Myanmar and Pakistan) Treasure Trove Act 1957 (Revised 1995) (Malaysia) Uniform Commercial Code (states in the United States) Uniform Partition of Heirs Property Act (New York)

Published online by Cambridge University Press

Published online by Cambridge University Press

u Introduction

The core of this book is a unique data set I have spent almost ten years compiling and updating. Laws in the jurisdictions studied here are current as of 2018, with the exception of China, whose 2020 new civil code is included.1 The data set chronicles the substance of property law for the issues covered in this book in at least 156 jurisdictions.2 That is, every chapter attempts to include an account of relevant law in the same 156 jurisdictions, which includes three U.S. states (NY, CA, and LA) and two Canadian provinces (Quebec and Ontario)3 (for coding methods and sources, see Method Appendix). On certain issues, this book covers some or all U.S. states and Canadian provinces, as well as looking into provincial civil codes in Mexico. The 247 jurisdictions studied (as shown on the book cover) are where more than 95% of the world population resides. Two hundred and seventy-nine dimensions of property law in each jurisdiction are surveyed by this book. This is, by far, the largest scale comparative property law tome. Each dimension is treated as a variable in my data set, and addresses a specific question regarding the substance of law (see Data Appendix). 1

2

3

The 2020 Puerto Rico Civil Code thus is not coded. Its 1930 code is studied instead. The 2020 Laos Civil Code (its first ever one) is not coded. Laos’ two statutes are studied instead. The 2020 Seychelles Civil Code is not coded. Its 1976 code is studied instead. The 2021 new Book 3 of the Belgium Civil Code regarding property is likewise not coded. Twelve South Pacific countries were coded as one jurisdiction because the only source available treats them as a collective unit (Farran 2013). Mexico is a federalist country with a federal civil code, and all the 31 states and federal district have their own civil codes, many of which are modeled after the federal one and are substantially similar. Except in Chapters 5 and 10, only Mexico’s federal civil code is considered. Although Australian provinces do not have exactly the same law, this book takes New South Wales law as the basis to code “Australian” property law. Most observations in our data concern nation states, but 10 other jurisdictions were also coded, each as one observation. In addition to the American states and anglophone Canadian provinces, I coded Quebec, Scotland, Puerto Rico, Hong Kong, and Macau.

1

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

2

introduction

I  Three Approaches As the book title suggests, this book fuses three analytical approaches. Each of Chapters 3–13 follows the same format. Part I is a comparative law overview and the law is often summarized quantitatively. Part II (sometimes broken into multiple parts) provides an economic analysis of the doctrines in question and often provides empirical examinations of the discussed economic theories. Chapter 1 and several other chapters use quantitative methods to describe the variations in property law.

A  Comparative Analysis The comparative analysis in this book focuses on the substance of property law. Put differently, black-letter legal doctrines take center stage. Prior comparative law studies are concerned mostly with the laws in certain developed countries; as a result, many interesting legal schemes addressing well-known problems have been ignored. One contribution of this book is to introduce to readers to interesting legal solutions adopted in under-studied countries.4 Chapters 3–13 each start with a typology of a single doctrine, or of multiple related doctrines. The typologies in each chapter are my own. Other scholars could reasonably come up with different ways to conceptualize and categorize the various legal schemes. Inevitably, this book may sometimes misclassify certain countries because laws on the books differ from laws in action.5 A “wiki” project that I would like to initiate following the publication of this book would hopefully reduce comparative law coding errors. That said, whether as a comparative exercise for its own sake, or as a springboard for economic analysis, the more important – and less error-prone – takeaway is the prototypes identified in each chapter. The prototypes are distinct real-world schemes addressing legal issues. Many of these prototypes have not been subject to legal analysis in English, not to mention economic analysis. The objects of my economic analysis are the prototypes, not laws in individual countries. Comparative analysis in this book takes account of, but looks beyond, the different “styles” of property law. Chang and Smith (2012: 4–5; 2016:

4 5

This book thus uses the new private law theory proposed by Grundmann et al. (2021: 2–3). In addition, the law in books in civil-code jurisdictions reflect legislatures’ policy choice, while the law in action there reflects courts’ law-making. A systematic analysis of legislature’s choice of designs in a particular doctrine should be of interest in and of itself.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

3

132–133) distinguish the “structure” and the “style” of a system of property law. The structure of property law refers to how property law groups ­problems so that they need not be treated in a fully articulated fashion. Serving the essential function of property – protecting uses in a world of positive institution costs – still leaves a great deal of freedom in terms of how to serve those objectives within a legal framework. Style describes a characteristic manner of doing things. In property law, one example of style would be the reliance on possession and a more implicit definition of ownership in common law systems versus the definition of dominion and departures from it in typical civilian systems. Likewise, a lease can be a contract given in rem protection or can be delineated as an in rem right of a limited scope (Chapter 3). To give another example, in civil law countries, rei vindicatio – the action to force someone to return possession of a thing to its owner (Brandsma 2015: 11) – is the major right of a property owner, while this expression is almost untranslatable into legal terminology in English (“revindication” is the usual English word, which means nothing to common-law lawyers) (see also Graziadei 2017: 73–76). Property owners in the common law, of course, are generally well protected – by different means with different labels (trespass, conversion, replevin, and so forth). Thus, in comparing law across jurisdictions, it is often necessary to pierce through the veil of style and examine the functioning of a particular legal rule within the property structure. As Chapter 3 shows, while many limited property forms have different names, coding them differently would be to mistake style for function. Chapter 6 also demonstrates that countries may use different doctrines to tackle similar legal issues. One takeaway point of the comparative exercise in this book is that common law and civil law are not that different if one looks beyond stylistic differences.

B  Empirical Analysis In terms of empirical analysis, this book is situated in the emerging trend of empirical comparative law, already thriving in constitutional law (e.g., Elkins et al. 2009; Gutmann et al. 2014; Law 2016; Chilton and Versteeg 2020), antitrust law (e.g., Bradford et al. 2019b), and corporate law (Armour et al. 2009; Spamann 2009a; 2009b). My property data set appears to be the only systematic coding of legal substance within private law. Chapter 1 uses unsupervised machine-learning methods to draw a legal family tree for the studied jurisdictions. In other chapters,

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

4

introduction

my property data set, and external data sets such as World Bank’s Doing Business indices, are also drawn on to demonstrate correlations of studied legal dimensions and explore potential explanations for observed divergence across countries.

C  Economic Analysis The second part of each of Chapters 3–13 analyzes which of the e­ xisting legal schemes is more efficient.6 Existing law-and-economic analysis often uses stylized mathematical models to identify an efficient mechanism, but traditional lawyers often find such models irrelevant – sometimes rightfully so. As Chapters 4 and 11 show, the prior stylized economic analysis missed important real-world features of law. Mathematical models grounded in realistic legal conditions will make a direct impact on the legal system. This book always starts with the economic analysis of real-world legal schemes. Even though these existing schemes may not be the best solutions, as they are the current law, it should be interesting in and of itself to know which legal schemes are more efficient. Each chapter closes with a conclusion on which legal systems have adopted more efficient rules. Sometimes, when none are clearly efficient, a theoretically more efficient and practically realistic scheme will be discussed. Comparative law and economics (De Geest and Van den Bergh 2004; Faust 2008; Eisenberg and Ramello 2016) brings to the table economic analysis of one national law may lack. Looking at a single national law and considering a possible economic rationale, analysts may subconsciously assume the efficiency of their own legal system. Chapter 10 offers such an example. Posner (1973) famously claims that the common law tends to be efficient (for a review, see Depoorter and Rubin 2017). Had the common law been compared head-to-head with rules in other countries – like what I do in this book – Judge Posner may have qualified his claim. The upshot of comparative economic analysis is that there are at least two legal schemes under scrutiny, and their relative costs and benefits must be 6

As Merrill and Smith (2020: 138) caution, only if a doctrine is relatively separable from the system does it make sense to ask whether it is efficient. This book will remind readers of relevant, supplementary, or alternative doctrines that could deal with the issues in question. Meta-law (see below) is often not embedded in specific property doctrines and thus will not always be discussed in the individual chapters. Readers should keep in mind that in some jurisdictions more than others, equitable provisions may intervene to replace inefficient rules as applied to specific contexts.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

5

examined. By contrast, analysis of one national law often implicitly compares the legal scheme in question with no legal scheme at all, thus giving the legal scheme, whatever its design, an edge. Using plain language instead of formal mathematical models, I aim to introduce to traditional property scholars the power of economic analysis. Chapter 2 will define efficiency and introduce the economic tools used in the following chapters. Providing a detailed description of real-world institutions, this book invites economists to further model them. As the following chapters show, some interesting topics have not yet caught the attention of economists. I hope that this book serves as a new starting point for future economic modeling. The economic analysis in this book is both positive and normative. It is positive, as each chapter demonstrates the allocative benefits and institution costs each legal scheme (i.e., prototype) entails. It is normative because based on the normative prior that efficiency is one important social value (Chapter 2), this book proposes that efficient property schemes should be a strong candidate for legislative adoption, especially when the most efficient scheme does not create more inequality of wealth or income than other possible schemes. The use of economic theories as an external approach to the study of property law makes this book a New Private Law study as defined by Gold et al. (2020). Economic analysis is often used to explain why lawmakers adopt certain legal rules or why certain legal phenomena emerge (Kornhauser 2022). As explained below, this book adopts an explanatory theory that uses whether a property doctrine is isolated from other parts in the legal system to explain why property law schemes sometimes converge and sometimes diverge. The whole book can be considered to offer dozens of empirical studies on theories on convergence and divergence of law. Chapter 2 emphasizes the role of third-party information cost in explaining the difference between property law and, that is, contract and torts law. These two explanatory perspectives are used throughout the book. In some chapters, an additional type of explanatory analysis is offered. Those chapters attempt to tease out whether differences in national cultures drive the variation of property laws across the globe.

II  Engagement with Existing Theories This book proposes no grand theories – that is deferred to my next book. Nonetheless, the book sheds light on several prominent theories, introduced below. In addition to reading each chapter separately, readers can

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

6

introduction

also read Chapters 3–13 together as preliminary empirical evidence for (or against) these high-level theories.

A  Convergence and Divergence of Property Law Is this the end-of-history for property laws (cf. Hansmann and Kraakman 2000a)? Meaning, do property laws converge over time? While this book is not able to empirically examine this question (due to the lack of panel data on property law), starting with certain reasonable premises, it provides rich empirical tests for the theory of Chang and Smith (2019) on whether and why some aspects of property law converge while others diverge.7 Again, the structure versus style framework is useful. Structure and style raise the issue of how tightly a given aspect of property law is integrated into the overall system. Private law doctrines that are most integrated into the overall system are the most difficult to change; doctrines that are easily treated in isolation, with fewer ripple effects, are conversely much easier to modify (Smith 2015: 2067–2074). In property law, the various doctrines and institutional features occasionally interlock. Those that are highly interconnected with the rest of the system, like possession (Chang 2015b), are difficult to change, in contrast to doctrines that can safely be treated in isolation, like the form of common-interest communities. Structures of property do appear to converge, if not being very similar to begin with. As this book will show, most jurisdictions address the same types of property issues. This suggests that the problem of serving property’s functions at positive information cost everywhere creates the same disputes. Uncertainty over ownership gives rise to the acquisitive prescription doctrine (Chapter 5). High costs of verifying true owners of 7

Amnon Lehavi pushed me to think about how and why this framework may or may not explain convergence and divergence in other fields, such as copyrights. My preliminary thought is that intellectual properties and antitrust regulations are less appropriate to be analyzed in this framework, as legal changes there have not been spontaneous. Rather, superpowers such as EU and the United States have interests in exporting their antitrust laws (Bradford et al. 2019a) or imposing their stronger IP protections. Areas that are subject to stronger jurisdictional competition, such as corporate law or trust law (Sitkoff and Schanzenbach 2005), may also be ill-suited for the framework here. Lehavi also pointed out that testing the framework by studying cross-country efforts to harmonize laws, such as Europe hypothec and DCFR. Chang (2016c) studies the provision on possession in DCFR and finds the provisions too complex and self-contradictory. These possession provisions are not adopted also because possession is one of the most interconnected concepts in property law. Full treatment of this issue is beyond the scope of this book.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

7

movables lead to the good-faith purchaser doctrine (Chapter 10) and the accession principle (Chapters 12 and 13). Use rights and security rights, while often bearing very different names, are staples in virtually all property systems (Chapter 3). The exact contents of property doctrines do not necessarily converge. Lawmakers around the world face the same issue of positive information costs, but the same problem does not always call for the same solution. Information costs only force legal systems to come up with a solution, but perhaps sometimes anything goes (Levmore 1987; Dari-Mattiacci and Guerriero 2019). Many, if not most, doctrines mix structural and stylistic aspects. When lawmakers for any reason settle on a solution, etched in civil codes or leading cases, they do not always have strong reasons to change the solution to become more like other jurisdictions. Solutions are more likely to converge if the doctrine in question is more isolated from other doctrines. This is true of structural and especially of stylistic aspects of law. In an interconnected doctrine, such as the definition of possession, convergence with other jurisdictions that require deviation from the status quo forces other pieces in the whole system to go along with the change. In civil-code jurisdictions, in particular, fear of unintended consequences arising due to changes to foundational doctrine in a civil code could kill any proposal for deviation. France and Germany each have their own conceptual system of possession, which is hard to uproot after hundreds of years of doctrinal interpretation. When European scholars proposed the Draft Common Frame of Reference (Perez and Liguerre 2019), they neither found common ground nor simplified the concept. Instead, they opted to maintain the two conceptual systems of possession – creating much confusion and contradiction (Chang 2016c: 11–23). By contrast, a more isolated “downstream” doctrine (Levmore 2019) will have more wiggle room, as, in the worst-case scenario, a failed experiment would not drag down the whole system. Co-ownership partition is a prime example of such a doctrine (Chapter 7). A majority of the studied jurisdictions prefer partition in kind and allow some forms of sales. Easement of necessity law also shows strong signs of convergence (Chapter 9). In addition, almost all jurisdictions with the specificatio doctrine use one (or both) of the two tests (Chapter 12). The widely adopted condominium form (70% of studied jurisdictions) is another example.8 8

Druey (2004: 100) notes that in Switzerland, the condominium form was statutorily sanctioned in 1965 due to economic need, even though earlier it had been considered inconsistent with other parts of the co-ownership law.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

8

introduction

To be sure, isolation and interconnection are not the only reasons for doctrines to converge or diverge. Many other factors – the benefits of convergence, for one – affect lawmakers’ decisions. That is, large benefits of convergence may push interconnected doctrines toward convergence. European efforts to streamline the registration of mortgages is a case in point (Erp 2002: 86). By contrast, small benefits associated with convergence will leave isolated doctrines untouched. Doctrines related to boundary encroachment (Chapter 6), accession (Chapter 13) and finders (Chapter 11) are three such examples. If there is a globally efficient solution to a legal issue, countries’ inclination to adopt it will be observed as convergence. When a legal issue only has locally efficient solutions, meaning what is efficient is contingent on other institutional features, countries’ inclinations to adopt efficient solutions will be observed as divergence. Prime examples of this point are seen in the adoption of the public faith principle (Chapter 4) and possession-based acquisitive prescription (Chapter 5), which depend on the capacities of local registries. Approaches to convergence and divergence are rooted in a combination of relative propensity to change and relative closeness of starting points. In a system like property, change over time will flourish or be cut off depending on the resultant fitness of the overall system. Alston and Mueller (2015) demonstrate that the various elements of the bundle of rights may be relatively isolated or show “epistatic” connections. In an epistatic connection, a change in one element will lead to an effect on a connected element. Thus, the right to draw water affects the value of the right to grow tomatoes, but the right to prevent airplane overflights is unconnected to either the right to draw water or the right to grow tomatoes. Once epistatic connections are in the picture, the implications of different patterns and densities of epistatic connections for the evolution of property rights are likely to be quite important. Along a spectrum, three types of scenarios can be distinguished. First, the elements in the bundle of rights might be wholly unconnected. If one gets the answer right for each element, then all one has to do is add up the effects of all the elements, and one can be assured that the entire bundle is optimized. If so, it is easy to change individual elements without the downside of severe, unrelated (by epistatic connection) negative effects emerging in the bundle. However, assuming epistatic connections away is unrealistic, even if convenient. At the opposite extreme is maximal epistatic connection: everything is connected to everything else. If so, the pattern of consequences to minor variations in one element of the bundle is random or chaotic and very difficult to predict. This pessimistic picture similarly fails to describe our

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

9

world. Some problems, like high-altitude airplane overflights, can indeed be treated in isolation. In between these two extremes of zero and total connectivity is what has sometimes been called “organized complexity” (Smith 2019; 2020; 2021b). Here, epistatic connections are important but far from universal. They can also be clustered. Innovation is promoted by the fact that interconnection is not complete and is semiorganized. Changes to part of the bundle can be made, and the overall effect can move in the direction of a local optimum more easily than under complete connection. Notions of essential function and interconnection allow us to form expectations, or even predictions, about the convergence and divergence of property systems. Structural aspects of modern property systems that solve the basic problem of managing use conflict, and avoiding intractability will cause some convergence on the exclusion-governance architecture (Smith 2002). To be sure, the relative emphasis on exclusion and governance will vary according to local conditions, and in particular, it will be easier to add, subtract, or modify governance rules than it will be for analogous changes to the exclusionary setup (Smith 2004a; 2004b; 2012b). Property systems will converge in having a mix of exclusion and governance and will diverge more in the area of governance than in exclusion, as exclusion is one essential element in property. Co-ownership management is a prime example. While everywhere in the world a single co-owner is entitled to act alone in evicting outsiders, countries differ in requiring a majority, super-majority, or unanimous vote to govern the coowned resources (Chapter 8). Because in general stylistic variation is more detachable from the system, divergence more easily arises in governance than in exclusion. Thus, the first proposition is that structural aspects of property law should show convergence and the structures in question should be stable over time. More tellingly, even if the initial condition of the property structure is no exclusion at all, this arrangement is unlikely to persist if open-access commons do not make sense on its own terms (i.e., it fails to provide benefits that exceed the costs, or compares unfavorably with other possible arrangements). A prime example is the people’s commune during the Cultural Revolution in China. Private property and individual farming had been the norm and practice, but during the revolution, the government mandated a shift to limited-access common property. As is well known, this social experiment did not last long (Coase and Wang 2012). The structure of property law, therefore, will converge to an exclusionbased system, regardless of the initial conditions.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

10

introduction

Regarding more stylistic features, those which are more or less i­nterconnected with the rest of the system of property law must be ­distinguished. The second proposition holds that, in any less interconnected aspect of the property system, more stylistic variation can be expected. The third proposition is that the less interconnected an aspect of the property system is, the more it could change over time. One reason for change is voluntary borrowing, or legal transplant, due to colonialism (Berkowitz et al. 2003a; 2003b; Klerman et al. 2011). Previous work has found that in the admittedly small number of mixed systems that have both ­common and civil law heritages, there is a tendency to borrow contract law more than property law (and never the latter without the former) (Palmer 2001: 57; Kim 2010: 711–714), and within property law to borrow more in the in personam than in the in rem aspects (Merwe 2003: 274–289). Thus, in terms of changes over time, stylistic features would converge or diverge, but more rapid changes will occur in less connected, rather than in more connected, areas of property law. Now, after decades, or even centuries, of evolution in property law, it is more likely to observe the convergence of isolated (less interconnected) doctrines, if at least one of the following conditions holds: (1) there is one or a few apparently dominant strategies; (2) convergence in a global market saves transaction costs and attracts investment and business; or (3) there have been conscious or subconscious, voluntary or involuntary borrowing or legal transplants, with or without explicit efficiency concerns. The key point is that less interconnected doctrines are more likely to vary (resulting in divergence or convergence depending on background conditions) than more interconnected ones. This book uses a snapshot of current property systems. The third proposition – that the interconnectedness of doctrines should correlate with rapidity of change – cannot be tested with the available data. As for the first two, assuming that systems are not subject to overwhelming pressures to converge, the reasonable conjecture of our world are that, first, the basic structure is highly systemic and so should show more convergence than do more structurally peripheral aspects. Second, among styles of property law, those that are connected to the rest of the law retain their greater diversity. This is based on the particular conditions of our world where the common and civil law have very different starting points (styles) in terms of property law (Chang and Smith 2012: 36–54). The civil-law system is inherently plural. Hence, what is more interconnected has different starting points and remains divergent, whereas what is less interconnected could diverge or converge due to one or more of the three forces laid out above. Table 0.1 summarizes the theoretical framework.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

11

Table 0.1  Convergence–divergence hypothesis Aspects in property system Structure

Structure Style, interconnected Style, interconnected Style, isolated

Initial conditions During evolution

Observed outcome

Same (exclusion- Remain governance based system) Different (e.g., Cannot sustain open access) Same Hardly change

Convergent*

Different

Hardly change

Divergent*

Same

Vary

Style, isolated

Different

Style, isolated

Different

Convergent Convergent

Divergent (by definition, cannot become convergent) Vary (e.g., borrowing, Convergent* transplanting, efficiency conformity) Divergent Vary (e.g., for national identity; cultural difference)

Note: * marks the situations in most jurisdictions.

B  Law versus Meta-Law Another theory that can tie together the comparative materials is law ­versus meta-law (equity) (Smith 2021a). As Smith (2003; 2012a) points out, ­private law in general, and property law in particular, employs the ­principle-exception structure, with exclusion being the principle, and governance the exception. Here, law is the principle and meta-law (equity) the ­exception. Put simply, ideally the baseline rule in property will be ­crystal clear, whereas exceptional applications of meta-law may be patternless, so as to deter opportunism. Chapters 4, 5, 6, 9, 10, 12, and 13 deal with doctrines where a key aspect is whether to distinguish between good-faith and bad-faith parties, and whether the legal protection afforded to them should differ. One

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

12

introduction

paradigm is to have mostly clear rules and leave all equity issues (including good faith versus bad faith) to a separate body of law – in common-law systems, equity law; in civil-law systems, perhaps unjust enrichment, the principle of good faith, or other untailored legal standards. The ongoing Restatement of Law Fourth, Property, is moving toward this direction. The other paradigm is to spell out all the related concerns and carve out exceptions within each doctrine. Which paradigm is a better fit for a certain country may depend on its legal culture. In addition, the two chapters on co-ownership, Chapters 7 and 8, also deal with meta-law. The new partition approach championed in Chapter 7 has in mind co-owners’ strategic bargaining and bidding behaviors and does the best it can to ameliorate such ill incentives. This is of course consistent with the core function of suppressing opportunism in meta-law (equity). Chapter 8 discusses when courts should have the power to delay partition request and to break no-partition covenants. These are equitable doctrines in common-law jurisdictions.

C  Are Judge-Made Laws Generally More Efficient? Posner (1973) famously claims that common law tends to be efficient while statutes tend to be inefficient. Civil codes are, of course, statutes. To address this claim from the current causal inference paradigm would be a challenge – there is no counterfactual. That is, unobservable is a parallel world in which everything is the same except that the rule in question is made by court rather than by legislature. Judge Posner’s approach is to demonstrate that judge-made law in the United States is generally efficient while U.S. statutes – governing, among others, public law issues – is often inefficient. This book’s approach is to observe the same property doctrines across the world and to ascertain whether judge-made schemes or statutory schemes are more efficient. The conclusions in Chapters 3–13, read together, do not support the view that common-law jurisdictions tend to have more efficient property law, casting doubt on Judge Posner’s claim.9

D  Legal Origins and Legal Families Economists discuss legal origins while lawyers discuss legal families. The two concepts are largely the same, but do these concepts accurately 9

For other recent empirical endeavors, see Niblett et al. (2010), Niblett (2017), Niblett (2020), and Chang (2020c).

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

13

describe the legal substance? The question is particularly acute for legal families because the concept does not refer to the past, but the present, suggesting that one may make an educated guess about similarities between the laws of two countries once it is known whether they belong to the same legal family. Chapter 1, using the property data set in unsupervised machine-learning algorithm, shows that the traditional legal family classification does not map onto the quantitative results based on my coding of property doctrines. Bradford et al. (2021) further show that the traditional legal family classification is utterly useless in antitrust law and only marginally useful in property law. To the extent that lawyers need the concept of legal family to have a sense of how similar a pair of countries’ laws are, the legal family concept must be rethought. Moreover, my empirical analysis suggests that notable variations exist even within a legal family. The traditional idea is that comparative lawyers can select one representative country from each legal family and simply compare these presumably prototypical countries (Zweigert and Kötz 1998). My analysis suggests that illuminating doctrinal designs may be unduly ignored under this approach. True, comparative law has evolved. The most recent trend is to move beyond the legal texts and compare the law-in-action across countries. Still, comparative law should be empirical (Spamann 2009c; 2015) no matter whether law on the books or law in action is studied, and comparative lawyers cannot safely ignore less studied countries.

III Terminologies Throughout the book, in addition to consciously using Hohfeldian terminologies (Hohfeld; Balganesh et al. 2022), I will use the following terms according to the defined meaning. 1. Bad faith = know or should have known = mala fide = has knowledge. 2. Good faith = do not know = bona fide = good faith = no knowledge. 3. Civil-code jurisdictions: 105 studied jurisdictions have a civil code, and all but California are considered civil-code jurisdictions. Bulgaria, Croatia, Estonia, Macedonia, Senegal, Serbia, and Slovenia do not have civil codes but have stand-alone statutes that cover nearly all property law (though sometimes only the immovable aspect). These seven countries are considered civil-code jurisdictions. In total, there are 111 civilcode jurisdictions and 45 no-civil-code jurisdictions.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

14

introduction

4. Immovables and movables: To save space, immovable and movable are used as countable nouns. (Im)movables will be used instead of (im) movable properties or (im)movable things. Immovables are often called realty, real estate, or real property, whereas movables are often called personality, personal estate, personal property, or chattel. But these alternatives will not be used unless necessary.10 Movables are limited to tangible (or corporeal) things. Intangible (or incorporeal) objects of property relations are called as such. Of course, things themselves and the (Hohfeldian) relationship regarding them should be distinguished, and this will be the topic of my next book. To simplify, this book often opts for the shorter descriptions. Also for brevity, this book uses land ownership to refer to fee simple absolute in common-law systems. 5. Registration: Registration is used as the umbrella term for the two major types of immovable information depository: registration-of-rights (including the Torrens version) and recording (elsewhere sometimes called recordation or registration of documents). A registration-ofright system “is always done in the form of a realfolium, i.e. ordered by the land registered.” By contrast, a recording system “is normally done in the form of a personalfolium, i.e. ordered by the name of the respective owner” (Schmid and Hertel 2005: 32). The personalfolium is called the grantor-grantee index in the United States. Some jurisdictions in the United States also have a track index, similar to realfolium. 6. Jurisdictions: It is impossible to be comprehensive. “No jurisdiction” in this book means “no studied jurisdictions,” and I write as if studied jurisdictions where I have not been able to find relevant stipulations surely have no relevant rules. The over-confident tone serves to save space and shorten the prose. Jurisdictions include both countries and subnational entities like Louisiana. 7. Restatement of the Law Fourth, Property: The Restatement is an ongoing project by American Law Institute (ALI) to describe American common law. ALI has granted me (as an Associate Reporter) permission to discuss the draft Restatement in this book. The cited sections and notes in this book have not yet been approved by ALI members. The book tries to strike a balance between coining new concepts and using existing terms as functional categories. As readers will find, statutes and codes in different jurisdictions often use the same term (sometimes 10

The term “real property” refers to land only in some countries, while in others this term means “relating to things” and thus includes movables.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

15

with linguistic variations) in somewhat different ways. Using existing terms brings familiarity to some readers but unavoidably confuses others. Therefore, this book attempts to define each key term functionally and readers should be careful not to take certain terms’ meaning for granted.

IV  Overview of the Book The book is divided into three parts. The first provides a holistic empirical view of the entire data set, lays out the economic framework, and looks into rules regarding voluntary transactions. The second contains five chapters regarding immovables, while the third includes four chapters on movables. The doctrines discussed in the latter two parts all involve ­noncooperation or outright intentional infringement of others’ property rights. Chapter 1 makes use of two empirical approaches. Its first part uses property law from 136 jurisdictions in an unsupervised machine-learning method (hierarchical clustering) that divide these jurisdictions into 10 legal families. Unlike the traditional wisdom that highlights the difference between common law and civil law, this chapter finds that, in terms of property doctrines, a trichotomy better describes the legal systems: one big group is jurisdictions affected by French property law; another big group is composed of jurisdictions that follow or resemble German property law; and the final group contains common-law jurisdictions, Nordic countries, and a number of socialist jurisdictions. The second part of Chapter 1 re-combines 156 jurisdictions into 149 countries, and computes the correlation coefficients among each country pair, to show dyadic similarities in property law. Chapter 2 defines efficiency. The efficiency criterion here is cost–­ benefit analysis, where cost is institution cost (including information and transaction costs) and benefit is what is called “allocative efficiency” in the literature. My efficiency criterion builds on 60 years of law-andeconomics research in property law, but I believe that this is the first time that efficiency has been formulated in this way. Chapter 2 positions efficiency as a first-order value, while welfare is a second-order value that includes efficiency and other first-order values such as distribution of wealth. In addition, Chapter 2 introduces the concepts of ex ante viewpoint and the property rule versus the liability rule, both of which will be drawn on in later chapters. This book’s comparative exercises start with Chapters 3 and 4, as both concern voluntary transactions (embodying the governance strategy) and the legal constraints. The numerus clausus principle and the mandatory

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

16

introduction

rules in property (such as the requirement to register land sales) and the legal constraints, and yet in the transaction context, the baseline is that the law does not compel any transaction; rather, the law only sanctions certain types of transactions. As said above, due to positive institution cost, the exclusion strategy is the foundation of all property systems. The book does not directly address exclusion, but exclusion is always the starting point of all the analysis. Thus, adverse possessors should be evicted, and encroaching buildings shall be torn down. Chapters 5 and 6 deal with the question of how to justify the acquisitive prescription doctrine and the building encroachment doctrine, which deviates from the exclusion strategy. Chapters 7 and 8 focus on co-ownership. Co-owners can of course manage properties within the legal constraints. These two chapters analyze rules that deal with the dire situation when collective agreement cannot be reached. Exclusion is the starting point but not always applied. Otherwise, the Book on Property in the civil codes could be very short. Instead, civil codes often spill ink on exceptions to the exclusion strategy. Chapter 9 deals with the involuntary governance strategy applied in the landlocked land context. Neighbors cannot completely shut down access by their landlocked neighbors. Chapter 9 explains why. Turning to movables. Movables have traditionally been viewed as less important than land, but the rise of intangible property rights and intellectual property rights have given rules regarding movables a more salient position. Chapter 10 focuses good-faith purchase. The doctrine’s highprofile application in stolen arts and analogy to copyright law (Balganesh 2016) provide self-evident explanation for inclusion in the book. This chapter circles back to Chapters 3 and 4. Here, two good-faith parties do not necessarily make a valid transaction. The substance of this doctrine, like the designs in Chapter 4, is instrumental in reducing institution cost. Although not many people have (literally) hit the jackpot, perhaps every person has found a lost item or two during her life. Chapter 11 discusses the finder doctrine, which is often mentioned in English textbooks on property when it comes to the section on the relativity of title. Yet many countries view finders in an entirely different way. This chapter analyzes how to design the finder doctrine so as to keep movables in the hands of higher valuers. Finally, Chapters 12 and 13 address the accession principle, which serves as the foundation of property law but has been rarely drawn on in court cases, as the legal doctrine is preempted by contracts. This book uses the accession to emphasize the import of putting bad-faith parties in check.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

17

More specifically, Chapter 3 focuses on the numerus clausus principle and the limited property rights. It first documents that about a quarter of jurisdictions explicitly adopt this principle, while many others do so implicitly. Chapter 3 argues that this principle is generally efficient. Chapter 3 will also show that property forms such as mortgage (called hypothec in civil law) and real easement are extremely popular. Chapter 4 focuses on how ownership of immovables and movables are transferred (i.e., whether registration is not needed, necessary, or creating opposability to third parties), whether registration creates absolutism (public faith principle), whether a real agreement is conceptually separate from a sale contract, and whether an invalid sale contract always leads to the invalidity of a real agreement (non-causa principle), and whether delivery or certain intentions are required to transfer ownership of personal properties or the sale contract itself is sufficient. This is where the traditional idea of legal families is conspicuous. Transfer doctrines involve how notice is given. The choice of registration system demonstrates how states, given path dependence, trade off transaction costs and third-party information costs. Which type of conveyance doctrine regarding immovables is efficient is contingent on factors outside of the law. It is easier to reform conveyance doctrine regarding movables, and lawmakers should provide alternative default rules (“menus”) more frequently and establish clear opt-out procedures (“altering rules”). Chapter 5 analyzes acquisitive prescription, a broader concept than adverse possession, and argues that registration-based acquisitive prescription with title and good-faith requirements can be justified by efficiency under certain conditions – Possession, however, is redundant, and may even give rise to undesirable outcomes. Given that boundary disputes can be left for another doctrine, possession-based acquisitive ­prescription – no matter whether possessors act in good or bad faith – can hardly be justified on an economic basis in countries with well-­functioning registrars if possessors do not have title. The possession-based acquisitive prescription can only be justified in jurisdictions with dysfunctional registrars. Chapter 6 discusses building encroachment, which is a topic closely related to adverse possession – at least in the common law. Chapter 6 documents the three different doctrinal approaches to dealing with encroachment over boundary and focuses on the building encroachment doctrine enacted in 52 jurisdictions. The prompt protest rule and encroachers’ not acting in bad faith are easy to justify economically (though not universally adopted). Encroachers’ good faith is increasingly unlikely given

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

18

introduction

the advance of mapping technology. A two-tier building encroachment doctrine (with safe harbor and sure shipwreck) is best. Even though this doctrine has been used as an example of a put option, it is not, and will be inefficient if treated as such. Chapter 7 summarizes the various partition approaches used around the world, finding that partition in kind is often preferred, with selling coowned things through public or internal auctions a common back-up plan. Perhaps as a result, the existing literature has focused on partition in kind and partition by sale, while ignoring intermediate partition approaches like partial partition that are prevalent in practice. Little attention has been paid to the use of revelation mechanisms such as self-assessment, nor to how judicial partition rules affect co-owners’ prejudicial-partition behaviors. Chapter 7 brings partial partition into the theoretical framework and proposes a new and feasible partition method that utilizes private information among co-owners and makes partition more efficient. Chapter 8 first provides an overview of the stipulations regarding how things held in tenancy in common (the most common co-ownership form of property around the world) should be administered and sold, as well as co-owner agreements not to partition. Then, Chapter 8 addresses whether the several types of rules lead to underuse or overuse – that is, whether tenancy in common may lead to tragedy of the commons or anticommons. The prevalent doctrine that provides one co-owner with a unilateral power to call for partition avoids a long-term tragedy but underinvestment and underuse of co-owned resources are still likely. This chapter ends with a proposed solution to ameliorate the underinvestment and underuse problems. Chapter 9 studies access to landlocked land and shows that most jurisdictions aptly use a mixture of ex ante and ex post viewpoints to design their doctrines. This chapter follows American parlance and divides the doctrine into “easements of necessity” and “statutory easements.” They have intuitive appeal: for statutory easements, owners of servient land should be compensated; easements should be necessary; and the location of the passage should cause the least damage to the servient land. As for easements of necessity, the landlocked owners can only gain access to land held by the grantor at the time of the conveyance. The prevalent scheme under statutory easement to solve this legal entanglement is neither the property rule nor the liability rule, but a “hybrid rule,” an unheralded mixture of the property rule and liability rule. Chapter 9 argues that the hybrid rule is efficient because it strikes a balance between facilitating voluntary transactions through reducing transaction costs, reducing

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

introduction

19

cost externalization, and preserving property value. More specifically, the hybrid rule stipulates that the extent of statutory easements should be set at where the marginal social benefit of prescribed passage sharply declines, and passage locations should be determined following the least damage rule. As for easements of necessity, the limited access rule and the gratuity requirement make economic sense from an ex ante viewpoint. Overall, the law leaves ample room for private negotiation, which is more likely to lead to the most efficient outcome than either ex ante rulemaking or ex post judicial adjudication. Chapter 10 identifies 21 variants of the good-faith purchase doctrine, which are often different combinations of several key factors. That said, none of the 21 schemes are the most efficient. Among the forms of goodfaith purchase doctrine currently in use, the market overt rule comes closest to ex ante efficiency because original owners, merchant dealers, and consumers all have incentives to spend close to optimal costs on verification and prevention, and the movables in question are more likely to be in the hands of higher valuers. Drawing on mechanism design literature, Chapter 10 argues that when both an original owner and a consumer are nonnegligent, the two parties can be assigned 50% shares of the movable in question, and an ensuing internal auction between them can ex post tease out who values the resource more. This internal auction design is inexpensive to administer. Chapter 11 observes that “finders, keepers” is in fact a minority rule globally. Finders often have to wait for several months to receive any benefit, and often cannot acquire ownership of the found items. Instead, they may receive rewards from the losers or from the state. The current economic analysis of this issue does not map exactly onto the doctrine itself. The better economic justification for the finder doctrine is that providing rewards induces finders not to misappropriate found items or leave them alone. Chapter 12 discusses the specificatio (mistaken improver) doctrine. About two-thirds jurisdictions have this doctrine, and the doctrinal structure is highly convergent. Most of these jurisdictions limit the application of the doctrine when the nonconsensual improvement is irreversible, and most assign sole ownership to either original material owners or improvers. Almost all jurisdictions adopt the disparity-of-value test and/or the transformation test, but there are eight ways that bad-faith improvers are treated. The disparity-of-value test, in and of itself, does not tend to assign ownership to higher valuers, however. While no ex ante rule-making can ensure allocative efficiency ex post, requiring both the disparity-of-value test and the transformation test is more likely to

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

20

introduction

increase efficiency. Lawmakers looking for a radical reform proposal may also adopt the internal auction mechanism to resolve the problem in specificatio. Besides, even good-faith improvers should not be compensated, as the nontransformative, low-value-increasing improvements are unlikely to be what material owners want. A clear rule of no compensation also decreases litigation cost. Chapter 13 focuses on the accessio and confusio doctrines, traditionally sibling doctrines to the specificatio doctrine. The accessio doctrine includes three types of combinations: immovables and immovables, movables and movables, and movables and immovables. Confusio ­concerns only mixture of movables. The big picture of these doctrines is that there is little sign of convergence, except perhaps in confusio. From an ­economic standpoint, it is quite clear when two things should be considered ­combined (thus the accessio doctrine applies) rather than separable: If (the value of attached thing) > (the value of the two post-separation things combined) + (the cost of separation), the two things should remain combined. The next question is who should own it. The key concern behind my analysis is still to deter opportunistic or careless interference with other’s property. The Conclusion recapitulates the lessons of the comparative, empirical, and economic analyses and points out future directions that enable us to even better test the grand property theories and understand property legal systems around the word.

https://doi.org/10.1017/9781009236553.001 Published online by Cambridge University Press

PART I Foundation

https://doi.org/10.1017/9781009236553.002 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.002 Published online by Cambridge University Press

1 Property Law around the World An Empirical Overview This chapter provides an empirical overview of property laws around the world. More concretely, all the property doctrines covered in this book – and beyond – are considered together in an unsupervised machinelearning algorithm called “hierarchical clustering.” The goal is to draw a legal family tree in a dendrogram that quantitatively summarizes degrees of similarity among the 136 studied jurisdictions. These jurisdictions are divided into ten legal families for expositional purposes. The study of legal families is what Zweigert and Kötz (1998: 46) describe as “international comparative law,” while legal scholars often engage in what Zweigert and Kötz (1998: 46) call “national comparative law” – that is, starting with a domestic law issue, studying how another country has dealt with the same issue, and then proposing reforms or new interpretations accordingly. In East Asia, at least, the compared countries are often those considered to fall within the same legal family, or those with a similar legal structure or legal substance – most notably, Germany. Thus, this chapter also provides several figures that show the correlation coefficients of property doctrines between several countries and all of the other ­studied countries.

I  Legal Families In all, my data set contains 156 jurisdictions. A total of 279 binary variables were included in the machine-learning analysis.1 All except North Korea that have fewer than 25 variables that take the value of 1 are excluded, as this suggests that either my sources are incomplete or the regulatory ­density in these countries is very low. (North Korea is included because it has a civil code; thus, the concern over incomplete sources is attenuated.) 1

Technically speaking, one included variable is a constant, as all the 136 jurisdictions left (but not 11 of the 20 excluded jurisdictions) all have the co-ownership form discussed in Chapters 7 and 8.

23

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

24

Twenty jurisdictions are thus excluded.2 If they were included in the hierarchical clustering analysis, they would form a supercluster themselves, as the missing information or lack of stipulation would be treated the same (coded as 0). This supercluster also affects the relations among other jurisdictions. Still, some jurisdictions that pass the arbitrary threshold and show up as similar in Figure 1.1 (Rwanda and Democratic Republic of Congo, for instance) have low regulatory density and their similarities should be interpreted with a grain of salt. In the following subsections, this chapter reverses the usual order of empirical works and discusses first, in Section A, the findings and their implications, while technical details regarding methods and their constraints are relegated to Section B. General readers may skip Section B.

A  Findings and Implications 1  Legal Family Tree Figure 1.1 shows the relative positions and percentages of disagreement of the 136 jurisdictions in a property-law family tree. The scale at the top shows the Gower distance, which is simply the percentage of disagreement. Among the 279 variables, if a pair of countries has the same values (both 0 or both 1) in 200 variables and different values in the remaining 79 variables, the percentage of agreement is 72% (=200/279) and the percentage of disagreement is 28% (=79/279). The Gower distance is thus 0.28. If two countries both copy, say, the Portugal Civil Code, their Gower distance is 0. Figure 1.1 quantitatively summarizes dissimilarities of the 136 studied jurisdictions, arranged vertically. At the bottom (the left side) of the dendrogram, each jurisdiction is considered its own cluster. The horizontal axis (the scale at the top) of the dendrogram indicates the Gower dissimilar coefficients. The “height” of the 135 nodes (or joining points), visualized as vertical line in Figure 1.1, thus represents the Gower dissimilar coefficients when two clusters merge. Figure 1.1 reports average-linkage clustering, under which the intercluster distance equals the average distance between all intercluster pairs of jurisdictions. As one moves up the tree (from left to right), the groupings of merged jurisdictions become 2

The 20 excluded jurisdictions are Benin, Bhutan, Bosnia and Herzegovina, Botswana, Brunei, Central African Republic, Liberia, Malawi, Maldives, Mali, Mauritania, Myanmar, Nepal, Oman, South Pacific countries (coded as one jurisdiction), Sri Lanka, Swaziland, Tanzania, Zambia, and Zimbabwe.

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

25

more dissimilar. Groupings continue until, at the far right of the dendrogram, all observations appear in a single supercluster. Figure 1.1 shows that the maximal dissimilarity between any dyad of jurisdictions in terms of property law is no more than 35%. My quantitative approach can produce any number of legal families. This chapter arbitrarily picks ten groups for expositional purposes. Ten, or any number of, groups can be identified by moving a vertical line from the right-hand side and moving leftward until ten horizontal lines appear to the right of the vertical line. Below, the ten groups are also labeled according to conventional wisdom to make it easier to refer to.3 The algorithm in no way knows or reveals that members in a certain group have been heavily influenced by, say, Germany or England. In Figure 1.1, from top to bottom, the first cluster is the English group (numerically labeled as Group 3 in the data released along with this book). The English group contains all the well-known common-law jurisdictions, plus Scotland and Israel, which are less similar with the core c­ ommon-law jurisdictions. The second cluster is China alone (Group 10). In my previous work (Chang et al. 2021), China’s 2020 Civil Code was found to be similar to Mongolia’s 1995 Civil Code. The similarity greatly attenuates after this book used the coding of Mongolia’s 2002 Civil Code. The following chapters demonstrate that China’s 2020 Civil Code is often idiosyncratic in content (though German and Taiwanese in structure) and stingy in spelling out many key property doctrines. Russian (USSR) legal scholarship was influential in China before the 1980s, though many Chinese scholars would be surprised that China is still closer to the Socialist family that contains many former USSR republics than to the German family. As conjectured in the following text, the similar lack of many standard property doctrines may explain the resemblance of the Chinese property law to the Russian property law. The third cluster is the Socialist group (Group 1). Civil codes in several former USSR republics have been influenced by the Russian Civil Code (Burnham et al. 2009: 353–354). Notable members also include the four Scandinavian countries: Denmark, Norway, Finland, and Sweden. They are closer to one another than to other countries. Note that many jurisdictions are grouped together in this cluster also because many variables take the value of 0, as the property doctrines do not exist. Inevitably, the design of my codebook is more affected by Western European, East Asian, and 3

Table 1.1 lists the jurisdictions in two, three, ten, and twenty groups.

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

Figure 1.1  Legal family tree https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

28

American laws, with which I am more familiar. Those doctrines may not exist in this group perhaps because (1) there has not been such disputes (e.g., the Scandinavian countries), (2) the regulatory density of property law is low (e.g., North Korea), or (3) disputes arising under those doctrines may be dealt with in completely different ways. The fourth cluster, the biggest, is the French group (Group 2). France and its former colonies constitute a subcluster. Spain and many of the Spanish-speaking Latin American countries form another subcluster. The Code Napoleon was forced upon Spain in the early nineteenth century, but Spain broke away from the French code later on. The inspiration of property law in Latin American countries is complicated and beyond the scope of this chapter. However, it is worth noting that Spanish-American colonies achieved independence before Spain adopted a French-oriented code; thus, they received the French civil law through voluntary transplant (Berkowitz et al. 2003b; Oto-Peralías and Romero-Ávila 2014: 576). Here it is shown that Latin American countries are close to one another, and closer to Spain, than to France. Besides, several jurisdictions from the Middle East and North Africa form another subcluster. These countries emulate French law without having any colonial history (Klerman et al. 2011), but do not buy the French property law wholesale. Notably, members of this group are the non-European part of the Ottoman Empire. Before its dissolution, the Ottoman Empire imported French commercial law and public law, though its private law is local and Muslim (Xu 2007: 303–304). This history partly explains its French inclination. Indeed, reading civil codes of, for example, Iraq and the United Arab Emirates, a private law scholar familiar with French law will certainly smell champagne and cheese. I call the fifth and sixth clusters “semi-French” (Group 9) and “quasiFrench” (Group 5), respectively, as they are part of the larger supercluster influenced by the Napoleonic Code. Portugal and its former colonies are here, as are a number of countries that have learned from non-French sources. Quebec is a famous mixed jurisdiction, also influenced by English law. Brazil has borrowed from French, German, and English laws (Pargendler 2012a: 810–812). The Netherlands is between (and beyond) the French and German systems. The seventh cluster has a sole member, another famous mixed jurisdiction – South Africa (Group 8), which has both a Roman-Dutch tradition and an English legal heritage. The final three clusters are all affected by German law. The Japanese cluster contains three other (South-)East Asian jurisdictions: Taiwan,

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

29

South Korea, and Cambodia. The affiliation of Japan is more complicated than commonly thought. The Japanese Civil Code of 1898, in its drafting stage, was heavily influenced by the French Civil Code. While Japanese private law scholarship later turned to German jurisprudence for inspiration – note that the German Civil Code did not go into effect until two years later, in 1900 – the Japanese Civil Code itself retains the French rules (Note 1906: 73; Belli 1959: 137). When South Korea’s civil code went into effect in 1960, it was indirectly affected by the German code through the Japanese code and Japanese scholarship (Kwon 2013: 114). The Taiwanese Civil Code, on the other hand, was enacted in 1930, when the Nationalist Party ruled mainland China. The Taiwanese Civil Code’s Book of Things is a hybrid of traditional Chinese law, Swiss law, and German law. Japanese scholars played an instrumental role in the drafting of the 1930 code (Chang et al. 2022: 23). When the Book of Things in the Taiwanese Civil Code was reformed between 2007 and 2010, the Japanese influence, partly due to the colonial experience in 1895–1945, is obvious (Chang 2016a: 228). The 2007 Cambodia Civil Code was drafted with assistance from Japanese scholars (Upham 2018: 109). Had I coded the 2020 Laos Civil Code, it would be likely to be grouped here as well, as Japanese scholars again played an instrumental role in the drafting stage of Laos’ code. The penultimate cluster is Quasi-German (Group 6). The members are the Eastern European neighbors of Germany. The Austria Civil Code predates the German Civil Code and has its own style. The final cluster is the German group (Group 4). Members contain several European neighbors of Germany, including Turkey. Turkey was the heart and soul of the Ottoman Empire, but in 1926, drafters of its new civil code under the Kemal administration adopted the Swiss Civil Code as their model (Xu 2007: 314). The big-picture, takeaway point is that common versus civil law divide is not the most salient dichotomy in property law. Chang et al. (2021), using 170 binary variables (most of which are included here), find that a French influence versus the lack thereof drives the classification. Here, the French supercluster is still salient. The (Quasi-/Semi-)German, Socialist, and English groups are parts of the large non-French supercluster in Chang et al. (2021). Here, with 279 binary variables, the German supercluster is deemed to be closer to the French supercluster. Still, standing on the other side of the German-French supercluster are civil-law Socialist countries spearheaded by Russia, Scandinavian countries, China, and English common-law countries. Civil law is plural, and common law is heterogeneous.

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

30

Legal families Socialist French English German Quasi-French Quasi-German Japanese South Africa Semi-French China Not coded

Figure 1.2  Legal families in world map

Figure 1.2 shows the ten groups in a world map. The three broadly speaking French clusters are in different shades of orange and yellow, whereas the three broadly speaking German clusters are in different shades of green (see also Table 1.1).

2  Distinctive Features What are the key variables that drive the legal families? A supervised machine-learning method called “sparse linear discriminant analysis” (Gaynanova et al. 2016) produces Figure 1.3. The rows are the key variables that contribute to the classification of legal families. Numbers in column heads represent the numeric group labels. The large positive numbers in the cells mean that variables taking the value of 1 contribute greatly to a certain grouping. The large negative numbers in the cells mean that when variables take the value of 1, it reduces the likelihood of being classified into a certain grouping. The numeric scale on the right-hand side shows the minimal and maximal coefficients contained in the cell. Blue numbers are positive, whereas red numbers are negative. The larger the number, the darker the color. The absolute values in the cell do not mean anything; what matters is the relative size of the values. According to Figure 1.3, the English group (Group 3) is featured by the voidable rule and entrustment rule in the good-faith purchase doctrine (Chapter 10). China (Group 10) has many one-of-a-kind rules, and the algorithm picks out the co-ownership quorum rule for sales – China requires supermajority votes rather than unanimity. The French group (Group 2) and the Quasi-French group (Group 5) have the right of use, disallow right of use and right of habitation to be

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

31

Table 1.1  Classification of jurisdictions into legal families Jurisdiction name Afghanistan Albania Algeria Angola Argentina Armenia Australia Austria Azerbaijan Bahrain Belarus Belgium Benin Bhutan Bolivia Bosnia and Herzegovina Botswana Brazil Brunei Bulgaria Burkina Faso Burundi California Cambodia Cape Verde Central African Republic Chile China Colombia Comoros Costa Rica Croatia Cuba Cyprus

Dichotomy

Trichotomy

French & German French & German French & German French & German French & German English & Socialist English & Socialist French & German French & German French & German English & Socialist French & German N/A N/A French & German N/A

French 2 French 9 French 2 French 5 French 5 English & Socialist 1 English & Socialist 3 German 6 German 4 French 2 English & Socialist 1 French 2 N/A N/A N/A N/A French 5 N/A N/A

2b 9 2b 5a 5c 1a 3a 6c 4b 2b 1a 2a N/A N/A 5a N/A

N/A French & German N/A English & Socialist French & German French & German English & Socialist French & German French & German N/A

N/A N/A French 5 N/A N/A English & Socialist 1 French 2 French 2 English & Socialist 3 German 7 French 5 N/A N/A

N/A 5c N/A 1b 2a 2a 3a 7 5a N/A

French & German English & Socialist French & German French & German French & German French & German English & Socialist English & Socialist

French 2 English & Socialist 10 French 2 French 2 French 2 German 6 English & Socialist 1 English & Socialist 3

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

10 groups 20 groups

2a 10 2a 2a 2a 6a 1b 3a

32 Table 1.1  (cont.) Jurisdiction name Czech Democratic Republic of Congo Denmark Dominican Republic Ecuador Egypt El Salvador England and Wales Equatorial Guinea Eritrea Estonia Ethiopia Finland France Georgia Germany Greece Guatemala Guinea Guinea-Bissau Haiti Honduras Hong Kong Hungary India Indonesia Iran Iraq Ireland Israel Italy Ivory Coast Japan

Dichotomy

Trichotomy

French & German English & Socialist

German English & Socialist

6 1

6a 1b

English & Socialist French & German

English & Socialist French

1 2

1a 2a

French & German French & German French & German English & Socialist

French French French English & Socialist

2 2 2 3

2a 2b 2a 3a

French & German

French

2

2a

French & German French & German French & German English & Socialist French & German French & German French & German French & German French & German French & German French & German French & German French & German English & Socialist French & German English & Socialist French & German English & Socialist French & German English & Socialist English & Socialist French & German French & German French & German

French German French English & Socialist French German German German French French French French French English & Socialist German English & Socialist French English & Socialist French English & Socialist English & Socialist French French German

9 6 9 1 2 4 4 4 2 2 5 2 2 3 6 3 5 1 2 3 3 5 2 7

9 6a 9 1a 2a 4a 4a 4a 2a 2a 5a 2a 2a 3a 6b 3a 5b 1a 2b 3a 3c 5a 2a 7

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

10 groups 20 groups

property law around the world

33

Table 1.1  (cont.) Jurisdiction name

Dichotomy

Trichotomy

10 groups 20 groups

Jordan Kazakhstan Kuwait Kyrgyzstan Laos Latvia Liberia Libya Liechtenstein Lithuania Louisiana Luxembourg Macau Macedonia Madagascar Malawi Malaysia Maldives Mali Malta Mauritania Mauritius Mexico Moldova Monaco Mongolia Morocco Mozambique Myanmar Nepal Netherlands New York New Zealand Nicaragua Niger Nigeria North Korea Norway

French & German English & Socialist French & German English & Socialist English & Socialist French & German N/A French & German French & German English & Socialist French & German French & German French & German French & German French & German N/A English & Socialist N/A N/A French & German N/A French & German French & German French & German French & German French & German French & German French & German N/A N/A French & German English & Socialist English & Socialist French & German French & German English & Socialist English & Socialist English & Socialist

French English & Socialist French English & Socialist English & Socialist German N/A French German English & Socialist French French French German French N/A English & Socialist N/A N/A French N/A French French French French German French French N/A N/A French English & Socialist English & Socialist French French English & Socialist English & Socialist English & Socialist

2 1 2 1 1 6 N/A 2 4 1 2 2 5 6 2 N/A 3 N/A N/A 2 N/A 2 2 5 2 4 2 5 N/A N/A 5 3 3 2 2 3 1 1

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

2b 1a 2b 1a 1a 6c N/A 2b 4b 1a 2a 2a 5a 6a 2a N/A 3a N/A N/A 2a N/A 2a 2a 5d 2a 4a 2a 5a N/A N/A 5d 3a 3a 2a 2a 3a 1a 1a

34 Table 1.1  (cont.) Jurisdiction name Oman Ontario Pakistan Panama Paraguay Peru Philippines Poland Portugal Puerto Rico Qatar Quebec Romania Russia Rwanda Sao Tome and Principe Scotland Senegal Serbia Seychelles Singapore Slovakia Slovenia South Africa South Korea South Pacific Spain Sri Lanka Suriname Swaziland Sweden Switzerland Syria Taiwan Tajikistan Tanzania Thailand

Dichotomy

Trichotomy

N/A English & Socialist English & Socialist French & German French & German French & German French & German French & German French & German French & German French & German French & German French & German English & Socialist English & Socialist French & German

N/A N/A English & Socialist 3 English & Socialist 3 French 2 French 5 French 5 French 2 German 6 French 5 French 2 French 2 French 5 French 5 English & Socialist 1 English & Socialist 1 French 5

N/A 3a 3a 2a 5c 5c 2a 6a 5a 2a 2b 5a 5a 1a 1b 5a

English & Socialist French & German French & German French & German English & Socialist French & German French & German French & German French & German N/A French & German N/A French & German N/A English & Socialist French & German French & German French & German English & Socialist N/A French & German

English & Socialist French German French English & Socialist German German German German N/A French N/A French N/A English & Socialist German French German English & Socialist N/A German

3b 2a 6a 2a 3a 6b 4b 8 7 N/A 2a N/A 5b N/A 1a 4b 2b 7 1a N/A 6a

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

10 groups 20 groups

3 2 6 2 3 6 4 8 7 N/A 2 N/A 5 N/A 1 4 2 7 1 N/A 6

property law around the world

35

Table 1.1  (cont.) Jurisdiction name Timor-Leste Togo Tunisia Turkey Turkmenistan Uganda Ukraine United Arab Emirates Uruguay Uzbekistan Venezuela Vietnam Zambia Zimbabwe

Dichotomy

Trichotomy

French & German French & German French & German French & German French & German English & Socialist English & Socialist French & German

French French French German German English & Socialist English & Socialist French

French & German English & Socialist French & German English & Socialist N/A N/A

French 2 English & Socialist 1 French 2 English & Socialist 1 N/A N/A N/A N/A

Figure 1.3  Key factors driving the grouping

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

10 groups 20 groups 5 2 2 4 4 3 1 2

5a 2a 2b 4b 4a 3a 1a 2b 2a 1a 2a 1a N/A N/A

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

37

(e.g., how long the prescription period is) and stylistic dimensions (e.g., whether mortgage is included in the Book of Obligation or the Book of Property), while I decide to use only categorical difference on legal substance in the analysis here. (5) Regarding certain variables like statutory lien and the right of first refusal, I am concerned that relying on civil codes only would be highly incomplete and running the risk of coding nominal differences (e.g., some countries opt to regulate statutory liens in tax statutes); thus, these variables have largely been omitted in the analysis here. Unless otherwise noted, the machine-learning methods used in this chapter follow Chang et al. (2021). For brevity, the method details are not repeated here. How robust is our result? How can readers be assured that the reported legal family tree is not merely one of the million very different family trees? We address these concerns by running a million bootstrap iterations (Hennig 2007) (for results, see Table 1.2) to measure the cluster stability of each group in the reported family tree. The stability value shows the percentage of cluster agreement. The goal of the cluster analysis is to group the individual countries into clusters such that every jurisdiction in a cluster is more similar to other jurisdictions in the same cluster than it is to jurisdictions in other clusters. A critical issue when evaluating clusters is whether a given cluster is substantively real and not an artifact of the clustering algorithm. Clustering algorithms often produce clusters that represent the actual structure in the data, but sometimes the algorithms produce a bucket that represents a “miscellaneous” cluster of data points that have no real relationship to each other and do not fit into any other cluster. More specifically, first, I run the hierarchical clustering analysis described above and derive 10 legal families from this analysis (Figure 1.2). Second, we use the “clusterboot” function from the “fpc” R package. The “clusterboot” function uses bootstrap resampling to evaluate the stability of a given cluster from the original data set of 136 jurisdictions. It then applies the same kind of average-linkage cluster analysis with Gower distance to the new data set. Bootstrapping is a general statistical methodology that relies on random sampling with replacement of subsets of the original data and measures accuracy for sample quantities. That is, the bootstrapping method, in each iteration, creates a new data set with the same number of jurisdictions (here, 136) but each data set may contain multiple country A and country B, while it does not contain country C. Third, for every legal family in the original clustering, the “clusterboot” function finds the most similar legal family in the new clustering and records the Jaccard similarity coefficient, which is computed by J(A,B) =

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

38 Table 1.2  Stability values for 10 groups Group name

Numeric label

Stability value

English Chinese Socialist French Semi-French Quasi-French South African Japanese Quasi-German German

3 10 1 2 9 5 8 7 6 4

0.90 0.55 0.78 0.76 0.44 0.56 0.54 0.47 0.51 0.56

Notes: If a dichotomy of legal family is used, Groups 1, 3, and 10 are one supercluster, while other groups are the other supercluster. If a trichotomy of legal family is used, Groups 1, 3, and 10 again are one super-cluster (English and socialist); Groups 2, 5, and 9 are another (French); and Groups 4, 6, 7, and 8 constitute the third (German). See Table 1.1.

|A ∩ B|/|A ∪ B|. Assuming that Group 1 in our reported family tree has 10 countries, and the group that is most similar to Group 1 in the first bootstrap iteration (Group 1′) has 9 + 2 countries (meaning nine countries are the same, and two countries are new), the Jaccard similarity coefficient is 9/(9 + 1 + 2). If Group 1′ is 10 + 3, the coefficient is 10/(10 + 3 + 0). Fourth, the second and third steps are repeated one million times and we compute the average of the one million Jaccard similarity coefficients for each legal family. As a rule of thumb, clusters with a stability value of less than 0.6 should be considered unstable. Values between 0.6 and 0.8 (some would say 0.75) indicate that the cluster measures a stable pattern in the data, but which points should be clustered together is not highly certain. Clusters with stability values above about 0.80 (some would say 0.85) can be considered highly stable. As Table 1.2 shows, most groups are stable or close to stable, suggesting that our reported result is not a random outcome. Given our data, in which many clusters merge at a relatively high Gower distance, groups with few members tend to be less stable. If jurisdictions are divided into 20 groups, so that the within-group heterogeneity is lower, stability values increase. As shown in Table 1.3, 17 of the 20 groups have stability values above 0.6, and again smaller groups tend to have lower values.

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

39

Table 1.3  Stability values for 20 groups Group name English Chinese Socialist French Semi-French Quasi-French

South African Japanese Quasi-German German

Numeric label

Stability value

3a 3b 3c 10 1a 1b 2a 2b 9 5a 5b 5c 5d 8 7 6a 6b 6c 4a 4b

0.97 0.63 0.63 0.63 0.72 0.53 0.70 0.97 0.72 0.71 0.64 0.80 0.72 0.59 0.73 0.66 0.45 0.62 0.81 0.69

Machine-learning is a fast-developing field. The hierarchical clustering algorithm used here is cutting-edge and powerful. Yet, clearly it has limitations. For one, depending on whether the studied jurisdictions are sorted in alphabetical, or reverse alphabetical, orders (the latter is used here), the dendrogram would be slightly different. If a parameter is tuned, the results reported in Figure 1.3 would change as well. More than one machine-learning methods exist, Ho et al. (2023, forthcoming) use the same data set and run additional algorithms that summarize the similarity and relationship among the studied jurisdictions. All the algorithms provide the same big picture, but classifications of certain jurisdictions vary.

II  Correlation of Property Law The property data set enables me to summarize how similar a pair of countries (often called a dyad) is in multiple ways. The hierarchical clustering

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

40

0

2

%

4

6

Dyadic correlation coefficients of property law, 2018

0

.2

.4

.6

.8

1

Figure 1.4  Distribution of correlation coefficients

method uses the Gower distance, which is the percentage of disagreement. Another method is to compute the correlation coefficients of each pair of countries. More specifically, for example, I have coded the same 279 dimensions of property law in China and Germany. Using the 279 zeroor-one values, I calculate the correlation coefficient. A value of 1 indicates a perfectly positive correlation, which happens when a pair of country has the same property doctrines (at least those studied). The Gower distance of these two countries is 0. A value of −1 means a perfectly negative correlation, which happens when a pair of country has made opposite decisions in all property doctrines. The Gower distance of these two countries is 1. A value of 0 means the property law in the two countries is uncorrelated. As Figure 1.4 shows, property laws in almost all country dyads are positively correlated. In this part, I intentionally use “countries” instead of “jurisdictions” because the unit of analysis here is indeed nation-states. Many international data sets collect information at the nation-state level. The GDP and population data are prime examples. Outside of private law, many legal fields operate at the nation-state or federal level; hence, there is no subnational variation. In a prior work (Bradford et al. 2021), in order to compare property law with antitrust law, which is operated and coded at

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

41

the nation-state level, the nine subnational jurisdictions in my property data set were adjusted in the following way: New York was used as a proxy for the USA, England was used as a proxy for the UK, and Ontario was used as a proxy for Canada. California, Hong Kong, Louisiana, Macau, Quebec, and Scotland were dropped. In addition, following Farran (2013), South Pacific countries were coded as one jurisdiction in the property data set, but left out in the cross-country comparison. Hence, there were 156 − 9 − 1 + 3 = 149 countries in the property data set. For each of the 149 × 149 = 22,201 country dyads, correlation coefficients were computed. The analysis in this part follows Bradford et al. (2021) in creating the 22,201 dyads and computing the correlation coefficients. With the dyadic data, new research questions can be explored. Bradford et al. (2021), using the previous version of the property data (based on 170 binary variables), inquire whether legal origins predict legal substance in property law and antitrust law. In short, in terms of property law, a dyad with a shared legal origin is to a small extent more similar with each other than a dyad without a shared legal origin. By contrast, whether a dyad shares a legal origin is unrelated to its similarity in antitrust law. The results suggest that legal origins may be an important predictor of legal substance in well-established legal fields, but do little to explain substantive variation in more recent areas of law. Countries with shared legal origins are not more likely to have similar antitrust regimes than countries without shared legal origins. This is likely for several reasons. First, countries’ antitrust laws have been shaped through regulators’ and policy makers’ engagement in various international organizations and trans-governmental networks. Another important factor is that antitrust laws are largely a more recent phenomenon, with most countries adopting them after 1990 (Bradford and Chilton 2019). By that time, these countries had many models to emulate. The EU in particular offered an attractive template to emulate given the specific and detailed nature of EU antitrust laws, as well as their availability in many languages (Bradford et al. 2019b). The EU’s active push to export its antitrust laws through trade agreements, and extend regulatory cooperation and technical assistance for new antitrust regimes likely further explains why the EU’s influence prevails over that exerted by legal traditions. EU law also diffuses through its member states. Moreover, as the law and development literature often proclaim that formal property rights are necessary in facilitating economic growth (but cf. Upham 2018; Chang 2022c), Chang (2023 forthcoming-b) uses the dyadic data to explore whether the similarity in property doctrines

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

42

Property correlation coefficients 4th quartile: 0.37–0.73 3rd quartile: 0.29–0.37 2nd quartile: 0.20–0.29 1st quartile: 0.03–0.20 No information

Figure 1.5  Correlation coefficients with German property law

Property correlation coefficients 4th quartile: 0.48–0.84 3rd quartile: 0.34–0.48 2nd quartile: 0.23–0.34 1st quartile: 0.02–0.23 No information

Figure 1.6  Correlation coefficients with French property law

Property correlation coefficients 4th quartile: 0.33–0.82 3rd quartile: 0.25–0.33 2nd quartile: 0.19–0.25 1st quartile: 0.07–0.19 No information

Figure 1.7  Correlation coefficients with American property law (proxied by New York law)

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

Property correlation coefficients 4th quartile: 0.26–0.40 3rd quartile: 0.21–0.26 2nd quartile: 0.17–0.21 1st quartile: 0.04–0.17 No information

Figure 1.8  Correlation coefficients with Chinese property law

Property correlation coefficients 4th quartile: 0.44–0.64 3rd quartile: 0.35–0.44 2nd quartile: 0.25–0.35 1st quartile: 0.02–0.25 No information

Figure 1.9  Correlation coefficients with Brazilian property law

Property correlation coefficients 4th quartile: 0.39–0.54 3rd quartile: 0.34–0.39 2nd quartile: 0.28–0.34 1st quartile: -0.05–0.28 No information

Figure 1.10  Correlation coefficients with South African property law

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

43

44

Property correlation coefficients 4th quartile: 0.36–0.48 3rd quartile: 0.27–0.36 2nd quartile: 0.19–0.27 1st quartile: 0.05–0.19 No information

Figure 1.11  Correlation coefficients with Israeli property law

Property correlation coefficients 4th quartile: 0.36–0.75 3rd quartile: 0.29–0.36 2nd quartile: 0.22–0.29 1st quartile: 0.03–0.22 No information

Figure 1.12  Correlation coefficients with Ukrainian property law

Property correlation coefficients 4th quartile: 0.45–0.60 3rd quartile: 0.37–0.45 2nd quartile: 0.29–0.37 1st quartile: -0.05–0.29 No information

Figure 1.13  Correlation coefficients with Dutch property law

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

property law around the world

45

is correlated with similarity in the pattern of economic growth and finds that, no matter whether in developed countries or developing countries, no such a correlation exists. The dyadic data can be used in a less ambitious way. As said, comparative lawyers may seek guidance only from countries with similar laws. My works in Chinese have criticized this practice. Nonetheless, given the practice, and it is certainly not wrong to consider a country’s most comparable fellow country, Figure 1.5 (Germany), Figure 1.6 (France), Figure 1.7 (the U.S.A.), Figure 1.8 (China), Figure 1.9 (Brazil), Figure 1.10 (South Africa), Figure 1.11 (Israel), Figure 1.12 (Ukraine), and Figure 1.13 (the Netherlands) show how similar in property law the ­chosen countries are to other countries. The correlation coefficients of each chosen country with the other 148 countries are divided into four quartiles and shown in each world map.

III Conclusion Based on the substance of property law, this chapter provides a classification of legal families that is not entirely the same as the previous comparative law literature has suggested. This is partly because the prior comparative law works did not focus on legal substance alone (rather, some include, say, legal culture), and they certainly did not focus on property law alone. Still, the gap may also be attributed to the difficulty of comparing sophisticatedly across so many jurisdictions across so many legal fields (not to mention when the temporal dimension is added to the mix) without a methodical means of recording what the law is in every jurisdiction and an objective way to analyze the legal materials at hand. This chapter demonstrates a new approach to comparative law – empirical comparative law – that is gaining traction. While the machine-learning tools are still in an early, rapidly developing stage, they already show promises to shed new light on our understanding of laws around the world.

https://doi.org/10.1017/9781009236553.003 Published online by Cambridge University Press

2 Economic Framework

As long as the conflict between wealth maximization and other goals is acknowledged, economic analysis can never claim final decision-making authority with respect to normative issues. (Reimann 2006: 844)

This book takes the ex ante viewpoint toward lawmaking and statutory ­interpretation. The sole normative goal accounted for is efficiency (rather than welfare, justice, or equality). Part I explains the difference between ex ante and ex post viewpoints and argue for the former. Part II discusses the weak normative position of pursuing efficiency as one of the goals ­important in law, and explains how efficiency of a legal rule is assessed. As many f­ ollowing chapters draw on the property rule versus the liability rule ­framework, Part III lays out the foundation and argues that in the real world what really matters is the choice between Rule 1 and Rule 2. Part III provides the f­ramework to evaluate the merits of these two rules, to be used in the chapters to come.

I  Ex Ante versus Ex Post Viewpoints An ex ante viewpoint enquires into how a statute or a legal interpretation affects future behaviors (Posner 2011: 9; Kelly 2014: 1141), whereas an ex post viewpoint focuses on how it impacts specific persons only (Merrill and Smith 2012: 60–64). In the court context, an ex post viewpoint guides judges to consider how their judgments affect the plaintiff and the defendant in the case. By contrast, an ex ante viewpoint requires judges to think through the ripple effect of their judgments on others down the road. To be able to do so consistently, judges and other decision makers better have good grasps of behavioral theories. Neoclassical economics, institutional economics, behavioral economics, and other branches in social sciences offer several competing behavior theories. These behavioral theories are employed to predict future behaviors in response to law. 46

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

47

A parallel distinction can be made between ex ante efficiency and ex post efficiency. Ex ante efficiency considers broader social cost and benefit, including those incurred in the future, while ex post efficiency considers only social cost and benefit within a narrow setting. Policymakers, when enacting regulations in the face of a lot of uncertainties and without much lobbying, could be inclined to take the ex ante viewpoint and enact legal rules that induce people to behave more efficiently. Judges, due to the nature of litigation process (Easterbrook 1984: 10), are inclined to consider only ex post inefficiency (Chang 2020c). Economic analysis of law focuses mainly on ex ante efficiency, and pays attention to ex post efficiency only when ex ante efficiency will not be compromised (Brooks and Schwartz 2005: 392; Smith 2009b: 134). Chapters 9 and 10 offer examples. This book interprets property doctrines in civil codes in ways that maximize ex ante efficiency. To achieve ex ante efficiency, this book holds the position that the best way is to facilitate voluntary transactions (often linked with the property rule, as it affords entitlement holders the power to refuse to deal). Bad-faith parties may be higher valuers, but they should prove that in proposing deals, rather than taking unauthorized actions (such as improving others’ movables or possessing others’ land). The ex ante and ex post concepts can be linked with the concepts of law and meta-law (see Introduction; Smith 2021a). From an economic standpoint, law (i.e., baseline rules created statutorily or judicially) should aim to promote ex ante efficiency as one of the first-order values. Due to positive institution costs, baseline rules could be incomprehensive and leave loopholes to be exploited. Meta-law (i.e., equity, the good-faith principle) deals with opportunism or socially undesirable behaviors by taking an ex post viewpoint – focusing on punishing opportunists. Because the nature of meta-law is not general, a meta-law ruling does not change the baseline rules, thus not affecting future behaviors in an ex ante inefficient way. As elaborated below, economic analysis recognizes that there is more than one value to achieve in every legal decision-making. Efficiency can be one major goal in designing law, whereas other values can be achieved in implementing meta-law. Elaboration on this topic is, however, beyond the scope of this book.

II  Efficiency versus Welfare This book adopts an economic framework that aims to maximize overall economic efficiency; in property law, this is measured by the size of

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

48

social benefits derived from allocating resources in a certain way relative to the institution costs required to realize such an allocation. An economically more desirable allocation of resources assigns them to parties who value them more (the “higher valuer”) in terms of “economic value” (see below). Allocative efficiency (in the sense of Pareto optimality) is achieved when such an allocation cannot be further improved (Cooter and Ulen 2012: 14). When a new policy introduces improved resource allocation, it is described in terms of higher allocative efficiency (in the sense of Kaldor–Hicks superiority) (Posner 1983: 91–92; Epstein 1994: 31; Smith 2004b: 1787; Merrill 2009: 488; Posner 2011: 66; Fennell 2013b: 1494). Allocative efficiency, however, focuses only on allocative benefits, while an efficiency criterion should also consider costs. A legal system that always perfectly assigns resources to their highest valuers could be very expensive. Allocative efficiency is not worth pursuing if prohibitively high institution costs are incurred in the process (Demsetz 2008: 109–114; Fennell 2013b). This book thus avoids using the term allocative efficiency in favor of “increasing allocative benefits” (except in Chapter 10, to be consistent with the prior literature). The most efficient (in the Kaldor–Hicks sense) property regime is thus one that increases or maximizes net social benefit: that is, social allocative benefits minus institution costs. The normative position of this book is that efficiency in this sense is one (though not the only) important social goal for law. Section A elaborates on the concept of institution costs. Section B digs deeper into allocative benefits. Section C distinguishes efficiency from welfare, a broader concept. Section D points out that in practice the efficiency analysis is carried out by a social cost–benefit analysis.

A  Institution Costs In a social cost–benefit analysis for a property doctrine, the cost ledger ­consists of institution cost,1 which is a combination of information costs and transaction costs.2 Information cost is the cost for acquiring knowledge about institutional facts. Transaction cost is the cost of establishing, maintaining, and using property rights (Barzel 1997: 4; Allen 2000: 898; Zerbe 2001: 12; Smith 2008: 446) that are not part of information cost already. 1

2

Many people may think that “institutional costs” read better. I follow Steven Cheung (1998: 515), who invented this term, in using institution costs. North (1990: 3) defines institution as “the humanly devised constraints that shape human interaction.” This framework does not apply seamlessly to all legal fields. Tort law, for example, considers the life loss and body harm, which are neither information cost nor transaction cost.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

49

Third-party information cost is the key to understand property law (Merrill and Smith 2001b: 386; 2001a: 848; Smith 2003: 1114; 2012b: 1706).3 Third-party information cost, a subset of information cost, means the expense for third parties (who are not parties to property transactions) to ascertain the changes to legal relations brought about by property transactions and to understand who owns what property rights. Chapter 3 recounts how the numerus clausus principle reduces the third-party information costs. Chapters 4 and 5 discuss how immovable registration systems affect third-party information cost. Land boundaries, a classic information cost issue, are covered in Chapter 6. Chapters 10–12 are concerned with issues that follow on uncertainties over who owns certain movables. Transaction cost, as the name suggests, is featured prominently when bargaining breakdown is a real possibility. Chapters 7–8 deal with negotiations among co-owners and the legal schemes that facilitate deals and rescue co-owners from deadlocks. Chapter 9 discusses the legal servitude doctrine that decreases transaction costs by reducing bargaining power asymmetry between parties.

B  Allocative Benefits Three types of property values are relevant to this book. Economic value equals market value plus subjective value (Blume and Rubinfeld 1984: 619; Fennell 2004b: 963–965; Miceli and Segerson 2007: 20). Subjective value is also known as subjective premium (Merrill and Smith 2010: 249) or the consumer’s surplus (Krier and Serkin 2004: 866). Market value is objective (Dana and Merrill 2002: 169), and thus the same for everyone, whereas economic value is subjective. Allocative benefits are measured by economic value. If otherwise, because market value is the same no matter who is assigned a particular resource, allocative benefits would be the same under any property doctrine. The thorniest problem in economic analysis of property law is that economic value is subjective and not easily (if at all) ascertainable by third parties. For lawmakers or judges taking the ex ante viewpoint, it is hard to know what legal scheme will always lead to ex ante efficiency. Rules, however complicated, will not always assign resources to higher valuers. This book draws on several proxies and rules of thumb that theoretically assign resources to higher valuers. Chapter 10 limits protection to good-faith 3

Still, as Smith (2021b: 30) emphasizes, economic analysis of property law is not just about information cost; rather, it uses a cost–benefit model.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

50

purchasers who paid reasonable consideration. Chapter 5 challenges the wisdom of giving away adversely possessed land for free. Due to the elusiveness of economic value, economic analysis of property law is often forced to take an agnostic position regarding what constitutes the best allocation. Instead, property law focuses on reducing institution costs so that higher allocative benefits can be achieved via voluntary transactions. Chapter 9 observes that lawmakers have given landlocked owners a limited extent of legal passage right, as it is better to level the playing field than to dictate who wins. Chapter 3 proposes streamlining the limited property forms for the same reason – lawmakers could not possibly know what limited property forms are good for whom; rather, lawmakers can only make transactions more feasible and let parties figure out what works best. By contrast, with proper designs, in certain settings, ex post efficiency could be attained. Chapters 10 and 13 propose fractional ownership and internal auctions to achieve ex post efficiency (i.e., maximizing ex post allocative benefits). Again, under this approach, lawmakers do not decide who wins – existing laws do, sometimes protecting good-faith purchasers and sometimes protecting original owners; instead, carefully designed auction mechanisms reveal in each case which party has higher economic values. Finally, this book uses several terms as coterminous with increasing allocative benefits: increase social wealth, decrease social resource waste, or the like. They all deal with the social cost and benefit of property doctrines.

C  Other Welfare Concerns Economists distinguish efficiency from welfare (e.g., Kaplow and Shavell 2002). Efficiency is only one of the values that social decision makers pursue. Distribution of income is another, of many, first-order values worth pursuing. Welfare is an all-encompassing second-order value that trades off the first-order values to reach a holistic compromise. Again, it is beyond the scope of this book to propose a full-blown social decisionmaking theory.4 The weak position this book holds is only that efficiency is one important first-order value that social decision makers should consider. This book demonstrates the efficiency implications of the studied property doctrines. Policymakers should account for the efficiency outcomes of these property doctrines as well as for other values. 4

Kornhauser (2022) contains many insights regarding this topic. Chang (2021: 91–127) and Chang (2020b: 11–112) elaborate my position in Chinese.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

51

D  Assessing Efficiency by Cost–Benefit Analysis The economic analysis conducted in this book (and almost all other lawand-economic writings) is nothing but a social cost–benefit analysis. The adjective “social” is necessary because the conclusions reached are not based on concerns for only certain types of private parties. Rather, all parties, including future parties, are considered. A cost–benefit analysis entails researchers analyzing how a new doctrine would change institution costs and allocative benefits and weigh whether, overall, implementing the new doctrine is worthwhile. A standard economics text would introduce the difference between Pareto and Kaldor–Hicks efficiency. Both, in their standard set-up, investigate the “willingness to pay” of everyone involved. Under Kaldor–Hicks efficiency, for instance, if some citizens are willing to pay $200 to switch to a new law, while the other citizens are willing to pay $190 to remain at the status quo, it is more efficient to switch to a new law, since 200 – 190 > 0. The new law, however, is not Pareto efficient, because some citizens clearly oppose the new law. In carrying out economic analyses of property doctrines, it is obviously impossible to survey each citizen in every studied country and inquire into their willingness to pay. What’s more, such questions are subject to respondents’ manipulation as well. Instead, what one sees in law journals are researchers’ thorough analyses of the relevant costs and benefits of several candidate legal schemes albeit absent any person’s willingness to pay. The foundation of this practice is that the social costs and benefits identified in the analysis are aggregates of citizens’ private costs and benefits. Researchers usually have little idea as to the distribution of private costs and benefits. Nonetheless, because Kaldor–Hicks efficiency criterion only measures whether total social benefit surpasses total social cost, a lack of knowledge concerning the distribution constitutes no problem. The implicit assumption of the preceding analysis is that people do not prefer certain income redistribution. As said above, this book will not address welfare issues other than efficiency; thus, that certain legal schemes create more inequality is a genuine problem for decision makers, but one bracketed in this book. However, if people are willing to pay for greater income equality (i.e., willingness to pay reflects people’s preference for equality), while the social cost–benefit analysis does not consider the distributional effect, social cost–benefit analyses would not always reach the same conclusion as the Kaldor–Hicks survey. This is why Adler and Posner (2006: 6) and Sunstein (2018) justify cost–benefit analysis as (only) a pragmatic decision-making tool. Still, given the distinction between efficiency

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

52

and welfare, if willingness to pay is reflective of people’s preferences for equality, the Kaldor–Hicks survey reflects not efficiency, but welfare. The difficulty in measuring people’s subjective willingness to pay again shows why welfare analysis is difficult to get off the ground, and why using social cost–benefit analyses to measure efficiency is the only practical choice.

III  Property Rules and Liability Rules Calabresi and Melamed (1972) advance a unifying framework of private law and started a half-century debate on the relative merits of the “property rule” versus the “liability rule.”5 Kaplow and Shavell (1996: 715) define the property rule in the following way: “[t]he state guarantees property right assignments against infringement through the threatened use of its police powers.” According to Ayres and Balkin (1996: 704), “[a] liability rule gives at least one party an option to take an entitlement nonconsensually and pay the entitlement owner some exercise price.” The literature centers around the assignment of entitlement and the efficiency of six prototypical entitlement protection rules. In the classic setting, with a “pollutee” resident versus “polluter” factory, as shown in Table 2.1, Rules 1 and 3 (the property rules) assign the entitlement to one of the parties who can determine how to use the resource. The other party can change this use only through voluntary exchanges with the entitlement holder. Rules 2 and 4 (the “call-option liability rules”) allow nonholders to take the entitlement if they compensate the loss of the holders. Rules 5 and 6 (the “put-option liability rules”) allow the holders of the entitlement to force others to purchase the entitlement. Consider the pollution example: Under Rule 1, residents can enjoin factories from polluting; put differently, to continue operating and polluting, Table 2.1  The conventional typology of Rules 1–6 Entitlement protection rules Entitlement holder Residents (the “pollutee”) Factory (the “polluter”) 5

Property rule

Liability rule: call option

Liability rule: put option

Rule 1

Rule 2

Rule 6

Rule 3

Rule 4

Rule 5

Following Kaplow (1992), this book consciously distinguishes between rules and standards.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

53

factories must acquire the consent of residents. Under Rule 2, factories can pollute without residents’ consent, if residents are compensated (generally at the court-specified amount). Under Rule 3, as the entitlement is assigned to factories, it has the privilege (not a Hohfeldian claim/right) to pollute (Smith 2004b; 2004a). To shut down factories, residents must buy factories out. Under Rule 4, factories can pollute, but residents can stop the operation of factories but must reimburse factories for their losses. Under Rule 5, factories can choose to stop polluting and request that residents ­compensate them for their losses. Under Rule 6, residents are entitled to clean air, but can request that factories pollute and compensate them. Calabresi and Melamed (1972: 1105) coin the term “property rule” to refer to Rules 1 and 3 because “much of what is generally called private property can be viewed as an entitlement which is protected by a property rule.” Rules 2 and 4–6 are the liability rules.

A  Rule 1 Is More Efficient than Rule 2 When Transaction Costs Are Low Calabresi and Melamed famously argue that when transaction costs are low, property rules are more efficient than liability rules. Judge Posner (2011: 86–89) generally follows this stance, and Cooter and Ulen (2012: 100) accept this maxim with some refinements. Hylton (2006; 2011) argues that when defensive actions and expenses are considered, property rules are generally superior to liability rules if transaction costs are low. Lewinsohn-Zamir (2001) makes a case for property rules over liability rules based on behavioral and experimental results. Property scholars generally agree that property rules should be the norm in the area of property law (Krier and Schwab 1995; Epstein 1997; 1998; Smith 2002; 2004b). Kaplow and Shavell (1996: 719), Ayres and Talley (1995: 235), Ayres and Goldbart (2001: 20) counter that as long as compensation under liability rules is set at the “average” (or “expected”) value of the non-option holder (i.e., the property owner), liability rules are, “on average,” superior to property rules, even in the field of property law,6 when transaction costs 6

Note Kaplow and Shavell (1996: 723) also favor the “use of property rules for protection of possessory rights in things.” Bar-Gill and Persico (2019) disagree with Kaplow and Shavell (1996), arguing that Rule 1 is superior to Rule 2 only if bounded rationality, instead of full rationality, is assumed. My view is that the game theory analysis in Bar-Gill and Persico (2019) work because the games assume that known three persons interact in known two periods. In the real world, even if all parties are fully rational, they do not have all the

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

54

are low. Carol Rose (1997) points out that the shadow examples in these analyses are contract law and accident law, not property law. In property law, because everyone’s economic value is hard to appraise, its average is elusive, and no one knows whether the liability rule compensation is set correctly. As courts generally use market value as the benchmark to assess compensation, the court-adjudicated damages are systematically below the average value of the non-option holder. Moreover, in property law, Rule 1 harnesses private information, not Rule 2.7 In a world with low transaction costs, information costs can be high. Specifically, ascertaining economic values is costly for nonowners such as potential buyers and courts. Under Rule 1, potential buyers must bargain with owners. An owner’s acceptance of an offer reveals the private information that the asset’s economic value is lower than the offered price, and no deal suggests that the offered price is not high enough. Under Rule 2, the prerequisite for harnessing the call-option holder’s private information is to set the exercise price of the call option correctly, but the court’s information costs are high. Therefore, Rule 1 can harness private information without judicial intervention, while Rule 2 can harness private information only if the court harnesses owner’s private ­information first – this assumes away the key hurdle, if not begging the question. Even the simplest single-chooser, first-order liability rule (Ayres 2005) has information cost problems. If, in practice, the expected value of non-option holders cannot be systematically ascertained accurately, the claim that Rule 2 is superior to Rule 1 is cast into doubt. If the expected economic value can be appraised accurately, but only with high information costs, increasing allocative benefits may be more likely under Rule 2 – nevertheless, the added allocative benefits may not be worth the information cost.

B  The Problem of Asymmetric “Cathedral” While Table 2.1 may leave the impression that the six rules are symmetrical, they are not. First, as pointed out above, Rule 1 gives the pollutee Hohfeldian claims, whereas Rule 3 does not; rather, the polluter merely

7

necessary information of how many infringers will show up at what time period. Bar-Gill and Persico (2019) also implicitly adopt a zero-transaction-cost assumption, as their model does not endogenize the costs of infringement and paying ransom. Bar-Gill and Persico (2019) also do not account for the role of investment. Avraham (2004) reaches the opposite conclusion because he assumes that voluntary transactions between parties are not feasible (i.e., transaction costs are very high) and that courts can accurately determine the correct damages. Neither assumption is wholly realistic.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

55

acquires a Hohfeldian privilege. Rule 3, understood through this classic example, may be useful in nuisance law, which locates at the intersection of property and tort laws, but is rarely used elsewhere in property law. Acquisitive prescription (Chapter 5) is not an example of Rule 3 because adverse possessors are subject to Rule 1 eviction before the statute of limitation runs and are protected by Rule 1 afterwards. Rule 3 and the famously rare Rule 4 have fatal flaws in comparison with Rule 2: nonowners (the polluters) do not reveal their economic value in any way. Rule 2, requiring compensation from nonowners, ensures that nonowners value the resource at least as much as the court-adjudicated compensation. By contrast, under Rules 3–4, no mechanism reveals any information about nonowners’ economic value. Merrill (1985: 1151) points out that if transaction costs are low, any entitlement allocation and protection rule will, through post-litigation voluntary transactions, lead to better allocations. That is, whether the court adopts Rules 2, 3, or 4, eventually resources will be owned by the highest valuers. Nevertheless, transactions are never costless. If a rule can save parties additional transactions, it may be worth considering. Finally, Rule 2 has other efficient traits vis-à-vis Rules 3–4. The prospect of losing the entitlements without receiving any compensation decreases owners’ incentives to invest (thus reducing property value) and increases owners’ expenses on prevention. In addition, Rule 2 is more likely than Rules 3–4 to incentivize nonowners to try to strike a voluntary deal with owners, as the compensation paid under Rule 2 could be just slightly lower than the price paid through voluntary transactions, and the litigation costs may be higher than the combination of bargaining costs and the price difference. Voluntary exchanges increase allocative benefits without judicial intervention. Thus, Rule 2 once again gains an edge over Rules 3–4 by facilitating more voluntary transactions. Besides, the typical Rule 6 derived from the seemingly symmetrical structure in Table 2.1 differs from the Rule 6 as delineated by Ayres (hereinafter “Ayresian Rule 6”). A typical Rule 6 allows anyone to force a purchase on any other person. This is symmetrical with typical call-option Rule 2, because under Rule 2, anyone can force another person to sell. By contrast, under Ayresian Rule 6, “[a] court might allow a Resident to enjoin pollution, but also give the Resident the option of waiving his injunctive rights in return for damages from the Polluter” (Ayres 1998: 797).8 8

Later, Ayres (2005: 29) appears to soften a bit and describe this type of put-option rule as “most closely analogous to Rule 6.”

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

56

Hence, the Ayresian Rule 6 is not a typical put-option rule, but rather a choice between Rule 1 (injunction) and Rule 2 (compensation).9 Moreover, the Ayresian Rule 6 is implicitly premised on the condition that polluters have polluted.10 Residents cannot randomly designate a person as a polluter and collect damages – if residents could, the Ayresian Rule 6 would be closer to the typical Rule 6. Given that polluters have “taken” the entitlement of residents, that is, have polluted, residents’ asking for payment of market value more resembles requesting compensation from an infringer under Rule 2 than forcing a stranger to purchase certain entitlements. Put differently, the Ayresian Rule 6 is in personam, not in rem. Chapter 6 will re-visit the shortcoming of Ayresian Rule 6.

C  A Framework for Preferring Rule 2 over Rule 1 It should be clear by now that optional law in the field of property is mostly about the choice between Rule 1 and Rule 2 (or a mixture of both rules), with the former as the default rule. The preceding section emphasizes that when transaction costs are low, Rule 1 is justifiably used as the default. When transaction costs or information costs are sufficiently high, Rule 1 is not always more efficient than Rule 2. This section lays out several variables that should be considered when the court or the legislature considers moving from Rule 1 to Rule 2.

1  Transaction or Information Costs Are High The prerequisite for even considering the use of Rule 2 is high transaction and/or information costs. Several important subtypes of institution costs in property law – for example, bargaining costs and verification costs – are worth considering in more detail. First, when bargaining costs are high enough to impede many potentially efficient trades (due to, for example, an owner’s monopoly power or strategic behavior), it is more likely that Rule 2 will become efficient. If bargaining costs are the reason to adopt Rule 2, the property doctrine need not enable option holders to acquire the 9

10

Epstein (1998: 842) interprets the Ayresian Rule 6 as Rule 1 because the property rule protection gives the original property owner the choice between damages and injunctive relief. This Epsteinian view may not be the mainstream interpretation of Rule 1. Nonetheless, even in the mainstream view, it would be conceptually clearer just to label the Ayresian Rule 6 as an either-or rule, a pliability rule (see Chapter 9), or a combination of Rule 1 and Rule 2, and leave the term Rule 6 to label a real put-option rule (if it ever exists). As Epstein (1998: 845) points out, the Ayresian puts are only given to property owners against wrongdoers.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

57

entire entitlement, or an entitlement of the optimal scale. Chapter 9 shows that a limited Rule 2 is sufficient to reduce bargaining costs and facilitate further voluntary transactions. Another example would be necessity (Epstein 1997: 2105). In cases like Vincent v. Lake Erie Transportation Co.11 and Ploof v. Putnam,12 ship owners and dock owners may not have had ample time to strike a deal before the storm hit, and dock owners’ strong bargaining power would have made bargaining costs high anyway. Case law holds that ship owners can moor in a dock without the consent of docks owner, but must ­compensate dock owners for their loss. Ship owners, however, only acquire temporary entitlement (during the storm). That is, Rule 2 here is limited. Second, other things being equal, if the identity of owners and the content of the property rights can be verified at low costs, Rule 1 is preferred. Rule 2 gains an edge when the verification cost exceeds its expected benefit (Sterk 2008: 1304). This might happen when verification costs are high; the probability of infringement is low; and the likely harm to the property owner is low or reversible. For example, regarding building encroachment cases (Chapter 6), in places where a “metes-and-bound system” (Libecap and Lueck 2011b; Brady 2018) is used, the content of the property rights (land size and boundary) often has to be verified at a high cost. By contrast, in places like Taiwan, where anyone can check the boundary of any land parcel on her smart phone, verification costs are much lower. Other things being equal, there is a stronger reason to use Rule 1 in the latter case, and Rule 2 could be adopted in the former to reduce verification costs and as a safety valve (Smith 2009a: 2128).

2  Transferring Entitlements Likely Increase Allocative Benefits In adopting Rule 2, the legislature or the court should have strong evidence for presuming that original nonowners value entitlements more than original owners. For example, in the context of accessing landlocked land (Chapter 9), landlocked owners usually value passages to public roads more than their neighbors. If owners are clearly higher valuing, the law should stick with the default Rule 1. Granted, a multiple-order liability rule (Ayres and Balkin 1996) may lead to the same result, but the costs involved in the re-takings are wasteful (Smith 2004b: 1789). Chapters  10  and 13 11 12

109 Minn. 456, 124 N.W. 221 (Minn. 1910). 81 Vt. 471, 71 A. 188 (1908).

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

58

discuss the merits of an internal auction (i.e., a full-blown multiple-order liability rule) when re-taking is not an issue. Courts can also look for other evidence indicating that nonowners value the resources in question more than owners. For instance, if nonowners conduct good-faith ex ante bargaining with owners, reaching certain agreements, but an honest error makes nonowners an infringer ex post, courts may adopt Rule 2 over Rule 1 as meta-law. Acquisitive prescription (Chapter 5) and building encroachment (Chapter 6) are prime examples.

3  Ex Ante Investment Not Important When ex ante investment is critical to enhancing the market value of the property or right-holders’ subjective value, including developing information about the asset (Smith 2004b: 1777), Rule 1 should be preferred, as it better protects the premium above market value, which is the usual baseline for court-adjudicated compensation under Rule 2.13 However, not all resources are constantly invested and not all contain unveiled secrets. Land value can increase with proper investments, whereas the value of a Swatch cannot. Ceteris paribus, there is a stronger reason to adopt Rule 2 in the latter case. 4  Courts Make Fewer Errors Courts are not always perfect appraisers of property value, and when courts systematically under-assess property value (Chang 2011), Rule 2 does not perform well (Ayres and Goldbart 2001: 63). Movables, especially commodities, tend to have clear market value, and their owners generally do not attach high subjective value to them.14 By contrast, the value of residential houses is not always easily identifiable, particularly in sparsely populated regions, and their owners usually attach a higher subjective value. Other things being equal, Rule 2 tends to perform better regarding movables than immovables. 13

14

Bar-Gill and Bebchuk (2010: 380) and Fennell (2013b: 1507) discuss the negative ex ante effects of Rule 2. Bar-Gill and Persico (2016) argue that, given that investment cannot be transferred to any future possessor of assets, and that both the current owner and the potential taker can invest on the assets in question, “property rules can rarely induce optimal investment.” They note that their argument is especially relevant in intellectual property. In physical property law, however, it is difficult to think of a real-world example that fits their descriptions and conditions. Of course, owners often attach subjective value to unique movables, like Van Gogh paintings.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

59

Judicial assessment costs are often positively correlated with transaction costs (Krier and Schwab 1995: 453). When this is the case, high transaction costs do not warrant the use of Rule 2, as erroneous property compensation would lead to resource misallocation. Nevertheless, where property values can be easily assessed by hedonic regression models ex post by courts (Chang 2010; 2011; 2013), but asymmetric bargaining power and other factors might impede bargaining (i.e., when judicial assessment costs are low but transaction costs between parties are high), Rule 2 could be preferable. Another variable that should be taken into account is whether the errors made by courts in assessing property value are correlated (or, put differently, the errors are common in assessing the values of conflicting uses) (Brooks 2002: 311).15 Rule 2 has an advantage over Rule 1 when the errors are negatively correlated or uncorrelated (Brooks 2002: 294). When the uses by both parties are distinct, the values of the resource for both parties are less likely to be correlated, and thus the judicial adjudication costs under Rule 2 are lower than those under Rule 1 – Rule 2 could be favored.

5  When Rule 2 Is the Only Choice When the things in question are damaged or ruined, Rule 2 is the best the law can do (Rose 1997: 2181; Fennell 2007: 1434). Rule 1, usually leading to an injunction, is only useful when the property rights can return to the status quo ante, but all the king’s injunctions cannot put Humpty Dumpty together again. The German property jurisprudence usefully distinguishes disposals into de jure and de facto. De jure disposal occurs when the disposed property can be recovered – for example, a nonowner sells a watch to a third party. By contrast, de facto disposal occurs when it cannot – for instance, a nonowner burns another’s car without the owner’s consent, extinguishing the owner’s property rights on the car. Generally, de facto disposal can only be dealt with by Rule 2. Granted, if Rule 1 is considered a call option whose exercise price is extremely high (Ayres 2005), the de facto disposer can be mandated to pay high punitive damages. High exercise prices, however, may over-deter and induce potential de facto 15

In Brooks’ example, if one is to judge whether A or B is taller, but only the upper half of their bodies are visible, it would be difficult to evaluate the absolute height of A and B – the errors are correlated. Nonetheless, it is not difficult to see who is higher, if one knows A and B stand on the same ground, and A and B stand shoulder-to-shoulder. Under the property rule, the judge only must compare the relative height, while under the liability rule, because of the need to assess compensation, the judge must know the absolute height.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

60

disposers to be too cautious. Thus, high punitive damages are not always desirable, at least when the de facto disposal is unintentional. Parchomovsky and Stein (2009: 1839) propose “propertized compensation,” which is a “damage measure that sets compensation equal to the owner’s pre-trespass asking price,” to be employed primarily in intentional trespass. Their proposal would narrow the domain of “inevitable Rule 2,” but under Parchomovsky and Stein’s strict evidentiary requirement, market-value compensation (the ordinary Rule 2) is still often the only feasible choice.

D  Radical Market: Limited Applicability This book does not subscribe to the radical market proposal of Posner and Weyl (2017; 2018). In Chapters 10 and 13, I propose to introduce an internal auction between the two disputing parties. The idea originates in the economic literature (Myerson and Satterthwaite 1983; Cramton et  al. 1987), and was introduced into law more than two decades ago (Ayres and Talley 1995; Ayres and Balkin 1996). Posner and Weyl’s Harberger tax idea, under which property owners have to publicize what they are willing to accept, and anyone else can pay the self-revealed price and take over, is based on similar economic logic. My endorsement of the internal auction mechanism, however, is limited. In the good-faith purchase and accessio contexts, internal auction is an innovative tool to solve the conundrum of which party to protect. Lawmakers have either given up and adopted the co-ownership solution, or made different decisions as to which party to protect. A 50–50 split is a reasonable, if not already adopted, compromise. Internal auction is only one step forward: using auctions to end co-ownership and ensuring that the high valuer becomes the full owner. In Chapter 7, regarding co-ownership partition, I do not champion the internal auction because co-owners’ fractional shares inevitably vary considerably and the mathematical features that ensure efficiency are unlikely to be met. In other contexts, such as building encroachment (Chapter 6), I do not propose internal auction because to make internal auction an engine for efficiency the two parties must be co-owners with roughly equal fractional shares – but handing encroachers half of the ownership of the encroached land (with or without compensation) is an unlikely political and legal decision, and giving the encroached a 50% interest in the encroaching part of the building (rather than the entire building) would create difficulties in registration if the encroached wins in internal

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

economic framework

61

auctions. This book only endorses innovative mechanisms when their real-world implementation is feasible. The radical market envisioned by Posner and Weyl is an across-theboard mechanism that, in my view, will not work in practice.16 Chang (2012c; 2013) reports my empirical work on a Taiwanese regime similar to the Harberger tax and shows that property owners did not honestly reveal their economic values; rather, they make strategic reporting decisions. In Posner and Weyl’s radical market, reporting what one is willing to accept is an even more complicated strategic task. It does not follow that most people will give up and simply report honestly. The more likely scenario is that the rational and the rich (assisted by experts) will report strategically, and the reported values will often deviate from the true ones. The informationally disadvantaged will fail to adjust their reported values timely and fall prey to activist investors. Picture the radical market during a pandemic. History may remember 2020–2022 as the COVID era, during which this book was completed. Many people suffered, and yet stock markets were often bullish and housing prices went up, followed by inflation. When property values greatly fluctuate, it is difficult for the disadvantaged to update reports of what they are willing to accept in a timely manner, and they may be displaced by activist investors. Granted, the radical market may occasionally solve the hold-out problem, but at too high a price.17 16 17

Fennell (2005; 2016) proposes callable fee, which rides along similar economic consideration. Callable fee, however, is a voluntary arrangement, and thus more feasible. For other critiques on Posner and Weyl (2017; 2018), see Rallo (2019), Fennell (2019a), Wyman (2019), etc. The radical market forces people to reveal their honest economic value, but as Ben-Shahar and Bernstein (2000) point out, parties’ secrecy interest (the desire to keep information private) may need to be taken into account. This secrecy interest, of course, potentially works against my proposed partition approach and the internal auction mechanism as well, though the concerns that Ben-Shahar and Bernstein (2000) point out are largely inapplicable in the property law context used in this book.

https://doi.org/10.1017/9781009236553.004 Published online by Cambridge University Press

3 Limited Number of Limited Property Rights Less Is More No one, I am sure, by the light of nature ever understood an English mortgage of real estate. Lord Macnaghten, Samuel v. Jarrah Timber & Wood Paving Corp. Ltd. [1904] AC 323.

One of the most difficult problems in a comparative exercise like this is avoiding confusion given the various labels applied to similar concepts and divergent meanings applied to similar terms across jurisdictions. For instance, many civil-code jurisdictions with Indo-European languages other than English use the term “hypothec” to refer to the security interest that in modern English law is a “charge on land” (Erp and Akkermans 2012: 534) and in modern American law a mortgage under lien theory (Serkin 2013: 142). In this book using the term hypothec runs the risk of confusing English readers unfamiliar with the term. Using the term mortgage, on the other hand, may confuse readers whose jurisdictions adopt the title theory of mortgage – that is, under the old common law, borrowersmortgagors transfer ownership of land to the lenders-mortgagees (Sagaert 2011: 1046–1047; La Forest 2017b: 33-2). This old usage of the term mortgage appears to be fading in immovable property law. While some US states still employ title theory (Serkin 2013: 142), California §2920 and New York (77 N.Y. Jur. 2d Mortgages §1) employ lien theory. Since 2002, in England, legal charge has become the only legal form of mortgage for registered land (Erp and Akkermans 2012: 534; Sparkes 2019: 222–223). In (formerly) common law jurisdictions that adopt the (Torrens) registration-of-right system, including Australia (Edgeworth 2017: 673, 794), Hong Kong (Mau 2010: 78), Ireland (de Londras 2011: 314), and some Canadian provinces (e.g., Ontario) (Ziff 2014: 440; La Forest 2017b: 33-3), New Zealand (Williams 2011: 88), and Singapore (See et al. 2018: 217) (though registered charge and registered mortgage still differ), a registered mortgage takes effect as a charge. 62

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

63

Malaysia (Maidin and Ali 2014: 207) and Uganda (Bakibinga 2018: 99–100) recognize both conventional title theory of mortgage and charge, whereas India, Myanmar, Pakistan (TPA) §58 recognize both simple mortgage (charge) and English mortgage (title theory). Given this, throughout this book, I refer to the aforementioned security interest on immovables as mortgage, but my typology below focuses on the lien form of mortgage (i.e., charge or hypothec), because this chapter will not elaborate on security arrangements that involve transfers of ownership.1 To overcome the terminological confusion, this book classifies limited property rights by their functions and key characteristics, and gives the terminology little weight.

I  Comparative Overview This part is divided into three sections. Section A discusses various ways that a limited or full-blown, explicit or implicit numerus clausus principle is recognized. Section B summarizes the major types of limited property rights (ius in re aliena). Section C gives an overview of future interests.

A  The Numerus Clausus Principle The starting point of property law is arguably the choice between a numerus clausus system, under which only the legislature can invent new property forms, and a numerus apertus system, under which transacting parties can invent new property forms. The typology is as follows. Type A. Adopting the numerus apertus principle are Scandinavian countries – Finland (Kuusinen 2010: 314; but cf. Hollo 2019: 72), Sweden (Lilja 2010: 46), Denmark and Norway (Faerstad and Lilja 2010: 217), Iceland2 – as well as Nicaragua,3 Panama (Espinosa González 2015: 29), Quebec (Emerich 2018: 217–218) (innominate dismemberments have been confirmed by cases), South Africa (Mostert and Verstappen 2015: 354) and Spain (Pacanowska and Soto 2010: 430). In practice, evidence suggests that in at least some of these jurisdictions, the number of property 1

2 3

Therefore, this chapter does not discuss chattel mortgage (on movables), which requires transfer of ownership and is explicitly allowed in, for example, Australia (Chambers 2013: 143), Canada (Ziff 2014: 463), England (Frisby and Jones 2009: 63), Ireland (Woods 2011: 126), Liberia (LCT) §10.142, and Singapore (See et al. 2018: 256). I thank Víðir Smári Petersen for this observation. P. 24 of http://repositorio.uca.edu.ni/3320/1/UCANI4431.pdf.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

64

forms is quite strictly closed (Chang and Smith 2019: 798; Smith 2021b: 28).4 The Statement of Motivation in the 2020 new Puerto Rico Civil Code explicitly endorses the numerus apertus principle. Type B1. Statutory provisions clearly reserve to legislators the power to create new property forms: Argentina §1884 (Alterini 2018: 200), Bulgaria (OA) §55 (Stoimenov 2011: 366), Burundi §II:I:1, China §116, Czech §977 (Tichy 2010: 269), Estonia (TLPA) §5 (Kullerkupp 2009: 231), GuineaBissau §1306, Japan §175, Kazakhstan §195, Kyrgyzstan §228, Louisiana §476, Macau §1230, Macedonia (AOORR) §4, the Netherlands §III:81, Paraguay §1953, Peru §881, Poland §244 (Brzozowski 2005: 54; Habdas 2018: 52), Portugal §1306, Rwanda §II:1, Serbia (LBOPR) §6, Thailand §1298, Timor-Leste §1226, and Ukraine (Maydanyk 2017: 45).5 Type B2. Not only statutes but also custom can invent new property interests: South Korea §185 and Taiwan §757. In practice, however, very few new forms have been recognized through custom (Chang and Smith 2015; Chang et al. 2022). Type B3. The numerus clausus principle is explicitly recognized in immovable properties only: Cyprus (Neocleous 2000: 587), Israel (LL) §161 (Deutch 1995: 161), and Uruguay (Barua 2018: 30). The list of registrable immovable property rights in Turkey §1008 is interpreted by Adal (2012: 160) as exhaustive. Type B4. Statutes provide an exhaustive list of sanctioned property forms. England (LPA) §1 is the prime example, effectively introducing the principle (Erp 2006: 1049; cf. Frisby and Jones 2009: 10); Brazil §1225 follows suit.6 DR Congo (LF) §109, El Salvador §567, Iraq §68, Morocco (LRRP) §8, and Tunisia (CDR) §12 list several types of property rights, but it is unclear whether their lists are exhaustive. Other countries may 4

5

6

In Norway, strong norms of legal practice and a high degree of consensus about background norms may serve to constrain the nonlegislative creation of new property forms – an event that rarely occurs (cf. Ørebech 2013; Stenseth 2019). In Spain, any real property right can be registered. In practice, the set of forms is not completely open, because the Spanish registrar ends up standardizing the forms – the Dirección General de los Registros y del Notariado, an office within the registrar, decides whether a right has the nature of immovable property. Nonetheless, there are examples where the Spanish registrar accepted new forms of real rights that had not had formal legal recognition. I thank Vanessa Casado Perez for this point. In South Africa, recognition of new property rights is also dealt with as a question of registration (Akkermans 2008: 475). Armenia §166 and Uzbekistan §168 prescribe that only statutes can create property forms regarding ownership by state, communes, and legal persons. This type of rule is not counted among the numerus clausus principle. See page 1 of this article: www.publicadireito.com.br/artigos/?cod=ec62f93b5e03666f.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

65

not list types of property forms in a single provision, but have a chapter for each recognized type. Thus, other than Brazil and England, countries grouped here may not have an iron-clad numerus clausus principle on the books. Type C. The numerus clausus principle is not clearly prescribed, but in practice followed by court jurisprudence and/or accepted by academia: Australia (Edgeworth 2017: 36), Austria (Koziol et al. 2005: §308 Rn.1; Faber 2009: 16), Belgium (Cauffman and Sagaert 2011: 202),7 Bolivia,8 Canada (Ziff 2014: 57), Chile §577 (Alessandri and Somarriva 1974: 47), Colombia §665,9 Croatia (Josipović 2014: 96), Cuba,10 Dominican Republic,11 Ecuador (Holguín 2010: 41), France,12 Germany (Kohler 2005: 228), Greece (Klaoudatou 2011: 201), Honduras,13 Hong Kong (Goo and Lee 2019: 724), Hungary (Szilágyi 2011: 439), Ireland (Gardiner 2009: 166), Italy (Greco 2009: 353), Latvia (Klauberg and Kolomijceva 2010: 552), Liechtenstein,14 Lithuania (Mikelenas 2011: 340), Malta (Galea 2010: 466), Mexico (Espíndola Bustillos 2009: 168) (its inclusion is debatable), New Zealand (Escrow Holdings Forty-One Ltd v District Court at Auckland, [2016] NZSC 167), Romania (Stanescu 2017: 49), Scotland (Gretton et al. 2017: 16), Singapore (Tang and Low 2019: 63) (comparatively lax in applying the principle), Slovakia (Petkov 2010: 364), Slovenia (Škerl and Vlahek 2020: 31), Switzerland (Foëx and Marchand 2010: 167), Venezuela (Domínguez Guillén 2017: 129) (though some advocate numerus apertus), and states in the United States (Merrill and Smith 2000: 9). I am not able to identify every instance of law-in-action recognition of this principle; thus, the preceding list is surely incomplete. Figure 3.1 shows the preceding categorization graphically. 7

Note that the 2021 September new Belgium Civil Code §3.1 & §3.3 explicitly adopts the principle. I thank the drafter of the new book on property, Vincent Sargaert, for this pointer. 8 See p. 23 of this thesis: https://repositorio.umsa.bo/bitstream/handle/123456789/20530/ TD-3275.pdf. 9 See pp. 33 & 43 of this thesis: http://vitela.javerianacali.edu.co/bitstream/handle/11522/ 10573/Redisenando_tradicion_juridica.pdf?sequence=1&isAllowed=y. 10 www.ecured.cu/Derechos_reales_en_Cuba. 11 www.delcarpio.com.do/blog/naturaleza-del-fideicomiso-dominicano/index.html#_ ftnref13. 12 In France, the leading opinion is that it also has an implicit numerus clausus principle, though one that is not as strict and clear as in elsewhere (Akkermans 2008: 167–168; Steiner 2010: 385; Akkermans 2017: 112). But compare Aynès (2008: 149). 13 www.iberoreg.org/sistema-registral/honduras/. 14 http://de.wikipedia.org/wiki/Sachenrecht_(Liechtenstein).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

66

Numerus clausus Unclear A: Numerus apertus C:Followed in practice B3:Explicit only in immovables B4:Implied by an exhaustive list B2:Only statutes+customs B1:Only statutes Not coded

Figure 3.1  The numerus clausus principle

B  Limited Property Forms My codebook enumerates 25 limited property forms.15 Use rights include: superficies, usufruct, the right of use, the right of habitation, restricted personal easement, profit à prendre, emphyteusis, the right to agricultural use, life estate, leasehold, the right of inheritable possession for life on state land, the right of permanent use on state land, the right for enterprises, and real easement. Security rights include: mortgage/hypothec, line-ofcredit mortgage, anticresis, pledge of immovable things, dian, Reallast / rentcharge, Grundschuld/land charge, pledge of movable things, pledge over rights, floating charge, and the right of retention. Notably, several types of limited property rights are not included in this list. (1) Non-possessory security rights on movables, like fixed charges, are not explicitly coded, as these arose in the modern secured transaction context and are often stipulated outside of civil codes and thus hard to systematically code. (2) Security transfer of ownership is not included because it is often developed in practice and hard to code accurately (Drobnig 2011a: 1030–1038). (3) Reservation (retention) of ownership (Frisby and Jones 2009: 136) is not included because of the various meanings (Erp and Akkermans 2012: 519) and because I conceptualize reservation of ownership as a modification of the traditio rule (Chapter 6), not a limited property right. (4) Statutory limited property rights (except the right of retention), such as lien, are not included. Needless to say, it is sometimes difficult to determine whether form A is a subtype of form B, or they are different forms. For instance, are ordinary 15

Future interests such as reversion and remainder in the estate system are excluded from tallying here. Ownership and co-ownership are not counted as limited property forms.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

67

mortgages and line-of-credit mortgages two separate forms of security interests? In most countries, the ordinary version and the line-of-credit version share many provisions. This chapter treats line-of-credit mortgages as a distinct security interest, as it is more flexible and less accessory to secured claims than the vintage mortgage. Use rights and security rights are introduced in turn below.

1  Use Rights on Immovables Use rights can be further divided into personal easement (alternatively, personal servitude or personal real burden) and real easement (alternatively, easement appurtenant or praedial servitude) (Gordley 2006: 82). Under the former, a parcel of land or a building is encumbered for the benefit of a person, whereas under the latter, a parcel of land or a building is encumbered for the benefit of a property right-holder of another immovable. Latvia §1130 defines servitudes to include real and personal servitudes, and the latter contains usufruct and right of habitation – consistent with this chapter’s classification. Real easement benefits dominant land and encumbers servient land. Right and duty holders are defined by their being owners of dominant and servient land, respectively. Jurisdictions with real easements adopt what is qualitatively the same definition with sometimes different labels. Thus, below, this chapter focuses on the more complicated personal easement (see classification in Figure 3.2). a. Scope of Personal Easement  Personal easement encumbers servient land but benefits specific persons, who may not own any immovables. Personal easement is sometimes suggested to be coterminous with common law easement in gross, though this chapter contends otherwise, to avoid confusion. In Australia (Hepburn 2002: 119), an easement in gross can only be created for the Crown or a statutory authority. In Singapore (See et al. 2018: 189) and Canada (Ziff 2014: 381), there can be no easement in gross. In New York,16 easements in gross are recognized only as nonassignable and noninheritable licenses – that is, not as full-blown in rem property rights. That said, profit à prendre, recognized in these jurisdictions,17 is functionally similar to easement in gross (Hovenkamp et al. 2016: 335), 16 17

49 N.Y. Jur. 2d Easements §9; Rascha and Dolana (2021: §18:9). In the United States, the draft of Restatement of the Law, Fourth, Property §6.1.E.3 makes this distinction: “Easements do not give the dominant tenant a right to remove anything from the servient estate. The right to remove substances such as timber, minerals, or game from land owned by someone else is a profit.”

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

68

Figure 3.2  Structure of use rights for immovables

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

69

and it is no doubt a type of personal easement as defined above. Thus, this chapter understands personal easement as broader than common law easement in gross. The term personal easement or its equivalent has also been used in five different ways: (1) Type D1: an umbrella term. In for example, Austria §478, Croatia (Josipović 2014: 117), Czech §1265, Estonia (TLPA Part 4 Chapter 2), Latvia §1131, Louisiana §534, Macedonia (AOORR) §209, Puerto Rico §467, Slovenia (Škerl and Vlahek 2020: 90), and South Africa (Van der Merwe et al. 2002: 441), personal easement includes specific property forms such as usufruct. This chapter uses the term “personal easement” as an umbrella term. (2) Type D2. Azerbaijan §255.8, Georgia §253, and Turkmenistan §266 have personal easement that is essentially the right of habitation, as other purposes are not specified. These countries are coded as having only the right of habitation. (3) Type D3. In Argentina §2165, Equatorial Guinea §531, the Philippines §614, Slovakia §151n, Spain §531, and Ukraine §401, personal easement is an independent property form, but recognition of personal easements is only in passing, with almost all stipulations in the civil codes devoted to real easements. (4) Type D4. Germany §1090, Greece §1188, Liechtenstein §253, Poland §296, and Switzerland §781 adopt an independent property form – following German law – called “restricted personal easement” (Gordley 2006: 93). Restricted personal easement is similar to the right of use (introduced below): Both forms’ mandatory or default rule is prohibition of transfer. In addition, the scope of restrict personal easement is by default determined by right-holders’ personal and familial needs,18 whereas such a scope is a default rule in 44 jurisdictions and a mandatory rule in 31, regarding the right of use. The only distinction between the two forms is that restricted personal easement is silent on whether right-holders can take natural yields, while in almost all countries with the right of use, the right to take natural yields is either explicitly recognized or implied from their application of usufruct rules mutatis mutandis. That said, the two property forms are hardly distinguishable. (5) Type D5. Personal easement in Scotland (Gretton et al. 2017: 211) is very different from other restricted personal easements and most of its 18

More specifically, the scopes are: in Germany §1091 and Greece §1189, personal need; in Switzerland and Liechtenstein, normal need; and in Poland, personal needs in consideration of social cooperation and local customs.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

70

subtypes can only be held by public authorities. Israel (LL) §92 allows “specific use” under personal easement but does not impose other restrictions. b. Classifying Use Rights  Ownership is a bundle of rights to possess, use, and enjoy things. Use rights contain a subset of these claims, powers, privileges, and immunities (collectively referred to as “Hohfeldian ablements”). Different use rights are characterized by distinct subsets of them. This chapter starts with usufruct and its three key characteristics.19 First, a usufructuary is entitled to use and enjoy things – for example, Switzerland §755 and Thailand §1417. Commonly, there are accompanying duties to preserve and manage well, if not to use immovables as is; for example, Bahrain §916, Egypt §988, and Louisiana §537. Second, a usufructuary is entitled to reap natural fruits and collect other proceeds. Third, a usufruct by default ends at the death of the original usufructuary.20 As this chapter categorizes limited property rights functionally, a use right with the three key characteristics but not the name usufruct was still coded as a usufruct; for example, Afghanistan §2299 right of benefit and Germany §1030 Nießbrauch.21 Usufruct and the right of use are differentiated in France (§578 and §625) and 77 other jurisdictions. The main differences are as follows: all but Argentina explicitly have default22 or mandatory23 rule that limits holders of the right of use to collect natural fruits for the needs of themselves and their families only24; holders in general cannot lease out the land, transfer their right of use, or pass the right to heirs25; for example, Mexico §1049 and §1051. 19 20

21

22 23 24

25

Of the 101 jurisdictions with usufruct, 89 explicitly or implicitly allow usufruct on movables. This chapter focuses on usufruct on immovables, though. United Arab Emirates is the only civil-code jurisdiction that remains silent on whether a usufruct extinguishes when the usufructuary dies and §1344 sets a default five-year effective period. Some countries like Kazakhstan (§195 stipulating a “right of land use”) do not spell out any detail of the right and the right does not bear the name usufruct. Some countries like Mali (ELC) §88 and Serbia (LBOPR) refer to usufruct, but do not spell out any detail. Neither types are classified as having usufruct in the following statistical analysis. Egypt §996, Peru §1028, Malta §392, and 37 other jurisdictions. Quebec §1172, Romania §749, Timor-Leste §1407, and 32 other jurisdictions. Some are mandatory rules and some are default rules, but it is unclear whether the scope can only be narrower, or parties may establish a right of use whose scope is beyond personal and familial needs. Algeria §855, Egypt §966, United Arab Emirates §1350 read like the scope can only be narrower, while Austria §504, Cosa Rica §367, Estonia (TLPA) §225, Haiti §514, Peru §1028, Tunisia (CDR) §§161–162, Uruguay §§544–545 reads like the scope can be broader than personal and familial need. Of the 78 jurisdictions, 13 allow leasing, 11 transfers, and 5 inheritance.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

71

The most narrowly tailored form of personal easement is the right of habitation (alternatively, the right of dwelling or the right of residence),26 a holder of which can use things for residential purposes only. Like the right of use, in 83 of the 92 jurisdictions that recognize the right of habitation, its scope depends on the need of right holders or their family (a default rule in 32 jurisdictions and a mandatory one in 51). In addition, holders of the right of habitation are prohibited from leasing out, transferring, or passing the right to heirs27; for example, Mexico §§1050–1051. The common law profit à prendre offers an interesting comparison to usufruct and the right of use. Profit à prendre enables its holders to collect natural fruits, one core feature of usufruct, but does not enable them to otherwise use and enjoy land. Unlike the right of use, holders of profit à prendre are not, even by default, limited to reaping natural fruits for personal or family needs. If a usufructuary wishes to cultivate land for agricultural purposes, is she allowed to do so? Interestingly, few countries with usufruct explicitly stipulate this. Planting literally falls within the ambit of “use and enjoy,” but not every type of farming activity preserves fertility. Parties are allowed to specify what is allowed in establishing usufruct, but it is unclear whether every country would, in practice, allow most kinds of farming. My hunch is that ordinary farming is allowed. Countries that explicitly address this issue include Poland §§267–268 & §275, which gives a usufructuary limited rights to erect buildings and to plant. Paraguay §2242 and Turkey §804 imply that a usufructuary may plant. Quebec §1140 explicitly stipulates that a usufructuary may conduct agricultural or silvicultural operations if the land is suitable. A type of use right designed for long-term agricultural activities is emphyteusis.28 Among the 30 jurisdictions that recognize this form, the shortest maximal period is 25 years, whereas the longest is 100 years. Four countries cap the duration at 50 years and six at 99 years. Eleven jurisdictions set no cap on its duration, whereas three extinguish the right with the deaths of the right holders. Eleven jurisdictions have emphyteusis but no usufruct, while 19 recognize both emphyteusis and usufruct, suggesting 26 27 28

Ireland has a right of residence (de Londras 2011: 241), which does not contain the key characteristics laid out in the text. Of the 92 jurisdictions, 15 allow leasing, 9 transfers, and 6 inheritance. The chengbao right in China §332 is middle- to long-term, and renewable. The Chinese Communist Party policy, thus far, is to renew the chengbao right for free, indefinitely. The chengbao right is coded as functionally an emphyteusis. I thank Prof. WANG Yang for this insight.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

72

that usufruct is not always considered sufficient for regulating the relationship between parties with regard to farming. Emphyteusis as a property form appears to be in decline. Taiwan, Greece (Karibali-Tsiptsiou 2012: 60), Finland (Hollo 2019: 68), and Portugal, among others, have abolished this form. Taiwan §850-1 replaced it with a shorter term right to agricultural use (capped at 20 years). Libya §1013 also adopts the right to agricultural use, but prescribes no maximal duration. No other jurisdictions appear to recognize a short-term right to agricultural use. Superficies enables a right-holder to use land to construct and own the buildings, at least until the superficies expires.29 Superficies is especially useful when land and the buildings atop it are always subject to separate ownership – Japan §265 and Taiwan §832, for instance, have such a system. Other countries, such as China §344 & §362, where land is owned by the state or the collective and buildings can be privately owned, necessarily have to allow superficies, real easement, or the like to structure the relationship between private buildings and public land. Superficies is also useful in jurisdictions where in principle buildings are subsumed in land according to the accession principle (see Chapter 13). Superficies thus serve as an exception to the accession principle and allow ownership of land and ownership of constructions to be separated and owned by different parties (Gambaro and Mattei 2002: 303; Josipović 2014: 119; Pilati 2016: 79; Sagaert 2017: 216; Alterini 2018: 213). Notably, the common law does not contain a form like superficies that enables the separation of building ownership from landownership.30 The following countries’ superficies contain only the right to build and own buildings but not the right to cultivation: Armenia §207, Austria (BG), Azerbaijan §250, Bolivia §201, Bulgaria (OA) §63, China §344, Croatia (AOORR) §280, Czech §1240, Estonia (TLPA) §241, Georgia §233, Germany (EV), Greece (Karibali-Tsiptsiou 2012: 111), Iraq §1266, Italy §952, Liechtenstein (SR) §57, Macau §1417, Moldova §443, Mongolia §150, Peru §1030, Russia §271, Serbia (Živković 2007: 18), Slovenia (Škerl and Vlahek 2020: 108), Switzerland §675, Taiwan §832, Turkey §826, Turkmenistan §245, and Ukraine §413. 29

30

In Quebec §1011, though, superficies is considered a special mode of ownership – ownership of the constructions and plantations. In Estonia (TLPA) §241 and Georgia §234, buildings are an essential component of superficies. Germany’s Erbbaurecht (Kohler 2005: 243) and Afghanistan §2323 (the right of Heker) are counted as superficies. Russia §271 is considered superficies, because a building owner has a use right over another’s land. The wording of the provisions, though, makes it sound like a statutory, rather than consensual, limited property right.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

73

The following countries’ superficies contain both the right to build and own buildings, and the right to cultivation and ownership of the produce: Afghanistan §2323, Argentina §2114, Belgium (Sagaert 2017: 216), Benin §53, Brazil §1369, Cuba §218, Democratic Republic of Congo (LF) §126, France (Terré and Philippe 2006), Indonesia §711, Japan §265, Jordan §1225, Lithuania §4.160, the Netherlands §V:101, Portugal §1524, Qatar §1029, Quebec §1009, Romania §693, Rwanda §II:76, South Korea §279, Spain (Aristi and Imbernón 2013: 64), Suriname §757, Syria §994, Thailand §1410, Timor-Leste §1414, and United Arab Emirates §1353. c. General-Purpose Use Rights  The aforementioned use rights are largely absent in common law jurisdictions, which rely on leasehold, a general-purpose use right. By contrast, a lease is a type of contract in all civil-code jurisdictions but Indonesia §720 and the Netherlands §V:85 (see also Miceli et al. 2001). With proper covenants, a leasehold can be tailoredmade for constructing buildings, cultivation, or providing a residence. Life estate is another general-purpose “use right” prevalent in common law jurisdictions. In common law parlance, it is a present possessory estate for the duration of one or more human lives. As compared to other use rights, the interaction between a life tenant and a bare owner (who either keeps a reversion or acquires a remainder from someone else) is minimal. Still, the latter can sue the former for “waste” (Krier 2006: 92).31 Finally, the former Soviet countries adopt unique general-purpose use rights. Belarus §217, Russia §265 (on state land), and Uzbekistan §165 stipulate the right of inheritable possession for life on land. Russia §266 explicitly enables right-holders to construct and own buildings, whereas §267 prohibits transfers of this right. This looks oddly like a medieval feudal English fee. Second, Belarus §217, Russia §268 (on state land), and Uzbekistan §165 stipulate the right of permanent use. A right-holder can possess and use land, including for construction purposes. From the civil codes themselves, it is unclear how these two property forms32 differ or interact. Third, Russia §§294–300 also stipulates rights held by state enterprises only (Kenjebayeva 2016: 111; Maydanyk 2017: 193–199). Similar rules are adopted in Belarus §276, Kazakhstan §196, Kyrgyzstan §230, Tajikistan §308, and Uzbekistan §176. These types of rights deal with 31

32

Civil-code jurisdictions that recognize usufruct may effectively create a right that is equivalent to life estate pur autre vie in the common law; for example, Argentina §2142: “The usufructuary may transfer his right, but it is his own life, and not that of the acquirer, that determines the maximum limit of duration of the usufruct” (Romañach Jr. 2015a). Kyrgyzstan removed the relevant provisions (§§234–236) from its civil code in 1999.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

74

the transition from a communistic ownership regime to a private ownership regime (Burnham et al. 2009: 353–354). Laos (LL) §59 adopts similar property forms for enterprises. Finally, do all jurisdictions contain a use right for agricultural purposes? If we count usufruct, emphyteusis, superficies with the right to plant, and the general-purpose leasehold as authorizing farming one way or another, few countries fail to specify farming-related use rights. These include former USSR republics (Belarus, Russia, and Uzbekistan), whose right of permanent use may include agricultural uses. Armenia, Kazakhstan, Kyrgyzstan, North Korea, Serbia, Slovakia, and Tajikistan have few, if any, limited property forms regarding use. Taiwan recognizes the right to agricultural use.

2  Security Rights on Immovables Figures 3.3 and 3.4 provide typologies of security rights on immovables and movables, respectively. An important terminological definition first. Some countries use “pledge” as an umbrella term for security rights, to the extent that “mortgage” is only a subcategory under “pledge” (e.g., Russia, Tajikistan, Uzbekistan). In this book, “mortgage” and “pledge” are considered distinct property forms under the broader category of “security rights,” with the key distinction that mortgage does not give creditors the right to possess collateral,33 whereas pledge does – and in some countries even a duty to possess collateral.34 Both types could recognize either movables or immovables as collateral. Here, only mortgage on immovables, pledge on immovables, and pledge on movables are discussed. a. Mortgage/Charge/Hypothec  The distinctive features of a mortgage/ charge/hypothec on immovables are (1) creditors do not have the right to possess; (2) mortgagors do not transfer landownership to mortgagees; and (3) mortgage is accessory to secured claims. In terms of (1), the right to possess refers to debtors’ possession of land as a principle; thus, it does not encompass scenarios wherein mortgagees can take possession when mortgagors seriously violate their duties; for example, the Netherlands §III:267. Nor does it consider the right to take possession of land upon

33 34

Collateral is any tangible or intangible asset belonging to a debtor over which the debtor grants a security interest to its secured creditor (White and Summers 2007: 4). Iran §772 prescribes that collateral must be transferred to the possession of creditors, but the validity of a secured interest does not depend on creditors’ remaining in possession. Iran is coded as having only pledge on immovables but not mortgage.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

75

Figure 3.3  Security rights on immovables: typology Notes: Security rights marked with * are not discussed in detail in this chapter.

default; for example, England’s legal mortgage (Smith 2013b: 322–323). The outlier in this matter is El Salvador §2157 (which appears to stipulate that parties can explicitly contract for possessory mortgage). Mortgages can be further classified in a variety of ways. This c­ hapter offers one which references whether the foreclosure clause (pactum

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

76

Figure 3.4  Security rights on movables: typology

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

77

commisorium or pactum de lex comisoria) is allowed. A foreclosure clause means that two parties agree that should debtors fail to repay in time, creditors may acquire landownership in lieu of repayment without recourse to judicial processes. Perhaps concerned with the typically asymmetric bargaining power between creditors and debtors, lawmakers in 60 jurisdictions have rendered foreclosure clauses null and void: for example, Algeria §903, Brazil §1428, Ireland (de Londras 2011: 320) (banned in 2009), Liechtenstein (SR) §288, Lithuania §4.170 (title theory of mortgage banned), Macau §690, Moldova (LoP) §13, the Netherlands §III:235, New Zealand (Williams 2011: 92) (banned in 2007), Nicaragua §2789, Nigeria (Smith 2018: 212),35 Panama §1550, Paraguay §2368, Peru §1111, Puerto Rico §1780, Qatar §1080, Romania §2433, Slovakia §151J, Slovenia (LoPA) §132, South Africa (Van der Merwe et al. 2002: 472), Spain §1859 (Aristi and Imbernón 2013: 130), Suriname §1207, Switzerland §816, Tanzania (LA) §124, Thailand §711, Timor-Leste §628, Turkmenistan §316 (Butler 1999: 105), Uganda (Bakibinga 2018: 101), Uruguay §2338, and Venezuela §1878. By contrast, some civil-code jurisdictions explicitly allow a foreclosure clause if parties so agree; for example, France §2459 (but collateral cannot be debtors’ main residence), Macedonia (AOORR) §225, Mexico §2916, Taiwan §873-1 (registered foreclosure clause is opposable), and Vietnam §§303–305. Some countries allow a foreclosure agreement to be entered into only after debts have come due; for example, Afghanistan §1861, Bahrain §964, Egypt §1052, Ethiopia §3060, and the Philippines §2087. In addition, in Belarus §331 (Butler 2000: 173), Kyrgyz §336, Russia §350 (Butler 2002: 136), Tajikistan §379, Ukraine §591 (Butler 2011: 162), Uzbekistan §281 (Butler 2007: 89), a mortgagee can enter into a foreclosure agreement with a mortgagor after an “unconstituted” public sale (i.e., with only one bidder). In contrast, in Kazakhstan §319, a mortgagee has the right to foreclose at the appraised value after an “unconstituted” public sale (Butler 2008: 118). The use of a foreclosure clause in common law or mixed jurisdictions that do not prohibit such is more complex. In these jurisdictions, the power to foreclose accompanies the mortgage/charge relationship, but the question is under what conditions are creditors allowed to exercise it. In Australia, courts only allow creditors to exercise foreclosure after an unsuccessful public auction (Hepburn 2002: 143). Similarly, in Scotland, mortgagees can seek a decree of foreclosure only after unsuccessful sales, which are rare (Gretton et al. 2017: 329). New York courts rarely allow 35

Note that in addition to a charge on land, Nigeria still recognizes title-theory mortgage.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

78

“strict foreclosure” (78 N.Y. Jur. 2d Mortgages § 481). A foreclosure clause is allowed but infrequently used in Hong Kong (Mau 2010: 83). The foreclosure process is long and cumbersome in Ontario, and mortgagors can transform a foreclosure action into one of sale (Ziff 2014: 452–454). Quebec §2781 imposes several conditions for foreclosure. In Singapore, foreclosure in charge is prohibited, and foreclosure in mortgage is rarely granted by court (See et al. 2018: 218, 237). Malaysia, by contrast, appears to allow chargees to foreclose (Maidin and Ali 2014: 207). b. Land Charge  The main difference between mortgage and land charge is that mortgage is accessory to a debt claim,36 while a land charge (or Grundschuld in German) is not accessory to a debt claim (Dannemann and Schulze 2020: 2053) – put differently, land charge only has “contractual accessoriness” (Stöcker 2012: 75). An accessory right follows the status of the secured claim (Erp and Akkermans 2012: 539) – a mortgage cannot be established without a claim to be secured and cannot exist with the secured claim already repaid in full; when a claim is assigned, the mortgage runs with it. Countries found to have land charge include Denmark (Olsen-Ring and Ring 2015: 113), Estonia (TLPA) §325, Finland (Ralli and Weckström 2010: 7) (rarely used), Germany §1191, Hungary (Sandor 2017: 223) (called “independent lien”; abolished and then re-introduced), Israel (ADA) §5, Liechtenstein (SR) §319, Norway (Simón-Moreno 2012: 202), and Switzerland §842. Slovenia (Škerl and Vlahek 2020: 103) had land charge previously, but repealed it in 2013. Sweden is said to have only land charge but not mortgage (Jensen 2010: 36–37; Nasarre-Aznar 2012: 95; Schaeferdiek 2015: 1282).37 c. Line-of-Credit Mortgage  Line-of-credit mortgage ensures a claim, the amount of which is not crystallized until the maturity date. Before the maturity date, a creditor has promised to lend to a debtor up to a certain amount of money, but the debtor does not have to use the entire credit line at once. Line-of-credit mortgages are known by many different names. In many Latin American countries, such mortgage ensures the fulfillment of “crédito abierto” (open-end credit); for example, Costa Rica §414, Guatemala §858, Honduras §2110, and Nicaragua §3783. Francophone countries that allow mortgage to ensure the fulfillment of “crédit ouvert” 36 37

For different levels of accessoriness in European mortgage law, see Stöcker (2012: 71). Georgia §286 is an example of the weak version of accessoriness. But cf. von Bar (1999: 534) observes that in Sweden mortgage is strictly accessory to a secured claim.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

79

(open-end credit), such as Algeria §891, Madagascar (OLRS) §36, Mali (ELC) §112, Morocco (LRRP) §180, and Tunisia (CDR) §206, were similarly coded as having line-of-credit mortgage.38 Other jurisdictions explicitly recognize line-of-credit mortgage: Bahrain §952, Cambodia §867 (called revolving hypothec), Canada,39 China §420, Egypt §1040, Germany §1190, Hong Kong (CPO) §45(1)(b), Iraq §1293, Israel (allowed in practice), Japan §398-2, Kuwait §981, Latvia §1281, Libya §1043, Liechtenstein (SR) §266, Lithuania §4.182, Louisiana (Willenzik 2015: 657), New York (RPL) §281, Pakistan (TPA) §79, Qatar §1068, Quebec §2688, Romania §2371, Slovenia (Škerl and Vlahek 2020: 99), South Korea §357, Spain (Garrido 2010: 36), Switzerland §794 (Druey 2004: 103), and Taiwan §881-1. France §2422 adopted a new “rechargeable hypothec” (Akkermans 2008: 143; Erp and Akkermans 2010: 47; Sagaert 2011: 1053), which is classified as a lineof-credit mortgage. Some provisions are classified as mere variations on the vintage mortgage, but not line-of-credit mortgage. Brazil §1487 stipulates that a mortgage can secure conditional or future claim-debt, as long as the maximal amount of credit is established. Chile §2431 stipulates that such mortgagebacked debt can be limited to certain amounts (and in any case no more than double the amount of the principal debt). Slovakia §151b requires parties to spell out the maximum amount to be secured when the value of the secured claim is unclear. This type of rule, in itself, is not considered line-of-credit mortgage because it appears to simply limit the scope of interests, prejudgment interests, and damages secured by mortgage.40 Admittedly, it is sometimes difficult to verify whether a similar provision would, in practice, allow line-of-credit mortgage. d. Possessory Security Rights  Possessory security rights on immov­ ables include pledge, anticresis, and dian. Under all three, debtors transfer possession, but not ownership, of immovables to creditors. 51 jurisdictions enacted anticresis (or antichresis or usufructuary mortgage),41 under which a creditor collects natural and/or legal yields which are used to offset interest on debt: for example, Brazil §1507, Cambodia §836, Chile §2443, Ethiopia §3124, Japan §§356–358, and Uruguay §2356. In some 38 39 40 41

Burundi §II:II:152 does not use this term, but appears to recognize this type of mortgage. www.canada.ca/en/financial-consumer-agency/services/loans/loans-lines-credit.html. Finland’s general pledge (yleispantttaus) (Ralli and Weckström 2010: 61) is not considered a line-of-credit mortgage either. Czech (Ronovská et al. 2020: 171), Mexico (Romañach Jr. 2003: 395), Serbia (Živković 2007: 9), and Slovenia (Škerl and Vlahek 2020: 100) have abolished anticresis.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

80

countries, yields may also offset the principal, for example, Algeria §956, Argentina §2215, Bahrain §1009, Belgium §2081 & §2090, Bolivia §1429, Colombia §2466, Ecuador §2345, Egypt §1104, El Salvador §2189, Equatorial Guinea §1881, France §2389, Haiti §1852, Honduras §2164, Iraq §1340, Italy §1960, Jordan §1393 (but uses by right-holders have to be preapproved by landowners), Kuwait §1038, Latvia §1362, Libya §1108, Malta §1987, Mauritius §2130, Nicaragua §3927, Panama §1622, Peru §1093, the Philippines §2132, Puerto Rico §1781, Qatar §1139, Seychelles §2085, Spain §1881, Syria §1065, Timor-Leste §606, United Arab Emirates §1474, and Venezuela §1855. The same is true in Burkina Faso, Comoros, Dominican Republic, Ivory Coast, Luxembourg, Madagascar, Niger, and Togo §2085 as well as India, Myanmar, and Pakistan (TPA) §58.42 By contrast, four jurisdictions are classified as having pledge on immovables but not anticresis. Afghanistan §1807, Iran §772, and Nepal (LA) §20 do not explicitly allow a creditor to use and benefit from yields in lieu of receiving interests on the loan.43 Vietnam §316, on the other hand, clearly stipulates that right-holders of pledge on immovables are not entitled to benefit from the yields. Anticresis is distinguished from dian in their attitudes toward the foreclosure clause. If a foreclosure clause is allowed, parties can specify in advance that if debtors default, their creditors acquire ownership of the collateral. Of 51 countries with anticresis, 40 make the foreclosure clause explicitly void, and 5 imply this rule or adopt it by default; for example, Chile §2441, Colombia §2464, Malta §1990, the Philippines §2137, and Seychelles §2088. By contrast, France §2458, since 2006, allows creditors to petition courts to foreclose collaterals to remain with them, if collaterals are not the main residence of collateral owners. India, Myanmar, and Pakistan (TPA) §67 set the foreclosure clause as the default rule. By contrast, under dian, interests, but not principals, are presumed to perfectly cancel out the proceeds during the dian relationship, and the foreclosure clause is perhaps the whole point of dian. In a dian 42

43

In TPA §58, anticresis is called “usufructuary mortgage.” TPA §58 defines six types of mortgages. There is the “simple mortgage” that is known in the lien theory of mortgage in the American literature. There is the “English mortgage” that is the title t­ heory of ­mortgage. “Mortgage by deposit of title-deeds” is allowed. The backbone of the “­mortgage by conditional sale” resembles dian. Finally, “anomalous mortgage” is a catch-all type. Mongolia §§153–157 appear to stipulate common provisions for pledge for movables and hypothec for immovables, but there is room to recognize pledge of immovables. §157.1.1 and §157.5.1 are also unclear whether possessing creditors may take natural yields and set them off against interests and principal of debts.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

81

arrangement, a landowner sells ownership to a purchaser with an in rem buy-back option to be exercised within multiple decades – in the case of Taiwan §912, 30 years is the maximal period that a dian can last, but historically a dian could last more than a century. The option strike price is the original dian price (Taiwan §924) (Chang et al. 2022). Dian was widely used in Chinese history (Ellickson 2012; Zhang 2017) and still exists as a formal property form in Taiwan §911 and as a customary property form in China (Chang and Xu 2018) and Hong Kong.44 e. Rentcharge and Reallast  Mortgage secures claims that are irrelevant to land, whereas rentcharge (interchangeably, rent charge) and Reallast (real burden) secure claims that come from land (Erp and Akkermans 2012: 270). Germany §1105 defines Reallast in the following way: “A plot of land may be encumbered in such a way that recurring acts of performance are to be made from the plot of land to the person in whose favor the encumbrance is created.” Germany §1108 makes landowners by default personally liable with all their assets.45 This personal liability rule is not universal. While Austria §530 (Koziol et al. 2005: 438) and Croatia (AOORR) §246 & §253 have also adopted it, and Macedonia (AOORR) §236 & §238 hinting at doing it, Switzerland §782, Liechtenstein §254, and Turkey §839 (Adal 2012: 166) explicitly stipulate that landowners are not personally liable. Belarus §§557–558 imply this.46 Other countries that adopt Reallast, such as Azerbaijan §321 (called “Hypothecation debt”), Czech §1303, Denmark (von Bar 1999: 57), Finland (Setten 2015: 272), Latvia §1260, Norway (Hegdal 2015: 916), Serbia (Živković 2007: 23) (called “real encumbrances”), Slovenia (Škerl and Vlahek 2020: 116–117) (called “encumbrance”), Suriname §782 (§784 hinting at no personal liability), and Thailand §1429 do not explicitly stipulate personal liability. By contrast, rentcharge is “the obligation to pay a sum of money periodically, usually annually, secured as a charge on land that does not belong to the charge holder. Once registered, the person who has the benefit of the 44 45 46

For Hong Kong law, see Tang Che Tai and Others v Tang On Kwai [2007] HKEC 674; Yau Kwai and Others v Yau Kar Siu and Others [2004] HKEC 927. Erp and Akkermans (2012: 542–543) distinguish Grundschuld, Rentenschuld, and Reallast. Belarus §554 uses rentcharge to secure the payment of annuities for life or permanently after a payee transfers ownership of things to a payer. The payment could be monetary or other forms of maintenance. If a payor further transfers the encumbered thing to a third party, the former’s obligation under the annuity contract pass to the latter (§557).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

82

encumbrance has the right to enforce the instrument in the same manner as a mortgagee under a registered mortgage” (Williams 2011: 72).47 The key distinction is that while Reallast can secure either monetary payments or positive duties to perform, rentcharge secures only recurrent monetary payments. Moreover, owners of burdened land are not required to be personally liable.48 Australia (Hepburn 2002: 363), Canada (Ziff 2014: 462), England (Sparkes 2019: 80), New Zealand (LTA) §2 & §§100–114, and Singapore (LA) §2 have adopted rentcharge.

3  Security Rights on Movables a. Pledge on Movables  The most prevalent type of security right on movables is pledge on movables, which gives creditors the right or even the duty to take possession of collateral. Type E1. 56 jurisdictions are coded as imposing pledgees-creditors a duty to possess if (1) once pledgors maintain or re-gain possession of pledged things, the pledge relationship is suspended or extinguished; or (2) the pledge relationship is effective only if pledgees remain in possession of pledged things. See, for example, Egypt §1100, Estonia (TLPA) §286, Georgia §269, Libya §1104, Paraguay §2308, the Philippines §2110, Portugal §677, and Taiwan §897. Type E2. 58 jurisdictions are coded as giving creditors a right, but not a duty, to take possession of collateral if their laws do not explicitly impose a duty on creditors to possess continuously; for example, Brazil §1433, Chile §2384, China §§425–431, and Italy §2786 & §2792. Right-to-possess jurisdictions include those stipulating that transfer of possession is necessary for pledge relationship to take effect, but remain silent on the outcome of creditors’ losing possession. Iran §772 is a rare example in explicitly prescribing that it is not a necessary condition for the validity of pledge that creditors remain in possession. Possession rules in other jurisdictions are more complicated49: Type E3. Guatemala §885, Kazakhstan §303, and Mexico §2859 prescribe that creditors have the right to take possession by default, but two parties can agree that pledgors remain in possession. 47

48 49

UK (RA) made a number of reforms. The creation of rentcharges other than estate ­rentcharges was prohibited. Existing (non-estate) rentcharges would be extinguished in 2037 (Sparkes 2019: 80). Estate rentcharges are a money charge on the land to cover the cost of works such as the cost of repairing a private street (Sparkes 2019: 90). This chapter only discusses consensual rentcharge. Malta §506 appears to allow only judicial rentcharge. The interesting variations on the theme of the Russian rule are worth noting.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

83

Type E4. Belarus §319, Russia §338, Tajikistan §367, and Uzbekistan §269 have the opposite default rule: creditors have no right to possess by default, but two parties can agree that creditors gain possession. Type E5. In Azerbaijan §§300–305 and Armenia §§254–258, “safekeeping” and “pledge of property in pawnshop” require transfers of possession, while “stationary pledge” allows locked or marked pledged things to stay with pledgors. Lithuania §4.198, similarly, allows both safekeeping and stationary pledge without such labeling. Type E6. In Finland (von Bar 1999: 187, 278–279), creditors have the right to possess, and possession is required to be opposable to other creditors of pledgors. Type E7. Under Mongolia §157.1.4, if pledgors do a poor job of safekeeping, creditors may demand possession. Twenty-six jurisdictions have certain rules regarding registration of pledge. Bosnia and Herzegovina (FPL) §4 and Brazil §1432 requires registration of pledge. Other countries do not make registration constitutive across the board. For instance, the Netherlands §III:237, Mexico §2859, Moldova (LoP) §6, and Panama §1554-A require registration only for non-possessory pledge. Greece §1214, Israel (PsL) §4 (Unknown 1969), and Romania §2482 stipulates that registration of pledge agreements can substitute for transfers of possession. France §2337 & §2340, Kyrgyzstan §325, the Netherlands §III:237, Moldova (LoP) §4, Peru (DL) §6, and Slovenia (Škerl and Vlahek 2020: 237–243) recognize both possessory and non-possessory pledge, without limiting the latter to locked or marked pledged things. Panama §1554-A also recognizes both types, but the latter can be used only in livestock. For the possessory pledge, Moldova (LoP) §21, the Netherlands §III:258, and Slovenia (Škerl and Vlahek 2020: 239) impose a duty to possess, while others provide creditors with a right to possess. b. Pledge over Right  Pledge/charge over right (such as using company shares as collateral in secured transactions) is widely used: for example, Georgia §254, Guatemala §887, Spain (Pacanowska and Soto 2010: 461), and South Africa (Van der Merwe et al. 2002: 791). In Commonwealth countries, a personal obligation can be charged (Bridge et al. 2018: 401). The United States (UCC) §9 recognizes pledge over right with other names and other types of security interests (cf. White and Summers 2007: 68–69). Here, as elsewhere, this chapter adopts a functional definition to include them all under the umbrella term of pledge over rights. Strictly speaking, assuming possession means taking physical control (Chang

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

84

2015b) because monetary claims and company shares are intangible and cannot be possessed, pledge over right is an oxymoron; charge over right would be the proper nomenclature. Many countries, though, use the term pledge to cover security interests on claims and company shares. Notably, Scotland (Gretton et al. 2017: 315) and Iran §774 explicitly refuse to recognize charge over right. c. Fixed and Floating Charges  Charges are non-possessory security rights. Fixed charges are created over specific assets, while floating charges are created over a fluctuating group of assets, such as inventory or machines in a factory. I did not explicitly code for fixed charges Floating charges (no matter whether the security interest is granted by corporate entities only, or by natural persons as well) are adopted in the following jurisdictions: Belgium, France §2333*, and the Netherlands* (Erp and Akkermans 2012: 454–459); California, Louisiana, and New York (UCC) §9; Armenia §259, Australia (Edgeworth 2017: 684), Azerbaijan §306, Belarus §321, Bhutan (MIPA) §4, China §396, Colombia (Ley 1676) §3, Costa Rica (LGM) §2, Croatia* (Josipovic 2013: 241), Cyprus (Synodinou 2020: 197), Czech* (Ronovská et al. 2020: 181), England (Frisby and Jones 2009: 144), Estonia (Kullerkupp 2009: 327), Finland* (Kuusinen 2010: 382), Greece* (Klaoudatou 2011: 322), Hong Kong (CPO) §56A, Honduras (MCL) §3, Hungary (Szilágyi 2011: 658), India (CA), Ireland (Gardiner 2009: 265), Israel (CL) §325, Italy* (Greco 2009: 367), Japan* (FCA) §1, Kyrgyzstan (LP) §4, Latvia* §1303, Lithuania* (Mikelenas 2011: 402), Malaysia (Maidin and Ali 2014: 207), Malta (Galea 2010: 534) (general hypothec as a functional equivalent), Moldova (LoP) §27, New Zealand (Williams 2011: 184), Nigeria (Smith 2018: 248), Norway* (Faerstad and Lilja 2010: 291) (but Denmark does not recognize it), Ontario (Ziff 2014: 463), Pakistan, Quebec §2715, Romania (Stanescu 2017: 98), Russia* §339, Scotland (Miller et al. 2009: 455), Seychelles §2071, Singapore (See et al. 2018: 140), Slovenia* (Rudolf et al. 2009: 620), South Korea* (FPA) §3, South Pacific countries (Farran 2013: 71), Sweden (Lilja 2010: 163–167), Uganda (Bakibinga 2018: 108), and Vietnam §295.50 50

I thank Prof. Nguyen Ngoc Dien for the observation of Vietnam law. Switzerland does not recognize floating charge, but Foëx and Marchand (2010: 238) point out arrangements can be made to approximate floating charge. South Africa has general notarial bond, which is close to, but still differs from, floating charge (Locke 2008). Germany allows floating charges on agricultural tenants’ inventory (Drobnig 1974).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

85

Identification of floating charges has proved difficult.51 Countries marked with an asterisk in the previous paragraph are coded as not recognizing floating charge by Djankov et al. (2008). Albania, Algeria, Bulgaria, Georgia, Macedonia, Sri Lanka, Tunisia, and Ukraine are identified as recognizing floating charge by Djankov et al. (2008), who do not provide citations that can be checked. In the American context, security interests are created consensually, whereas liens typically arise pursuant to common law rules or statutes which create the liens (Picker 2009: 496).52 To avoid confusion, the nonconsensual, noncontractual lien discussed in this book is labeled a statutory lien, and the term security rights include both security interests and statutory liens.53 d. Possessory and Non-Possessory Statutory Lien  Statutory liens on movables can be divided into possessory and non-possessory statutory liens. The prime example of possessory statutory lien is the right of retention (ius retentionis), which, by operation of law, enables creditors of matured claims who possess things belonging to their debtors to withhold the things until claims are paid, or until corresponding collaterals are provided. The right of retention, understood as such, is alternatively labeled as a mechanic’s lien (See et al. 2018: 264), a possessory lien (Sheehan 2011: 342), or a special lien (California §3051). Note that the right of retention in some countries is also used to refer to the Roman law maxim exceptio non adimpleti contractus (Cauffman and Sagaert 2011: 287; Gretton et al. 2017: 293), which is a defense in a reciprocal contract to withhold one’s performance until the other party is ready to perform. This is related to, but can be distinguished from, the right of retention discussed here. Most, if not all, countries recognize non-possessory statutory liens, without which, say, tax authorities would not be authorized to impose liens on private property. It is beyond my capacity to systematically code non-possessory statutory liens, because many of these liens are likely prescribed in tax statutes, thus falling out of my ambit. Also, under the 51

52 53

Kazakhstan §301, Tajikistan §361, and Uzbekistan §272 allow things owned or created by pledgors in the future to be collaterals. The stipulations are not specific enough, so I count them as not having floating charges. Djankov et al. (2008) code Kazakhstan as not recognizing floating charge. California §2873 clearly stipulates that a lien can be created by contract, and §2877 lists mortgage, pledge, bottomry, or respondentia as types of liens. Statutory liens in this chapter includes both “judicial liens” and “statutory liens” under 11 U.S. Code §101 (Bankruptcy Code).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

86

umbrella term of non-possessory statutory liens, there are vastly different contexts under which such liens arise, and often with different labels, such as statutory mortgage and statutory pledge. Many countries affected by French law do include non-possessory, statutory liens (privilèges) in the civil codes (51 of them have a separate chapter on privilèges), but not all countries adopt such a terminology. Limited by space and due to the complexity of non-possessory statutory liens, this chapter does not go into detail. A more meaningful coding exercise would be to categorize the scenarios under which statutory liens would arise and to describe the ranks of liens vis-à-vis voluntary security interests.

C  Future Interests Only 13 jurisdictions recognize any general form of future interest, that is a non-possessory interest capable of becoming possessory (Krier 2006: 104). These are Australia, California, England, Hong Kong, Ireland, New York, New Zealand, Ontario, India and Pakistan (TPA) §14 (Singh and Kaur 2020), Singapore (Tang and Low 2019: 101), Tanzania (LA) §83, and Uganda (Bakibinga 2018: 22).54 The land tenure system in Scotland was abolished in 2004 (Gretton et al. 2017: 503). Hence, where civil-code and common law jurisdictions part ways is not in present interests, but in future interests. Granted, the two legal families conceptualize present interests differently, but, as shown above, sometimes the difference is merely a matter of style, not of structure or function (Chang and Smith 2016: 134–140). Still, civil-code jurisdictions do appear to take a generally negative attitude toward future interests. In France §896 & §898, for instance, future interests are explicitly prohibited, with only a few exceptions in succession law to preserve certain properties within families, not facilitate property transactions (§§1048–1049). Interestingly, though, 29 jurisdictions allow consecutive usufructs, so long as future usufructuaries exist at the time the first usufruct becomes effective, which allows parties to hold a future interest in using land. Adopters include, for example, Albania §236, Equatorial Guinea §469, Eritrea §1210, Guatemala §705, Iran §45, Libya §989, Louisiana §546, Macau §1375, Malta §331, Mexico §982, Moldova §395, the Netherlands §III:203, Panama §454, Peru §1022, the Philippines §564, Portugal §1441, 54

It is debatable whether Israel recognizes future interests. I thank Yael Lifshitz and Ronit Levine-Schnur for their observations. Besides, it appears that Malaysia no longer recognizes future interests.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

87

Puerto Rico §398, Qatar §1015, Quebec §1122, Spain §469, Suriname §791, Timor-Leste §1364, and Venezuela §584. Iran §§44–45 is idiosyncratic – though consecutive usufruct is allowed, and even a person not yet born can be a grantee, a usufruct by default terminates when the bare owner dies.55 By contrast, Argentina §2132, Chile §769, Colombia §828, Costa Rica §336, Ecuador §783, El Salvador §773, Honduras §749, Nicaragua §1481, and Uruguay §498 explicitly ban this.56 Many other civil-code jurisdictions, while not prohibiting future interests explicitly, do not recognize any such interests.

II  Economic Analysis This part very briefly reviews and updates the optimal standardization theory that provides an economic justification for the numerus clausus principle, and then empirically describes how the limited property forms currently adopted are highly standardized, with several widely used forms even bearing the same labels. The last two sections argue that limited property forms are better if single-purpose rather than multipurpose, and for the recognition of future interests.

A  Why Close the Number of Property Forms? 1  Optimal Standardization Since Merrill and Smith (2000), a standard account of the numerus clausus principle is that closing the number of property forms is a means to attain the goal of optimal standardization (but cf. Hansmann and Kraakman 2002). Optimal standardization theory explains that more property forms bring not just social benefits but also costs. Specifically, additional social benefits are derived from incremented increases in asset values due to improved use opportunities. This is the allocative benefit discussed in Chapter 2. The reduction in frustrations facing transacting parties is counted among the social benefits. Framed in a manner more consistent with the framework 55 56

Scotland has “improper liferent” that is not entirely the same, but a named third party will acquire ownership after a usufructuary dies (Gretton et al. 2017: 339). Guillermo Gapel Redcozub, a property professor from Argentina, observes that usufruct has not been widely used in Argentina, and its main use is for fathers to transfer the “nude” ownership to sons while keeping a usufruct during their lives – partly to save successionrelated expenses. The nude ownership that sons receive is a future interest, not a present interest. Here, again, even though consecutive usufruct is banned in Argentina, usufruct is still essentially used as a device for transferring future interests!

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

88

in Chapter 2, costs incurred due to ­frustration could be framed as a social cost of not allowing more property forms. There are also additional types of institution cost – system cost – imposed on third ­parties, including operating expenses of registries and costs due to the adjustment of taxation rules in response to idiosyncratic new p ­ roperty forms. The key insight of optimal standardization theory is that as the number of property forms increases, the marginal (i.e., additional) social benefit tends to not increase, and may decrease, whereas the marginal social cost tends to increase. For instance, if only two types of security interests are allowed, transacting parties must work around the restrictions to close a deal. It is difficult, however, for transacting parties to create an arrangement equivalent to mortgage without the legal infrastructure that sanctions mortgage – why should a good-faith transferee be bound by a preexisting mortgage if it is just a mere contract between the transferor and a bank? Because of the “essential role” (Hansmann and Kraakman 2000b; Rauterberg 2020) of property law, if the number of limited property forms is small, the frustration cost will be high and resources may not be allocated in value-maximizing ways due to inability to structure the best deal. Once there are, say, ten types of security interests allowed, creating more adds little for transacting parties. On the other hand, transacting parties may find it difficult to comprehend the implications of, say, five types of security interests encumbering a parcel (cf. Rose 2011; Smith 2011b).57 As more and more property forms are allowed, lawmakers will at some point find themselves in a position where the marginal cost of creating one more property form is higher than its marginal benefit, meaning that an optimal number of property forms have been created. The optimal number could vary by country, depending on the types of forms recognized and other institutional conditions, such as the type of registration system adopted. As Chang and Smith (2015) have elaborated and updated the optimal standardization theory elsewhere, this chapter will not repeat the detailed analysis of optimal standardization. It is, however, worth noting that a more comprehensive analysis must distinguish between a registrationof-right system and a recording-of-deed system, as changes in institution cost under these two registration systems differ greatly. Table 3.1 summarizes my arguments. The transaction costs in column (3) refer to costs associated with the registration and legal systems mentioned above, whereas the third-party information cost concerns the expenses incurred 57

Another way to control third-party information costs is through a registration-of-right system (see Chapter 6).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

89

Table 3.1  Changes in marginal (social institution) cost Types of registration (1)

Creation of new Type of institution property forms (2) cost (3)

Marginal cost curve (4)

Registration of rights 2 Registration of rights 3 Recording of deeds 4 Recording of deeds

NC or NA

3rd-party information cost Transaction cost

Flat at >0 or rising Rising

3rd-party information cost 3rd-party information cost

5

NC or NA

Rising exponentially Flat at =0 (but fixed costs greatly increase) Flat at >0 or rising

1

Recording of deeds

NC or NA NC NA

Transaction cost

Notes: NC = numerus clausus; NA = numerus apertus.

by potential land purchasers (e.g., conducting title chain searches and understanding the implications of the registered information). Under both registration systems, additional property forms increase legal ­system costs (Rows 2 and 5). Under the registration-of-right system, the registration system cost also increases, as registries adjust to accommodate new forms (Row 2). Marginal third-party information cost under a ­registration-of-right system is always larger than 0, if not ever increasing, as the more types allowed, the more difficult it is for potential purchasers to understand the encumbrances (Row 1). Marginal third-party information cost under a recording-of-deed system with the numerus clausus principle would rise exponentially. Consider a potential purchaser who has to check six previous deeds to ensure the purported seller has ownership and the types of encumbrances that may exist: the information cost in a world with ten property forms is more than twice than that in a world with five because purchasers have to figure out, for each deed, whether any of the ten forms has been established. The number of permutations is 106 versus 56 (Row 3). Marginal third-party information cost under a recording-of-deed ­system with the numerus apertus principle would be – perhaps counterintuitively – always zero, but fixed costs become very high. The marginal cost is zero because in a country with the numerus apertus principle, no

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

90

one knows how many property forms there are at any given time, and thus any new addition to the list of property forms is unnoticeable. The flip side of the zero marginal cost is very high fixed cost – all transactions regarding immovables with some history are very expensive, as a slightly different description of an interest in a deed would be considered a new form; all idiosyncrasies matter (Row 4). Being zero, third-party information cost under a recording-of-deed system with the numerus apertus principle does not affect the assessment of the optimal number of forms.

2 Externalization Optimal standardization theory itself does not inform us whether it would be best to adopt the numerus clausus or numerus apertus principle. Whether the numerus clausus principle would produce an optimal number of property forms depends on the legislature. If forced to choose between recognizing too many and too few forms, the legislature should over-shoot. If 30 or 40 property forms are recognized (think of the abundant present and future interests recognized in American property law), the market would converge on the most useful forms – life estate in trust being the most prevalent interest in the United States (Krier 2006: 90). In contrast, if only three or four property forms are enacted, transacting parties are barred from creating more useful forms. Advocates of the numerus apertus principle often draw an analogy from contract law – if freedom of contract is the basis of contract law, should not the freedom of property form be the basis of property law? The analogy breaks down if the social costs and private costs associated with creating new property forms diverge. Put differently, as contracts are in personam, new forms of contracts do not affect noncontracting parties, whereas property rights are in rem and bind third parties. If some of the costs of creating new property forms are externalized, too many property forms will be created. Hence, the key question to ask is whether a substantial portion of the cost is externalized. If not, the numerus apertus principle would be efficient. My arguments are summarized in Table 3.2.58 My starting point is that the social benefit of creating new property forms under the numerus apertus principle is hardly externalized and can be safely ignored.59 In addition, unlike Merrill and Smith (2000), I do 58 59

Limited by space, I cannot elaborate my arguments – see my book in Chinese (Chang 2021: 211–229). Even if there are external benefits, as long as internalized benefits are large enough, new property forms will still be created (Fennell 2019b: 64–83).

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

91

Table 3.2  Externalizing costs and benefits under the numerus apertus principle

1 2 3

Types of registration

Types of cost and benefit

Externalization of costs and benefits?

Registration of rights Registration of rights Registration of rights

Social benefit

Barely

3rd-party information cost Transaction cost

Barely

4 5

Recording of deeds Recording of deeds

6

Recording of deeds

Social benefit 3rd-party information cost Transaction cost

Externalization of registration system cost; externalization of legal system cost Barely No external cost No externalization of registration system cost; but externalization of legal system cost.

not think that third-party information cost (to avoid violating another’s property right or to verify property rights held by transacting opponents) is externalized to a great extent as long as proper notice is required. Regarding Row 5 in Table 3.2, it is important to note that creating a new property form does not impose marginal social cost, nor an externalized one (see also Row 4 in Table 3.1). The numerus apertus principle is inefficient for two reasons. First, as pointed out above, it greatly increases the fixed cost of property transactions. Second, an excessive number of property forms will be created due to the externalization of transaction costs. In Row 3 in Table 3.2, the registration system cost is externalized because the parties who create a new property form and first register it are unlikely to be asked to pay astronomical fees to cover the expenses incurred by registries to overhaul their computer systems and train staff regarding new forms. In Rows 3 and 6 in Table 3.2, legal system costs are externalized. Most countries levy property (holding) taxes and/or property transaction taxes. Free creation of new property forms provides ample room for creative lawyers to help clients avoid taxes. Lawmakers, of course, can respond by closing loopholes, but this just starts another round of tax-avoidance efforts. The social cost of this back and forth is just one aspect of the legal system that will be challenged by an open number of property forms.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

92

3  Highly Standardized Across countries, limited property forms are highly standardized. First, most countries have around ten forms (see Figure 3.5).60 While mortgage over movables and a few idiosyncratic limited property rights are not included in my coding, even if all are included, the total number of limited property forms will not increase by more than five in most, if not all, countries. This is based on my thorough reading of the civil codes. Of course, if many limited property forms have been enacted outside of the civil codes, the line-up of property forms would be longer. Second, certain property forms are core, with high adoption rates. Figure 3.6 shows that, most notably, real easements and mortgages are widely used. Several other forms dating to the times of Roman law are also still adopted.

Civil-code jurisdictions

0

10

Frequency

20

30

Not civil-code jurisdictions

0

5

10

15

0

5

10

15

Number of Limited Property Forms

Figure 3.5  Distribution of limited property forms Notes: N = 156. North Korea Civil Code adopts zero limited property forms. I cannot identify any limited property forms in Mauritania and Swaziland. 60

Of course, if a country adopts a loose numerus clausus principle, market participants may create new property forms that will not be reflected in the statutory sources I consulted.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights Not civil-code jurisdictions

Civil-code jurisdictions

Real easement

Real easement

Usufruct

Usufruct

Use

Use

Habitation

Habitation

Superficies

Superficies

Mortgage

Mortgage

Anticresis

Anticresis

Pledge (personal)

Pledge (personal)

Pledge (right)

Pledge (right)

Retention

Retention

Floating charge

Floating charge 0

20

40

60

93

80 100

0

20

40

60

80 100

% of adoption

Figure 3.6  Most frequently used limited property forms Notes: X-axis shows the % of jurisdictions that adopt a certain limited property form. “Mortgage” refers to mortgages of real properties. “Pledge (personal)” means pledges of personal properties, whereas “Pledge (right)” refers to pledges over rights.

B  Limited Property Rights as Lego Bricks and Modules Property forms have been described as having a Lego-like interface (Smith 2012b: 1708; Akkermans 2017: 110) – or, Lego on the outside, Play-Doh on the inside (Fennell 2011c: 56). While Merrill (2012a) points out that the Lego metaphor works less well between neighbors and between property owners and the state, it is nonetheless helpful in further thinking through limited property forms. Consistent with this metaphor, below I describe a three-level structure: on top, ownership as the full bundle (Chang 2015b); in the middle, limited property rights as “modules” (Smith 2006; 2011a); on the bottom, Hohfeldian ablements as Lego bricks. At the Lego brick level are four types of Hohfeldian jural relationships embodying different substances. The foreclosure clause in anticresis and dian is a conditional Hohfeldian power to transfer ownership. Many civil codes stipulate that ownership of a thing entails the right to use, enjoy, possess, and transfer it. The rights to use, enjoy, or possess are a “jural composite” of a number of in rem Hohfeldian ablements. More

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

94

specifically, a right to use includes a privilege to utilize the space, a claim to fend off intruders, and immunity from interruption when bare ownership changes hand. As such, the rights to use, enjoy, and possess are better conceptualized as situated in the mezzanine level between the bottom and middle level (making it a four-level structure, if you will), while the right to transfer is a simple Hohfeldian power. Limited property rights as modules combine certain Hohfeldian ablements. Hypothec does not contain the ablement combinations known as the right to possess, while all personal easements come with the right to possess to a certain extent. Limited property rights are not Lego bricks because sometimes they do not fit together – a landowner who has established a superficies for another to build a high rise is unable to establish a real easement for neighbors to enjoy ample sunshine. All the Hohfeldian ablements add up to ownership, but an owner never loses all of the ablements when decomposing (démembrement) ownership into a number of limited property rights. What is called owners’ residual authority (Merrill 2012b: 2067) are those Lego bricks that bare owners keep, most notably, the power to transfer bare ownership. Within the modular interface, the relationship between bare owners and limited property right-holders is customized, even personalized. Under real easement, a wide variety of servitudes, from negative to positive, can be tailored-made. While the right of use and the right of habitation limit their scope to the needs of right-holders and their families, right-holders apparently have different needs (explicitly recognized in, e.g., Austria §505, Bolivia §250, and Brazil §1412). Thus, the intensity of use varies from one right of habitation to another. These two forms are thus personalized even though their modules appear narrow in scope. Modular structure affects third parties to a different degree. Most ­people outside of a property module do not have to know what complicated arrangements have taken place within the module; they simply have to respect boundaries and keep off (Smith 2002; 2004b). Potential transactors sometimes have a more difficult task (Merrill 2012b: 250). A person seeking to acquire an anticresis on Whiteacre has to make sure whether a usufruct has been carved out. Two reforms of limited property forms would reduce third-party information costs for potential transactors, and the transaction costs of owners. First, streamlining the purpose of each limited property form. As summarized in Part I, in some countries, superficies enable holders to construct buildings and grow plants, and a usufruct can be used to grant another party the right to collect natural fruits, use and enjoy land as is,

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

95

and even to grow plants. Leaseholds and former Socialist countries’ idiosyncratic general-purpose use rights may allow right-holders to do everything or tailor the scope in contracts. By contrast, for example, the right of habitation and profit à prendre are single-purpose use rights. Simple modules are easy to combine, whereas complex modules could clash with one another. Hence, limited property rights are better single-purpose vehicles. More specifically, if superficies is relabeled to include only the right to construct and own buildings without the right to plant, a superficies can easily be combined with an emphyteusis, the right to agricultural use, or a profit à prendre in a transaction between a landowner and an ambitious user. By contrast, if a transactor only intends to build and own a house, only a superficies needs be established. In the future, when a potential transactor seeks to acquire a profit à prendre, it will be clear from the entry of superficies in the registration record alone (without delving into the superficies contract) that the superficies holder does not have the right to collect ­natural fruits. Lawmakers may be tempted to bundling a few modules into a large module, which will look like usufruct and superficies in many countries. As the functional analysis above suggests, there are only one or two dozen limited property forms that make sense – lawmakers cannot create a usufruct without the right to possess, for instance. If all limited property forms are single-purpose, the total number of forms will not be too large to comprehend. Bundling by law would not be necessary. The more important task for lawmakers is to provide a menu of options (Ayres 2006; Listokin 2009; Lin and Chang 2018) to reduce the institution cost for transacting third parties – this is the second reform approach. Menus in law, like menus in restaurants, provide a set of labeled choices. The contents of a Big Mac Combo Meal are known to patrons, counter staff, kitchen staff, and all those concerned. Applying the metaphor to the context of real easement, if the law provides only a barebones module of real easement and requires transacting parties to pin down a specific purpose, it will be difficult for most transactors to not over- or under-specify the scope, and difficult to register. Austria §§475–502 offers a good example, providing a menu of more than a dozen real easement “combos” so as to make it easy to refer to and use a specific type of real easement – for hunting, for passage, what have you. Regarding security interests, some countries, such as Iran, use a catch-all provision to enact a huge module of security interest, allowing the parties to decide whether creditors or debtors possess collateral. The benefit of giving each typical arrangement a distinct label should be

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

96

obvious. Lawmakers can start with the identified key characteristics listed in Figures 3.3 and 3.4 and give each “combo” a distinct label. Packaging within a property module to create menu options is much more important than bundling property modules.

C  Future Interests Not Necessarily Create High Information Cost The unwillingness of most jurisdictions to allow future interests is puzzling. As evidenced by countries that allow future interests, property owners highly value the temporal division of their rights. Future interests do create some information costs for third parties, who may not be aware of such rights, as holders of future interests are not possessory. However, especially in countries with registration-of-rights system, establishing future interests on land creates minimal third-party information costs for potential transactors. No matter how many life estates and remainders have been established, the terms and holders can easily be reflected in registration records. Hence, the benefits of allowing future interests are likely to be higher than the costs, at least in jurisdictions already with a functioning registration system (see Chapters 4 and 6). Perhaps distaste for feudalism and fear of dead-hand controls explain the legislative choice. If these are indeed the real concerns, simple forms of future interests could be allowed without compromising those values. The consecutive usufruct is one. Owners could also be allowed to sell to a third party the reversion interest, while other forms of future interests remain unavailable. The reversion interest is a possessory right in the future when the current possessory right expires. For instance, Abe grants a 30-year usufruct of Blackacre to Bob; the right to use Blackacre 30 years from now is a future interest held by Abe as a reversion. My proposal is that Abe should be able to sell the reversion to Chris. Reversion interest is particularly valuable in an aging society. Elderly and perhaps childless couples often would like to exploit the value locked into their houses while they are alive. A reverse mortgage or trust, if legally sanctioned, enables the elderly to enjoy financial flexibility, but alternative routes should be considered. Under my proposal, elderly couples could, through a straw man, establish a life-time usufruct, a life estate, or a right of habitation for themselves, and sell the reversion to gain access to a lump-sum amount of money – which they may need for medical treatments for which monthly payments of a reverse mortgage would be insufficient.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

97

In a way, civil-code jurisdictions do not have to recognize future interests as such because a work-around exists. In a common law jurisdiction, selling a reversion can be done via two routes. (1) “To Bob for life, then revert to Abe” creates a life estate with a reversion in Abe (the original owner). Abe can further sell the reversion to Chris. (2) “To Bob for life, then to Chris” gets the deal done in one deed. In a civil-code jurisdiction, “to Bob a usufruct, until Bob dies” creates a usufruct in Bob. Abe cannot sell the reversion which is not a recognized property form. As a workaround, Abe can sell ownership to Chris but set up a usufruct for Bob or himself (Terré and Philippe 2006: para. 860). Depending on how transaction taxes are levied, the work-around may or may not be sufficiently attractive for most people. Putting tax issues aside, in this latter scenario, Abe essentially sells Chris a future interest. Given the legitimate workaround, why not conceptually and formally allow it? In addition, while a work-around is feasible in the aforementioned simple transaction, it is not feasible in more complicated settings. For instance, an elderly couple may like to take care of a cousin who is 20 years younger. Thus, their preferred deal is to keep a life estate or life-time usufruct, to grant their apartment to their cousin for life, and then to sell the reversion at some point. Unless a country allows registrations of two usufructs that take effect sequentially (some explicitly do not), no workaround may be available. At least under a registration-of-right system, a transaction like this is not difficult to register. The potential benefit is not small while the institution cost is tiny. Concerns over dead-hand control can be alleviated by something like the common law “rules against perpetuities.” Thus, it appears inefficient to flat-out ban future interests.

D Recursiveness Smith (2011b: 152–153), in the context of justifying the numerus clausus principle, observes that the recursive nature of property forms and the existence of the trust to a large extent make unnecessary a significant number of forms. Nonetheless, recursiveness and trust do not guarantee that, if lawmakers enact suboptimal number of property forms, the market can make up for the deficiency via other channels. The recursiveness of property forms is most useful when future interests are allowed. However, many civil-code jurisdictions are hostile to future interests. Holders of a restricted personal easement, the right of use, and the right of habitation, to name a few, cannot transfer their rights, let alone establishing subrights. Recursiveness is much limited in civil-code jurisdictions.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

98

Moreover, very few countries outside the common law world allow property-form trust (Chang and Smith 2012: 41–42).61 According to the global survey of trust laws by Hofri-Winogradow (2020), only 40 of the civil-code jurisdictions (39%) – and 58 of 156 overall – recognize trusts. Even civil-code jurisdictions that recognize a trust often frame it as a contract, not a property,62 and the trust form allowed is much less flexible than its common law counterpart (for one, no power of appointment). To compensate for the lack of flexibility, one would expect that countries with no trust law at all would enact more limited property forms. Twosample t-tests show that the opposite is true: Countries with trust law have more limited property forms (counting those regarding movables or not), and the difference is statistically significant (p < 0.001).

E  Registry Capabilities Arruñada (2012: 61) opines that it is more sensible to “define a stricter numerus clausus and a lighter workload for less capable registries.” My data enable us to ascertain whether lawmakers across the globe follow this advice. The jurisdictions marked as “unclear” or “numerus apertus” in Figure 3.1 are labeled as “No NC principle” in Figure 3.7, while those marked otherwise (except “not coded”) in Figure 3.1 are labeled as “NC principle” in Figure 3.7. NC stands for numerus clausus. Registry workload can be measured in a number of ways. Here, the number of immovable property forms is used as a proxy. For measurement of registry capability, this chapter relies on World Bank’s 2017 Doing Business data (see also Chapter 4), where a higher score implies higher capabilities. If the advice by Arruñada (2012: 61) is followed, with or without the numerus clausus principle, lower registry scores should correlate with fewer immovable property forms. As Figure 3.7 shows, this is not the case. This is perhaps not surprising, as the number of immovable property forms is largely determined by legal transplant/diffusion (see generally Berkowitz et al. 2003a; 2003b; Spamann 2009b), regardless of registry performances. As Arruñada and Garoupa (2005) point out, the registration-of-right system may be too much of a good thing, as it is expensive. Developing countries may not be able to afford (and it does not necessarily make efficiency sense to maintain) registries with heavy 61 62

Ramseyer and Nakazato (1999: 27) note that a fee simple determinable in Japan binds the parties only contractually. Ho and Lee (2013) survey contract-form trust laws in Asian jurisdictions.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

limited number of limited property rights

99

0–4 immovable property forms

No NC principle

5–6 immovable property forms

7–9 immovable property forms

0–4 immovable property forms

NC principle

5–6 immovable property forms

7–9 immovable property forms 0

20 40 60 80 100 Score: registering property

Figure 3.7  Registry capability, workload, and the numerus clausus principle Notes: NC = numerus clausus. N = 30, 30, 20, 9, 34, and 33 from top to bottom.

workloads that a registration-of-right system entails. Arruñada et al. (2022) find out that in Benin, incomplete land registration due to budget constraints led to a surge in land-related litigation. Lawmakers should account for the resource allotted to registries when determining whether to enact a large number of immovable property forms and introduce the registration-of-right system. That said, my proposal to make limited property forms single-purpose should alleviate registries’ burden under a registration-of-right system, as it is easier to ascertain whether limited property rights conflict when each form has a narrower scope.

III Conclusion Economic analysis suggests that lawmakers should (1) explicitly endorse the numerus clausus principle; (2) make limited property forms singlepurpose; (3) enact a sufficient number of limited property forms to cover various uses and purposes; (4) provide menus within modules where appropriate; and (5) recognize future interests.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

100

No country appears to follow all of the prescriptions. Common law countries recognize future interests but often do not endorse the numerus clausus principle explicitly; they have ample property forms, but leasehold and life estate are multipurpose use rights. Among civil-code jurisdictions that prescribe that the number is close, many stick to the Roman limited property forms that are not necessarily the best fit for modern societies, and the popular usufruct and superficies are often multipurpose.

https://doi.org/10.1017/9781009236553.005 Published online by Cambridge University Press

4 Transfer of Ownership Transaction Cost versus Information Cost This chapter focuses on how ownership of movables and immovables is transferred – specifically: is immovable registration constitutive or ­declaratory? How many jurisdictions adopt the German “abstract principle” under which two distinctive agreements (first a sale contract and then a real agreement) are involved in a sale? Does the validity of the sale contract affect the real agreement (the non-causa principle)? Is the public faith principle recognized? Is delivery required to transfer ownership of movable things, and if so, in what manner? This chapter is structured as follows: Part I provides an overview of these doctrines and principles in the world. Part II analyzes its economic costs and benefits. Part III concludes.

I  Comparative Overview This chapter deals with transfers of ownership of immovables and movables, not the creation of limited property rights. The distinction is particularly important in terms of immovables because some countries adopt different rules for the transfer of landownership and creation of limited property rights, such as mortgage. This book uses registration as the umbrella term for the recording-of-deed system and the registration-ofright system. Section A classifies countries based on whether registration is constitutive or declaratory. Section B identifies countries that adopt the public faith principle from those that do not. Section C focuses on the transfer of movable ownership, especially the role of delivery therein.

A  Ownership Transfer Rules for Immovables In terms of transferring landownership between a current owner qua seller and a buyer, the key questions to ask are: first, do sale documents (deeds) have to be recorded, or the conveyance of rights registered, to make the transfer of ownership legally effective? If yes, registration has a 101

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

102

part i: foundation

Transfer Ownership of Real Property Government approval Constitutive+Real+Non-causa Constitutive+Real+Causa Constitutive Opposable Contract itself No information

Figure 4.1  Transfer ownership of real property

constitutive effect. If not, sale contracts in and of themselves should suffice to transfer ownership, subject to occasional additional requirements such as delivery. Second, if registration is not constitutive, does it have a declaratory (or, opposable) effect (Schmid and Hertel 2005: 33)? For a legal system whose registration creates opposability, sale documents that are not registered are still valid between the transacting parties, but are not valid against third parties, in particular good-faith parties. In some countries, government approval of land sales is necessary. More specifically, the ownership transfer rule for land (in the case of state ownership of land, buildings) can be classified into several groups. See Figure 4.1. Type A. A sale contract itself (in some countries, plus delivery) transfers ownership, and the role of registration is not specified. Cuba §178, Iran §140, and North Korea §38 are prime examples. A number of other countries adopt the original Code Napoléon approach. That is, registration is opposable in terms of mortgage creation (Arruñada 2012: 48), but not ownership transfers. Thus, these countries are classified under Type A as well. They are, for example, Algeria §165 & §792, Burundi §II:II:176, Guinea §671 & §840, and Monaco §1426,1 as well as §1138 and §1583 of Burkina Faso, Comoros, Ivory Coast, Madagascar, Mauritius, Niger, Seychelles, and Togo. Type B. A sale contract itself transfers ownership, but the transfer is opposable to third parties only if registered. That is, registration has only declaratory effect. Jurisdictions belonging to Type B are Belgium, 1

Note that Monaco §1426 stipulates that building sales in the Principality are not effective until registration.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

103

Denmark, Luxembourg, Poland, Portugal, Spain, Sweden (Schmid  and Hertel 2005: 34, 37, 48), as well as Benin §186, Bolivia §521 & §1538, Bulgaria §§112–113 (Santisteban and Sparkes 2015: 247), California §1214 & §1217, Costa Rica §480 (Romañach Jr. 2014a), Equatorial Guinea §606, El Salvador §667, Ethiopia §1645, Finland (Hollo 2019: 138) (but every buyer is required to file for registration within six months of acquiring ownership), France (Erp and Akkermans 2012: 901), Guatemala §1141 & §1148 & §1808, Honduras §713, Hong Kong (LRO) §3,2 Indonesia §623, Italy §1376, Japan §§176–177, Lithuania §1.94 & §6.393 (Smaliukas 2017: 48–49) (registration of servitude, however, is constitutive!), Louisiana §§517–518, Macau (Ai 2013: 96), Malta §996 & §1347, Mexico §3012, New York (RPL) §291, Nicaragua §3948 & §2575, Norway (LRA) §20, Ontario (La Forest 2017b: 30-4), Panama §1761, Paraguay §1968, Peru §2022, the Philippines §709, Puerto Rico §546, Quebec §2941, Romania §§1674–1676, United Arab Emirates §1277, and Venezuela §1917 follow this rule. Type C. Registration is a prerequisite for valid transfers of real property ownership. That is, registration has constitutive effect. Type C1 is the modal rule, adopted by Afghanistan §2212, Albania §83, Argentina §1892, Armenia §135 & §176, Australia (Hepburn 2002: 169), Azerbaijan §139, Bahrain §859 & (PRL) §17, Belarus §165 & §224, Brazil §1227 & §1246, Cambodia §135,3 Chile §686, China §§208–209,4 Colombia §756, Croatia §119 (Josipovic 2013: 91), Cyprus (Synodinou 2020: 125), Dominican Republic (RERL) §90, DR Congo (LF) §59, Ecuador §702, Egypt §934, England (LRA) §27 (Schmid and Hertel 2005: 36),5 Eritrea §1079, Georgia §183, India (IRA) §17, Iraq §508 & §1126, Ireland (Schmid and Hertel 2005: 38), Israel (LL) §7 (Unknown 1970), Jordan (Ziadeh 1979: 45), Kazakhstan §118 & §155, Kuwait §890 (Karam 2011), Kyrgyzstan §255, Liberia (LCT) §8.121 & §8.131, Libya §938 (Libya and Ansell 1972), Macedonia (AOORR) 2 3 4

5

There is no freehold interest in Hong Kong. Long leases instead are used (Goo and Lee 2019: 4–5). Cambodia §134 stipulates that registration of limited property rights is opposable to third parties. That is, only registration of ownership of immovables is constitutive. China §208 stipulates that in terms of real properties, in principle, property rights of real properties cannot be established or changed without registration, but §209 carves out exceptions. A right to farming rural land (chengbao), for example, is established once the contract goes into effect. Local governments, though, should note the rights in their books, rather than in the ordinary land registry, to confirm the existence of such rights (§333). The HM Land Registry in England states that “[a]nyone buying or selling land or property, or taking out a mortgage, must apply to us to register.” See www.gov.uk/government/ organisations/land-registry/about. Cf. Hayton and Matthews (2007: 31–33) (describing the complicated rules regarding registered and unregistered estate in England).

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

104

part i: foundation

§148, Malaysia (Maidin and Ali 2014: 157), Moldova §499, Mongolia §§109–110, Myanmar (TPA) §54, New Zealand (Williams 2011: 44), Oman (LL) §25, Pakistan (IRA) §17, Qatar §246 & (LRERS) §4, Russia §223 & §551, Serbia (LBOPR) §33, Singapore (LTA) §45, Slovakia (Schmid and Hertel 2005: 37), South Korea §§186–188, Suriname §670, Syria (Ziadeh 1979: 44), Tajikistan §143, Thailand §1299, Timor-Leste §348 & §809, Tunisia (CDR) §305, Turkmenistan §206, Uganda (Bakibinga 2018: 33), Ukraine §182 (Maydanyk 2017: 235), Uruguay §1664, Uzbekistan §111, Vietnam §106, Zambia (LDRA) §6, and Zimbabwe (DRA) §14. Type C2. In addition to registration and valid sale contracts, a real agreement (dingliche Einigung in German) (Drobnig 2011b: 1004), that is, a conceptually distinct “contract,” is also required. This third feature distinguishes this group from Type C1. Real agreement, a German concept, describes the meeting of minds at the time of registration (for real properties) and delivery (for personal properties) as a separate, thingrelated contract. That is, it takes two types of contracts, and registration, to transfer ownership of real properties. Many members of the C2 group are countries influenced by German jurisprudence, but their civil codes often do not make the requirement of a real agreement crystal clear. Put differently, the existence of a real agreement in all property sales is a matter of judicial and/or academic construct. My list of Type C2 countries, below, may be under-inclusive: Austria §431, Czech (Ronovská et al. 2020: 91–92), Estonia (TLPA) §64-1 & §120, Germany §873 & §935, Greece (KaribaliTsiptsiou 2012: 62), Hungary §V:38 (Szilágyi 2011: 578–583) (discussing the disputed nature of real agreements), Latvia §993 (Klauberg and Kolomijceva 2010: 563–564), Liechtenstein §1045, the Netherlands §III:16 (Salomons and Haentjens 2017: 110–112) (discussing the disputed nature of real agreements), Scotland (LRA) §50, Slovenia (Škerl and Vlahek 2010: 119), South Africa (Van der Merwe et al. 2002: 49; Merwe 2004: 216–217), Switzerland §656, Taiwan §758 (Chang et al. 2022: 34), and Turkey §705 & §1021 & §1024. Among the Type C2 countries, Germany, Taiwan,6 Scotland, South Africa, and Turkey (Type C2b), for instance, adopt the non-causa principle,7 which means that the validity of the transfer of ownership is judged 6 7

Real agreements are not absolutely non-causa in Taiwan, but most often they are (Chang, Chen, and Wu 2022: 91). The degree of non-causa differs in Germany, Scotland, and South Africa, as exceptions were created to different degrees and contexts (Van der Merwe et al. 2002: 49; Erp and Akkermans 2012: 842–843). In Turkey, scholars also debate on whether and to what extent the non-causa principle is adopted. I thank Prof. Eyüp Ipek for this point.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

105

independently of a sale contract.8 Thus, while a sale contract can be annulled by the previous owner for, say, fraud, ownership does not automatically revert to the previous owner. Instead, the previous owner can only sue the buyer qua current nominal owner in unjust enrichment (Erp and Akkermans 2012: 831). If the buyer becomes bankrupt, the previous owner cannot simply take possession; rather, she will be regarded only as an ordinary creditor (Erp and Akkermans 2012: 836). If the buyer further sells the movable thing in question, the good-faith purchaser doctrine does not apply, as the first buyer is still entitled to sell as the owner of the movable thing in question. That is, even though the second buyer is aware of the invalidity of the previous sale contract, the second buyer is nonetheless the legitimate owner because the real agreement between the seller and the first buyer is still legally intact (Erp and Akkermans 2012: 835). Type D. In Bhutan (LA) §§159–164, Brunei §23, Nigeria (Smith 2018: 139), Sri Lanka (Rupesinghe and Rupesinghe 2007: 79), and Tanzania (LRA) §41, transfers of real properties are not completed before the government approves the sales.

B  Public Faith Principle The public faith principle – also called absolutism or the principle of publicity (Mattei et al. 2009: 933) – has two prongs. The first prong is that good-faith transferees acquire “a real right free from an existing burden” that is not registered (Erp and Akkermans 2012: 908) – sometimes called the negative effect of registration (van Vliet 2017: 159). Registration creates notice. No matter in countries with the constitutive or the declarative system, third parties are bound by the registered rights. Put differently, third parties are not bound by rights that are not registered. For instance, if landowner L and bank B entered into a mortgage but failed to register it, good-faith transferee T will not be bound by the mortgage and may acquire clean ownership without encumbrance.9 8

This is in contrast to the causa principle, under which a transfer of ownership is void when the underlying sale contract is annulled. Ownership automatically reverts to the previous owner. The previous owner thus has an action of revindication (rei vindicatio) (Erp and Akkermans 2012: 830–831). 9 If registries do not contain deeds or information regarding certain encumbrances (say, mortgage), but a bad-faith third-party purchaser is aware of the encumbrance, will the third party acquire ownership without the encumbrance? If yes, the legal system does not inquire into what information a third party has. Rather, the information contained in the registry is what matters. If no, the legal system consciously protects only good-faith third

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

106

part i: foundation

The second prong of the public faith principle is that good-faith transferees acquire a “non-existing real right” that somehow appears in the registry (Erp and Akkermans 2012: 868–869) – sometimes called the positive effect of registration. In the example above, one may tend to think that the unencumbered ownership is the “non-existing real right,” and thus consider the first and second prongs as the same. However, the second prong does have an independent function when, for example, Y is the actual owner, but due to a registrar’s mistake or some other reasons, Z is registered as the owner. A good-faith transferee, T, who purchased from Z will be protected under the second, but not the first, prong of the public faith principle. Below, this chapter classifies a jurisdiction as adopting the public faith principle only if the relevant stipulations are broad enough to include both prongs of the public faith principle. Some jurisdictions that adopt a constitutive or declaratory system explicitly adopt the first prong, and it will be classified as only so, whereas the others arguably implicitly so – if good-faith third parties are not protected against unregistered rights, what do opposability and constitutivity mean? The public faith principle and the constitutive (or opposability) effect of registration afford protections in different scenarios. A buyer of land who had registered her property rights uses the constitutive (or opposability) effect against third parties who registered their rights later. By contrast, a buyer of land uses the public faith principle against third parties who have acquired property rights earlier but did not register their rights. The constitutive effect of registration and the public faith principle are naturally a pair, and at least in the practice in Europe, they are (Schmid and Hertel 2005: 38). That said, Figure 4.2 shows that not all constitutive systems adopt the public faith principle while quite a number of declarative systems do. Overall, 54 jurisdictions (35%) adopt both prongs of the public faith principle. Note that some jurisdictions limit the application of the public faith principle to good-faith transferees who have given reasonable consideration.10 In the survey and coding of the public faith principle,

10

parties, and both the knowledge of the third parties and the constructive notice provided by registries count. It is not always clear whether bad-faith purchasers would be bound by the encumbrance. Argentina §1893, California §1214 & §1217, China §311, Estonia (TLPA) §56, Germany §892, Macedonia (AOORR) §151, Mexico §3009, to name a few, clearly stipulates that bad-faith transferees are bound by the encumbrance. Russia §302 (Butler 2002: 120–121) stipulates that the public faith principle applies only if acquirers have paid for the real properties. China §311 requires purchasers to pay reasonable price to enjoy the benefit of the public faith principle.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

107

Absolutism of immovable registration Declarative+no absolutism Declarative+absolutism Constitutitve+no absolutism Constitutitve+absolutism No information or other combinations

Figure 4.2  Absolutism in real estate registration

I  thus  focus on whether good-faith purchasers who have paid market price are protected. More specifically, countries can be divided into three large groups: Types E, F, and G. Type E. No public faith principle regarding ownership. All the countries that allow transfers of land ownership upon consummation of sale contracts naturally disallow the public faith principle. In addition, some countries adopt the first prong of the public faith principle only regarding certain security rights such as mortgage and pledge. Algeria §885 (mortgage) & §950 (pledge), Mauritius §2201-2, and §2198 of Burkina Faso, Comoros, Ivory Coast, Madagascar, Niger, Seychelles, and Togo fall into this group. In addition, these stipulations do not make clear whether transferees will be bound if they are aware of the encumbrance. Type F. Jurisdictions of this type adopt only the first prong of the public faith principle regarding ownership. That is, unregistered property rights do not bind good-faith third-party transferees. In these countries, constitutivity or opposability is clear from statutes, but the first prong of the principle is in some countries implied while in others it is explicit. Countries are classified under this type because there is no explicit rule about the second prong of the public faith principle. Afghanistan, Albania, Armenia, Bahrain (PRL) §18,11 Belarus, Eritrea §1081, Lithuania §1.75, etc. are prime examples. In addition, Van der Merwe et al. (2002: 283–287) discuss how the South African registry does not guarantee the accuracy of registered rights, and how actual owners are preferred over good-faith third parties. Ziff (2014: 484) discusses how in Ontario, 11

Bahrain §946 prescribes the second prong of the public faith principle only regarding mortgage.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

108

part i: foundation

good-faith purchasers who buy from registered owners who had acquired ownership by forgery will not be protected. Also notably, Brazil §1247 Sole Paragraph is a rare example of declaring the lack of the second prong of the public faith principle. Belgium (LH) §129 (formerly §2198), France §2451 (formerly §2198) (Erp and Akkermans 2012: 908), and Luxembourg §2198 should be Type-F instead of Type-E. These countries stipulate that a transferee acquires ­ownership free of mortgage and privilege if the encumbrance is not registered. This limited first-prong public-faith principle related to only mortgage and privilege was part of the original Napoleonic Code; their opposability system was a much later addition. Now, even without their aforementioned civil code stipulations, the first prong can be derived from the opposability effect and is applied beyond the context of security rights.12 Type G. Clearly, public faith principle is adopted. Schmid and Hertel (2005: 36–38) list the following European countries as having the publicfaith principle: Austria, Croatia (AOORR) §122 (Josipović 2014: 102), Czech, Denmark, England, Estonia (TLPA) §56, Finland (Hollo 2019: 144), Germany §892, Hungary §5:173, Ireland, the Netherlands §§3:24– 3:26, Norway (LRA) §25 & §27, Poland (Brzozowski 2005: 68), Portugal §291, Scotland,13 Slovakia, Slovenia, Spain, Sweden, and Switzerland. In addition, the public faith principle is adopted in Argentina §1893, Australia (Hepburn 2002: 68), Azerbaijan §140, China §311, Costa Rica §456,14 Cyprus (Synodinou 2020: 152) (good-faith transferees for value are preferred over adverse possessors who have not registered ownership), Dominican Republic (RERL) §99, Georgia §185, Greece §1204 (Taliadoros 1982),15 Indonesia,16 Israel (LL) §10, Liberia (LCT) §8.133, Liechtenstein §§554–555, Macau §235 & §284, Macedonia (AOORR) §§150–151, Malaysia (Maidin and Ali 2014: 76–77), Mexico §3009, Moldova §497, 12 13 14 15

16

Egypt (despite §1034), Ethiopia (despite §3051), and Qatar (despite §1062) are classified as Type F for a similar reason. Scotland (LRA) §86 prescribes a sophisticated rule, under which a seller (together with an acquirer) must be in possession openly and peaceably for a year. Costa Rica’s public faith principle is idiosyncratic in that §456 does not explicitly limit its protection to good-faith transferees. Greece’s classification in this group is controversial. §1204 appears to stipulate the public faith principle, and Greece has a constitutive system. But Mattheou (2010: 18–19) reports that even a good-faith transferee is not protected from inaccuracy in registration. Leks (2018: 32) described the qualified public faith principle in Indonesia, as good-faith purchasers will be protected if actual owners fail to object within five years following the issuance of a certificate of land.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

109

Mongolia §109.3 & §114.1, New York (RPL) §399 (a limited public-faith principle), New Zealand (LTA) §§182–183, Nicaragua §3949, Panama §§1762–1763 (Romañach Jr. 2009), Peru §1135 & §2014, Romania §901, Serbia (Živković 2007: 29–30), Singapore (See et al. 2018: 96), Taiwan §759-1, Thailand §§1299–1300, Tunisia (CDR) §305, Turkey §§1023–1024 (for an exception, see Adal 2012: 159; see also Ansay 2020), Turkmenistan §208 & §326, Uganda (RTA) §64, and Zambia (LDRA) §59. In all, only 54 jurisdictions explicitly stipulate absolutism. It could be that absolutism elsewhere is prescribed in registry-related statutes and thus elude my coding effort. Or courts in practice may recognize absolutism. However, it is worth emphasizing that it may very well be the case that some of these countries simply do not enable absolutism as it makes sense only if registries contain relatively complete and reliable information.

C  Ownership Transfer Rules for Personal Properties Several paradigms explain how ownership of personal properties can be transferred (Figure 4.3).17 Coded here are the rules regarding sales of specified items (such as a particular oil painting).18 Some countries, like Albania §164, adopt different rules for specified items and unspecified items (e.g., a brand new iPhone 14). This chapter focuses on the rule for the former. Additionally, some countries provide a menu of choices with one being the default rule for transferring ownership. This chapter codes the default rule. Type K is a consensual system (Erp and Akkermans 2012: 788) under which consummation of a sale contract itself transfers ownership of movables. Countries which are members of this group include, for example, §1138 and §1583 of Burkina Faso, Comoros, Dominican Republic, France, Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, Seychelles, and Togo, as well as Afghanistan §2210, Albania §164, Algeria §165, Azerbaijan §181.4, Belgium §711, Bolivia §584, Bulgaria (Stoimenov 2011: 406), Burundi §III:264, Cambodia §133–134, Costa Rica §480, Cuba §335, Czech (Ronovská et al. 2020: 92–93), Egypt §§932–933, 17

18

Some countries require registration to transfer ownership to certain movable properties. (I do not systematically code this.) One notable example is China §225, in which registration of sales of vessel, aircraft, and automobiles is the prerequisite for the ownership transfer to be opposable to third parties. In addition, the Netherlands §§III:21–III:31 is famous for drawing the line between registrable and non-registrable things. Mattei (2000: 106) notes that in practice, courts sometimes qualify the general rule in application.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

110

part i: foundation

Transfer ownership of movables Functional approach When intended Delivery Contract itself No information

Figure 4.3  How to transfer ownership of personal property

Equatorial Guinea §1450, Finland (Kuusinen 2010: 344), Guinea §671, Haiti §1368, Iraq §1126, Japan §176 (§178 stipulates that delivery creates opposability), Jordan §§1146–1148, Kuwait §889, Libya §207, Louisiana §518, Macau §402, Malta §1347, Mexico §2249, Monaco §1426, North Korea §38, the Philippines §§1496–1501 (many ways to transfer ownership, entering into a public instrument being one way; that is, delivery not always required), Poland §155 (Negbi 1975), Portugal §1317 & §408, Qatar §246 & §919, Quebec §1453, Romania §1674, Senegal §277, Syria §894, Thailand §458, Timor-Leste §343 & §1238, Tunisia (CDR) §22, and United Arab Emirates §1275. Type H is an intention-based system. When ownership of movables is transferred depends on transacting parties’ intention. England (SGA) §18 provides a prime example for this approach, laying out five complicated rules to ascertain intentions of buyers and sellers. Adopters of this rule also include Hong Kong (SGO) §3, India (SGA) §19, Ireland (SGA) §17, Malaysia (SGA) §19, Malawi (SGA) §20, New Zealand (SGA) §20, Nigeria (SGA) §17, Ontario (SGA) §18, Pakistan (TPA) §8, Scotland (SGA) §17, Singapore (SGA) §17, South Pacific countries (Farran 2013: 78) (governed by common law principle and Sale of Goods statutes), Tanzania (SGA) §19, and Uganda (SGSSA) §25.19 Type I. Scandinavian countries adopt a functional approach (Ross 1957; Haug 2017) and do not have a unitary approach to identify a magic moment when ownership of movables is transferred. While Denmark (Millung-Christoffersen 2019), Norway (Faerstad and Lilja 2010: 215) and 19

von Bar and Drobnig (2004: 325) note that the English Sale of Goods Act is “heavily influenced by the French code.”

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

111

Sweden (Lilja 2010: 14) follow the functional approach, Finland, as noted above, adopts a consensual system. Type J. The Roman Law tradition requires traditio (delivery).20 In its simplest form, delivery is fulfilled when movables change hands, or when a sales contract comes into effect if the buyer already has actual control of the thing in question (traditio brevi manu). Argentina §1892, Armenia §176, Austria §426, Bahrain §858, Belarus §§224–225, Brazil §1267 (Romañach Jr. 2015b: 234), California §1054, Chile §684, China §224, Colombia §740, Croatia (AOORR) §116, Ecuador §686, El Salvador §665, Eritrea §1080, Estonia (TLPA) §92, Ethiopia §1186, Georgia §186, Germany §929, Greece §1034, Guatemala §1803, Honduras §711, Hungary §V:38 (Vékás 1995: 85), Indonesia §1475, Iran §47 (Shahabi 2007), Israel (SL) §33, Italy §1376, Kazakhstan §238, Kyrgyzstan §255, Laos (PL) §28, Latvia §987, Liechtenstein (SR) §172, Lithuania §4.49, Macedonia (AOORR) §145, Moldova §321, Mongolia §111.2, the Netherlands §III:90, New York (UCC) §2-401, Nicaragua §2575 & 2578, Panama §1220A, Paraguay §2062, Peru §947 (Romañach Jr. 2014b: 148), Puerto Rico §549, Russia §223 (Butler 2002: 93–94), Serbia (LBOPR) §34, Slovakia §133, Slovenia (Škerl and Vlahek 2020: 171), South Africa (Van der Merwe et al. 2002: 250), South Korea §188, Spain (Aristi and Imbernón 2013: 105), Suriname §666, Switzerland §714, Taiwan §761, Tajikistan §§247–248, Turkey §763, Turkmenistan §209, Ukraine §334, Uruguay §775, Uzbekistan §185, Venezuela §1487, and Vietnam §161 belong to this type. These jurisdictions, however, make different policy decisions regarding whether to acknowledge certain types of “constructive delivery” for sales purposes. The two most common forms of constructive delivery are constitutum possessorium (in which original owners turn themselves into mere possessors for acquirers of the movable in question) (Erp and Akkermans 2012: 817) and attornment (in which sellers agree to assign to buyers their right to recover the thing in question from a third party). Table 4.1 lists the jurisdictions that allow one or two types of such constructive deliveries.21 20

21

van Vliet (2017: 151–158) further distinguishes between a causa traditio system and an abstract traditio system. Countries that adopt the abstract system have been mentioned above. El Salvador §665 and Honduras §711 may allow constitutum possessorium, but it is unclear. Guatemala §1810, among others, recognizes real, symbolic, or legal delivery, and legal delivery may include attornment and constitutum possessorium, but it is unclear.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

part i: foundation

112

Table 4.1  Constructive delivery in movable sales Jurisdictions

Source

Argentina Austria Bahrain Brazil China Colombia Croatia Eritrea Estonia Georgia Germany Greece Hungary Latvia

§1892 §428 §887 §1267 Sole Paragraph §§227–228 §754 (AOORR) §15 §973 (TLPA) §§93–94 §186 §§930–931 §977 & §1035 §V:3 n/a (Klauberg and Kolomijceva 2010: 567–568) Liechtenstein (SR) §503 Macedonia (AOORR) §145 Mongolia §111.2.3 Netherlands §3:90 & §3:115 Serbia (LBOPR) §34 Slovakia §590 Slovenia §60 (Škerl and Vlahek 2020: 172) South Africa (Van der Merwe et al. 2002: 269–276)22 South Korea §§189–190 Switzerland §717 & §924 Taiwan §761 Turkey §764 & §979 Turkmenistan §209 Uruguay §767

Constitutum possessorium Attornment Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Debated

No No No Yes Yes No Yes Yes Yes Yes Yes Yes Yes No

Yes Yes No Yes No No Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No

Note: Only Type-J jurisdictions are included here. Other countries, such as Costa Rica §482, Malta §§1380–1381 and Finland (Faerstad and Lilja 2010: 242–244), may recognize attornment or constitutum possessorium, but they are not included. 22

See also Cooper and others v. Dabbs and another (92/2003) [2004] ZANWHC 29 (1 November 2004), paragraph 10, at www.saflii.org/za/cases/ZANWHC/2004/29.html for more info; Page Automation (Pty) Ltd v. Profusa Properties CC t/a Homenet O.R. Tambo and Others (2011/11966) [2012] ZAGPJHC 251, paragraph 20, at www.saflii.org/za/cases/ ZAGPJHC/2012/251.html.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

113

II  Economic Analysis A  What Type of Registration Is More Efficient Several strands of literature show that registration systems that provide more accurate and easier-to-digest information increase property value. First, the economic development literature suggests that formal property rights are important (if not necessary) for economic development (de Soto 2000).23 Registering land, in an important way, makes property rights more formal. As Field (2004) shows, in Peru, receiving a legal property title led to a 48% decrease in the fraction of households that locate entrepreneurial activities inside the home and a 36% reduction in the fraction of households that report keeping individuals at home to protect property. In addition, Koo (2011) shows that most land in Taiwan was not registered during the Qing Dynasty (1662–1895), as landowners attempted to evade land taxes. When the Japanese started to colonize Taiwan (1895–1945), a thorough land survey was made, and all surveyed land became registered. Land prices increased significantly after property rights became formally defined. Moreover, farmers became more willing to apply green fertilizers and to grow crops that took a long time to produce, such as tangerines – that is, to make long-term investments. Arruñada et al. (2022) use a natural experiment in Benin to show that if the registration process is not sufficiently comprehensive, it leads to the demand for complementary litigation aimed at perfecting property rights. Second, Libecap and Lueck (2011a; 2011b) show that, other things being equal, the rectangular system (as compared to the metes and bounds system) lowers institution costs and increases property value (but compare Brady 2018). But in rugged terrain, the rectangular system is costly (Libecap et al. 2011). Third, and more directly related to the surveyed doctrines in this chapter, Miceli and Sirmans (and co-authors) in a series of articles argue for the superiority of the Torrens registration system over the recording system. Miceli and Sirmans (1995b) argue that as long as the subjective value of current owners (who purchased from purported owners) is higher than that of claimants (who were the actual owners), the Torrens registration system is superior to the recording system because in high transaction costs setting the recording system assigns property rights to claimants and no following transactions would take place. The problem is that, to be exact, Miceli and Sirmans (1995b) do not compare 23

But compare Upham (2018), Upham (2013), Qiao (2017) who are skeptical of the role of formal property rights. I review Upham (2018) in Chang (2022c).

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

114

part i: foundation

registration of rights versus recording of deeds, but compare with or without the public faith principle. As shown in Figure 4.2, adoption of the public faith principle does not perfectly correlate with the use of the registration-of-right system. In a country with recording and the public faith principle, current owners are still protected. In a country with registration of rights but without the public faith principle, current owners may not be protected. Miceli et al. (2002) use the same Torrens registration versus recording theoretical comparison while considering the effect of adopting the public faith principle. Following their logic, the takeaway point should be that in a system with the public faith principle, the equilibrium land price is higher.24 As emphasized in Chapter 2, the discussion of efficiency cannot leave out the institution cost (see also Arruñada 2012: 7). Arruñada and Garoupa (2005: 712) point out that the registration-of-right system incurs higher operating costs than the recording-of-deed system, and argue that recording is more efficient when privacy (i.e., land not recorded) is considered. Their argument is based on the observation that the registrationof-right system assures all registered parcels while the recording-of-deed system assures only the more valuable parcels whose owners voluntarily opt out of privacy to record their transactions. The formal model in Arruñada and Garoupa (2005: 713) assumes that landowners are free to decide whether to record deeds or register rights. While this assumption may be realistic in some countries, their model does not consider the constitutive effect and opposability effect of registration. Some countries with the recording-of-deed system stipulate the constitutive effect; for example, the Netherlands and Greece.25 Some jurisdictions with the recordingof-deed system stipulate the declarative effect; for example, California, France, Italy, and New York.26 Even though real property transacting parties have the option of keeping transactions and ownership status private, 24

25

26

Using the same set-up, while varying the conditions of whether and what kind of title insurance are available, Miceli et al. (1998) conclude that registration of rights is superior to recording of deeds. For more on the same conclusion (the superiority of registration), see Miceli et al. (2000). Baker et al. (2002) point out that optimal title searching does not require buyers to search all available records. Schmid and Hertel (2005: 32) and Yiannopoulos (2008: 163–164) report that Greece adopts the recording-of-deed system with constitutive effects. Arruñada (2012: 159) and Kleijn et al. (2006: 117–118) report that the Netherlands adopts the recording-of-deed system with constitutive effects. Schmid and Hertel (2005: 32) report that France and Italy adopt the recording-of-deed system.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

115

buyers are strongly incentivized to record deeds to reduce risks. In France and Greece, for instance, some 95% of land has been registered (Schmid and Hertel 2005: 31).27 Still, the broader point in Arruñada and Garoupa (2005) holds. Even if almost all real property transactions are registered, no matter under the registration-of-right or recording-of-deed systems, the institution cost of establishing and maintaining the registration-of-right system is higher than that of establishing and maintaining the recording-of-deed system. The former’s higher institution cost has to be balanced against its higher property value and lower institution cost at the time of each transaction. More specifically, leaving the absolutism aside for a moment, the thirdparty information cost is higher under the recording-of-deed system, whereas its transaction cost of running functioning registries is lower. The upshot for the recording-of-deed system is that the expense of registries usually comes from general tax revenues, and taxpayers do not benefit equally from an accurate registry. In addition, the third-party information cost (surveying, title searching, buying insurance, etc.) is incurred by the transacting parties themselves. Hence, users pay under the recording-ofdeed system, which, generally speaking, leads to less socially wasteful policies. However, if land is traded frequently, every potential purchaser has to incur the same information cost, which may add up to a considerable sum (together with the low fixed cost of setting up recording) that surpasses the high fixed cost of setting up registration plus the low variable cost (i.e., the third-party information cost) under the registration-of-right system. It is difficult to come away with a general conclusion regarding whether the registration-of-right system is superior, once policymakers take into account additional legal designs, such as the public faith principle, whether and to what extent constructive notice can be provided outside of registries, marketable title acts (Arruñada 2012: 63–64), purging actions and quiet title suits (Arruñada 2012: 155), the mandatory use of notaries, whether registries are part of the court system. Some of these important designs are discussed below.

B  Absolutism Not Always Realistic The first prong of the public faith principle incentivizes registration while the second prong may dis-incentivize information gathering by third 27

But compare Oda (2009: 168) who observes that in Japan, “people often fail to register transactions because of tax consequences.”

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

116

part i: foundation

parties (as they are protected as long as they genuinely believe the registration information) – though the second prong may also incentivize property owners to double check on their status in registries to correct errors. That is, after introduction of the first prong into the system, property owners are encouraged to register, thus keeping the registration information up-to-date, and the higher accuracy enables the adoption of the second prong, which gives property owners an extra layer of incentives to keep an eye on the registration information, making the registration even more accurate. Third parties are thus economically justified in paying attention to only the registration information. Absolutism, however, is not free. The first prong is cheaper than the second prong from the state budget perspective. With or without the first prong, disputes arise between two private parties. By contrast, the second prong is likely to be accompanied by a compensation fund or a state liability regime. If X is registered as the landowner while Y is the actual right holder, and Z believes in the registration and buys from X, Z will be protected. X may be long gone or judgment proof, and the error may be attributed to mistakes by registrars. Y would be compensated by the state or a compensation fund. A state may be unwilling to adopt the public faith principle if the aforementioned virtuous circle of increasing accuracy is hampered for any reason. Adoption of the absolutism may be a signal for (lawmakers’ faith in) reliability of registration. Figure 4.4 is consistent with this conjecture. The top boxplot shows the distribution of the quality of the land administration index (DB 2016 methodology) by whether absolutism is adopted. The bottom boxplot shows the distribution of the registering property score (DB 2017–2019 methodology) by whether absolutism is adopted. The latter considers more aspects of registration than the former. For both indices, t-test by whether absolutism is adopted is highly statistically significant (p < 0.001), suggesting that countries that adopt absolutism are associated with higher quality registries.

C  Constructive Notice Outside of Registration As described above, in countries with the second prong of the public faith principle, buyers with actual notice of the identity of actual owners (who are not registered as such) will not be protected. This is not problematic, as the information cost for these buyers is zero, since they already know. A  more difficult question is whether buyers should be protected when

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

117

10 20 Quality of the land administration index (0–30)

30

No absolutism

Absolutism

0

No absolutism

Absolutism

0

20

40 60 80 Registering property score (0–100)

100

Figure 4.4  Absolutism and World Bank’s doing business (DB) data Notes: 50 countries with absolutism and 97 countries without absolutism are included in this figure. Only countries (but not sub-country jurisdictions) with Doing Business data are included. Each plot is a boxplot. The thick line shows the median, whereas the box marks the 25 and 75 percentiles. Interquartile range (IQR) is the distance between the 75 and 25 percentiles. From the 75 percentile, a distance of 1.5 times the IQR is measured out and a whisker is drawn up to the largest observed point from the dataset that falls within this distance. The same applies to observations outside of the 25 percentile. Observations out of reach of the whiskers are outliers shown in circles. Top boxplot: the quality of land administration index (0–30) (DB 2016 methodology) is composed of five indices: “the reliability of infrastructure, transparency of information, geographic coverage, land dispute resolution and equal access to property rights. Data are collected for each economy’s largest business city.” Bottom boxplot: “ScoreRegistering Property (DB 2017–2019 methodology)” is “calculated as the simple average of the scores for ‘Procedures (number)’, ‘Time (days)’, ‘Cost (% of property value)’, and ‘Quality land administration system (0–30) (DB 2017–2019 methodology)’. The score ranges from 0 to 100, where 0 represents the worst regulatory performance and 100 the best regulatory performance.”28 28

For Data: www.doingbusiness.org/en/data/exploretopics/registering-property. For more explanations of the methodology: www.doingbusiness.org/en/methodology/registeringproperty.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

118

part i: foundation

they do not have actual notice but constructive notice may come from outside of registries. Similarly, in countries with only the first prong of the public faith principle, the difficult question is whether buyers with constructive (but not actual) notice of unregistered property rights should be protected. If such constructive notices make for bad-faith buyers, registration information is then just a starting point for potential transacting parties. Title searches are insufficient, as not all encumbrances are recorded in registries. Title insurance, information middlemen, or guarantees from sellers or third parties may be necessary to consummate real property deals. That is, if a buyer cannot take the lack of encumbrance shown in registries at face value, institution costs skyrocket. Below, I use American law as an example to discuss the pros and cons of allowing constructive notice outside of registration. The ongoing Restatement of the Law, 4th, Property §5.4.3.2. summarizes why there is constructive notice if there is visible possession or visible use (e.g., unregistered short-term lease); property interests are apparent from reasonable inquiry (e.g., prescriptive or implied easement); and a lawsuit involving title to a parcel of real property is pending (the lis pendens doctrine) – purchasers acquire properties during the pendency of related litigation take title subject to the outcome of the litigation. Also excluded from the operation of recording acts are “conveyances made by or to the United States government; devises by wills and transfers by intestate succession; transfers by operation of law, such as title by adverse possession; adjustments of boundaries by various techniques of practical location, etc.” As the preceding paragraph should make clear, the scope of constructive notice is pretty large. The law (or, at least, the draft Restatement) does not often take to task holders of unregistered interests. While a person who has acquired ownership via adverse possession but failed to record the declaratory judgment will not be protected against a good-faith third party, a person in a similar situation but who has, for instance, construed a driveway that could alert third parties will be protected against third parties. The scope for constructive notice under the Restatement appears to be inefficiently broad. It may be justifiable to allow constructive notice if the encumbrances are short-term, with compensation, and not very valuereducing. Short-term leases and implied easements by landlocked land (see Chapter 9) fall into this group. Nonetheless, my hunch is that between adverse possessors who have acquired declaratory judgments and thirdparty purchasers, the former is the least cost avoider in terms of recording

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

119

declaratory judgments in registries, to provide record notice. In addition, between parties who have filed lawsuits regarding title of certain plots and third-party purchasers, the former is the least cost avoider in appending a note of pending litigation in registries.

D  Causa versus Non-Causa Theoretically and practically, the conceptual construction of a real agreement is not necessary and even normatively problematic. It is thus not surprising that it is a minority rule. This theoretical construction is harmless, however, because it does not create any extra institution cost for transacting parties. By contrast, the non-causa principle, a byproduct of the concept of real agreements, is not merely a conceptual construction but carries real-world risks for transacting parties, especially sellers. This principle chooses to afford less protection to sellers qua original owner and more protection to third-party creditors and purchasers. The noncausa principle, however, is unlikely to incentivize transacting parties to behave more efficiently ex ante, if it affects behaviors at all. Purchasers will not verify more under the non-causa principle, as they are better protected. Sellers should be aware of the heightened risks, but it is hard to analyze their responses without a realistic example in which the non-causa principle matters – that is, a sale contract is void or voidable for certain reasons but the following real agreement is neither void nor voidable. Moreover, because the non-causa principle, as compared to the causa principle, extends greater protection to the purchaser, it is a normative decision that cries for justification in terms of its efficiency and redistributive effects. I have not encountered, nor can I come up with, convincing justifications. Thus, I call into question the desirability of the non-causa principle.

E  Default Rules and Menus in Movable Transfer Designs Lawmakers should reconsider how to re-structure delivery as a default rule or a menu. A default rule is applied when transacting parties remain silent on the issue (Ayres and Gertner 1989). A menu is a legislature- or courtprovided rule that is clear enough for transacting parties to opt in easily (Ayres 2006). An altering rule is a procedural protocol that stipulates how default rules can be opted out and menus can be opted in (Ayres 2011). At least in jurisdictions that value possession for its public notice function and emphasize actual control by purchasers as a key element in

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

120

part i: foundation

ownership transfer, constructive delivery (attornment and constitutum possessorium) is simply a different animal. In attornment, purchasers receive a claim that may never be practically fulfilled, and yet sellers’ duty is considered observed. In constitutum possessorium, purchasers become owners while perhaps never lays their hands on the object. The continuous possession of original owners poses a threat of double sales to buyers. Of course, constitutum possessorium and attornment are common in some transactions, perhaps standard in certain industries. This does not mean that these constructive deliveries should be default rules for everyone. Nonetheless, the jurisdictions that recognized either attornment or constitutum possessorium (listed in Table 4.1) as a way of delivery in sales all use them as default rules. That is, if contracts are silent on what counts as delivery, constructive delivery can be used. This chapter argues that with the development of artificial intelligence and reduction of institution costs, different default rules can be applied by type of transactions, or even personalized (Porat and Strahilevitz 2013; Casey and Niblett 2016; BenShahar and Porat 2021). Hence, constitutum possessorium and attornment should not be one-size-fit-all default rules for ownership transfers. This chapter argues that constitutum possessorium and attornment should be menus that parties can opt in, rather than default rules that parties can opt out. Setting constructive delivery as menus rather than default rules prevents the informationally superior sellers from tricking the informationally inferior buyers. To use constitutum possessorium and attornment to replace traditio or traditio brevi manu, sellers have to explicitly acquire the consent of buyers, and thus buyers are more likely to be aware of the pitfalls behind constructive delivery. This design is a majoritarian default rule (Ayres and Gertner 1989; 1999) because presumably many buyers (at least in certain industries/contexts) do not prefer to use nonphysical transfer of possession as delivery and may demand lower prices (to reflect higher risks) if constructive delivery is used. To receive higher prices, sellers would then prefer to use physical delivery (which is often easy to do and consistent with social understanding) by default. In certain types of transactions where stakes are high and information asymmetry is grave, a civil code may impose an altering rule, such as the agreement to use constructive delivery must be in writing. To the extent that the law allows parties to stipulate what constitutes traditio, parties can of course agree by themselves to use constitutum possessorium or attornment, even though the law is silent on it. This is also undesirable, as lawmakers can easily stipulate constitutum possessorium and attornment as menus so that transacting parties can easily opt in.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

transfer of ownership

121

To consider the issue more broadly, we can adopt either intention or actual delivery as default rules or menus. That is, a statute can stipulate actual delivery as the default rule, whereas laying out ways in which transacting parties can easily opt into an intention-based method of transferring ownership. Conversely, the ownership transfer rule regarding movables can be, by default, intention-based whereas parties can easily opt into an agreement that prescribes either actual or constructive delivery. Lawmakers around the world, however, so far only stick to one set of default rules, without considering menus. While it is hard to make a case for the superiority of any default rule for transferring ownership of movables, it is likely suboptimal not to provide any menu option.

III Conclusion In some issues, such as when and how ownership of movables change hands, the lawmaking decisions may not matter that much in practice. In such scenarios, it is not surprising that path dependence is a major factor in determining what kind of doctrines is applied. In other issues, such as the requirements and effects of real estate registration, it largely depends on the legal infrastructure of a particular country. While it is easy to copypaste the civil code of France or Germany, it is hard to import the notary tradition in France, and, needless to say, building up and maintaining a registry like that in Germany takes an enormous amount of resources. In short, there may be no one-size-fit-all registration system. Due to the lack of accurate measurements of the legal infrastructure in each country, this chapter will not evaluate whether adoption of the public faith principle or not, whether registration-of-right or recording-of-deed, or whether the constitutive effect or the opposability effect is necessarily economically superior. That said, this chapter holds the view that, to a large extent, registries should be an information-clearing house, and property-rightrelated information should be pooled to registries. Constructive notice outside of registration should be avoided as much as possible.

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.006 Published online by Cambridge University Press

PART II Immovable Property

https://doi.org/10.1017/9781009236553.007 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.007 Published online by Cambridge University Press

5 Acquisitive Prescription Hardly Justified in Modern, Developed Countries

This chapter surveys and analyzes the peculiar property doctrine known as “acquisitive prescription”1 in civil law,2 and “adverse possession” under common law. It is peculiar, at least in the case of adverse possession (a subtype of acquisitive prescription in the scheme of this chapter), because adverse possessors do not have to pay “erstwhile owners”3 to acquire ownership – by definition, there are no valid sale contracts between them; erstwhile owners and adverse possessors may not even know of one another’s identities or existence until litigation is initiated. Acquisitive prescription (which will be used as the umbrella term for reasons explained below) is exceptional in private law because ownership changes hands without the consent of the erstwhile owners – a practice which goes against the general policy in favor of voluntary transactions.4 What justifies 175 of the 202 surveyed jurisdictions (87%) in maintaining acquisitive prescription in modern times? Is path dependence from Roman law (called usucapio) to blame? China’s new civil code, enacted in May 2020, notably contains no stipulations regarding acquisitive prescription – a conscious decision by its drafters. Is this absence an illadvised innovation in private law with “Chinese characteristics” or a wise trailblazing decision? 1

2 3

4

Prescription, according to Black’s Law Dictionary, refers to “the effect of the lapse of time in creating or destroying rights.” Acquisitive prescription thus means acquiring rights (here, property rights) after a certain period of time. For instance, the Quebec §2917 and the Louisiana §3446 use this term. This chapter follows the draft Restatement of the Law Fourth, Property, in calling the original owner whose land has been occupied by a possessor for a long period of time the “erstwhile owner,” even though in some illustrations the owner would not lose ownership. But Peñalver and Katyal (2007: 1143–1144) justify the re-distributive value created by property outlaws such as adverse possessors. Alexander (2008: 789–790) echo Peñalver and Katyal’s argument. Underkuffler (2011) contrast homeless adverse possessors and strategic adverse possessors.

125

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

126

Part II: Immovable Property

This chapter is the most thorough survey of acquisitive prescription law in the literature to date,5 and contributes to better understanding of a legal doctrine that has existed for two millennia. As to policy, this chapter reconsiders the justifications offered for acquisitive prescription. Much prior literature in English focuses on the adverse possession doctrine in American common law. The analytical findings, therefore, are not readily applicable to acquisitive prescription doctrines in other countries. This chapter argues that acquisitive prescription based mostly on possession, rather than registration (called adverse possession in this chapter), as is the case in most civil and common law countries, cannot be justified on efficiency grounds in modern times,6 whether the legal system has a registration-of-right or recording system for land rights. More efficient regimes should require an attempted transfer of ownership (that turns out to be defective), the registration of such a title (or ownership), and good faith. Lawmakers should update their antiquated legal doctrine or cite a noneconomic justification for acquisitive prescription. In this regard, American common law is largely inefficient. In the United States, adverse possession is often used to resolve boundary disputes (sometimes called building encroachment) (Singer 2010: 156)7; this type of dispute is better left to a separate doctrine8 that incorporates considerations specific to boundary disputes (see Chapter 5). With boundary disputes excluded, the scope of the adverse possession doctrine regarding good-faith possessors is limited only to situations involving color of title (defective written instrument). This chapter is structured as follows: Part I provides separate typologies for registration-based and possession-based acquisitive prescription regimes. Part II explores the economic justifications for the several main types. Part III concludes. 5 6 7

8

For another large-scale comparative project on adverse possession law, see Dari-Mattiacci et al. (2016). Cf. Peñalver and Katyal (2010) justify adverse possession on non-efficiency grounds. For American cases that do not involve boundary disputes, see, for example, Howard v. Kunto, 3 Wash.App. 393, 477 P.2d 210 (1970); Paine v. Sexton, 88 Mass.App.Ct. 389, 37 N.E.3d 1103 (2015). I thank Eduardo Peñalver for bringing these two cases to my attention. Howard v. Kunto is a particularly interesting case, as due to systemic surveying errors, many landowners in the neighborhood held deeds that describe their neighbors’ land. Technically speaking, the possessors did not have even apparent title, though they have valid title to an adjacent plot. This case, I would argue, does not defeat my argument that as a general matter (in terms of law) purely possession-based acquisitive prescription is not warranted, but it reminds us of the import of having equity (as meta-law) to resolve outlier cases like Howard v. Kunto. Meier (2015: 51) treats mistaken boundary encroaching as a separate category. Miceli and Sirmans (1995a: 162–164) note the rent-seeking problem in boundary encroachment cases.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

127

I  Comparative Overview This part surveys acquisitive prescription law around the world.9 Acquisitive prescription regimes are divided into two major types: ­registration-based and possession-based. Conceptually, the two types are mutually exclusive, but the legal system of a country such as Germany can contain both types. Almost all registration-based acquisitive prescription regimes also require possession, while some possession-based acquisitive prescription requires registration of possession, which means the fact of someone’s possession is reflected in the registration records. Registration of possession is different from registration of ownership or title for ­registration-based systems. If a country requires registration of ownership or title, even if it also requires possession, this chapter classifies it as registration-based. Thus, in my methodology, no possession-based system requires registration of ownership or title, but a registration-based system may (and, in fact, often does) require possession. Adverse possession should not be conflated with the property doctrines covered in this chapter because possession is a precondition for adverse possession, as its name suggests. In my survey, at least one country has adopted a registration-based acquisitive prescription that does not require possession at all. This chapter will further challenge the role of possession in acquisitive prescription. The adversity requirement in adverse possession is defined as nonpermissive use (Singer 2010: 149), but many countries require, for instance, a conveyance (that turns out to be defective) from the erstwhile owner qua seller to the buyer qua acquisitive prescription claimant (more on this later) that purportedly gives title. In this situation, the buyer’s use is permissive, at least before the seller finds out about the defect and reclaims ownership. The buyer in this situation does not, strictly speaking, possess adversely. This chapter uses the term 9

This chapter ignores the technical differences listed below. First, after the statute of limitations runs, some jurisdictions like Quebec §2918 and Taiwan §769 require a judicial application, whereas in some countries, adverse possessors can automatically acquire ownership after the prescription period – for example, South Africa (Van der Merwe et al. 2002: 230) and some states in the United States. Second, in many countries, possessors acquire ownership, while in some countries, erstwhile owners can no longer recover possession of their land, but possessors do not acquire ownership (Stake 2001: 2422 fn.15; Hayton and Matthews 2007: 46–47; Jansen 2012). DCFR VIII.–4:301(1) stipulates that unless a possessor acquires ownership of movables, an original owner does not lose ownership. The draft Restatement of the Law Fourth, Property, §1.2.2.1, §1.2.2.2, and the accompanying comments emphasize that the statute of limitations and the requirements for adverse possession are aligned. That is, if any of the requirements for adverse possession are not met, the statute does not run. If the statute of limitations does not run, adverse possession will have no chance to apply.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

128

Part II: Immovable Property

adverse possession as an alternative label for possession-based jurisdictions that have no title requirement. The concept of (just) title is complicated and worthy of further explanation. A good start to understanding it is through the lens of the Louisiana §3483: “A just title is a juridical act, such as a sale, exchange, or donation, sufficient to transfer ownership or another real right. The act must be written, valid in form, and filed for registry in the conveyance records of the parish in which the immovable is situated” (emphasis added). Argentina §1902 also provides a useful definition: Just title for acquisitive prescription is the one having as a purpose the transfer of a principal real right that is exercised by possession, with the formal requirements required for its validity, when its grantor is not capable or does not have authority therefor. The good faith required in a possessory relationship consists in not having known, nor to have been able to know, the lack of right thereto. When recordable things are involved, good faith requires the previous examination of the registry documentation and proof, as well as performing the pertinent acts of verification established in the respective special regime. (emphasis added) (cf. Aristi and Imbernón 2013: 110; Romañach Jr. 2015a)

The gist of (just) title, therefore, is that an acquisitive prescription claimant must have been a transferee in a prior conveyance. The claimant comes to claim acquisitive prescription because it turns out that a legal defect exists in the conveyance. If, for example, a seller, due to temporary mental illness, did not have the legal capacity to transfer her ownership to a transferee; alternatively, a middleman made a mistake and as a result an ownership transfer did not become effective in the eyes of the law. Under US common law, a similar rule is known as “color of title”10 or “apparent title,” according to Black’s Law Dictionary. In this chapter, I opt for the term “apparent title” because “color of title” is not intuitive to civil lawyers and “just title” may be misunderstood as a normative concept.11 Apparent title conveys the idea that a transferee, based on a prior conveyance, succession, or legacy, has reason to believe that she is (entitled to be) the owner, though it turns out her ownership is defective. As analyzed below, apparent title and possession, in addition to knowledge (good or bad faith), are the key elements in understanding the similarities 10

11

Merrill and Smith (2017: 171–172) explain color of title as “a person has some document – usually a deed, will, or judicial decree – that purports to convey title but does not in fact do so because of some legal defect.” DCFR uses “putative title.” Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference (DCFR), Full Edition, at 4192.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

129

and variances in acquisitive prescription around the world. For simplicity, below I often use “with(out) title” to mean “with(out) apparent title.” In most countries, there are acquisitive prescription regimes for both immovables and movables. This chapter focuses on those for immovables only. In Japan and Taiwan, among others, acquisitive prescription rules for land and buildings may differ. For simplicity, this chapter only summarizes rules for land. Oftentimes, within a jurisdiction, a possessor can prescribe both ownership and limited property rights.12 This chapter focuses only on acquisitive prescription of land ownership.13

A  Registration-Based Acquisitive Prescription Thirty-seven jurisdictions have registration-based acquisitive prescription,14 for which three elements are critical – registration of apparent ownership or apparent title, knowledge (good faith or bad faith), and possession. (To avoid repetition, title and ownership below means apparent title and apparent ownership.) To count as registration-based, a jurisdiction must require registration of either title or ownership. Registration of mere possession, under my scheme, does not count as registration-based. The difference between registration of title and registration of ownership is that, under my scheme, the latter requires that an acquisitive prescription claimant has been registered as the owner for a certain period of time, while the former requires that the cause of registration of ownership – the title – has to be the underlying legal instrument. That is, under registration of ownership, a random registration mistake may anoint a registered owner as de jure owner after expiration of the prescription period.15 12

13

14

15

82 of the 172 jurisdictions (48%) with possession-based acquisitive prescription for ownership also have rules regarding acquisitive prescription for limited property rights. Only one jurisdiction that does not have acquisitive prescription for ownership has rules regarding acquisitive prescription for limited property rights – Pakistan – for prescriptive easement. For simplicity, this chapter will not discuss acquisitive prescription regarding inheritance claims and waqf. Qatar §968 and Afghanistan §2280 have possession-based prescriptive acquisition regarding inheritance claim for 33 years. Qatar §967, Jordan §1183 (Hashem 1990: 179), Libya §974, and United Arab Emirates §1319 have special stipulations for waqf (prescription period being 30 or 33 years, much longer than that for ordinary objects). For an introduction to the concept of waqf, see Tang (2018: 2277–2281). Mongolia §104.2 stipulates rules for acquiring ownerless immovables after 15 years in possession in good faith and being registered as owners. As it limits the objects to ownerless immovables, it is not coded as recognizing registration-based acquisitive prescription. The German Federal Supreme Court explicitly held that registrants do not have to explain how they became nominal owners (BGH, Oct 29, 1971, docket number V ZR 122/68 at Beck-Rechtsprechung 31124133) (Picker and Herrler 2019: §900 Rn. 14).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

130

Part II: Immovable Property

Table 5.1  Registration-based acquisitive prescription: two dimensions Types of registration Registration of apparent title Knowledge Good Type A0. and bad Portugal and its former colonies: faith both Angola, Brazil, Cape Verde, allowed Guinea-Bissau, Macau, Mozambique, Sao Tome and Principe, and Timor-Leste. Good Type A1. faith only Argentina, Bolivia, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Equatorial Guinea, Finland, Honduras, Italy, Nicaragua, Panama, Puerto Rico, Romania, Scotland, and Spain.

Registration of apparent ownership Type B0. Estonia, Georgia, Germany, Sweden, and Turkmenistan.

Type B1. Azerbaijan, Liechtenstein, Slovenia, South Korea, Switzerland, and Turkey.

The difference between registration of title and registration of ownership may arise due to differences in registry type. All countries identified as requiring registration of ownership in Table 5.1 appear to have a registration-ofright system, whereas many (though not all) countries identified as requiring registration of title in Table 5.1 have a recording system. What type of registration system a country uses is often not specified in the civil code. My coding of the registration system is thus unreliable; nor does the World Bank Doing Business Report (discussed below) inquire into the registry type directly. I code a country as requiring registration of title or ownership based on the language in the acquisitive prescription doctrine. For instance, Spain and the Latin American countries that emulate Spain’s acquisitive prescription rules are coded as registration-based, and, in particular, requiring registration of title, despite the fact that Spain has adopted a registration-of-right system. This is because the Spain §1949 stipulates: “Ordinary prescription of ownership or rights in rem to the detriment of a third party shall not take place against a title registered in the Property Registry, unless it is pursuant to another title which has also been ­registered, and the time shall begin to run from registration of the latter” (Técnica 2009: 220) (emphasis added). Panama §1686, Colombia §2526, Chile §2505, El Salvador §2244, Puerto Rico §1849, and Equatorial

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

131

Guinea §1949, among others, have the same stipulation. Costa Rica §861 and Nicaragua §898 have the same stipulation albeit expressed differently. Table 5.1 uses knowledge and types of registration to divide all jurisdictions with a registration-based system into four types. Almost all jurisdictions require possession. Each type is further explained. Type A1. Spain §1949 & §1957,16 Equatorial Guinea §1949 & §1957, and five Latin American countries – Argentina §1898, Honduras §2286, Nicaragua §888 & §898, Panama §1694, and Puerto Rico §1857 – have very similar rules. In their ordinary acquisitive prescription regimes, four elements – title, registration of title, good faith, and possession – are all required. This is registration-based acquisitive prescription. In addition, a possessor who is either bad faith or without title must possess for a longer time than the ordinary acquisitive prescription regime requires. This is possession-based acquisitive prescription under my scheme. Italy §1159 is close to the Spanish model in requiring title, its registration, and good faith, but §1159 does not explicitly require possession. Nonetheless, because §1159 is located in the code chapter on possession, and both §1158 and §1159-bis on acquisitive prescription rules require possession, the possession requirement should arguably be implied in §1159 to warrant Italy’s inclusion in Type A1. Chile §2505, §2507 & §2510, Colombia §2526, §2529 & §2531, Ecuador §2406 & §2410, and El Salvador §2244 & §§2249–2250 developed slightly different regimes. Their registration-based acquisitive prescription laws also require all four elements – title, registration of title,17 good faith, and possession. Their “extra-ordinary prescription” (as it is called) is ­possession-based. Those without registered title, good or bad faith,18 may acquire ownership by possession-based acquisitive prescription (10 years in Chile and Colombia; 15 years in Ecuador; and 30 years in El Salvador). Other countries that fall into Type A1 include Bolivia §134, Costa Rica §853 & §§860–861, Finland (CRE) §XIII:10, Romania §§930–931 (Stanescu 2017: 117), and Scotland (PLSA) §1. 16

17 18

The relevant stipulations in Spain have been implicitly replaced by the Mortgage Act, according to the court in Sentencia TS (Sala 1.ª) de 21 enero 2014, Rec. 916/2011. I thank Prof. Vanessa Casado Pérez for this observation. Note here that while Ecuador has a registration-of-right system, its civil code uses the term “registration of title.” In Ecuador §2410 & El Salvador §2249, a custodian is presumed bad faith and thus unqualified to acquire ownership by prescription, unless custodians can establish themselves as possessors by meeting the following two requirements: (1) erstwhile owners cannot prove that possessors during the prescription period recognize the former’s ownership, and (2) possessors have possessed continuously without violence.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

132

Part II: Immovable Property

Type A0. Registration-based acquisitive prescription in Portugal §1294 and its former colonies – Brazil §1242 Sole Paragraph, Angola, Cape Verde, Guinea-Bissau, Macau §1220, Mozambique, Sao Tome and Principe, and Timor-Leste §1214 – require title, registration, and possession. A possessor who is good faith (10 years), as compared to bad faith (15 years), saves 5 years in the prescription period. Type B0. Registration-based legal systems that do not require title and its registration appear to be those affected by German property law (Chang et al. 2021). Germany §900 emphasizes the importance of an acquisitive prescription claimant’s being registered as the owner and being in possession. Georgia §167, Estonia (TLPA) §123, South Korea §245, and Turkmenistan §188 follow this model. In the registration-based system in Sweden (LC) §XVI:1, which adopts a registration-of-right system with only the opposability effect, a shorter, 10-year prescription period applies if a possessor in good faith with title has registered ownership, whereas a longer, 20-year prescription period applies if a possessor either lacks good faith or is without title has registered ownership (von Bar 1999: 532, 597; Jensen 2010). Type B1. Slovenia (Škerl and Vlahek 2010: 136–137), Switzerland §661, Liechtenstein (SR) §42, and Turkey §712 are close to the German model but require good-faith possession. Azerbaijan §178.5 is idiosyncratic in that it is close to the Swiss model but does not explicitly require that an acquisitive prescription claimant be in possession. Azerbaijan (and perhaps Italy) is the rare country that does not require possession.

B  Possession-Based Acquisitive Prescription Costa Rica, Finland (Hollo 2019: 153), Georgia, Sweden (Santisteban and Sparkes 2015: 309), and Turkmenistan have registration-based, but not possession-based, acquisitive prescription. Another 32 countries listed in Table 5.1 have an acquisitive prescription regime that is both registrationand possession based. A total of 97 jurisdictions have only possessionbased acquisitive prescription, whereas 22 jurisdictions – for example, China (Chang 2012b), Singapore (See et al. 2018: 297), and Pakistan19 – have neither. See Figure 5.1. 19

In Pakistan, adverse possession of ownership is against the principles of Islam. Plea of adverse possession has been declared against the injunctions of Islam by the judgment of Hon’ble Shariat Appellate Bench in a case titled “Maqbool Ahmad vs. Govt. of Pakistan” (1991 SCMR 2063).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

133

AP classification Both Possession-based Registration-based No AP Not coded

Figure 5.1  Adoption of registration- and possession-based acquisitive prescription Notes: A total of 235 jurisdictions are included in the graphs. “No acquisitive prescription” means that no acquisitive prescription rule is found in the jurisdiction. “Registration-based” means that the jurisdiction only adopts registration-based acquisitive prescription. “Possession-based” means that the jurisdiction only adopts possession-based acquisitive prescription. “Both” means that the jurisdiction adopts both registration- and possession-based acquisitive prescription.

Unlike in a registration-based system, where apparent title is either required or not, in a possession-based system, apparent title could be required, prescription-period-shortening, or not required. American common law is a prime example. Some states require apparent title as an element of adverse possession, but most do not – although in 15 states, the statute of limitations is shorter with apparent title.20 Table 5.2 shows the several functions apparent title serves in possession-based acquisitive prescription. (Again, to avoid repetition, title below means apparent title.) Finally, some statutes explicitly allow only unregistered immovables to be subject to acquisitive prescription; they are marked with * below. More specifically, all but one Type-E0 jurisdictions allow goodfaith possessors with title to enjoy a shorter prescription period. That is, good-faith possessors with title enjoy a shorter prescription period, while bad-faith possessors with title, bad-faith possessors without title, and good-faith possessors without title face the same longer prescription period. They include Albania §§168–169, Algeria §828, Brazil §1238 & §1242, Democratic Republic of Congo (GLC) §§I:647–648, Egypt 20

The 15 states are Alabama, Alaska, Arizona, Colorado, Georgia, Illinois, Kentucky, Louisiana, Michigan, North Carolina, North Dakota, South Carolina, Texas, Washington, and Wisconsin. See Draft Restatement of the Law Fourth, Property, §1.2.2.3 (Reporters’ Note).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

Part II: Immovable Property

134

Table 5.2  Typology of possession-based acquisitive prescription Apparent title requirement Reduces Spain et al. and Knowledge Not required Required prescription period Chile et al.* Total Good faith only Good faith < bad faith Good faith = bad faith

Type C2. 19 Type C1. 15 Type C0. 39

Type D2. 5 Type D1. 0 Type D0. 0

Type E2. 6 Type E1. 0 Type E0. 33

Type F2. 0 Type F1. 0 Type F0. 11

Total

73

5

39

11

30 15 83 128

Notes: * See Part I.A for explanations of the 11 Type F0 jurisdictions (Argentina, Chile, Colombia, Ecuador, El Salvador, Equatorial Guinea, Honduras, Nicaragua, Panama, Puerto Rico, and Spain) under Type A1.

§§968–969, France §§2272–2275, Greece §1041 & §1045, Guinea §784 & §787, Haiti §2030 & §2033, Jordan* §1182, Libya §§972–973, Louisiana §3475 & §3486, Malta §2140 & §2144, Mauritius §§2261–2263, Monaco §2094, Paraguay §1989–1990, Peru §950, the Philippines §1117 & §1137, Syria* §§917–918, Tunisia (CDR) §§45–46, United Arab Emirates* §§1317–1318, Uruguay §1204 & §1211, and Venezuela §1977 & §1979, as well as §2262 & §2265 in Belgium, Burkina Faso, Comoros, Dominican Republic, Ivory Coast, Luxembourg, Niger, Seychelles, and Togo. Hungary §V:45 is an outlier because it does not explicitly require possessors who have title to be good faith. That is, possessors with title, good or bad faith alike, enjoy a shorter prescription period, while possessors without title, good or bad faith alike, must wait longer. The six Type-E2 countries, Croatia (AOORR) §18 & §159, Indonesia §1963, Iraq* §1158, Macedonia (AOORR) §124, Serbia (LBOPR) §28,21 and Suriname §1984, allow only good-faith possessors to prescriptively acquire ownership, and those with title enjoy a shorter prescription period. Type D2 includes Cuba §186 (urban immovables only), Czech §1090, Guatemala §620 & §633, Latvia §999 & 1024, and Scotland (PLSA) §2. Type C1 contains 15 countries that require a longer prescription period for bad-faith possessors. Notably, several countries among them require the registration of possession. Portugal §1295 and its former colonies – for 21

Note that the Serbian law and the Macedonian law, at least in their English translation, are very similar in wording and structure.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

135

example, Timor-Leste §1215 and Macau §1220 – treat registration of possession as a precondition for possession-based acquisitive prescription, and registration of possession is a judicial procedure, before which a claimant must be in possession for at least five years. Another model is Mexico §1152, where registration of possession appears to be an independent means of acquisitive prescription.22 Also included in this group are Bulgaria (OA) §79, Cambodia §162, Japan §162, Poland §172, Taiwan* §§769–770, and Ukraine §344. Epstein (1986: 685–689) advocates a twotiered statute of limitations requiring longer prescription periods for badfaith claimants.23 This mechanism has been realized in these countries. The 39 Type-C0 jurisdictions, by contrast, treat good- and bad-faith possessors equally in terms of prescription period. Equal treatment means these jurisdictions basically omit the requirement regarding knowledge. That is, the distinction between good faith and bad faith, unlike in the many other jurisdictions summarized above, is not embedded in the acquisitive prescription doctrine. These jurisdictions include Afghanistan §2279, Australia (Hepburn 2002: 109–110), Azerbaijan* §178.6, Bahrain* §903, Benin (Loi 2013) §30, Bolivia §138, Bosnia & Herzegovina (UN-HABITAT 2005: 54), Burundi §§III:647–648 (the notes in its Civil Code explicitly state that the design is not the same as the Napoleonic Code), California §1007 & (CPC) §318 & §325, Cyprus* (Neocleous 2000: 599), England (Sparkes 2019: 160) (almost impossible to acquire registered land through prescription), Ethiopia §1168, Germany §927, Hong Kong (LO) §7 (Goo and Lee 2019: 326), India (Mitra 2007: 376), Ireland (de Londras 2011: 377), Israel (PL) §5 (acquiring registered land requires a longer prescription period) (Israel Law Resource Center 2007), Italy §1158, Kuwait §935, Malawi (RLA) §134, Malaysia* (Maidin and Ali 2014: 170), Mongolia §105, New Zealand (Williams 2011: 73–74), Nigeria (Smith 2018: 155), Oman (LL) §§13–14 (Keesee 1982), Ontario* (LA) §51 (Ziff 2014: 140), Qatar §966, Quebec §2918, Romania (Stanescu 2017: 116–117) (explicitly pointing out that possessors need not be in good faith), Slovakia §134, South Africa (Van der Merwe et al. 2002: 231), South Korea §245, Sri Lanka (Rupesinghe and Rupesinghe 2007: 63), Switzerland §662, Tajikistan §258, Thailand §1382,24 Uganda (Bakibinga 2018: 25), and Zimbabwe (PA) §4. Among them, notably, New York (RPAPL) §501 stipulates that if record owners 22 23 24

Notably, almost no Mexican states’ civil codes contain this design. I thank Rafael Ibarra Garza and Cynthia Alejandra Sánchez Torres for this information. Ellickson (1986: 733–734) discusses and agrees with the merit of Epstein’s proposal. §1383 may require bad-faith possessors to possess for a longer period of time before being qualified as owners.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

136

Part II: Immovable Property

AP classification F0 E2 E0 D2 C2 C1 C0 No AP Not coded

Figure 5.2  Types of possession-based acquisitive prescription Notes: A total of 202 jurisdictions are included in the graphs. This map has 33 fewer observations than the previous one, because only 17 US states are included (they are California, New York, and 15 additional US states mentioned in footnote 20). For the classification, see Table 5.2.

cannot be ascertained in the registry, possessors can be bad faith, as claim of right is not required. If record owners can be ascertained, possessors must have a claim of right. In the case of New York, a claim of right is like a combination of good faith and apparent title.25 This is not entirely the same as seen in California §1007 and (CPC) §318 & §325. Type C2 includes 19 jurisdictions: Armenia §187, Austria* §1468, Belarus §235, Denmark (von Bar 1999: 32) and Norway (Feys 2006: 20),26 Eritrea §979,27 Estonia (OA) §111 & §124, Kazakhstan §240, Kyrgyzstan §265, Laos (PL) §42, Liechtenstein* (SR) §43, Lithuania* §§4.68–4.69, Moldova §332, the Netherlands §III:99, Russia §234, Slovenia (Škerl and Vlahek 2010: 135), Turkey* §713, Uzbekistan §187, and Vietnam §236. In these jurisdictions, only good-faith possessors may acquire ownership by prescription, and titles are not required. Figure 5.2 summarizes the aforementioned classifications in a world map. Figure 5.3 gives an overview of the typology of possession-based acquisitive prescription. 25

26 27

This chapter agrees with the draft Restatement of the Law Fourth, Property, that a few existing concepts related to adverse possession could be overlapping and that they have been often manipulated to reach the right results in particular cases. It is thus important to streamline the preconditions for adverse possession. I thank Víðir Smári Petersen for the observation on Norway and Denmark. §985 is ignored in the coding, as I do not understand its meaning.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

Figure 5.3  Typology of possession-based acquisitive prescription

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

137

138

Part II: Immovable Property

II  Modern Justifications for Acquisitive Prescription Part I surveys laws on the books around the world. The prevalence of acquisitive prescription is not hard to explain. It has been a venerated doctrine since the Roman law, and received by developed European countries when they started to build modern private law systems. The doctrine was later inherited by other countries through colonization or voluntary imitation. Lawmakers have kept or borrowed it because, to this day, a civil dispute may occasionally be resolved through the acquisitive prescription doctrine. How can acquisitive prescription be justified in modern times?28 Adverse possession may have been efficient in premodern times, but in the 21st century – the age of GPS, block-chain, Google Earth, etc. – acquisitive prescription must be rejustified, if possible, on new grounds. This part is divided into several sections, each dealing with a particular type of acquisitive prescription regime explored in Part I. In the following sections, when I refer to a system that requires title, I mean either (1) a country with a recording system that requires property deeds that convey ownership to be recorded or (2) a country with a ­registration-of-right system that requires a claimant of acquisitive prescription to be registered as an owner due to a conveyance of ownership. This title requirement is meant to broadly include other legitimate sources of ownership, such as through inheritance, but to exclude parties who record forged deeds or who have been registered as owners due to pure mistake by registries. I define title in this way due to my view that acquisitive prescription in modern times is better reserved and used to redress hardship caused by defects in an otherwise legitimate transfer of ownership, but not to give irrelevant persons a free pass to ownership.

A  Registration-Based, with Good Faith and Apparent Title Third-party purchaser protection is one, but not the main, function of registration-based acquisitive prescription. Protecting registration-based acquisitive prescription could reduce third-party information costs. If purchasers check the information in registries but are still bound by unregistered adverse interests, registration-based acquisitive prescription could give purchasers a fresh start after the statute of limitation runs. 28

Lovett (2017a) defends the adverse possession doctrine in the U.S. from the late nineteenth century to the late twentieth century. Lovett (2017b) discusses the adverse possession doctrine in the specific context of Louisiana Civil Code.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

139

Nonetheless, given that the limitation period is usually quite long, purchasers can hardly count on registration-based acquisitive prescription; thus, they still must expend information costs. In this regard, much more useful is the public faith principle (see Chapter 4), under which purchasers relying on information contained in a registry regarding who has property rights and how a piece of land is encumbered by adverse interests will be protected against holders of unregistered adverse property interests. Yet, the public faith principle has not been universally adopted. In the United States, for instance, purchasers are bound by adverse interests of which they have actual or constructive notice. A recorded deed is only one way to give constructive notice. An adverse interest that can be discovered with “reasonable inquiries” or “standard due diligence”29 still binds subsequent purchasers. Purchasers, or the professionals they hire, may have reasonably attempted to discover such adverse interests but fail. Registration-based acquisitive prescription is a long shot to have a fresh start, but it is better than nothing for purchasers. The registration-based acquisitive prescription regime is better justified in its protection of nominal owners against the party from whom ownership is conveyed. (Here, a nominal owner is better described as a “second party,” not a “third party,” because a dispute arises between a pair of transacting parties, like a buyer and a seller.) Especially in jurisdictions where registration is “constitutive” (see Chapter 6), buyers, ex ante, are usually aware of the importance of having their names entered in the registry. Buyers often do not have the expertise to verify whether all aspects of the transaction are sound, but they can usually tell by common sense whether their names are in the registry book. A registry in a constitutive system usually issues an ownership certificate to the registered owner.30 If buyers can never rely on the official ownership certificate or registration records as a definitive sign that the sale has been consummated, transaction costs of conveyance will increase significantly.31 This function of protecting a 29 30

31

See Draft Restatement of the Law Fourth, Property, §5.4.3.2. Registries in 8 out of the 15 jurisdictions with available World Bank data and with this acquisitive prescription regime, however, do not deliver legally binding documents that prove ownership. This does not decrease transaction and information costs. This feature of their registries suggests that this acquisitive prescription regime is not the best fit in these jurisdictions. Good-faith registrants must return the land before the prescription period runs. Depending on the contents of other doctrines, these registrants may or may not be compensated for improvements made. If not, the prospect of losing investment value may lead to underinvestment on the land.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

140

Part II: Immovable Property

second party is different from that of protecting a third party, because, unlike in the latter case, it does not concern the inaccuracy of information regarding property rights in the registries, but rather defects that arise typically in a conveyance. If this is indeed the case, why not make registration of ownership an immediate cure for any transactional defects? That is, why not set the statute of limitations for good faith, with-title, registration-based acquisitive prescription at zero days? In jurisdictions with well-functioning ­registration-of-right systems, land sellers must be registered owners (and very often actual owners). Conveyances by self-proclaimed owners, whose names are not in the registry, will be rejected by registrars. The most likely scenario in which this regime is applicable appears to be that a seller in a deal lacks the legal capacity or authority to sell, or a deed is not properly acknowledged by a notary. Here, a prescription period of zero days would entirely defeat the point of making the conveyance void or voidable in the first place. At zero days, a buyer may not have relied on the fact of being the new owner, whereas a seller or her representative does not have an opportunity to redress the problem. For instance, a textbook example would be that a seller becomes temporarily mentally ill and the contract she makes is void, making the buyer’s title only an apparent one. A prescription period of zero days would render the stipulations regarding legal capacity and authority useless. Zero days are too short, and decades too long, to protect registered owners. Lawmakers must weigh two considerations, but the balance chosen by the jurisdictions strikes me as too long. The minimal32 prescription periods stipulated by the 16 jurisdictions with good faith only, with-title, registration-based acquisitive prescription are 5 years (5 jurisdictions) and 10 years (11 jurisdictions). If any party is at fault, it is not buyers, as they are good faith, that is, the law already considers them as not questioning the validity of title for good reasons. Buyers have engaged in a genuine transaction and gone through the salient registration procedure. By contrast, sellers convey their ownership without legal capacity or authority. Sellers may do so knowingly; sometimes, however, as in the case of mental illness, sellers are not at fault. No studied jurisdiction varies the prescription 32

Prescription periods could be inflated by other factors. Here the minimal length is used. One such factor is whether an erstwhile owner is absent. The original France §2265 double the length of the prescription period when erstwhile owners are absent. This was transplanted by, for example, former French colonies, Spain §§1957–1958, Ecuador §2408, Colombia §2529, and El Salvador §2247.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

141

period by whether sellers are at fault, whether third parties on the sellers’ side are affected, or whether sellers receive reasonable consideration (such as market value). Long acquisitive prescription periods are problematic because goodfaith registrants may sell “too early” and fail to meet prescription requirements. Suppose a good-faith registrant has sold the land in question after, say, three years of being registered as the owner. At the time, the statute of limitations has not run. The erstwhile owner learns about the title defect after the statute of limitations runs, say 12 years after the defect occurs. When the erstwhile owner sues the good-faith registrant for unjust enrichment,33 the latter cannot draw on the acquisitive prescription doctrine as a defense, as the registrant has not been registered as the owner for a long enough period. This may suggest that in countries with registration-based acquisitive prescription, the rule better hinges on extinctive prescription instead of acquisitive prescription – erstwhile owners cannot claim against good-faith registrants after a certain number of years. Nonetheless, even when the law has switched focus from how long the good-faith registrant is registered as the owner, to how much time has elapsed since the erstwhile owner was not registered as the owner, goodfaith registrants still will not be protected if the acquisitive prescription doctrine continues to require possession. Suppose the law were changed to prescribe extinctive prescription after ten years. The first purchaser, with apparent title, sells to the second purchaser after three years.34 An additional nine years afterwards, the erstwhile owner sues the first purchaser for unjust enrichment. The first purchaser cannot draw on the protection afforded by the acquisitive prescription doctrine because she has been in possession for at most three years. As a result, the first purchaser must disgorge the presumably higher price (paid to her by the second purchaser) through an unjust enrichment action brought by the erstwhile owner. The ex post disgorgement is simply a redistribution of wealth and is irrelevant for efficiency. Nonetheless, ex ante, good-faith purchasers/ registrants have incentives to increase verification efforts to reduce the 33

34

A successful unjust enrichment claim essentially gets the erstwhile owner the difference between the higher prices the registrant received from the third party and the usually lower prices the registrant had paid. In jurisdictions with the public faith doctrine, the erstwhile owner cannot re-claim landownership when good-faith third parties have acquired land from registered owners. As Arruñada (2017) and Arruñada et al. (2019) point out, a distinctive feature of property law is that it involves sequential exchange. When thinking about property law, one has to go beyond a two-party setting and take into account a third transacting party.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

142

Part II: Immovable Property

probability of losing their profits, or to withhold sales before the statute of limitations runs. The additional verification and wait create social waste.35 If this is considered a normatively undesirable result, there are three possible solutions. First, the prescription period should be much shortened, as argued above. In this scenario, the prescription could still be acquisitive prescription and the possession requirement may not inflict much harm and arguably bring some social benefits occasionally. Second, the law should recognize reverse-tacking. A total of 116 jurisdictions recognize tacking of possession, but none allows reverse tacking. Tacking means that subsequent possessors can combine the possession period of previous possessors in calculating the needed prescription period.36 Reverse tacking means that previous possessors can combine the possession period of subsequent possessors. For the first purchaser in the aforementioned hypothetical example, acquisitive prescription would be a valid defense against the original seller if reverse tacking were allowed. Here, as in the first solution, possession can still be required. Third, the law should not require possession in registration-based acquisitive prescription. As described above, only Azerbaijan (and perhaps Italy) does not require possession, but what purposes do possession serve in a registration-based acquisitive prescription regime? If decreasing genuine buyers’ transaction cost is sufficient to justify this regime in economic terms, the possession requirement imposes unnecessary social cost in verification and waiting. The problem is whether the social costs are justified by the benefit brought by the possession requirement. The best argument for social benefit is as follows: if neither the original seller nor the purchaser with apparent title possesses the land, but rather a third person possesses the land, allowing the purchaser to acquire ownership via prescription does not necessarily enable her to use the land. Requiring possession ensures that the party with the apparent title will be able to use the land, thus increasing efficiency. That said, the expected social benefits of the possession requirement are small. First, in the above scenario, keeping the land in the non-possessory original seller’s hands does not enable him to use it, either. Eventually, 35

36

Note that the logic here does not apply to possession-based acquisitive prescription, as possessors are not able to sell their occupied land before becoming owners, whereas, before the legal defect is exposed by the erstwhile owner, registrants are the (nominal) owner and can transfer their property rights. Tacking is one of the key issues in Howard v. Kunto, supra note 7.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

143

the actual possessor may acquire ownership via the p ­ ossession-based regime. Second, though the concept of possession varies in intension and extension across jurisdictions (Chang 2015b; 2016c), it is conceivable that many good-faith registrants under most definitions of possession will still fulfill the continuous possession requirement when they are the registered owners. Put differently, good-faith registrants, believing that they are the true owners, will possess anyway. Yet in cases where good-faith registrants have, say, sporadic possession, they will not be entitled to prescribe ownership. In sum, title-required, good-faith, registration-based acquisitive prescription serves an important function in reducing transaction costs on the buyers’ side. Nonetheless, two fault lines persist. First, a lengthy prescription period does not seem to strike the right balance, and may consequently fail to achieve the goal of saving transaction costs by imposing a lot of risk on good-faith registrants. Second, the possession requirement appears to create more social costs than social benefits. The possession requirement, thus, should be taken out from the registration-based regime. All jurisdictions presumably would prefer to reduce transaction costs to facilitate voluntary exchanges. In those with registration-of-right systems, a title-required, good-faith, registration-based acquisitive prescription doctrine can reduce transaction costs. Yet, only 16 jurisdictions have adopted such. While possession-based acquisitive prescription could protect some of the good-faith registrants, that is not always the case. Moreover, as of now the possession-based acquisitive prescription doctrine usually requires an even longer prescription period; thus, it is not a substitute for the registration-based doctrine.

B  Registration-Based, with Bad Faith and Apparent Title As a general matter, for reasons of efficiency, this chapter is in favor of abolishing acquisitive prescription for bad-faith possessors altogether. Portugal, for example, allows bad-faith registrants with title, after a longer period, to become permanent owners. The explicit accommodation of bad faith, registration-based acquisitive prescription is puzzling. Given that registrants have title, they must know who the sellers are, at least at one point in time. Once they are aware of the defect, should property law encourage registrants to wait for the prescription period to pass? Probably not, because bad-faith registrants know that their legal ownership is precarious and thus would refrain from investing in the land. If bad-faith

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

144

Part II: Immovable Property

registrants cannot prescriptively acquire ownership, nor be compensated for improvements made, they are likely to reach out to erstwhile owners to renegotiate as soon as possible, either to get their money back or secure outright ownership through a new agreement – both of which are routes to eliminate uncertainty. Bad-faith registrants’ conscious decisions not to bargain (Merrill 1985: 1135) will lead to doubts as to whether they value the land more than erstwhile owners.37 Of course, good and bad faiths are not given – possessors who fail to actively verify title would be considered good faith (Kim 2004: 2–3). Here, as in other property doctrines studied in this book, bad faith should mean “know or should have known,”38 rather than being coterminous with “know.” A registrant who is simply ignorant of the title defect, but who could have easily discovered it, should be classified as bad faith. Such an interpretation of bad faith should filter out most intentionally innocent registrants. In court, debating whether a possessor is in good or bad faith increases litigation cost (Depoorter 2010: 186). Although the litigation cost of proving good or bad faith is a social cost, under-investment during the limbo of the prescription period is another kind of social cost. Moreover, presumption of good or bad faith in practice resolves (or kills) much of the evidentiary issue and keeps litigation costs in check.39 If title defects are not attributable to registrants, it seems right to presume good faith, whereas the presumption of bad faith is right when title defects are attributable to registrants. The draft Restatement of the Law Fourth, Property, goes in another direction. Good and bad faith will not be distinguished in the adverse possession doctrine as such a consideration is deferred to the laws of equity and restitution. In terms of results, bad-faith adverse possessors are still unlikely to become owners simply after the passage of time (Helmholz 1983).40 For the many jurisdictions that do not have a separate equity system, this approach does not fit into their civil code structure. To be sure, there are many general principles in civil codes, and unjust enrichment is often an integral part. Not all civil-law courts are willing or able to carve 37 38

39 40

Epstein (1986: 686) puts it more colorfully: bad faith possessors “are both bad people in the individual cases and a menace in the future.” All doctrines that incorporate the element of good faith have the same problem of requiring and assessing optimal good faith. The good faith purchase doctrine is a prime example and shares the same concern (see Chapter 10). One more reliable proxy is to use adverse possessors’ paying taxes to presume good faith. Lovett (2017a) and Peñalver and Katyal (2010: 250) recount the debate surrounding Helmholz’s claim.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

145

out bad-faith adverse possession from the general principles or the law of unjust enrichment.41 Where the case against bad-faith adverse possessors is strong enough, it is more desirable for lawmakers in countries without an equity tradition to explicitly exclude bad-faith possessors and registrants from gaining ownership.42

C  Possession-Based, with Good Faith and Apparent Title Protecting good-faith possessors with title, like protecting those in a ­registration-based system, can be well justified. These regimes provide a “safety valve” (Smith 2009a: 2128–2129) for the usually rigid formality requirement in land conveyance. In jurisdictions with a recording rather than registration-of-right system, a buyer may commission a professional title search and yet fail to uncover a broken chain of title. A possessionbased regime with the apparent title requirement is a valuable safety valve for a genuine, faultless purchaser.43 The case for this type of acquisitive prescription is a little weaker than the registration-based counterpart because registration of title is not required. As shown in Table 5.2, five jurisdictions require title in their possession-based regimes and do not enact a registration-based system. Scotland, already with a registration-based acquisitive prescription regime, adopts an additional possession-based regime. Latvia and Czech have had a constitutive regime but have chosen this possession-based system over a registration-based system, which is puzzling. Six countries allow only good-faith possessors to acquisitively prescribe ownership, and good-faith possessors with title enjoy a shorter prescription period. Similarly, 33 countries allow both good- and bad-faith possessors to acquisitively prescribe ownership, and the good-faith ones with title enjoy a shorter prescription period. As explained above, favorable treatment to possessors with title makes sense. Nonetheless, even in this case, the prescription periods – 5, 7, 10, 15, or 20 years – appear too lengthy. Why do these jurisdictions not have title registration as a 41

42 43

At the very least, civil law courts should draw on general principles such as abuse of right to thwart certain bad-faith possessors from becoming legal owners. Smith (2021a) calls this approach meta-law. The preceding logic applies to bad-faith possessors with title but failure to register it. Title insurance will not disappear because before the statute of limitations runs, buyers qua possessors are not protected by the acquisitive prescription regime (Fennell 2006: 1083). On the other hand, title insurance is not a panacea, as it is unlikely to cover all the expected profits on buyers’ side.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

146

Part II: Immovable Property

necessary condition? In 21 of the 37 jurisdictions with World Bank’s 2016 Transparency of Information Index,44 the registries do not deliver legally binding documents to prove ownership. With weak registries, it is pointless to require title registration. Still, many countries could better ensure that acquisitive prescription is used wisely if possessors have to register their title. Registration types, and how well registries function, lead to different efficiency judgments regarding which acquisitive prescription regimes are best. In a country with constitutive registration and a well-functioning registry, a possession-based regime should be upgraded to registrationbased, with good faith, apparent title, and its registration required. In a country with a dysfunctional registry, a possession-based regime may be locally efficient. In between are countries with well-functioning registries plus declaratory registration, discussed above. These countries did not choose to induce all land sales to channel through registries and do not always explicitly make registration opposable to third parties. It is reasonable for lawmakers in these countries to choose to protect buyers who have title but do not register by a possession-based acquisitive prescription regime with the title requirement.

D  Adverse Possession: Possession-Based, without Apparent Title From an economic standpoint, at the most general level, any acquisitive prescription regime, especially a possession-based one without any title requirement, may be justified in the following ways. First, because adverse possessors value the land in question more than do erstwhile owners (Gordley 2006: 141; Depoorter 2011: 1113; Richardson 2015: 1403),45 an acquisitive prescription regime that awards ownership to possessors increases allocative benefits (Fennell 2006: 1064). Second, an acquisitive prescription regime can be sustained with low institution cost. The institution cost incurred by possessors (who want to utilize the land), erstwhile owners (who need to fend off trespassers), and third-parties (who are potential purchasers of land from either adverse possessors or erstwhile owners) must all be considered. As compared to registrationbased regimes and possession-based regimes with title requirements, possession-based regimes without title requirements are more difficult 44 45

See Table 5.4 for information about the data. Cf. Katz (2010: 67–70) discusses from agenda-setting vacancy to an individual with agendasetting authority.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

147

to justify. The three subsections below thus adopt this more structured framework to guide the analysis. Subsection 1 gauges the costs and benefits of possession-based regimes without title requirements in a country with well-functioning recording systems, and takes the United States as the prime example (Arruñada 2012: 44). Subsection 2 explores those in a country with well-functioning registration-of-right systems. Germany and Taiwan are the shadow examples (Arruñada 2012: 44; Chang et al. 2022: 31). Subsection 3 discusses countries with dysfunctional registration or recording systems.

1  In Recording Systems a. Allocative Benefits  Do adverse possessors always value the land more than erstwhile owners? It is unlikely that either adverse possessors, or erstwhile owners, are consistently the higher valuer. If neither party has legal ownership and a decision-maker is thus choosing between two claimants of equal footing, perhaps a “more likely than not” standard is sufficient to favor one party systematically. Nonetheless, given that erstwhile owners are legitimate owners and adverse possessors are trespassers to begin with, one may demand that adverse possessors meet a heightened standard to justify the acquisitive prescription regime on allocative benefit grounds. The evidence from social science that adverse possessors are more often the higher valuers is weak and inconclusive. First, research in the field of behavioral law and economics points out that adverse possessors may experience an “endowment effect” (Kahneman 2011: 288–299)46 after occupying and using the land for several years. Psychological studies, however, often do not distinguish between possession and ownership; thus, it is hard to tell whether the endowment effect comes from possession or ownership (Lewinsohn-Zamir 2015: 128–148). Those who attempt to tease out the effects of possession versus ownership find inconsistent results (Zamir and Teichman 2018: 209–212). Moreover, ownership is “imputed” in the lab, while possession is “felt”; thus, lab experiments, due to their constraints on this issue, may never provide sufficiently strong evidence to pass the heightened standard. Finally, even if adverse possessors all experience strong endowment effects, they will not necessarily value the land more than the erstwhile owners who themselves experienced endowment effects while possessing the land in question. 46

Peñalver (2009: 830) argues that adverse possession honors a person’s psychological attachment.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

148

Part II: Immovable Property

If a general case that adverse possessors are usually higher valuers cannot be made,47 whether it is the case in a narrower setting can still be explored. Again, an assessment of allocative benefits, in this context, is a comparison of erstwhile owners’ willingness to accept and possessors’ willingness to pay (or, equivalently, their economic values). While the former is revealed in erstwhile owners’ purchase and their refusal to deal at market value,48 the latter is never revealed. Tom Merrill has advocated for the application of the liability rule, at least when possessors are bad faith – that is, possessors have to pay erstwhile owners to gain ownership. (No jurisdictions adopt anything close to this.) If adverse possessors in particular cases are not willing to pay market value at the time of the dispute, they are likely lower-valuers. Even if possessors are willing to pay market value, it is still unclear whether they value the land in question more than do erstwhile owners. Nevertheless, policy makers can rest assured knowing that the law is not facilitating a blatantly inefficient transfer. The indemnification requirement could be criticized as unnecessary (Merrill 1985: 1151–1152) if transaction costs are low enough, as the two parties can bargain with each other to redress any allocative inefficiency.49 This Coasean critique of Merrill’s proposal, however, is open to criticism in at least two ways. First, Farnsworth (1999; 2016) argues that animosity between parties generated in litigation is a deal-breaker.50 Litigating parties will not bargain over the land in question. Thus, without the liability rule design, evicted adverse possessors will not become legal owners even when they are higher valuers. Second, the mechanism design literature in economics has demonstrated that a two-party bargaining scenario with one party (an erstwhile owner) owning 100% of the thing in question 47

48

49 50

But cf. Peñalver and Katyal (2010: 129) who note that when the distribution of property is extremely skewed, the adverse possessors may be the higher valuer. This argument of lawbreakers’ placing higher value on properties can only justify bad-faith adverse possession but not good-faith adverse possession. In addition, while this argument is highly relevant in the adverse possession scenario that this chapter focuses on (an illicit possessor occupying another’s land), this argument is less applicable to the modal adverse possession cases in the United States – building encroachment – as the two parties in such disputes are neighboring landowners, and their wealth may not be extremely different. Landowners may check out market prices from time to time and decide whether to unload their assets. Third parties may also contact them to offer to purchase land in question at around market prices. If the Coase theorem holds, the law need not re-assign entitlements via acquisitive prescription. But see the empirical critique of the Farnsworth idea in Chang and Lin (2021) and Chang et al. (2019).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

149

while the other party (an adverse possessor) owns 0% will not ensure increases in allocative benefits (see Chapter 10). Some efficient trades will not be consummated. In short, Merrill’s proposal to add an indemnification requirement increases, but does not ensure, increases of allocative benefits. Fennell (2006) has argued that bad-faith adverse possessors could be (but are not necessarily) “efficient trespassers.” To tease out the true higher valuers, adverse possessors should demonstrate that market transaction is not feasible and erstwhile owners are aware of adverse possessors’ occupation and their interest in the land in question (Fennell 2006: 1040–1041; Peñalver and Katyal 2010: 148–152). As said above, this chapter is, in general, against bad-faith adverse possession, but I am willing to entertain the idea of “efficient trespass” in Fennell’s well-confined scenario.51 Still, one wonders how many bad-faith adverse possession claims could survive in this scenario (this is an empirical question). In terms of results, Fennell’s world (which does not welcome good-faith adverse possession) may be entirely without successful adverse possession. Moreover, even if a few cases pass muster, it is doubtful whether the allocative benefit gain in these few cases could justify the transaction cost and information cost incurred by the adverse possession regime (see Subsection b). That is, as compared to a world where no adverse possession is allowed, a world with an adverse possession regime under which only few possessors acquire land ownership may not produce a net gain. The only way to tease out whether erstwhile owners or adverse possessors value property more, according to the mechanism design literature in economics, is through an internal auction in which only an erstwhile owner and an adverse possessor, assigned (roughly) equal shares of ownership of the land in question, participate (see Chapter 10). The design of such an internal auction will induce both parties to bid honestly, and thus the auction can reveal which party is the higher valuer and should gain full ownership, while having to compensate the other party. This proposal, however, begs the question of why adverse possessors should be awarded (any) ownership. As argued above, bad-faith adverse possessors should be discouraged. Perhaps good-faith adverse possessors? 51

Fennell’s analogy of efficient knowing speeding to bad-faith efficient trespassing, however, may not be entirely appropriate. Unlike adversely possessing land, speeding does not involve long-term investment, and the gain from speeding itself cannot be taken away, while bad-faith trespassers may refrain from investing as, say, the crops they grow could belong to the erstwhile owners.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

150

Part II: Immovable Property

Recall that good-faith adverse possessors here do not have title, and boundary disputes are excluded from analysis in this chapter. Hence, good-faith adverse possessors have to be, on the one hand, very mistaken, and, on the other hand, to have spent reasonable efforts in verification, to avoid being described as “should have known” and thus bad faith. The set of such good-faith adverse possessors may be quite small. Hence, this chapter does not advocate awarding adverse possessors 50% or so shares. Without a roughly equal share, an internal auction cannot work magic in inducing honest bidding and facilitating Pareto optimal trading. Maximal allocative benefits cannot be ensured as a result. The prior literature has also discussed whether erstwhile owners’ slothfulness should be “penalized” and adverse possessors’ diligence should be rewarded (Merrill 1985: 1128–1131; Merrill and Smith 2010: 37–38). The allocative benefit aspect of this argument implies that the latter is the higher valuer, partly because they develop the land in question. Existing works have countered that the implicit pro-development mentality behind adverse possession regimes is not the most environmentally friendly (Sprankling 1994; Stake 2001: 2433; Klass 2006: 292–294). Conservation may very well be more valuable than development (Shavell 2004: 73–74). Even if adverse possession regimes are regeared to serve as a development machine, it is a blunt tool, as landowners can simply evict possessors without doing any development (Stake 2001: 2436). A tax imposed on vacant land may be more effective in promoting development. Then again it all boils down to whether “slothfulness” and “diligence” translate to systematic differences in economic value. In sum, no argument is strong enough to explain and justify the broad adverse possession regime adopted now. Merrill’s indemnification requirement improves the regime, but the outcome may still fail to meet a high-bar standard which requires compelling evidence that acquisitive prescription will very likely lead to higher allocative benefits. Fennell’s reform proposal helps achieve the goal of increasing allocative benefits, but in effect is close to getting rid of the regime altogether. b.  Institution Costs  Acquisitive prescription regimes should minimize information costs and transaction costs while maintaining the allocative benefits derived. In net, it is unclear whether a legal system with an adverse possession regime produces lower information and transaction costs. The prior literature notes that adverse possession can clear stale claims, quiet title, and induce information about the identity of landowners – that is, the possibility of losing ownership due to adverse possession will force

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

151

erstwhile owners to sue possessors, which enables the latter to identify the whereabouts of passive and absentee owners and negotiate with them (Merrill 1985: 1128–1131; Epstein 1986: 678; Netter et al. 1986: 219; Merrill and Smith 2010: 37–38). It is questionable how useful this is. Given the strict conditions of adverse possessions, this regime is unlikely to clear stale claims or quiet title in many parcels. Rather, only a tiny number of plots would have a fresh start. Marketable title act, quiet title action and even rules against perpetuities are more useful in eliminating ancient interests, and reduce information cost regarding who owns what property rights. It is also questionable (though an empirical answer is needed) how often a serious potential buyer cannot locate the current owner with the help of professional middlemen. In addition, whether drawing out current owners through the process of adverse possession is a recipe for successful trades between adverse possessors and owners is also doubtful.52 Moreover, as existing works already illuminate, in the United States, because adverse possessors gain ownership automatically once the statute of limitations runs (Klass 2006: 287), adverse possession itself introduces unrecorded interests in land and increases information cost (Fennell 2006: 1063). Third-party information cost would be lower had the law been that subsequent parties would be bound by adverse possession if an adverse possessor had brought a quiet title action and recorded the declaratory judgment. The draft Restatement of the Law Fourth, Property, is inclined to take the position that subsequent parties would not be bound only if an adverse possessor has brought a quiet title action but failed to record the declaratory judgment. Hence, third parties who have checked the public records could be surprised – and legally defeated53 – by adverse possessors who have not brought a quiet title action. True, potential buyers who have undertaken a field trip to the land may be able to identify adverse possession, but this hinges on real-world cues which are not always easy to recognize. In addition, once adverse possessors meet the requirement for adverse possession and become owners, they are not required to continue to possess in an open fashion. A field trip may even be misleading under this type of circumstance. In any way, this investigation of potential existence of adverse possession increases information costs across the 52

53

Also, because in fact the adverse possession doctrine in the United States is used mostly to resolve boundary disputes, it can hardly clear stale claims (as they still exist in the remainder of the neighbor’s land). Difficulty in identifying a transacting partner is also not applicable, as the partner literally lives next door. See Draft Restatement of the Law Fourth, Property, §1.2.2.1.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

152

Part II: Immovable Property

board – any land could have been adversely possessed and sellers may have incentives to hide the information, so all potential buyers have to figure out whether an adverse possessor is around. Therefore, the adverse possession doctrine decreases information costs (in one aspect) in a small number of deals but increases information costs (in another aspect) in every transaction. This does not seem to be a bargain worth striking. Assume that genuine good-faith possessors exist. (The draft Restatement of the Law Fourth, Property, contains many illustrations, most of which are adapted from real cases, but almost all of which are in the boundary dispute context, not in this context.) Is awarding ownership for free through adverse possession necessary to give good-faith possessors an optimal incentive to invest? I do not think so. If acquiring ownership through prescription is a necessary condition, any substantial prescription period will fail the goal of optimal investment, because every landowner will think that before the statute of limitations runs she may lose everything. A strict construction of continuous possession, where a week of renouncing possession will restart the clock, is not tailored made for the purpose of inducing optimal investment. Good-faith adverse possessors’ diligence and reliance interests (Miceli and Sirmans 1995a: 161) have to be protected, because, ex ante, every landowner in a recording system will, to some degree, worry about the reliability of surveyor and lawyer reports. Optimal investment and low transaction cost cannot be achieved if good-faith adverse possessors’ investment cannot be recouped. To do so, the law of equity in commonlaw countries and possession-related doctrines in, for instance, Germany §996 and Taiwan §955 can readily take care of compensating good-faith adverse possessors for their necessary expenses and the value they have created. This kind of indemnification requirement should suffice to induce adverse possessors to invest more. Possession-based acquisitive prescription, therefore, is neither necessary nor useful for promoting efficiency in countries with a recording system.

2  In Registration-of-Right Systems a. All or Most Real Properties Have Been Registered  In a country with a registration-of-right system, where most, if not all, land parcels are registered, or at least the land in question is registered, it is even more difficult to justify possession-based acquisitive prescription. In terms of allocative benefits, all the preceding analysis applies. Moreover, the question of whether there are any genuine good-faith adverse possessors looms larger because in a registration-of-right system, no matter whether

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

153

registration has a constitutive or opposable effect, the most updated rightholding and right-holder information is easily retrievable in registries. Who would possess a parcel for 20 years without ever bothering to confirm whether she is the registered owner? When possessors do check and realize that they lack ownership, they become bad faith. In addition, in a registration-of-right system, the contained information likely serves as the basis for levying property taxes. If everyone else pays their property taxes every year, what kind of adverse possessors could be unaware of this well-known duty while still being genuinely good faith?54 Institution cost rationales are also weaker in a registration-of-right system. First, a registration-of-right system does not have “stale” claims. Registration is very often constitutive or opposable, and the realfolium (track index) makes clear to any concerned party who owns what. Thus, there is, in general, little need to quiet title, nor recognize adverse possession to quiet title. Besides, adverse possessors do not have to take possession in a hostile way to force erstwhile owners to identify themselves. In some countries, the former could easily locate the latter via information saved in registries. In other countries, registries can be revamped as an information clearing house – the identity of current owners need not be revealed (for privacy or other reasons) and registries could relay potential purchasers’ offers and contact information to current owners. As the World Bank’s Doing Business data show (see Table 5.3), at least in the largest business city, to the extent that the registration is comprehensive and reliable (which is not always the case, as shown in Table 5.4), most jurisdictions have policies that enable potential adverse possessors to identify current owners. In registration-of-right systems, adverse possession laws can either require adverse possessors to register their ownership once the statute of limitations runs, or make registration the precondition for opposability.55 Third parties, therefore, can simply rely on the information provided by registries and do not have to launch an on-site investigation. With such a regime, adverse possession neither increases nor decreases information costs for third parties, but that means adverse possession cannot be justified on the ground of saving institution costs.

54 55

Some US states thus require adverse possessors to have paid taxes during the prescription period (Stake 2001: 2422). Otherwise, the information cost-reducing function of registries will be discounted (Stake 2001: 2442–2443; Posner 2011: 98).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

154

Part II: Immovable Property

Table 5.3  Transparency of land ownership information Who is able to obtain information on land ownership at the agency in charge of immovable property registration in the largest business city? Freely accessible by anyone Anyone who pays the official fee Only intermediaries and interested parties Only intermediaries (notaries, lawyers, etc.) Records are not publicly available Total

Frequency % 26 73 39 3 6

18 50 27 2 4

147

100

Notes: N = 147. This table includes all jurisdictions that are both coded by me and have available World Bank data. Cuba, Liechtenstein, Monaco, North Korea, and Turkmenistan are not included in the World Bank data sources. Louisiana, Macau, and Scotland are subnational jurisdictions that are omitted. South Pacific countries are included in World Bank data sources but are omitted because there is only one holistic coding of law for all the countries. Data Source: The World Bank Doing Business Data “Transparency of information index” released in November 2016.

b. Many Real Properties Have Not Been Registered  In some such countries, not all land parcels have been registered. Lack of complete coverage is attributable to various reasons, such as the high cost of surveying all parcels. It is thus hard to engage in a general discussion as to whether and how registered and unregistered properties should be treated separately for economic reasons. Still, a few comments are in order. First, in general, bad-faith adverse possession should be discouraged. Second, a citizen with common sense in a country with the registrationof-right system would more or less know that the information provided by registries is authoritative and more reliable than the fact of possession. Third, only 14 of the 128 jurisdictions (9%) with possession-based acquisitive prescription explicitly limit the object of such acquisitions to unregistered land.56 Other jurisdictions, if in fact allowing adverse possession over registered land, should have second thoughts on the underlying justification. A registration-of-right system is expensive – a w ­ ell-functioning one especially so. Limiting adverse possession to unregistered land only 56

Taiwan is one of them, but Taiwan now has no unregistered land! Consequently, while the Taiwan Civil Code has rules for adverse possession of land ownership, adverse possessors are doomed to fail.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

155

would hardly increase anyone’s transaction cost and information cost by much, as anyone could easily check with low cost and get authoritative answers on whether she owns the land she possesses.

3  Dysfunctional Registries Many countries’ registries, however, are dysfunctional. There are 123 jurisdictions that (1) are surveyed in this chapter, (2) are surveyed by the World Bank Doing Business Data, and (3) have any type of acquisitive prescription defined above. Table 5.4 shows how they fare in keeping track of land right information. Apparently, the level of functionality of these registries varies, but about a quarter to a third of these jurisdictions have arguably dysfunctional registries. In these jurisdictions, registration-based acquisitive prescription does not make sense, as registration information is not reliable and is incomplete. A registration-based regime may even spawn corruption, as strategic persons find their way into registries through bribery. By contrast, possession-based acquisitive prescription may at least reduce information cost. In jurisdictions with well-functioning registries, possession-based regimes are used to replace current owners with adverse possessors. In those with dysfunctional registries, possession-based regimes instead identify who real owners are. Think of two parties in such a jurisdiction dispute in court, each claiming to be the legal owner: One party has occupied the land for 10 or 15 years and behaved like an owner. The other party has been absent but presents documents issued by registries or registry records that suggest that she is the owner. Here the case for favoring the former is stronger, as there are many ways in which the records or documents are unreliable, while peaceful and continuous possession for many years strongly suggests that the community (which is generally more closeknit in countries with dysfunctional registries) accepts the possessor as the owner. Given that land transaction markets are not that efficient, due to the dysfunctional registries, actual use is more strongly correlated with highervaluing. Information and transaction costs, in general, will not increase due to recognition of ownership acquired through prescription because transacting parties may be relying on actual possession in physical space.57 57

An anecdote is in order here. I hired a Cornell LL.M. student from Colombia, who informed me that the registry in Colombia is not functioning very well. Legal owners’ property rights could be transferred without their authorization and then registered. To prevent this from happening, many landowners put up huge signs in front of their houses, stating that “this property is NOT for sale, for lease, or for any type of transfer.” She was shocked when I told her that no developed countries experience this phenomenon.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

156

Part II: Immovable Property

Table 5.4  Functioning of registries Surveyed results Registered all private land in the largest business city† Registered all private land† Mapped all private land† Mapped all private land in the largest business city† Cost of registering property is ≧5% of the property value* (N = 121) Law requires all property sale transactions be registered at immovable property registry to make them opposable to third parties.‡ A specific compensation mechanism exists to cover losses incurred by parties who have engaged in good faith in property transaction based on erroneous information certified by an immovable property registry? ‡ Immovable property registries commit to delivering a legally binding document that proves property ownership within a specific time frame?**

Number of Percent of jurisdictions jurisdictions 54

44

29 35 63 64

24 29 52 53

115

94

36

29

63

52

Notes: N = 122. Cuba, Liechtenstein, Monaco, North Korea, and Turkmenistan are not included in World Bank data sources. States in the United States, Hong Kong, Macau, and Scotland are omitted subnational jurisdictions. South Pacific countries are included in World Bank data sources but are omitted because there is only one holistic coding of law for all the countries. Data Source: World Bank Doing Business Data. * is from Registering Property; † is from the Geographic Coverage Index. ‡ is from Land Dispute Resolution Index. ** is from the Transparency of Information Index. The 2016 data were used. Data are available at www.doingbusiness.org/content/dam/doingBusiness/excel/ Historical-data---complete-data-with-scores.xlsx.

In these jurisdictions, adverse possessors are more likely to be good faith than elsewhere. A good-faith adverse possession regime will not be pointless. Whether bad-faith adverse possession should be allowed is a more difficult question. Here, someone knows that she is not the owner of a plot she is interested in cultivating or developing. Due to the dire situation of the registration system, she simply cannot find the legal owner. Even so, a possession-based regime with a reasonable prescription period

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

157

still creates under-investment. As argued above, optimal investment can be attained by compensating possessors with the necessary expenses and value created. One might imagine that in most countries, such compensation for bad-faith possessors is less generous than good-faith possessors. Here, a Fennell-proposal-like regime may be established: bad-faith possessors should notify registries and give public notice of their adverse possession. Erstwhile owners’ knowledge should probably not be required. The upshot for these bad-faith possessors is not landownership for free, but equal treatment as good-faith ones when it comes to compensation for necessary expenses and created value.

E  Registration-Based, without Apparent Title Type B adopts registration-based acquisitive prescription without the apparent title requirement. I would imagine that, at least in the cases of Germany, Switzerland, and South Korea, buyers without title can hardly get their names into registries.58 That is, registrants who need acquisitive prescription are most likely to have title. Indeed, it is difficult to come up with a concrete example where long-term possessors registered as owners do not have title. Rare cases may arise due to rare mistakes, but this may be better dealt with by a land registration compensation fund used in the Torrens version of the registration-of-right system. (Germany, for instance, already adopts this.) The fund comes from fees levied along with each registered transaction, so it is like mandatory insurance among landowners. A compensation fund, but not acquisitive prescription, will not disrupt the normal operation of property law and everyday land transactions.

III Conclusion In modern times, efficient acquisitive prescription regimes are very different from what most countries have enacted. In countries with wellfunctioning registration-of-right systems, a registration-based regime is warranted. Good faith, apparent title, and registration of title should be required, but not possession. The prescription period should be shorter 58

The Münchener Kommentar and the Staudinger Kommentar for the German Civil Code §900 contain only five cases decided by a German federal court or state court (regarding registration-based acquisitive prescription), and four involve wrongful registration that took place in the nineteenth century! See citations in Chang (2022a: 423).

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

158

Part II: Immovable Property

than that currently in place. Given that boundary disputes can be resolved in a separate doctrine, a possession-based regime is unnecessary. As compared to countries with well-functioning registration-of-right systems, countries with reliable recording systems are more likely to have genuine good-faith adverse possessors. A registration-based system that requires good faith, apparent title, and its recordation is also warranted. A possession-based regime is unnecessary and counter-productive. In countries with dysfunctional registries, requirements of registration of title may not be ideal, as even serious transacting parties may not have checked the registry or may treat registration as unnecessary. Good faith, however, should still be a necessary condition. Possession-based acquisitive prescription may be warranted in these countries, because ownership information cannot be reliably retrieved anywhere. Good-faith adverse possessors, who neither know nor should have known that they lacked title, but have failed to register or record their (apparent) title, may appear sympathetic to some. This chapter, however, contends that there is no efficiency reason to award ownership for free simply because they are good faith and in possession. In most studied countries, the law of equity or unjust enrichment doctrines are sufficient to protect these faultless adverse possessors. Finland and Costa Rica have acquisitive prescription regimes that are closest to efficiency for countries with well-functioning registries. They both have registration-based systems that only allow good-faith registrants who have registered their apparent title for a certain number of years. In addition, neither allows acquisitive prescription solely based on possession, without good-faith registration of apparent title. Their doctrinal requirements are not perfect, as possession, in addition to registration, is still required, and the prescription period is arguably too long. According to the World Bank’s Doing Business Report data, Costa Rica’s registries are not of the highest quality, but they are in the second quartile. Whether its close-to-efficient law lives up to its promise depends on how registries work on the ground. One of the contributions of this chapter is to call attention to the fact that how registries work affects the efficiency level of private law doctrines. A number of countries that are in the lowest quartile of the Doing Business ranking have acquisitive prescription rules that are unsuitable given their dysfunctional registries. It would be great if all could become like Finland, with well-functioning registries and efficient laws, but before such developmental miracles take place, acquisitive prescription doctrine should be adjusted to the reality of registries as they currently exist.

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

acquisitive prescription

159

Epstein (1986: 678) comments, in the context of adverse possession, that “protection of the guilty is not an end in itself, but the inevitable and necessary price paid in discharging the primary function of protecting those with proper title.” This chapter suggests that all countries can improve their laws and thus reduce the “price … of protecting those with proper title.”

https://doi.org/10.1017/9781009236553.008 Published online by Cambridge University Press

6 Building Encroachment In Search of an Efficiency Justification

When a construction encroaches on and extends over a land boundary, the trespassed neighbor’s legal remedy, by default, is an injunction (under common law) or a claim of rei vindicatio (under civil law) to remove the encroaching constructions. That is, the baseline rule, embodying  the exclusion strategy (Smith 2002), is that encroachers must tear down the encroaching construction on their neighbors’ land, if neighbors so demand. Nonetheless, the baseline is not always strictly adhered to. Part I discusses three legal doctrines that provide exceptions to the baseline. The focus of this chapter is the first of those approaches – the building encroachment doctrine. As for the two other approaches, adverse possession is discussed in Chapter 5, while accession is covered in Chapter 13. The building encroachment doctrine starts from the proposition that the wronged neighbor can demand the removal of encroaching structures. If encroachers are willing to comply with such a request, and compensate neighbors for any damage done and unjust enrichment for their use of the property during the period of encroachment, the story ends. Encroachers may voluntarily opt for this solution when it is less costly than continuing the occupation and paying neighbors. Oftentimes, especially when the encroached upon part of the adjacent land supports a valuable construction, encroachers would prefer to not tear down the encroaching part. Encroachers and neighbors will then bargain in the shadow of law (Mnookin and Kornhauser 1979). Building encroachment doctrine, in particular, tends to give encroachers more bargaining power – this chapter examines whether that is justified. Figure 6.1 shows land boundaries, with the encroachers’ land on the left and neighbors’ land on the right. D is the encroached land, and B+D is the entire adjacent land. The construction in question sits on C+D. The part of the construction standing on D is the encroaching construction. 160

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

161

Figure 6.1  Illustration of the issue

I  Three Roads Diverge in the Doctrinal Woods There are three different exceptions to the baseline exclusion strategy. Section A summarizes the first approach, a tailor-made doctrine of building encroachment. Section B describes the second approach, dealing with the building encroachment issue with the adverse possession doctrine. Section C delineates the third approach, dealing with this issue with the accession doctrine.

A  Building Encroachment Doctrine Following the framework of Smith (2013a; 2021a), this chapter describes the building encroachment doctrine in terms of the law versus metalaw structure. The starting point in all studied jurisdictions for building encroachment is removal – the unsurprising result of property law’s baseline exclusion strategy. Taking the exclusion strategy to its logical extreme may create opportunism. For instance, neighbors can wait upon the completion of the encroachers’ development and later seek a larger side payment (as the encroachers’ investment has increased) or more compensation (as the encroachment may extend further). The building encroachment doctrine is thus second-order – a meta-law over the regular law of rei vindicatio. More specifically, the doctrine is often a combination of “ex ante tailored rules” and “ex post tailored standards” (for the typology, see Smith 2021a: 1080). The best example of the ex ante tailored rule is Portugal §1343, which stipulates that neighbors have to oppose to encroachers within three months of the latter’s occupation or be barred from exercising rei vindicatio. This prompt protest rule is summarized in

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

162

Part II: Immovable Property

Section 4. Almost all jurisdictions require encroachers to be good faith. When bad faith includes “should have known” (as this book advocates), it borders on being an “ex post tailored standard,” rather than an “ex ante tailored rule.”1 A number of countries incorporate requirements such as “removal causing significant harm” or “preservation introducing considerable losses” or ask courts to weigh the gains and losses of the two parties. These are typical “ex post tailored standards.” Sections 1 and 2 give an overview of the types of rules and standards that countries use to constrain the rei vindicatio right held by neighbors (encroached parties). That is, if the rules and standards are met, neighbors can only ask encroachers to purchase ownership (Section 1) or an easement (Section 2) of the encroached land.2 Section 3 introduces another type of building encroachment doctrine, which embodies only “ex post tailored standards.” Some countries instruct courts to take an all-factors considered approach, which borders on being “ex post untailored standards.” The differences in the designs in countries classified in Sections 1 and 2, on the one hand, and Section 3, on the other, differ in the following way. Regarding the former, the rules and standards are the necessary and sufficient conditions for preservation. Courts have to consider statutorily prescribed factors, though when it comes to ex post tailored standards, courts still have room for interpretation. When the conditions for preservation are met, courts have to rule for preservation. Regarding the latter, in some countries, statutes only spell out the necessary conditions for preservation (e.g., good faith and prompt protests). Failing them, encroachers must remove overhang constructions; meeting them, encroachers are still subject to the discretion of the court. In other jurisdictions, like New York state, courts are asked to balance various interests, or account for all circumstances. That is, the law does not list specific, necessary conditions for preservation. For brevity’s sake, this chapter refers to courts in the latter category as having “discretionary powers.” 1

2

Ex ante tailored rules often suffer from exploitation by opportunists. Assume that good faith means only “already know the boundary,” developers can meet the requirement by being willfully ignorant. Depending on how the law is designed, encroachers could be awarded with a claim to buy ownership or easements from neighbors. Oftentimes, it will be neighbors who seek redress in the form of forced purchase. Encroachers, however, if already aware of the encroachment, may seek to acquire ownership or easement of the encroached land when planning to sell constructions in question to a third party. In Croatia (Josipovic 2013: 100–101) and Macedonia (AOORR) §123, for instance, both good-faith encroachers (if removal causes substantial damages) and neighbors are explicitly each given a claim.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

163

Table 6.1  Typology of building encroachment doctrine The encroached, under certain conditions, only has a claim to sell easement or use rights to encroachers The encroached, under certain No conditions, only has a claim to sell land ownership Yes to encroachers

No Yes Type A. Type B. 19 jurisdictions. 4 jurisdictions: Georgia, Mongolia, Thailand, and Turkmenistan. Type C. Type D. 24 jurisdictions. 6 jurisdictions: Estonia, Germany, Moldova, the Netherlands, Poland, and Turkey

Notes: Courts in the 19 Type-A jurisdictions have discretionary powers, introduced below. Courts in the two Type-C jurisdictions, Argentina and Taiwan, also have discretionary powers.

Boundary encroachment Type D Type C Type B Type A No BE doctrine Not coded

Figure 6.2  Typology of building encroachment doctrine Notes: Argentina and Taiwan have both the features of Type A and Type C but are shown as Type C only in this map.

In total, 52 jurisdictions follow the building encroachment doctrine. See Figure 6.2 and Table 6.1 for the typology.3

1  Claims to Sell Land Ownership More specifically, in Type C and D jurisdictions, if certain conditions are met, neighbors are not allowed to demand removal of encroaching 3

In Spain, while the civil code sticks to the baseline position, the law in action deviates from the law in books (Rizzolli 2009: 680).

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

164

Part II: Immovable Property

construction and can only demand encroachers purchase ownership of the encroached part of land to resolve boundary disputes. The requirements for the rule to be implemented are shown in Table 6.2. A few countries are idiosyncratic: Czech only requires good faith of encroachers, while the Netherlands is the only country that does not explicitly require good faith. Brazil has a complicated rule that involves a clear threshold (1/20 of the neighboring land) and damages that are twice or ten times the market value (Romañach Jr. 2015b: 233). Table 6.2  Conditions for neighbors’ losing claims for removal Encroachers’ good faith

Neighbors’ prompt protest Two parties’ losses

Argentina §1963 Brazil §§1258–1259

✓ ✓



Croatia (AOORR) §155 Cuba §175

✓ ✓



Czech §1087 Estonia §148 Germany §912 Hungary §§V:28–V:29 Liechtenstein (SR) §56 Macau §1263 Macedonia (AOORR) §123 Malta §571 Moldova §390 Netherlands §V:54

✓ ✓ ✓ ✓

✓ ✓ ✓





✓ ✓



✓ ✓

✓ ✓

Nicaragua §631 Paraguay §1961

✓ ✓

✓ ✓

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

✓ (building value>land value) ✓ (removal causes substantial damage) ✓ (preservation does not cause considerable damage)

✓ (removal causes substantial damage) ✓ (justified by the circumstances) ✓ (removal causes substantial damage)

✓ (removal causes disproportionately more harm for encroachers than preservation causes harm for neighbors)

building encroachment

165

Table 6.2  (cont.) Encroachers’ good faith

Neighbors’ prompt protest Two parties’ losses

Peru §944 Poland §151

✓ ✓

✓ ✓

Portugal §1343 Rwanda §II: 25

✓ ✓



Serbia (LBOPR) §24 Switzerland §674









Taiwan §796 Timor-Leste §1263 Turkey §725

✓ ✓ ✓

✓ ✓ ✓

✓ (preservation does not cause considerable damage) ✓ (removal causes notably more harm for encroachers than preservation causes harm for neighbors)

✓ (justified by the circumstances)

✓ (justified by the circumstances)

Note: Angola, Cape Verde, Guinea-Bissau, Mozambique, and Sao Tome and Principe follow Portuguese law. Brazil also considers the scope of infringement.

The preceding analysis focuses on claims to sell the encroached part of the adjacent land. Under certain conditions, neighbors have claims to sell the entire adjacent land, including the non-encroached part. Argentina §1963 (remaining part unusable), Croatia (AOORR) §155 (no extra conditions), Hungary §V:28 (remaining part unusable), Macedonia (AOORR) §123 (no extra conditions) (Josipovic 2013: 100–101), Paraguay §1961 (remaining part unusable), Peru §944 (remaining part unusable), Poland §151 (“losing economic value”) (Habdas 2018: 205–206), and Taiwan §796 (remaining part unusable) are examples. In addition, or alternatively, neighbors may demand compensation for the depreciation in value in the remaining part of land. Brazil §1258, Greece §1010, Macau §1263, Timor-Leste §1263, Portugal §1343 and its former colonies are examples. Serbia (LBOPR) §25 is idiosyncratic. When either encroachers are bad faith or neighbors protest promptly, neighbors are awarded with a call

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

166

Part II: Immovable Property

option to purchase the construction (in its entirety, it appears). The statute even specifies the compensation criterion: the amount of average construction price of a construction in that location, at the time of issuance of the court decision.

2  Claims to Sell Easement or Use Right In Type B (Georgia §179, Mongolia §137, Thailand §1312, and Turkmenistan §201), the neighbors can only sell a real easement or a certain use right for the encroached land to resolve the boundary dispute if encroachers are good faith and neighbors protest promptly (good faith is not required in Mongolia, whereas prompt protest not required in Thailand). Georgia does not specify whether an easement or use right is legally prescribed, but require encroachers to pay annual compensation. I take the liberty of classifying the Georgian rule here because of the payment requirement. Countries that use forced purchase of easement or use rights to resolve building encroachment (Types B and D) often have explicit rules that require or allow periodic compensation rather than lump-sum payment. They include Estonia §148, Georgia §179, Germany §912, Moldova §390, Mongolia §137.2, and Turkmenistan §201. 3  Discretionary Power Table 6.3 lists the 21 jurisdictions (Type A or Types A+C) that give courts discretionary power to preserve encroaching construction when neighbors otherwise have claims for removal. Many jurisdictions, such as New York (RPAPL) §8714 and California (CPC) §871.5,5 specify factors that courts should consider, while others, like Louisiana §670, merely stipulate that courts have discretion (Bell and Parchomovsky 2012b: 911).6 Countries including Australia (PLA) §185 and New Zealand (PLA) §325 prescribe very elaborate standards. As shown in Table 6.3, some other countries lay out specifically important factors, such as encroachers’ being good faith, neighbors’ prompt protest,7 or scope (encroachment being minor), as preconditions. All but Italy require ordinary compensation if encroaching constructions are preserved. Italian Civil Code, by contrast, demands encroachers to pay double damages (Rizzolli 2009: 680–681). 4 5 6 7

See also Marsh v. Hogan, 81 A.D.3d 1241. See also Salazar v. Matejcek, 245 Cal.App.4th 634. In American common law, it is called the “relative hardship” doctrine (Sterk 2012: 2146). Neighbors who do not protest timely lose their claims to demand removal of encroaching construction, and can only demand compensation. Thus, there is no need for courts to exercise the discretionary power to preserve the construction in question.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

167

Table 6.3  Courts that have discretionary power to preserve construction

Jurisdiction Algeria §788 Argentina §1963 Australia (PLA) §185 Azerbaijan §173.3 Bahrain §854 California (CPC) §871.5 DR Congo (LF) §24 Egypt §928 Greece §1010 Israel (Naama v. Turkia, 2011) Italy §938 Kuwait §885 Libya §932 Louisiana §670 New York (RPAPL) §871 New Zealand (PLA) §325 Qatar §916 Slovenia (Škerl and Vlahek 2020: 193) South Africa (Van der Merwe et al. 2002: 261) Taiwan §796-1 Tunisia (CDR) §37

Balance various interests/ Good Prompt Scope of judged by all faith protest encroachment circumstances ✓

✓ ✓

✓ ✓ ✓ ✓ ✓

✓ (“manifestly abusive”) ✓ ✓



✓ ✓

✓ ✓ ✓





✓ ✓ ✓ ✓









✓ ✓ ✓

✓ ✓ ✓ ✓







4  Protest: How Prompt Is Fast Enough? Thirty-one jurisdictions incorporate prompt protest in their building encroachment doctrine, described in the three preceding sub-sections. Ten jurisdictions precisely define promptness: 9 (Portugal §1343 and its former colonies), 3 months; Turkey §725, 15 days. Seven jurisdictions prescribe immediate protest, while eight require protest within some reasonable time. In practice, these two standards may be the same.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

168

Part II: Immovable Property

Several countries stipulate an absolute, rather than a relative, type of protest deadline. Estonia (TLPA) §148 specifies prompt protest as opposing “before commencement of construction or, at the latest, at a time when removal of the part did not involve excessive expenses.” Hungary §V:29 prescribes that a neighbor protests against the encroachment at a time when restoration of the original state would not have caused unreasonable harm to the builder. Greece §1010 requires protest be lodged before a large part of construction is completed. Notably, Louisiana §670 adopts both types of deadlines: a relative deadline (reasonable time after neighbors have known about the encroachment) and an absolute deadline (after buildings are substantially completed, regardless of whether or not neighbors were aware of the encroachment). A few remaining countries give no specific timeline.

B  Acquisitive Prescription Doctrine Presumably, without a building encroachment doctrine, the ­possession-based acquisitive prescription doctrine applies. Specifically, scholars in the United States (Draft Restatement of the Law Fourth, Property), Australia,8 Hong Kong (Chiang 2016: 62), and Ireland (Woods 2016), among others, have indicated that the adverse possession doctrine is (mainly!) used to solve boundary disputes. Note, however, that in my classification, New York, California, and Louisiana have rules that give courts discretionary power in dealing specifically with building encroachment disputes. In addition, Armenia, Belarus, Benin, Bosnia and Herzegovina, Denmark, Finland, Hong Kong, India, Kazakhstan, Kyrgyzstan, Laos, Lithuania, Malawi, Norway, Oman, Russia, Sri Lanka, Tajikistan, Uganda, Ukraine, Uzbekistan, and Zimbabwe have the possession-based acquisitive prescription doctrine, but neither building encroachment doctrines, nor accession doctrines, regarding two immovables.9 Under an adverse possession regime, encroachers have a significantly harder time acquiring ownership or using rights in order to retain the encroaching construction, as the prescription period is long. 8

9

O’Connor (2006: 45) notes that building encroachment disputes may be dealt with under the adverse possession doctrine, but there is a trend in Australia, Canada, and England to make separate provisions for building encroachment disputes. Madagascar (OLRS) §82 does not have a full-blown building encroachment doctrine, but reduces the prescription period for prescriptive acquisition from 20 years to 10 years in the building encroachment context.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

169

C  Accession Doctrine Sixty-four jurisdictions, for example, Austria §418, do not stipulate a building encroachment doctrine but contain a doctrine regarding accession of construction to land (see Chapter 13).10 In particular, some, for example Costa Rica and Singapore, have neither a building encroachment doctrine nor a possession-based acquisitive prescription doctrine, but they do have accession doctrines regarding two immovables. In France, scholars debate whether the accession doctrine should be applicable. The highest court in France, however, has clearly ruled that neighbors have full rights to demand removal of encroaching constructions, even when encroachers are good faith. The court reasons that neighbors’ exerting their owner power cannot constitute abuse of right. In addition, only the state has the eminent domain power, so neighbors cannot be forced to sell the encroached land to encroachers (Strickler 2006: 379–380; Terré and Philippe 2010: 225–227; Mackaay 2021: 300). Israel (LL) §§21–26 adopts the accession approach, and its law does not limit the application to building encroachment, but to any erection of structures and plants on another person’s land. A 2011 civil appeal case Naama v. Turkia (8661/10) shows that the court may exert equity power in boundary disputes. The court ruled that the neighbor in the case, Naama, will be allowed to tear down the encroaching kitchen when Turkia, the current owner of the kitchen (and the house annexed to it) who did not build the encroaching kitchen, transfers the construction to a third party.

II  Economic Analysis How encroaching buildings are dealt with, as shown in the previous part, is by no means universal. All jurisdictions start with the baseline exclusion strategy. In countries that follow the accession approach, encroached neighbors gain rights to the encroaching construction; thus, the exclusion strategy is maintained. In countries that adopt the adverse possession approach, encroachers are in a tough position, as prescription periods are generally multiple years, even multiple decades. That is, the exclusion strategy is largely maintained. Only in the 52 jurisdictions that carve out the building encroachment doctrine do encroachers have a shot at 10

Slovenian law considers construction encroachment as an exception to the accessio rule (Škerl and Vlahek 2010: 140).

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

170

Part II: Immovable Property

keeping encroaching constructions, defying the baseline exclusion strategy. This part analyzes whether the building encroachment doctrine makes economic sense.

A  Safety Valve: Good Faith Will Lose Its Import The baseline exclusion strategy which embodies the property rule (see Chapter 2) is generally more efficient in property law. The ­law-and-economics literature has debated whether the property rule (here, removal no matter what) or the liability rule (here, encroachers compensate neighbors in lieu of removal) is more efficient in facilitating voluntary transactions, or whether the two rules are equally efficient (Ayres and Talley 1995; Ayres and Balkin 1996: 704).11 Property scholars, however, have come away with the conclusion that the property rule is more efficient (Brooks and Schwartz 2005: 392; Smith 2009b: 134). In the building encroachment context, Sterk (2008: 1295) argues that any advantages the liability rule has in overcoming ex post strategic bargaining are less than the ex ante advantages that the property rule generates. The key point is that due to information costs, courts can at best award to neighbors ­market-value compensation, which is often lower than neighbor’s economic value. Therefore, while the ideal-type liability rule may have information-forcing advantage over the property rule in ex post bargaining, such an ideal type has not been implemented in the real world. Further, from an ex ante viewpoint, encroachers are in the best position to anticipate conflicts with owners, so the property rule, which induces encroachers to disclose conflicts, is more likely to generate efficient resource uses (Sterk 2008: 1294–1295). Further consider the following example. The market value for a certain part of Blackacre is $100, while Phil the owner values it at $120. Dora, who can create economic value of $110 on Blackacre, may decide to encroach under the liability rule because Dora only has to offer Phil $100 compensation for doing so. The result is that the part of Blackacre in question creates $110 instead of $120 in economic/social value. By contrast, under the property rule, Dora, in anticipation of the potential removal of her encroaching construction, will bargain with Phil in advance to gain his consent to build. The lack of voluntary agreement in this stylized example suggests efficiency – after all, Phil is the higher valuer. If we tweak the facts a bit, 11

Rose (2000: 9–10) also discusses the liability rule in the building encroachment context.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

171

and let Dora value the land at $130, a voluntary agreement of sale is generally expected under the property rule. Again, higher allocative benefit is achieved. While in this revised fact pattern the liability rule would lead to the same outcome as the property rule, we can readily see that the property rule in general creates higher allocative benefit via voluntary transactions, regardless of specific facts. By contrast, whether the outcome under the liability rule is desirable depends on the fact pattern. In any event, the building encroachment doctrine is not a typical liability rule, as encroachers cannot disrespect land boundary at will and to any extent they like. Oftentimes, neighbors can restore order to property rules by promptly protesting. That said, the building encroachment doctrine shows that neither is the prototypical property rule at work here. A prompt protest rule requires neighbors to act promptly upon knowing of encroachment, otherwise losing the rei vindicatio claim. Under a prototypical property rule, no such duties are imposed on landowners. Yet the prompt protest rule makes efficiency sense as it reduces social waste, because without the prompt protest rule, neighbors may strategically delay notifying encroachers. First, the resources spent by encroachers in the interim would soon be wasted upon removal of encroaching constructions. Second, the more resources invested by encroachers in the encroaching constructions, the higher the price encroachers will be willing to pay. Rent-seeking neighbors could thus sit on their valuable information for a while, with encroachers accumulating site-specific investments (Rizzolli 2009: 687). A counter-argument is that if eventually all rent-seeking neighbors sell their encroached land to encroachers, the higher price encroachers pay only amounts to a re-distribution of wealth and has no efficiency implications. Moreover, the higher price encroachers may pay provides an incentive ex ante to carefully verify land boundaries. Nonetheless, should bargaining fail, again the additional resources invested during the delay are social waste. While the prompt protest rule and the encroachers’ good faith are easy to justify economically as necessary conditions (Brooks 2002: 314), surprisingly, it is more difficult to make an efficiency argument for the good faith of encroachers being a sufficient condition. As Fennell (2006: 1066–1067) points out, being good faith (ignorant of encroachment) is not necessarily correlated with an encroacher’s ability to use the land more efficiently. Moreover, the chance of freely acquiring ownership via the adverse possession doctrine, or gaining ownership at below-economic value via the building encroachment doctrine, provides positive incentives to remain

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

172

Part II: Immovable Property

in ignorance. Being ignorant is privately efficient for encroachers but may be socially inefficient (Fennell 2006: 1071). Besides, the substance of the building encroachment doctrine would have no effect on the behaviors of genuine good-faith encroachers (Fennell 2006: 1072–1073), but bad-faith encroachers may pretend to be good faith. Hence, there is no ex ante efficiency reason to protect good-faith encroachers. Yet, as the previous part documents, all but one jurisdiction adopting the building encroachment doctrine requires encroachers to be good faith. Smith (2009a: 2128–2129), Merrill and Smith (2007b: 54–56) and Merrill (1986: 88–89) offer a “safety valve theory” that points out why only good-faith encroachers can benefit.12 One rationale is that an injunction for removal gives neighbors too much bargaining power. I will return to this point in the next section, but I note here that the too much bargaining power point is not an efficiency argument, but perhaps a fairness argument, as the bargaining edge neighbors enjoy is no greater when encroachers are good faith as compared to when encroachers are bad faith. The safety valve theory clearly does not consider problematic the asymmetric bargaining power neighbors have vis-à-vis bad-faith encroachers (Smith 2009a: 2129). The other rationale for protecting good-faith encroachers from a removal order is that they, aware of the probabilistic loss, may conduct excessive boundary surveys, which is socially inefficient (Sterk 2008: 1322; Smith 2009a: 2128). Throughout this chapter, I argue that good faith (in all countries that incorporate this requirement) should be interpreted as being without knowledge that can be acquired ex ante with justified costs. That is, for encroachers, if the cost of acquiring additional information is larger than the reduced probability of building encroachment, multiplied by the social loss created by encroachment, encroachers are good faith. Encroachers may reasonably worry that courts may not follow this proposed efficiency formula, or adopt it but lack accurate information.13 Thus, if good-faith encroachers will face an injunction filed by neighbors, encroachers have incentives to conduct excessive boundary surveys. Still, the concern for excessive boundary survey does not support treating good faith as a sufficient condition for preservation (as Czech law prescribes). 12 13

Parchomovsky and Stein (2009: 1828) argue for “market-value compensation, not propertized compensation, in good-faith encroachment.” As Miceli and Sirmans (1995a: 162) point out, it is difficult to distinguish good-faith and bad-faith encroachments. The legal regime must be ready to “punish” good-faith encroachment, so as to deter bad-faith trespass and provide incentives for developers in general to verify land boundary.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

173

Moreover, excessive boundary surveys are not a universal problem. Surveying costs less in countries with a rectangular system than in those with a metes-and-bound system. Surveying is quite cheap in, for example, Taiwan where a government agency is in charge of surveying (upon application) and the ascertained boundary is definitive. Developers will not survey boundaries excessively, as a one-time, official survey is sufficient to ensure that their constructions will not be removed for encroachment. Furthermore, with the advancements of Google Earth, GPS, and other information technology and applications, there is little room for developers to claim their good faith for new construction projects, as ascertaining land boundaries has become more and more accurate and low-cost. Hence, an efficiency-minded crystal ball would predict that the building encroachment doctrine, if preserved at all, would move toward emphasis on prompt protest, and good faith would one day lose its entire import, as every developer is bad faith under my proposed formula.

B  A Two-Tiered Structure This section discusses whether the building encroachment doctrine should be structured, and argues for a two-tier system that has not been adopted. From an ex ante economic standpoint, as discussed above, there is less and less room for any kind of exception to the baseline exclusion strategy, as the cost of verifying land boundaries has greatly decreased. The building encroachment doctrine, in its nature as an exception, can be economically justified from an ex post viewpoint – it creates social waste to demolish useful resources (encroaching constructions). Hence, the building encroachment doctrine attempts to strike a balance between ex ante efficiency and ex post efficiency. Which design does a better job?

1  Ex Post Inefficiency When the encroachment is ex post inefficient (i.e., the encroached land could be put to more valuable use than serving as ground for the encroaching construction), Type A may have an edge over Types B–D. If the encroachment is ex post inefficient and by definition ex ante inefficient, sticking to the baseline exclusion strategy and ordering removal are clearly efficient. Under Types B–D, however, encroachers may be good faith and thus neighbors lose the rei vindicatio claim. That is, good faith should be a necessary but not sufficient condition for preservation. If transaction cost is low, post-litigation Coasean bargaining would still lead to the

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

174

Part II: Immovable Property

(ex post efficient) result – removal – by neighbors’ paying off encroachers. Nonetheless, if transaction cost is high due to bilateral monopoly (Sterk 1987: 57–58) or animosity between two parties (Farnsworth 1999; 2016), the ex post inefficient construction may persist. One plausible solution under Types B–D is to add more conditions, such as lack of prompt protest, like many countries already did. Even better, statutes can make ex post efficiency as a necessary (but not sufficient) condition for deviating from the baseline exclusion strategy. After reading the property laws of more than 150 jurisdictions, however, I am pessimistic about lawmakers’ making economic consideration explicit in statutes. By contrast, most Type-A jurisdictions (see Table 6.3) ask courts to balance all factors or consider interests of both parties. An efficiency-minded court can rule for removal when the encroachment is ex post inefficient and rule for preservation only when the encroachment is ex post efficient, the encroacher is good faith, the neighbor fails to protest promptly, and the circumstances suggest that the ex ante effect on future potential encroachers is small. The downside of such designs is that if courts are not efficiency minded, an all-things-considered standard often fails to provide guidance.14

2  Ex Post Efficiency If the encroaching construction is ex post efficient, courts should consider the ex ante effect on efficiency. That is, judicial ruling of preservation may induce potential encroachers to spend less on boundary surveys and thus increase the probability of building encroachment.15 Again, Types B–D would ignore the ex ante effect on efficiency, while Type A can accommodate it, if done properly. If encroachers’ good faith and neighbors’ prompt protest are always accounted for in the correct way, a Type A design is more likely than other designs to lead to removal, as in this scenario, given the requirements of good faith and prompt protest are met, Types B–D generally results in the deprivation of neighbors’ rei vindicatio claim, whereas courts under Type A have discretion to order removal in any case. If we stick to law-and-economics’ general emphasis on ex ante efficiency, Type A is to be preferred again. 14 15

The concern is particularly acute in developing countries (Schäfer 2019). My empirical study (written in Chinese) on Taiwanese courts of first instance in 2009– 2020 finds that in 20% of the observations in which courts use the discretionary power to preserve encroaching constructions. Chang (2014)’s empirical study observes Taiwanese courts of first instance in 2009–2012.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

175

3  Safe Harbor and Sure Shipwreck To strike the best balance between ex ante and ex post efficiency, the best design seems to be a two-tier system. The first tier contains two ex ante tailored rules, whereas the second tier is an ex post tailored standard. One of the ex ante tailored rules is what Morse (2016) calls a sure shipwreck – courts have to order removal if encroachers are bad faith. The other rule is that courts have to order preservation if encroachers are good faith and neighbors fail to protest promptly. It is not a typical shipwreck for neighbors because if neighbors procrastinate but encroachers are bad faith, courts should still order removal. This is not a safe harbor for encroachers, as they cannot control whether neighbors protest in time. When encroachers are good faith and neighbors oppose without delay, the ex post tailored standard kicks in. Courts can consider other factors, notably the relative and absolute loss suffered by the two parties in removal and preservation. Here, a clear rule, such as more than 10 square meters or larger than 10% of the encroached land, could be gamed by encroachers. It is better to leave it to the court to make a holistic decision.16 C  Put-Option Rule Efficient? Building on the discussion of call-option and put-option liability rules in Chapter 2, this chapter uses the building encroachment scenario as a springboard to discuss whether a put-option rule exists and, if so, whether it is efficient. In the building encroachment context, a landowner with a typical Rule 6 is entitled to force any adjacent landowner to purchase his land. A landowner with an Ayresian Rule 6 can either request removal or request compensation against his encroacher. As shown in the previous part, no studied jurisdiction enacts a typical Rule 6 – for obvious reasons. In addition, no countries except Serbia (LBOPR) §25 appear to stipulate an Ayresian Rule 6 because there is no need to do so. First, a claim of injunction implies that its holder can threaten to exercise it in order to force the other party to enter a deal that pays the holder.17 Second, if an encroaching construction is worth more than the 16

17

If the encroached party has the power to remove, and uses it to extract unreasonably high payment from the encroaching party, the court may notify the tax assessors of the high values the encroached party places on her land. As Ayres (1998: 816) recognizes, the claim for purchasing encroached land here has little value for neighbors, as they can increase their payoff by demanding an injunction and then bargain for a larger payment with encroachers.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

176

Part II: Immovable Property

land underneath, an encroacher has every reason to strike a deal with his neighbor in order to avoid removal, and the neighbor, who may strategically hold out, would be the source of high transaction costs that impede a trade. Giving the neighbor an extra claim to “force” a purchase (exactly what the encroacher plans to do voluntarily and desperately) does not make economic sense. The critical difference between Rule 1 and Rule 2 is not that an entitlement holder protected by Rule 1 cannot request damages instead, but that an entitlement holder protected by Rule 2 cannot request an injunction. Notably, a pure Rule 1 and an Ayresian Rule 6 have nuance differences when encroachers are willing to tear down encroaching constructions (because encroaching constructions are worth less than the compensation required). Here, under a pure Rule 1, neighbors cannot use their injunctions to bargain for a voluntary sale of encroached land, as injunctions are not threating to encroachers who rather remove encroaching constructions. Neighbors would still have unjust enrichment claims against encroachers for their benefits during the encroachment, but no more. By contrast, under an Ayresian Rule 6, neighbors are entitled to force a purchase of encroached land even if encroachers would rather remove encroaching constructions.18 There does not appear to be any efficiency advantage in forcing such an exchange. On the other hand, a claim for forcing a purchase of the entire plot owned by neighbors, not just the encroached part, makes more economic sense. If courts use market value to calculate the sale price due, and neighbors can sell the unencroached part at or above the adjudicated prices in the market themselves, this legal design does not put neighbors in a better position. Nonetheless, if due to smaller sizes or odd shapes, the unencroached part would become unmarketable, a claim for forcing a purchase of the entire land internalizes more losses by neighbors. This expected extra payment to buy the entire land also incentivize encroachers to be more (i.e., close to optimally) careful. Note that countries that have this rule do not provide it as an alternative to the claim for injunction; rather, it is a substitute for the claim for injunction. Hence, these countries cannot be described as adopting an Ayresian Rule 6. 18

Whether encroachers will be forced to purchase the encroached land, no matter whether they have torn down the encroaching constructions before a court verdict, depends on how the law is designed and interpreted. Ayres (2005: 28), citing Pile v. Pedrick, argues that courts would force encroachers to purchase encroached land even if encroachers prefer to remove encroaching constructions.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

building encroachment

177

Lee Fennell raised an interesting point to me: why is “buying the entire land” efficient while “buying the entire time (of only the encroached land)” is not? My argument focuses on the unencroached part. The unencroached part may be rendered useless once the encroached part is bought out, so neighbors should be able to demand encroachers to buy the unencroached part to reduce losses (caused by encroachment!). By contrast, the unencroached “time” (i.e., since removal to time eternity) is highly unlikely to be rendered valueless. Hence, neighbors’ put option to sell the entire land to encroachers may be efficient.19 The building encroachment doctrine is more accurately described as switching from Rule 1 to Rule 2 under certain conditions. Ayres & Goldbart use the building encroachment doctrine as an example of a “dual chooser rule,” under which the plaintiff (encroacher) has a call option and the defendant (neighbor) has a put option (Ayres and Goldbart 2001: 35–37). It should be clear by now that the doctrine does not work as indicated by their model. First, in Ayres & Goldbart’s words, the encroacher only “signals a willingness to buy property” but does not actually “strike” and reveal that her value is higher than the exercise price. Moreover, in the “intentional encroachment” cases discussed by Ayres & Goldbart, no country awards bad-faith encroachers with a call option, as neighbors can insist on removing encroaching constructions. Finally, neighbors’ entitlement is not protected by a typical Rule 6. The claim for forcing encroachers to purchase the encroached land is a replacement for the more powerful claim for injunction.

D  Institution Cost The final factor to consider is the institution cost of the building encroachment doctrine. The preceding analysis only accounts for the benefit side of the cost-benefit analysis; that is, which party (neighbors or encroachers) values the encroached land more. Nonetheless, from the efficiency standpoint, even if the building encroachment doctrine could overall speaking achieve higher allocative benefits, policymakers still have to consider its cost. The building encroachment doctrine, for instance, deviates from 19

The same analysis applies to the partial takings context (Bell and Parchomovsky 2017). The de minimis exception, under which condemnees should not have claims against the state to condemn the entire land, should be thought of from the viewpoint of the remaining part. That is, property owners should be allowed to demand condemnation of the entire plot when what the government has not condemned is rendered useless. Whether the condemned part is physically small or barely intrusive in itself is not the key.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

178

Part II: Immovable Property

the baseline exclusionary strategy. Encroachers may refuse to go away or settle; rather, they may litigate and try to prove themselves good faith. In countries that give courts the discretionary power and courts’ jurisprudence is unpredictable, many lawsuits may be filed. By contrast, under the clear-cut exclusionary strategy, parties may bargain privately under a “clear shadow” of law. Whether the institution cost would be high depends on each country’s institutional environment (litigiousness, predictability of court ruling, etc.) and a general claim cannot be made.

III Conclusion The building encroachment doctrine, if designed properly, is better fitted to deal with this type of boundary disputes than the adverse possession doctrine and the accession doctrine. This chapter argues that encroachers’ good faith is a necessary but not sufficient condition to preserve encroaching constructions. The prompt protest rule reduces opportunism. An ex post tailored standard under which courts have discretionary powers but some important factors are laid out is less likely to lead to inefficient preservation, if courts are efficiency-minded. Still, if courts are instead trespass-friendly, judicial discretionary powers may lead to too much preservation. The two-tier structure in Taiwan and Argentina is better. At the first tier (at the “law” level, using ex ante tailored rule), encroachers’ bad faith surely leads to removal, and if they are good faith, neighbors’ failure to oppose in time upon knowing the encroachment surely leads to preservation. Then, at the second tier (at the meta-law level, using ex post tailored standard), when encroachers are good faith and neighbors protest promptly, courts can take other factors into account and decide whether to preserve or to remove. Conceptually, the building encroachment doctrine is not a put-option rule, but a conditional limit on the property rule. The law-and-economics literature has used the boundary dispute as a springboard for advanced analysis but important details of how the doctrine is actually designed are lost. Using real-world designs of the doctrines, this chapter reveals why characterizing – or reforming – the doctrine as a put option is ill advised.

https://doi.org/10.1017/9781009236553.009 Published online by Cambridge University Press

7 Co-Ownership Partition Proposing a New Auction-Based Design This chapter deals with the most prevalent type of concurrent ownership, referenced in 145 jurisdictions (93%) and known as “tenancy in common” in common law. It is difficult to identify the defining characteristics of such a form.1 One key element is perhaps that each co-owner has her own fractional share and does not require co-owners’ consent in alienating the share – notwithstanding other co-owners’ right of first refusal— for example, Bulgaria (OA) §33 and Mongolia §108. In addition, as tenancies in common generally do not come with a separate purpose for maintaining the co-ownership (a business partnership would be such an example), in 117 of the 131 jurisdictions (90%) with a clear judicial partition doctrine, any co-owner is explicitly given the power to break up the co-ownership at any time (e.g., France §815) – though some countries disallow co-owners doing so at inconvenient times, and some (e.g., Austria §830) give courts discretion in postponing partition for a certain period. Hence, I take separate and alienable shares and unilateral power to partition as candidate key elements of tenancy in common. These elements are important because there are numerous ways that co-ownership is described in civil codes and other statutes – indivision, the French term, comes to mind. To avoid confusion, hereinafter this chapter will refrain from using the term “tenancy in common” whose feudal connotation makes the term a poor fit for use in comparative law. Instead, this chapter will simply use the term “co-owner(ship)” to describe this legal relationship. “Concurrent ownership” is used as an umbrella term for all types of legal relationships under which two or more persons share ownership. The Conclusion chapter further discusses other types of concurrent ownership. 1

Common-law countries traditionally view joint tenancy as having four unities while tenancy in common having only the unity of possession. The four unities are not the way many civil law countries distinguish between their two (or more) types of co-ownership. From a modern common-law view, perhaps the most distinctive feature of joint tenancy is the right of survivorship (Ziff 2014: 338), but in many countries none of the form of co-ownership has the feature of the right of survivorship.

179

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

180

part ii: immovable property

The chapter starts with how co-ownership can be terminated upon a coowner’s striking her unilateral power to partition. Then, it briefly recounts the rule regarding whether, and for how long, the law allows co-owners to enter covenants to maintain co-ownership. Finally, this chapter summarizes the quorum rules for establishing management plans and selling the entire co-owned thing.

I  Comparative Overview The co-ownership partition doctrine exhibits more variety than I imagined.2 While a majority of jurisdictions prefer “partition in kind” (i.e., physical division of a plot with N co-owners into N subplots whose sizes are exactly proportionate to co-owners’ share percentages), and pragmatically allow other partition approaches when partition in kind is infeasible or too costly, secondary schemes cannot be simply grouped under “partition by sale.” First, many countries employ public auctions (i.e., anyone can bid) while others use internal auctions (i.e., only co-owners can bid), and some allow both. Second, some countries give co-owners call or put options to transfer shares. Third, some countries utilize “partial partition” which itself includes a variety of solutions. For instance, a court may decide to maintain co-ownership of a part of the land to serve as a passage (see Chapter 9). In addition, a court can partition the land to auction off a piece while physically dividing the rest. Partition by sale, even as an umbrella term for the first two (or even just the first) types of legal arrangements is too vague. This chapter thus will use the term partition by sale only as a synonym for partition by public auction. As partial partition comes in a variety of forms (e.g., Dukeminier et al. 2010: 345–346), the comparative overview below omits the details. Among the 65 jurisdictions that allow partial partition, 23 explicitly allow coowned things to be assigned to some, but not all, of the co-owners (e.g., Estonia (TLPA) §77), while 18 require co-owners’ consent or application to do so (e.g., Brazil §1322 and Peru §988) – very likely because co-owners who receive disproportionately large subplots have to compensate other co-owners and may not have the means to do so. Partial partition does not include those types of legal schemes that give co-owners a call or put option. Partial partition occurs at a court’s discretion in settling disputes. A court may order one, or some, of the co-owners (each) to become sole owners and compensate other former co-owners – the 2

Bulgaria (OA) §34 and Libya §845 defer the detailed design to other laws.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

181

Partition approach Types E and F Type D Type C Type B Type A No information or no rule

Figure 7.1  Typology of partition approaches

compensation is called owelty in the United States (Casagrande 1985: 764). By contrast, a call or put option is a co-owner’s Hohfeldian power. Exercising a call option often leads to one co-owner becoming a sole owner. Exercising a put option enables some, or all but one, co-owners to exit the co-ownership relationship. Two hard cases are Moldova and Romania, which were coded as having partial partition, rather than awarding call options, because the courts have discretion, when partition in kind is infeasible, to choose between attributing the whole plot to one or a few co-owners, and ordering a public auction. If co-owners have call options, when they strike the calls, courts cannot opt for public auctions. Albania §207 is another example of partial partition – co-owners do not appear to have Hohfeldian powers to force others to sell their shares. Below I describe the six prototypes of partition approaches, one by one, and distinguish Types A, B, and C into numerous subtypes. Figure 7.1 provides an overview of the distribution of these approaches around the world.

A  Prefer in Kind to Public Auction Type A1. Partition in kind is preferred. If partition in kind is impossible or causes great damage, public auctions should be used: Afghanistan §1967, Argentina §2374, Austria §843, Brazil §1322,3 Cambodia §212,4 Canada (Ziff 2014: 351–353),5 China §304, Costa Rica §§272–273, Croatia 3 4 5

If the bids are the same, co-owners, relative to outsiders, have rights of first refusal. Courts can also order partition by sale if “proper grounds” exist. In Canada, courts in general have some discretion as to partition. In Ontario, partition in kind is preferred (Ziff 2014: 352 fn. 89).

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

182

part ii: immovable property

(AOORR) §50, Cuba §166, Cyprus (Neocleous 2000: 607; Synodinou 2020: 97), Ethiopia §1272, Equatorial Guinea §404, Finland (Hollo 2012: 87), Georgia §963–964, Germany §§752–753, Greece §§799–801, Guatemala §494, Honduras §§2224–2225, Hong Kong (PO) §§4–6 (courts have large discretions), Hungary §V:84 (Gocza 2013), Iran §598, Iraq §1073,6 Ireland (de Londras 2011: 168–173), Israel (LL) §§39–40, Japan §258, Liberia (LCT) §80, Louisiana §§810–811, Macau §1312 & (CPC) §946, Malaysia (Maidin and Ali 2014: 153), Malawi (RLA) §99, Mexico §940, Moldova §361, Monaco §§703–704, Morocco (Gignoux 2015: 14) (law unclear), New York (RPAPL) §915, Nicaragua §1704, North Korea §43, Norway (CA) §15, Panama §412, Philippines §498, Poland §212 & (CPC) §625, Portugal (Passinhas 2019: 98), Puerto Rico §338, Romania §676, Scotland (Gretton et al. 2017: 150), Slovakia §142, South Korea §269, Spain §404, Taiwan §824, Turkey §699, Uruguay §1756, and Venezuela §§769–770 & §§1070–1071. Type A2. Partition in kind is preferred. If partition in kind is impossible or causes great damage, public auctions or other means the courts see fit should be used. The courts’ discretionary power under Tanzania (LA) §164 distinguishes Type A2 from Type A1. Type A3. Partition in kind is preferred, unless the court finds that public auction is more equitable; for example, California (CPC) §872.810 & §872.820. Type A4. Under Vietnam §219, partition in kind is preferred, and the back-up plan is for co-owners who wish to partition to sell their shares (unclear whether this is a put option).

B  Prefer in Kind to Internal or Public Auction 1  Internal Auction as Default Type B1a. Partition in kind is preferred; if impossible or value-reducing, courts hold internal auctions among co-owners, but any co-owner may request that auctions be open to outsiders to bid. Internal auctions as the default are implied, as can be inferred from the rule that any co-owner can request that the auction be open to outsiders. Included in this group are Bolivia §§169–170, Burundi §§III:350–351, Chile §2313 & §1337, Ecuador §2213 & §1353, El Salvador §1217, Guinea §903, Haiti §§1459–1460, Malta §§515–517, and §§1686–1687 of Burkina Faso, Comoros, Dominican 6

Iraq’s design appears to contain a call or put option, but it does not. The Iraq design lacks a call option as no co-owner can buy shares without the consent of others; neither is it a put option as no co-owner can sell shares without others’ being interested in purchasing.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

183

Republic, France (but cf. Dyson 2003), Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, and Togo. Type B1b. Partition in kind is preferred; if impossible or value-reducing, then internal auction among co-owners is preferred. If internal auctions do not work out, courts should order public auctions. For example, Pakistan (PPIP) §§9–11.

2  Public Auction as Default Type B2a. Partition in kind is preferred; if impossible or value-reducing, then auctions (presumably public ones) – but co-owners could unanimously agree to put co-owned things to internal auctions: Algeria §728, Bahrain §797, Egypt §841, Kuwait §836, Qatar §869, Syria §795, and Tunisia (CDR) §120. Belgium §827 appears to have the same rule: If the things cannot be conveniently partitioned in kind, any co-owner can demand a public auction. However, parties, if all adults, may agree that auctions be made before a notary of their choice – presumably internal auctions.7 Type B2b. Partition in kind is preferred; if impossible or valuereducing, then auctions (presumably public ones). Courts may decide to use internal auctions. Czech §1144 & §1147 is the only example. Types B2a and B2b differ because, in the former, internal auction is used only when co-owners all agree. Type B2c. Partition in kind is preferred; if impossible or value-reducing, then public auctions. If it is impossible to transfer the property to third parties, then internal auctions. Mongolia §108 & §489 is the only example. 3  No Default Type B2d. Partition in kind is preferred. If partition in kind is impossible or causes great damage, public or internal auctions should be used. Liechtenstein (SR) §30, Switzerland §651, and Thailand §1364 are three examples. Type B2d differs from the Types B2a and B2b because neither public nor internal auctions are preferred. C  Prefer in Kind, with Call or Put Options 1  Public or Internal Auctions, plus Put Option Type C1. Partition in kind is preferred. If it is impossible or valuereducing, then auctions (presumably public ones), but all co-owners 7

Belgium §§1686–1687 is the same as the French counterpart. The rule stipulated in § 827 is essentially the same as that in §§1686–1687.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

184

part ii: immovable property

could unanimously agree to put co-owned things to internal auctions. In addition, a co-owner may petition to sell her shares to the other co-owners. United Arab Emirates §1166 is the sole example. Type C1 is similar to Type B2a, but the latter does not have the put option design.

2  Public Auction, plus Put Option, without Internal Auction Type C2. Partition in kind is preferred. If it is impossible or valuereducing, then auctions (presumably public ones). In addition, a co-owner may petition to sell her shares to other co-owners. Jordan §1044 is the sole example of Type C2. Type C2 is similar to Type B2a and Type C1. Compared to Type C1, this type does not have the internal auction design. Compared to Type B2a, this type has the put option design. 3  Public Auction, with Call Option Type C3a. Partition in kind is preferred. If physical division is not easy, then public auctions are used. Co-owners who do not apply for public auctions can buy the rights of other co-owners who apply for public auctions: for example, Albania §207 and Colombia §2334 & §2336. Type C3b. Partition in kind is preferred. If not easy, then public auction. Co-owners can buy the shares of other co-owners at tax-assessed or court-determined value, to avoid public auction. Italy §720, Peru §§988– 989, and Slovenia (Škerl and Vlahek 2020: 123). Type C3c. Partition in kind is preferred. If not easy, then public auction. A co-owner can buy the shares of other co-owners if there are “serious reasons” for co-owned things to belong to her as a whole. Macedonia (AOORR) §52 & §56 is the only example. 4  Partition of Share, with Put Option Type C4a. Partition in kind of a co-owner’s share is preferred (other coowners presumably remain in co-ownership).8 If impossible or valuereducing, then the co-owners who demand partition can request other co-owners buy the former’s share. Armenia §197, Azerbaijan §220.3, Belarus §255, Kyrgyzstan §271, Laos (PL) §25, Lithuania §4.80, Russia §252, Tajikistan §300, Ukraine §364, and Uzbekistan §223 adopt this type and do not have public auctions. Type C4b. Kazakhstan §218 adopts this type but courts could order public auction. 8

New York (RPAPL) §916 also allows for such partitions of shares.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

185

Type C4c. Quebec is idiosyncratic.9 If one co-owner wants to exit coownership, the remaining co-owners may by unanimous decision activate §1033, under which the former can either partition her share in kind or sell her share to the latter (resembling a put option). If co-owners cannot reach agreements, and partition in kind is inconvenient, §863 allows courts to order sales. Partition in kind of the entire co-owned thing is not the most likely outcome. By way of comparison, put options here are exercised by active coowners who seek partition, while call options, as in the case of, for example, Albania, are exercised by passive co-owners.

D  Prefer Public Auction to in Kind Type D. Public auctions are preferred. Nonetheless, upon application of an interested party, and “the property may be conveniently and profitably divided in kind amongst those entitled,” the court may order partition in kind. Seychelles §§819–821 appears to adopt this design.10 Sweden (CA) §7 sets public auctions as the default but allows co-owners to demand partition in kind.

E  Trustees Decide Type E. England (LPA) (Sparkes 2019: 69) imposes a trust in all cases of co-ownership. Partition in kind has become extremely rare. This rule, functionally, is similar to the Seychelles §818 rule that imposes a fiduciary in all cases of co-ownership of immovables.

F  Judicial Discretion Type F. Both partition in kind and public auction are allowed: Australia (Hepburn 2002: 151; Edgeworth 2017: 286–288), Eritrea §1099, Estonia (LP) §77, India (Mitra 2007: 344–347), Indonesia §§1066–1085,11 Latvia §1075, the Netherlands §III:185, New Zealand (PLA) §339, Singapore (See et al. 2018: 88–91) (courts would not order partition in kind if the government did not approve it), South Africa (Van der Merwe et al. 2002: 324–325), and Sri Lanka12 (PL) §26. 9 10 11 12

I thank Prof. Hugo Tremblay for explanations. The rule in Seychelles is unclear, as §§1686–1687 also have the rule that countries in Type B1a have. The Indonesian rule is unclear from the civil code and commentaries (Leks 2018: 40–41). Rupesinghe and Rupesinghe (2007: 100) suggest that in practice partition in kind is preferred.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

186

part ii: immovable property

II  The Fragmentary Land Problem Fragmentary land is land parcels that are too tiny to be used in a more valuable way that would be feasible if the land parcels were larger. Thus, strictly speaking, the fragmentary land problem is not merely a problem peculiar to co-ownership, but also one faced by owners of tiny plots – whether these plots are, or were, solely or jointly owned does not matter. This part, therefore, focuses on the fragmentary land problem arising from co-ownership. Partition in kind and partial partition, but not public or internal auctions, may create fragmentary land. The fragmentary land problem exists only when (1) the size of the coowned land parcel is too small to be economically viable; (2) there are so many co-owners that decision-making costs are very high; and (3) the coowned land has undergone physical division. That is, if the plot is vast, or the number of co-owners small, partition in kind or partial partition still would not result in fragmentary parcels of land. If the first two conditions hold, the question is whether physical division will be adopted when it should not. That is, will courts, in handling partitions of co-owned land, create fragmentary land? As Part I shows, all studied jurisdictions allow partition in kind and most, at least according to the law on the books, prefer it.13 Courts, thus, may order partition in kind and inadvertently create anticommons, merely in deference to the law’s expressed preference. Heller (2008: 121) observes that fragmentary land will become a significant problem if “we give landowners strong reasons to divide land down to big-inch level and provide no countervailing mechanism to bundle the pieces back into productive use” (emphasis added).14 If landowners exhibit this preference (or lack of awareness) for fragmentary land, and courts often do not realize its cost, inefficient fragmentary land will be created. Not all real-world examples of fragmentary land that Heller (2008) has discussed, however, follow from the preceding dynamic. Heller (2008: 130) has argued that “the African American, Native American, and Irish cases all share a tragedy of the anticommons thread. The law 13

14

Bell and Parchomovsky (2005: 601) claim that the courts in the United States “have long exhibited a preference for partition in kind.” However, Merrill and Smith (2007b: 643) and Stoebuck and Whitman (2000: 223) observe that courts in the United States have increasingly favored partition by sale. Heller (2008) does not specifically indicate that courts are to blame. I examine judicial partition in kind nonetheless because it is more likely to lead to fragmentary land.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

187

encouraged farmers to fragment their resources down to the big-inch level and offered them no tool with which to aggregate land back to a usable scale.” These three cases, however, would not be pooled together under my framework. African Americans may leave wills to assign their estates as they wish and avoid fragmentary shares and fragmentary land altogether. Native Americans have neither the right to transfer their shares, nor right to petition for partition; hence, they face a “fragmentary share problem” (difficulty in administering co-owned resources; see Chapter 8) that never leads to a fragmentary land problem. By contrast, because Irish law stipulates that each male descendent has an inheritance right, Irish farmers must equally divide land among their sons; thus, they suffer from a serious fragmentary land problem and the tragedy of the anticommons.15 Theoretically, none of the partition laws in the studied jurisdictions necessarily lead to fragmentary land if courts avoid using partition in kind in the appropriate cases. Ultimately, empirical studies of court practices in each jurisdiction are warranted to clarify the actual situation.16 Chang (2012d) studies partition cases in the courts of first instance in Taiwan and finds that courts tend to order public auctions, rather than partition in kind or partial partition, when ordering the latter would lead to fragmentary land. Similarly, plaintiffs in those cases tend to ask for partition by sale, rather than partition in kind or partial partition, when they would otherwise receive tiny plots. This finding, granted, is not readily generalizable to other countries, but without solid empirical studies finding the opposite, we should not lose faith in courts (or, for that matter, plaintiff co-owners) too soon. Empirical studies from Taiwan inform us that, while courts and coowners are not inclined to create fragmentary land, this does not mean that they do not occasionally do so. Some courts follow the literal meaning of partition law and order partition in kind, no matter what. As Part III will discuss, disfavoring partition in kind over public auctions is not a panacea, as public auctions have their own problem. More importantly, both physical divisions and auctions fail to consider the different subjective values held by co-owners. Part III thus proposes a new partial partition approach that has yet to be used in any jurisdiction. 15 16

Langer (1975: 56), the source relied on by Heller (2008: 237) to demonstrate Irish problem, indicates that the small size of Irish farms was due to subleasing, not inheritance, though. Nonetheless, there appears to be no dataset regarding judicial partitions in the United States (Mitchell 2005: 598) or elsewhere.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

188

part ii: immovable property

III  A Proposal for a More Efficient Partition Approach This part first considers more generally the implications of different partition approaches (summarized in Part I) on the bargaining dynamics that precede resorting to judicial partition (see also Waldeck 2011: 753–754).17 Second, this part considers the potential impacts on efficiency of intermediate and partial forms of partition, which are prevalent both in the law on the books and law in action. Considered together, these two branches of analysis show how a new partial partition approach advances the efficiency of co-ownership by policing bargaining behaviors and inducing revelation of honest valuation.18 I have developed with Lee Fennell the new partial partition approach (hereinafter the “CF approach”) (Chang and Fennell 2014b; 2014a). The challenge in this part is to design partition approaches that allow subjective values to be considered in deciding how to partition property without encouraging socially wasteful struggles over surplus in earlier periods. A useful partition approach must also deal reasonably well with the liquidity and coordination shortfalls that can hamper revelation of coowners’ subjective value in the co-ownership context. No first-best partition solution is achievable under real-world conditions. The internal auction championed by Cramton et al. (1987), discussed in Chapter 10, does not guarantee efficiency because co-owners do not necessarily have roughly equal shares. Hence, the CF approach, while also failing to ensure efficiency, is a significant improvement over the existing legal arrangements (Part I). The considerations relevant to the choice of a new partition mechanism are laid out below.

A  Beyond a Binary Ex Post Analysis 1  Subjective Valuations and Economies of Scale Subjective valuation and economies of scale feature prominently in existing economic analyses of the choice between partition approaches (Miceli and Sirmans 2000: 789–793). In essence, their argument is as follows: If partition in kind does not decrease the economy of scale, partition in kind is always efficient. If partition in kind does decrease economy of scale, partition in kind is sometimes efficient because it preserves 17 18

Baucells and Lippmant (2001) examine the role of court delays in generating bargaining leverage for co-owners independent of the substantive rule that will be (eventually) applied. For discussion of non-efficiency concerns, see Craig-Taylor (2000) and Mitchell et al. (2010).

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

189

subjective value, whereas partition by sale is sometimes efficient because it prevents a fall of market value (MV) due to fragmentation. To be more exact, when the aggregate subjective value of the nonconsenting owners is larger than the premium derived from economy of scale, partition in kind should be used. Bell and Parchomovsky (2005: 601) disagree with this analysis, favoring partition by sale across the board (but see Claeys 2011: 36). Their argument is that as long as co-owners participate in the auction, co-owners with high subjective values can preserve their subjective values by winning the bid. Participation of co-owners with high subjective values in the auction does not ensure efficient results, however, if more than one co-owner has subjective value in the property. The key is that subjective value is idiosyncratic and nontransferable. Separate co-owners may be unable to coordinate to bid their joint valuations. Perhaps they are too numerous or contentious to coordinate easily, or they hold idiosyncratic attachments to different but overlapping subsets of the land. Liquidity often presents another obstacle. In American jurisdictions, for instance, the winning bidder has to pay the auction price in cash, and many co-owners will not be able to do so (Reid 1986: 872). This problem could be mitigated if banks could provide a quick loan to winning bidders. As the property’s assessed value determines the amount that a bank would be willing to lend, however, speedy loans may be insufficient to enable financially constrained co-owners with idiosyncratically high subjective valuations to protect their interests in the property. It is also worth noting that properties are often sold at a price lower than MV (Mayer 1995; 1998). Such an auction discount does not necessarily signal inefficiency. Most likely, the initial purchaser will be a middleperson who then resells. If the ultimate purchaser is the high valuer, the only question is whether transferring the property to her in two transactions rather than one adds net costs. Due to the problems of liquidity and coordination, however, highest-valuing co-owners may fail to end up with the property. These explain why partition in kind could sometimes be more efficient than partition by sale. However, as argued below, partial partition in kind may do even better.

2  Partition’s Shadow The analysis above assumes that the efficiency of a partition rule can be evaluated based solely on the partition event itself (Miceli and Sirmans 2000: 788). That said, partition rules carry implications for earlier, ex ante decisions (Bebchuk 2001: 633–634). The analysis focuses on the

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

190

part ii: immovable property

ex ante impact of judicial partition rules on pre-partition bargaining, but it is worth noting that a series of earlier decisions, including the decision to become a co-owner or to make investments in developing subjective attachments to the land, may also be influenced by the partition doctrine. Central to the analysis is an understanding of judicial partition as a private taking (Bell 2009: 565–566). One or more of the co-owners will be coercively dispossessed of her undivided fractional share in the property and given either land or money. Depending on the applicable partition rules, a co-owner might use the judicial process to “take” the property of other co-owners for less than its value.19 As co-owners are also potential takees, they may engage in costly stratagems to stave off the taking or to bring it on, depending on the level of expected compensation. These concerns, well-recognized in the eminent domain context, are heightened in the partition context because the parties themselves have considerable control over whether the taking will occur and considerable insight into the compensation that will be provided in the event the taking occurs. Type C4 enables co-owners to be apportioned only their share, without the power to break up others’ co-ownership. That is, Type C4 does not allow private takings. Its put option design instead constitutes private forcings (Fennell 2014). Neither an across-the-board rule of partition by sale, nor a blanket rule of partition in kind, can reliably control pre-partition strategizing and encourage efficient voluntary partitions. Either rule may offer some co-owners an attractive strategic opportunity to pursue the less socially valuable type of partition in a given case. Suppose there are two types of co-owners, some with high subjective values (HSV co-owners), and some whose valuations merely track the MV of the property (MV co-owners). Many co-ownership situations will involve a mix of these co-owner types. Consider, first, partition by sale. HSV co-owners will be fearful of the judicial partition process if they lack the financial ability to emerge as the highest bidder in a sale. Knowing this, MV co-owners may attempt to extract side payments from the HSV co-owners for agreeing not to seek judicial partition. These efforts (and the efforts of the HSV co-owners to resist making such payments) will be socially costly, regardless of the distributive result. 19

The point is stronger here than in the context of eminent domain. When the government is the taker, it is not entirely clear whether and how monetary payments influence incentives to take (Levinson 2005: 916, 969; Chang 2013: 14–46, 75–89, 138–139, 158–160). For private parties, the relationship between monetary payoffs and incentives is far more straightforward.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

191

The bargaining situation is not necessarily improved if the HSV coowners are able to bid in the partition sale. MV co-owners may resist an efficient voluntary agreement to partition the property in kind if they believe they will get a greater surplus from a partition sale. This could occur if the partition sales procedure will induce an HSV co-owner to bid her true valuation and will require her to disgorge equal shares of that winning bid to her co-owners. This procedure will effectively transfer some of the HSV co-owner’s subjective surplus in the property to the MV co-owners, and it will do so at positive cost. Interestingly, this problem becomes more severe the better the judicial sales procedure is at eliciting honest revelations; a blunter procedure would reduce the incentive of the MV co-owners to strategically seek judicial partition. An inverse problem exists when bargaining in the shadow of judicial partition in kind. Here, the HSV co-owners could stand on their rights and threaten to invoke the judicial partition procedure even when the economy of scale of the intact plot would garner a higher price for MV coowners. In such a case, the HSV co-owners might threaten to seek judicial partition in kind unless they are paid off by the MV co-owners. Indeed, some MV co-owners might masquerade as HSV co-owners in an effort to glean more of the surplus that a sale of the property would bring. These efforts would again be socially costly, even if unsuccessful. Moreover, the HSV (or faux HSV) co-owners could proceed to invoke judicial partition in kind in order to gain ownership of fragments that could enable them to operate as holdouts in any later-attempted reassembly of the land. If the parties were always certain that the efficient judicial partition choice would be made by courts – whether partition in kind or partition by sale – none of the co-owners could threaten the other with an inefficient procedure merely to extract surplus. To make the efficient choice, the court needs to ascertain whether the value derived from economy of scale is larger than the subjective value lost through auctioning the coowned properties off. While the extent of the economy of scale can be assessed objectively by hedonic regression models, subjective value is notoriously unobservable by third parties, such as courts (Chang 2012a: 64). Miceli and Sirmans (2000: 793) suggest a rough proxy: that partition in kind should be preferred when the land parcel is large and the number of co-owners is small. However, this proxy is insufficient to ensure efficient results and may not successfully control strategizing.20 20

For example, Chang and Fennell (2014a: 12–13) analyze situations involving diseconomies of scale that may also be suitable for partition by sale.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

192

part ii: immovable property

In all jurisdictions, partition law leaves open the possibility of either partition procedure. On one view, this lack of clarity may impede the voluntary partition bargaining process. Parties may be too eager to go to court, each believing he or she can convince the court that economies of scale outweigh subjective valuations, or vice versa. On the other hand, the uncertainty of the judicial partition outcome could reduce certain forms of strategizing and improve bargaining (Ayres and Talley 1995). Ideally, a legal rule would accommodate co-owners’ efficiency-enhancing reasons for requesting or resisting a particular partition approach (e.g., that it will preserve or destroy subjective value, or that it will realize or undermine economies of scale) without encouraging costly efforts to wrest surplus from other co-owners.

3  Intermediate Partition Approaches The existing economic literature on partition primarily focuses on the choice between partition in kind and partition by sale (Bell and Parchomovsky 2005: 600). That said, Part I shows that there are other types of partition approaches. A variety of other intermediate approaches exist in practice. For example, Types C1, C2, and C4 couple a presumption in favor of partition in kind with a put option that lets any co-owner be bought out upon her demand. Type C3 and some US jurisdictions (Kuperman 2011: 290–295) offer an allotment system that gives co-owners who desire continued possession the call option to buy out the co-owners who have petitioned for a partition sale. Below, this chapter considers a new intermediate solution – a new type of partial partition – using the following example. A decedent leaves Homeacre to his two children, Ann and Burt, and to a longtime family friend, Casper, in equal undivided shares. Homeacre contains two structures; a family home that sits on two-thirds of the lot, and a detached garage apartment that sits on the remaining third, as shown in Figure 7.2. Ann and Burt grew up in the family home, and Casper, a musician, has been living rent-free in the garage apartment for the past 20 years. Ann 1

2

3

Family Home Garage Apartment

Figure 7.2  Homeacre

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

193

Table 7.1  Valuations of Homeacre’s areas Subjective increment Total economic Market value [held by] value Portion with home $150K (areas 1 and 2) Portion with garage $50K apartment (area 3) Whole parcel (kept intact) $250K (areas 1, 2, and 3)

$0

$150K

$105K [Casper]

$155K

$50K [Ann]

$300K

wants to live in the family home, but only if she can have all of Homeacre, including the garage apartment and surrounding yard. Burt, who never liked Homeacre, just wants his share of the money from sale. Casper desperately wants to continue living in the garage, which occupies a central location in a community to which he is deeply attached and has many features that he has tailored to his highly idiosyncratic tastes. The MVs and the subjective values that the co-owners attach to the areas in Homeacre identified in Figure 7.2 are set out in Table 7.1. In this case, the total economic value is greater when splitting the property into two pieces ($150K + $155K = $305K) than when keeping it together as a whole ($300K). It would also be economically destructive to order partition in kind, full stop, as this would require physically splitting up the family home, which now occupies two-thirds of the property. Ann and Burt, each left with a portion of a house, would eventually cooperate to sell it, but doing so would entail a significant degree of hassle. Partition by sale would maximize value if Casper could be the highest bidder; he could then keep the garage apartment portion and sell the balance, producing the most valuable use of the land. Casper is illiquid, and hence not a good candidate to be the highest bidder. Ann could be the highest bidder at $300K and then sell the two pieces separately, but again, Casper will be unable to bid his valuation of $155K. A sale will at most yield the parties a total value of $300K. Enter partial partition. The court could let Casper keep his portion of the land and order partition by sale for the balance. As Table 7.2 shows, this solution maximizes the property’s total economic value. Significantly, a partial-partition alternative will not be pursued if the court follows a rule that seeks to maximize the MV of the land; MV is higher if the property is maintained as a single unit. Moreover, if the

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

194

part ii: immovable property

Table 7.2  Payoffs for Ann, Burt, and Casper Ann’s payoff (1) Full partition in kind

Burt’s payoff

Casper’s payoff Total

$55 ($75 less $55 ($75 less $155† $20 in hassle) $20 in hassle) (tie) $83.33 $83.33 $83.33

(2) Partition by sale, and Ann, Burt, and Casper do not participate in the auction (3) Partition by sale, and Ann $100 ($300 less $100† is the highest bidder (at her $200 to pay reservation price) off others) $83.33 (4) Partition by sale, and Ann $133.33 ($300 is the highest bidder (at MV) less $166.66)† (5) Partial partition: Partition in $75 $75 kind for the garage portion (area 3); partition by sale for the home portion (areas 1 and 2)

$265 $250

$100

$300

$83.33

$300

$155† (tie)

$305†



marks the best partition approach for a co-owner and in total. Note: Gray shading indicates the party retains physical possession of all or part of the property. Values are given in thousands.

partition doctrine requires or allows courts to move directly to partition by sale if full partition in kind is infeasible, then the court would again choose an inefficient outcome, given liquidity constraints. Recall that only 62 jurisdictions explicitly allow some form of partial partition. Adding the alternative of more flexible forms of partial partition can improve the efficiency of judicial partition outcomes. As this example shows, partial partition can address liquidity problems that  prevent HSV co-owners, like Casper, from being the highest bidder. Partial partition can also be valuable when multiple co-owners hold high subjective values in different portions of the property. Rather than requiring all to bid against one another, and then engage in further transactions among themselves to distribute each section of property to its highest valuer, a partial-partition procedure can simply allocate the property to those who wish to remain in possession (so long as the areas in which they hold subjective value do not overlap) while cashing out the shares of the other co-owners. Carrying out a partial-partition plan requires establishing rules for how payments will be collected from, and disbursed to, the co-owners.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

195

More fundamentally, it requires some method for determining when partial partition is appropriate. The associated design challenges are explored in Section B.

B  Concrete Proposal 1  Self-Assessment-Based Partition Rules The efficient resolution of a partition action may depend on private valuation information inaccessible to a court. One possibility is to use the coowners’ own self-reported valuations to inform the decision. Self-assessed valuation mechanisms have received primary attention in the eminent domain context (Levmore 1982; Bell and Parchomovsky 2007; Chang 2012c), but in some ways the partition context is a more promising setting in which to apply these ideas.21 The basic idea behind self-assessment is that the owner of a piece of property is in the best position to know its value. The challenge is to devise a mechanism that will induce her to reveal that knowledge. Such a mechanism must be capable of checking both overstatements and understatements, and this requires that two consequences tending in opposite directions attach to any statement of value that an owner submits (Plassmann and Tideman 2011: 38–98). Three types of consequences might attach to a co-owner’s statement of value. The first is simply whether the co-owner gets her way with respect to the dissolution of the co-tenancy. If she wishes to stay on the land, does she get her wish? Second, there is the question of what, if anything, she must pay to her co-owners. Third, there is the question of what, if anything, she will receive from her co-owners. These last two consequences together determine the positive or negative price associated with the coowner getting, or not getting, her way in the partition proceedings. These same consequences will influence both valuation statements and the decision to seek or resist judicial partition in the first place. To return to the private taking analogy, a co-owner who stays on the land, or who is a winning bidder, is cast in the role of a taker who must compensate others, while a co-owner who loses out finds herself in the role of a takee who receives compensation. A party will want to take when she will have to pay out compensation less than what she will gain in kind, while a party will hope to have her property taken if she will be compensated beyond the level of her loss. Likewise, a party will not want 21

Self-assessment has been used or proposed in a variety of other situations (Fennell 2005; Epstein 2014; Landeo and Spier 2014; Posner and Weyl 2017; 2018).

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

196

part ii: immovable property

to be cast in the role of a taker if she will have to compensate beyond her gain, and will not want to be a takee if she will receive less compensation than she loses in kind. The valuation statement itself can determine whether one will be a taker or a takee, and what price one will receive or pay out in that role. Under a first-price auction, if the valuer gets her preferred solution, the court will use her stated subjective value to ensure that all parties receive equal shares from the partition action. Consider again the example of Ann, Burt, and Casper. If the parties submitted the valuations implicit in Table 7.1, this information would reveal to the court that the optimal solution would entail breaking the property into two parts, one of which would go to Casper and the other of which would be sold. The total economic value (counting Casper’s subjective premium) of $305K would then be split three ways, with each receiving $101.66K. As the property that Ann and Burt receive will bring each of them only $75K, each would be entitled to receive a payment of $26.66K from Casper. This is no different from the distributive result that would have obtained had there been a partition auction in which Casper was induced to pay his reservation price for the whole property. The only difference is the revelation mechanism. A concern is that parties like Casper will understate their valuations because the payments they must make are tied to their bids. If they have full information about the valuations of the other parties, they would try to state a value that is epsilon above the next highest bid. Of course, other co-owners might attempt to push up Casper’s valuation statement (as it determines what they will receive from Casper) by threatening to place overstated bids of their own. As Casper attempts to avoid stating his full subjective value and as his co-owners attempt to force him to do so, understatements and overstatements may place the property in the hands of the wrong party. Yet even if this alternative did a reasonably good job of checking overstatements and understatements, the fact that the amount of one party’s subjective valuation determines the payouts to other co-owners could be problematic from an ex ante perspective. It could lead other co-owners to strategically seek judicial partition simply to receive transfers of subjective premia from other co-owners. A procedure that instead ties payouts by a winning co-owner to another metric, such as the amount that the other co-owners lose when the court adopts her solution over another, could help to address this issue, although it would do so imperfectly. For example, a court might require a party like Casper who wins partial partition in kind as a result of his valuation to compensate the other parties

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

197

for any difference between the value of the shares of the sales proceeds they will realize from the balance of the property, and the shares they would have received had the entire property been sold as a consolidated unit. This, of course, raises the question of what auction or sales mechanism would have been employed to sell the unit as a whole. Depending on the procedure used, the sales price could be anywhere from Ann’s full reservation price to something less than MV. Regardless of the metric chosen, Casper’s payments to the others would not be benchmarked to his own valuation statement, but rather to someone else’s valuation of the entire parcel that is elicited through an auction or other revelation procedure. Will such an approach induce more honest valuations? Casper may be wary of overstating his subjective surplus when it could produce a result – partition in kind plus a duty to compensate – that he likes less than simply getting a share of the proceeds from selling the parcel as a whole. If he is sure he will be better off with partition in kind (after compensating the others), he might well overstate his valuation to be assured of winning partition in kind – and this mechanism would allow him to do so with impunity.22 But his overstatement would not produce inefficiency relative to partition by sale; by hypothesis, Casper is the highest valuer.23 While it is always possible for Casper to miscalculate and overbid in error, he is less likely to make this mistake when there are clear reasons to doubt that his preferred solution is the efficient approach – as where there is a large economy of scale associated with keeping the property whole.24 Casper has no obvious incentive to understate his subjective surplus because the amount he has to compensate the others does not depend on how high or low his valuation is, but rather on how high or low the hypothetical sales price would be for the whole property. Of course, the same liquidity problem mentioned above might recur here, causing Casper to simply accept partition by sale rather than submit any valuation statement at all, but this problem is significantly buffered in this context, and presumably easier to solve. Instead of having to go to an auction and bid the 22

23 24

As there is no penalty tied to the magnitude of his statement once it is over the threshold that wins partition in kind and triggers his compensation obligation, he would not be further constrained from making an overstatement. As the CF approach does not seek to use self-assessed value in the future (such as in levying property taxes), it does not matter that the highest valuer exaggerates his or her true value. This is because he would run the greatest risk in such situations of getting his way and then having to (over)compensate the others. Depending on how his compensation to the others will be determined, he might also be deterred from overstating his value if he knew another co-owner, such as Ann in the running example, also held significant subjective value in the property, or if he feared strategic bidding by one or more of his co-owners.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

198

part ii: immovable property

full amount of his subjective value (which is far above the property’s MV), he need only come up with a fraction of that amount to pay the others.25 How, under alternative regimes, the payouts from Casper to the other parties would be calculated will introduce new concerns, including the possibility of strategic bidding by Casper’s co-owners. Suppose Casper’s payouts were keyed to Ann’s valuation of $300K. If Casper were required to bring everyone to the level they would have occupied had Ann bid this amount and distributed the proceeds equally, then he would have to pay Ann and Burt $25K each.26 As this result compensates Burt beyond the baseline of MV, it might cause Burt to seek judicial partition and resist a voluntary solution. Perhaps more worrisome is the fact that it might cause Ann or Burt to strategically overstate their valuations to increase the share they will receive from Casper.27 They run the risk of increasing their bids if they are uncertain about Casper’s true subjective valuation, as they might accidentally outbid him and then have to compensate the others. If Casper fears their overbids will dispossess him of the property, he might overstate his own bid, though again, he does so at the risk of having to compensate for a second bid that is above his own true reservation price. The full dynamic must be left to formal modeling and empirical testing, but the risk does exist that the parties’ strategic behavior will sometimes produce inefficient results. As elaborated in Chapters 10 and 13, if co-owners’ shares deviate too much from equality, an internal auction mechanism (close to the CF approach here) cannot always guarantee efficiency. Another option would merely require the winner to compensate his coowners for their share of the whole property’s MV. This approach has the advantage of controlling gaming by the MV co-owners who would otherwise try to increase their compensation. The disadvantage is that there would be no check on overstatements by HSV co-owners like Casper and Ann. They would each put in infinite bids in an effort to outbid the other, knowing that they would only be obligated to compensate the other co-owners at MV. 25 26

27

If Casper cannot access equity in the property immediately, an alternative would be to place a lien on the property (Porat 2009: 212). Areas 1 and 2 are expected to sell for $150K, which will give Ann and Burt $75K each. An additional $25K would bring each to the $100K level that would result from an even split of Ann’s bid of $300K. The procedure described in the text equates to a second-price or “Vickrey” auction. As Vickrey (1961: 22) himself recognized, collusive bids by sellers designed to “jack the price up” can undermine the procedure’s truth-revealing properties. Co-owners who expect to lose the auction and receive the proceeds of the second bid are effectively in the position of a seller – yet unlike ordinary sellers, they are entitled to bid on the property.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

199

A final set of possibilities would break the link between the amount that Casper must pay and the amount that his co-owners receive.28 For example, suppose Casper had to pay an amount that would be sufficient to give his co-owners their shares of the total economic value of the property ($305K, on the numbers above), but that his co-owners actually received only their share of the property’s MV. This approach would limit overbidding while avoiding gaming by MV co-owners (including in the initial decision to seek judicial partition). It would likely produce underbidding (like any other first-price auction), but at least there would not be any concerns about MV co-owners attempting to force up the bid. The downside of this approach is that other HSV co-owners, like Ann, would receive payouts that are significantly less than their subjective valuations. This prospect could open the door to ex ante strategizing or lead to overbidding to avoid bearing a loss. A variation on this theme, which will be used as the basis for the CF approach below, would allow co-owners who elect partial partition in kind to recover their shares of their own synthesized bids, while limiting co-owners who opted for a partition sale to their share of MV. Additional design features might be added to control inflated partial-partition bids by MV co-owners. As this brief survey suggests, it does not seem possible to design a fully incentive-compatible mechanism given the constraints specified and the goals pursued. Which of the options canvassed above will perform best is an empirical question and one that could benefit from formal mathematical modeling. This chapter closes with a brief summary of how such a procedure might be operationalized under the last alternative discussed above, recognizing that a different set of rules for payments and payouts might ultimately prove superior.

2  Three Steps a. Step One: Electing a Partition Method  A co-owner seeking judicial partition would first be required to state his or her preference for one of two alternatives: full partition of the property by sale, or partial partition in kind. A co-owner who seeks full partition of the property by sale would be requesting the liquidation of the property at MV. A co-owner who wished to remain on the land could petition for partial partition in kind.

28

Delinking of this sort has featured in some past auction proposals (Wiseman 1974; Tideman and Tullock 1976; Bell and Parchomovsky 2014: 24–25).

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

200

part ii: immovable property

To invoke this procedure, a co-owner might have to meet additional criteria, such as demonstrating a connection to the property, satisfying a holding period, or posting a bond that will be forfeited if she wins the right to remain on the property but sells within a particular time frame. b. Step Two: Valuation  The court would ask each party petitioning for partial partition in kind to submit the following: (1) a diagram of the property with any area in which the individual holds subjective value clearly marked, (2) a statement of the total value that the individual places on the marked area, and (3) an independent appraisal of the balance of the property. Following the values given in Table 7.1, Ann would submit a diagram with the entire property marked and a value of $300K indicated. Casper would submit a diagram with the garage apartment area (area 3) marked and a value of $155K indicated, and would also submit an appraisal for the balance of the property (areas 1 and 2). When multiple HSV co-owners have attachments to nonoverlapping pieces of the property, they may elect to file a combined submission with the court, or they may file separate submissions. Any party filing a separate submission should indicate to the court whether she wishes to have her submission combined with those of the other co-owners seeking partial partition, where it is possible to do so.29 If the submissions reflect conflicting plans for partitioning the property, the court will order an auction at which bids will be automatically placed on behalf of the co-owners seeking partial partition in kind. Each such bid would be constructed from the reservation price stated for the portion of the property that the co-owner wishes to retain plus the MV of the balance of the property. To return to the earlier example, a bid of $305K would be submitted on behalf of Casper ($155K valuation of area 3, plus $150K MV of areas 1 and 2), and a bid of $300K would be submitted on behalf of Ann (reflecting her subjective valuation of the whole property). c. Step Three: Deciding and Settling Up  In the running example, Casper would win the bid. He would get to keep area 3, and areas 1 and 2 would be sold for their MV ($150K). Casper would be required to make a payment into a fund that would be sufficient to give his co-owners their 29

Space does not permit tracing all the wrinkles associated with these “bundling” alternatives. The basic intuition is that a group of HSV co-owners might have a combined valuation that exceeds the MV of the entire parcel, even though no single HSV co-owner would have a high enough valuation on her own.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

201

shares of the full economic value revealed by his bid. Casper would thus have to pay $53.33K, the difference between Ann and Burt’s collective $203.33K share of $305K and the MV sales price of $150K for areas 1 and 2. Ann and Burt do not actually receive this full amount, however. Instead, Ann (as a co-owner who also elected partial partition in kind) would receive enough to make up one-third of her $300K bid (here, $25K)30 and Burt (who did not elect partial partition in kind) would receive enough to make up his share of the $250K in MV (here, $8.33K).31 The balance of the funds paid by Casper (that is, the portion that was not disbursed to Ann and Burt; here, $20K)32 goes into a fund earmarked for co-owner mediation efforts or for administering the system.33

C  Why the CF Approach Is More Efficient In essence, the CF approach can be described as the following: if no cotenant petitions otherwise, partition by public auctions; if one or more co-tenants want a part of the parcel, courts host internal auctions in which co-tenants submit sealed bids based on their subjective value. A typical partition in kind will not be ordered unless all co-tenants declare that they prefer a proportional piece of the parcel and are indifferent as to its location in the original co-owned plot of land.

30 31 32

33

This amount is calculated by subtracting her share from the sale of areas 1 and 2 ($75K) from her share of her own bid ($100K). This amount is calculated by subtracting his share from the sale of areas 1 and 2 ($75K) from his share of the property’s MV ($83.33K). $20K = $53.33K − $25K − $8.33K. For co-owners, the $20K that goes to the fund is money left on the table. If they have good information as to how much each other would bid, they may reach an agreement of voluntary partition. Facilitating ex ante bargaining is an advantage of the CF approach. Note that because voluntary partition has to be consensual, no coowner would agree to a deal that makes her worse off. That is, the $20K would only make all or some of the co-owners better off in voluntary agreements, as compared to judicial partition. Court costs also represent savings that parties could realize through a voluntary arrangement. Baucells and Lippmant (2001: 1207–1209) model the effect of legal costs on bargaining dynamics. A potential difficulty here arises if appraisals of the MV of the entire plot (or, areas 1 and 2) turn out to be overly optimistic or pessimistic. As pointed out above, properties are likely to be sold below MV at court auctions. With the availability of large data sets on land sale prices, a hedonic regression model can predict MV with high accuracy (Chang 2010; 2011; 2013), alleviating this concern. Moreover, additional design features could provide opportunities for co-owners to bid on the relevant portions of the property so that a negotiated sale can be conducted at a later date. The fund could also be used to fill in the gap if the sale price of the sold portion falls below the appraised value.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

202

part ii: immovable property

The CF approach is likely to be more efficient than the partition approaches summarized in Part I. First, many countries set partition in kind as the default. It is highly likely that courts would sometimes order partition in kind simply because that is the default rule and perhaps preferred by one co-tenant, while auctioning off the majority of the land can avoid future fragmentary land problems (Part II). That said, as most jurisdictions in Types A–C (Part I) contain statutory language broad enough to account for economy of scale, and 55 of the 113 Types A–C jurisdictions explicitly allow partial partition, adopting the CF approach would not be inconsistent with the current law for these studied jurisdictions. For the two Type-D countries that do not explicitly allow partial partition, their laws risk too often disregarding co-tenants’ subjective value. Second, no studied jurisdictions explicitly allow a co-owner to bid on only part of the land. This would not be a problem if the land equity market functions well – investors and co-owners like Casper can join forces to place a bid in Casper’s name. In Type C4, a co-owner like Casper could acquire a part of the land. If the premia from economies of scale outweigh co-owners’ subjective value, courts could block the partition of shares and allow the co-owners who want out to force other co-owners to buy the shares. Nonetheless, such designs do not induce co-owners to reveal their true subjective value; thus, the partition outcome may not increase allocative benefits as much as the CF approach would. Third, under Type C, co-owners are awarded either call or put options to either buy or sell shares from their co-owners. The price (compensation) is most likely assessed according to MV, if not below MV.34 The allocative outcome may not maximize social wealth. Put options are then merely an unsatisfactory exit option. If some co-tenants need to cash out quickly, other co-tenants may hold out and do not agree to voluntary partition plans, so that they can acquire the former’s share cheaply. Type C3, using the call option design, aggravates the strategic behaviors as some cotenants can actively request an opportunity to buy others’ shares without having to compete with outsiders in public auctions, nor offer to buy at their economic value. Fourth, Types B–C choose between public and internal auctions, or prefer one to the other. Assuming that (1) some or all co-tenants have liquidity constraints; (2) the equity markets do not always function well; and (3) unlike the CF approach, current laws do not allow co-tenants to 34

My own empirical works have shown the tendency of courts to under-assess MV of land (e.g., Chang 2010; 2011; 2017).

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

co-ownership partition

203

bid on part of the land; as compared to internal auctions, public auctions have the advantage of bringing in sufficient cash to purchase entire plots. In cases where cash-poor co-tenants’ economic value is higher than outsiders’ economic value, internal auctions have the advantage of allocating resources efficiently, though the likely lower auction price is distributively worse for co-tenants who lose bids. In addition, internal auctions enable co-owners to keep properties in the original families or ethnic groups – avoiding the massive farmland loss among African Americans that have resulted from partition actions (Craig-Taylor 2000: 772; Dagan and Heller 2001: 551; Mitchell 2001; Heller 2008: 121–125). As compared to these prototypical public and internal auctions, the CF approach presumes neither co-owners’ liquidity nor the functionality of the equity market. Outsiders’ evaluations are incorporated in the MV assessment. Co-tenants with high subjective value are likely to keep the land, whereas those without (much) subjective value receive compensation that commensurate with their own valuation. Fifth, trustees and courts in Types E–F have large degrees of discretion in deciding how to divide up properties, but the laws do not provide much guidance. It should not be difficult for these countries to adopt the CF approach wholesale. Finally, the CF approach is not costly to administer. It is more complicated than the already familiar public auction, but courts should take advantage of the complexity to convey a message to co-owners: do not try to game the system; just submit an honest bid. The CF approach, as compared to the partition regimes currently in force, may or may not lead to more frequent physical divisions of land – it depends on how often partition in kind has been ordered. For countries that have rarely used partition in kind, the CF approach may lead to increase in its use, if co-owners’ subjective value outsizes the premium from economy of scale. This raises a question as to whether the more frequent fragmentation of land could be inefficient for future generations.35 The whole economic point in favor of partition in kind is the subjective value of co-owners. What if heirs or transferees do not share such subjective value? The institution cost of land assembly may not be overcome by later generations. Indeed, in a land market where current co-owners are not able to gauge accurately the economic values of their heirs or transferees, co-owners may err on the side of demanding partition in kind, while 35

Compare Heller and Salzman (2021: 207–208) who propose to use family farm corporation or give right of first refusal to current owners in partition actions to preserve family farm.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

204

part ii: immovable property

the value-maximizing plan is to take advantage of the economy of scale. This would not be a problem should MV already reflects the highest and best use in the future. Whether land market is efficient is an empirical question and the answer is likely to vary by developmental stages.

IV Conclusion This chapter starts by showing that jurisdictions around the world utilize several legal schemes in structuring their co-ownership partition law: partition in kind, public auctions, internal auctions, put options to sell shares, and call options to buy shares. Also, a number of countries defer to trustees or courts to make partition decisions. This chapter submits that a variant design, based mainly on internal auctions, is more efficient, in the sense that this intermediate partition approach can harness private information (economic values of all coowners) and induce co-owners, to a large extent, to bid honestly in internal auctions for partitioning. The CF approach also eliminates strategic and inefficient behaviors that could be prevalent under certain designs (so that a number of jurisdictions give courts discretionary power to postpone partition actions for some years; see Chapter 8). Despite the availability of this concrete, low-cost, and practicable proposal, no jurisdictions have adopted the most efficient partition approach.

https://doi.org/10.1017/9781009236553.010 Published online by Cambridge University Press

8 Managing Co-ownership Tragedy of the Common-Ownership?

Communio est mater rixarum (Co-ownership is the mother of disputes.) (Van der Merwe et al. 2002: 484)

Following on the previous chapter, this chapter again deals with coownership. Chapter 7 focuses on how co-ownership ends, whereas this chapter deals with how co-owners make collective decisions and whether statutory decision-making rules create inefficiencies – more colorfully known in the literature as “tragedy of the commons” and “tragedy of the anticommons.” More specifically, Part I first recounts the rule regarding whether, and for how long, the law allows co-owners to enter covenants to maintain and/or administer co-ownership. Part I then summarizes the voting rules for setting up administrative plans and selling co-owned things in their entirety. Part II analyzes the theoretical implications of the different rules for co-owners. Put differently, this chapter deals with co-ownership governance rules. Elsewhere, I have argued that “collectively owned” rural land in China is, in essence, a type of concurrent ownership with Chinese characteristics. Governance rules for collectively owned land are sometimes lacking, and sometimes overly complicated. The best solution is to make each rural village a public corporation and give villagers nontransferable shares, streamlining the governance structure (Chang 2018a; 2020d). Even though corporate governance is a hot topic, and political scientist Elinor Ostrom (1990) won the Nobel Prize in Economics for studying the commons, co-ownership governance has garnered insufficient attention.

I  Comparative Overview This part is divided into three parts. Part A deals with a special case of co-ownership administration: co-owners’ agreements not to partition. 205

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

206

part ii: immovable property

This is an extension of the legal issues discussed in Chapter 7. Part B summarizes general rules for co-ownership administration. Part C lists rules in countries where a sale of the entire co-owned thing does not require unanimous consent.

A  Covenant Not to Partition Eighty-seven jurisdictions explicitly allow co-owners’ agreements (covenants) to stop partition (see Table 8.1). The modal cap for the freezing period is five years. Under American common law, co-owners can agree not to partition and such agreements are binding on their successors if their duration is limited to a “reasonable time” (Stoebuck and Whitman 2000: 216; Singer 2010: 359).1 Eighteen jurisdictions do not prescribe a clear maximal period; for example, Scotland (Zimmermann et al. 2005: 728). Among 87 jurisdictions, 66 have an explicit cap, of which 42 allow co-owners to renew the covenant not to partition; for example, Poland §210. Two jurisdictions explicitly disallow renewal; the rest make no stipulations regarding renewal. As Table 8.1 shows, 23 of these 87 jurisdictions give courts equity powers to terminate covenants before expiry. While one would expect that this power would be given when the embargo period is long or when the maximal period is not clearly specified, it is not the case. The four jurisdictions with the longest embargo period do not award courts such an explicit power: Liechtenstein (30 years), Quebec (30 years), Taiwan (30 years), and Switzerland (50 years). 12 of the 18 jurisdictions without an explicit cap do not give courts the power to terminate covenants before their expiry. On the other hand, while 118 jurisdictions allow one co-owner alone to file for partition at any time (in the absence of no-partition agreements), in 33 jurisdictions the partition procedure may be postponed. In 13 jurisdictions, the law bans co-owners from demanding partition at an inconvenient time. Ten jurisdictions cap the court-ordered postponement at one, two, three, five, or ten years, five jurisdictions leave it entirely to the court to decide the length of postponement. In six other jurisdictions, courts have discretion to dismiss partition requests. Figure 8.1 shows that in Argentina, Guatemala, Honduras, Italy, Kuwait, Moldova, the Netherlands, Senegal, and Singapore, courts have the power both to 1

See also 37 A.L.R.3d 962; Mitchell et al. (2010: 616).

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

207

Table 8.1  Covenants for not partitioning

Cap for no-partition period (in years)

Covenant terminated by court before expiry?

3 4 5 10 15 30 50 No explicit cap Reasonable

No 0 0 38 7 1 3 1 12 2

Yes 1 1 11 3 0 0 0 7 0

Total 1 1 49 10 1 3 1 19 2

Total

64

23

87

Judicial power regarding partition Neither powers Both powers Postpone partition Terminate no-partition covenant No information or not coded

Figure 8.1  Judicial power to postpone partition and terminate non-partition covenant

postpone partition and to terminate a non-partition agreement before its expiry. Do covenants not to partition bind successors to shares? If not, a coowner who no longer wishes to be bound by the covenant could simply transfer 0.1% of her share to a trusted person who would file for judicial partition of the property. Of the 88 jurisdictions with covenants not to partition, 37 have rules regarding whether covenants among co-owners run with shares. They can be divided into nine types, as shown in Table 8.2. Type A stipulates whether non-partition covenants run with the land. It is

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

Always run

Run if registered

Run if giving other notices

c

b

a

Type A1. Afghanistan §1953, Iraq §1070, and Macedonia (AOORR) §50.

The conditions are that acquirers could have easily found out about the administrative plans, or the plans were decreed by courts. Farran (2013: 56) observes that covenant laws in South Pacific countries are similar to those in New Zealand and the common law. Farran (2013) does not explicitly mention the role of registration or other notices. The notice requirement in Ukraine is to notarize administrative plans.

Partition only

Type A2. Belgium §815, Croatia (AOORR) Type A3. South Africa. §47, Peru §993, Portugal §1412, Quebec §1014, and former Portuguese colonies such as Macau §1311 and Timor-Leste §1332. Type B2. Germany §1010, Greece §1115, Type B3. Poland Both partition and other Type B1. Algeria §716 & §722, Bahrain Israel (LL) §29 & §37, Liechtenstein (SR) §221,a and administrations §780 & §790, Egypt §828 & §834, Italy §26, Netherlands §III:168, and Taiwan §1107, Kuwait §821 & §830, Libya American §826-1. §837 & §843, Qatar §855, Switzerland common law. §649a, and Syria §783 & §800. General administrations Type C1. Georgia §958, Mongolia Type C2. Austria §828, Estonia (TLPA) §79, Type C3. Australia, (unclear whether §488.4, Slovenia (Škerl and Vlahek Hong Kong, Ireland, Lithuania §4.81, and Nigeria, Ontario, no-partition 2020: 123), and Turkey §695. New Zealand.b and Ukraine agreement included) §358.c

Contexts in which running covenants are recognized

Table 8.2­  Typology of running covenants between co-owners

managing co-ownership

209

unclear whether covenants regarding other aspects of the administration of co-owned land would run with the land.2 Type B leaves no room for doubt that running covenants regarding administration of co-owned land in general, and no-partition agreements in particular, are recognized. The no-partition agreements and general administrative agreements share the same applicable rules (like the maximal duration). Type C stipulates running covenants in the general context of co-owned land administration, and it is plausible that the rules apply to no-partition agreements as well.3 The second dimension in the typology is the notice requirement. In some countries, covenants run with shares only if registered; in others, other types of notice are sufficient. In other countries, no notice requirements are specified. Civil-code jurisdictions generally particularize non-partition agreements and co-ownership administration in the co-ownership section in the civil codes. By contrast, while common law countries usually have laws regarding non-partition agreements, their rules on co-ownership administration are more obscure. In the general administrative context excluding partition, the answer for these jurisdictions often boils down to interpreting the (restrictive) covenant law, which is broader than co-ownership administration. The summary below, to focus on the research questions in this chapter, greatly simplifies the law and largely ignores the (sometimes important) distinction between law and equity. Some common requirements for covenants to run – such as that covenants “touch and concern” the land, and original covenanting parties intend to bind successors – will not be repeated. Instead, I pay attention to what kind of notice is sufficient for covenants to run, and to whether co-tenant agreements are an

2

3

In South Africa, the sole example of Type A3, co-owners can enter into non-partition agreements for some time, and agreements to partition bind third parties who have notice (Van der Merwe et al. 2002: 323). I infer from the above that non-partition agreements are likely to bind successors in interests if they have notice. However, Georgia §962, a Type-C country, prescribes that a no-partition agreement terminates upon the death of a co-owner, while Georgia §958 stipulates that administrative plans bind successors. It is unclear whether a no-partition agreement by default runs to a transferee (i.e., before any co-owner dies). Slovenia (Škerl and Vlahek 2020: 122) and Estonia (TLPA) §76 & §79 are of another type – no-partition agreements are stipulated separately. Slovakia §143a limits such running agreements to those between married couples, so it is not included in Type C. Mongolia §489.4 stipulates that covenants not to partition do not run to successors, but it is unclear whether they run to transferees other than successors.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

210

part ii: immovable property

example of restrictive covenants, as co-tenants’ c­ ovenants regard only one piece of land, rather than two parcels. 1. American law: Restrictive covenants include co-tenant agreements and notice can be given by recording (Hovenkamp et al. 2016: 350–353).4 2. Canadian law understands restrictive covenants in a way that accommodates agreement between co-tenants (La Forest 2017a: 16-2), but no authorities explicitly state the inclusion of co-tenant covenants or those regarding partition (that said, no-partition agreements are certainly allowed, but discussed in separate contexts – whether and how non-partition agreements run are unclear). Covenants run to successors in title (Ziff 2014: 410–418), if there is notice. 3. I cannot find a clear definition of covenants in Nigeria, but covenants run if there is notice – and covenants may be registered to provide notice (Smith 2018: 107–111). 4. New Zealand (PLA) §§301–302 defines covenants in a way that accommodates agreements between co-tenants, but no authorities explicitly state the inclusion of co-tenant covenants or those regarding partition. Covenants run to successors only if registered (Williams 2011: 85–86). That is, unlike the United States and Canada, New Zealand allows only registered covenants to run. 5. Hong Kong (CPO) §41 clearly stipulates that covenants include ­co-­tenant agreements, and covenants run to successors if registered (Goo and Lee 2019: 536–537, 543). Here Hong Kong is similar to New Zealand. 6. Ireland understands restrictive covenants in a way that accommodates agreement between co-tenants,5 but no authorities explicitly state the inclusion of co-tenant covenants or those regarding partition. Covenants run to successors if there is notice (de Londras 2011: 251–262) – in practice, the doctrine of notice has been replaced by the requirement to register deeds that create restrictive covenants (Woods 2011: 78). 7. Australia (RPA) §81 defines restrictive covenant in a way that accommodates agreements between co-tenants, but no authorities explicitly state the inclusion of co-tenant covenants or those regarding partition. Covenants run to successors, if there is proper notice. Unlike in other aforementioned common-law jurisdictions, in Australia, registration of covenants is not sufficient notice (Hepburn 2002: 130–134). 4 5

I thank Henry Smith and Tom Gallanis for their help here. Ireland (LCLRA) reformed its covenant law, but LCLRA’s definition of covenants covers only those regulating dominant land and servient land. For co-tenant agreements on the same land, the common law and equity apply (de Londras 2011: 262).

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

211

8. In England, since 1925, a trust is imposed on all co-ownership (Sparkes 2019: 69), so management and sales of such trust properties (what would have been co-owned property) are left for the two to four trustees to decide. 9. In Scotland, successors are not bound by covenants on land (Zimmermann et al. 2005: 728). 10. Singapore (LTA) stipulates that a restrictive covenant must involve two plots of land: the dominant land and the servient land. Co-tenant agreements relating to the same land are a simple contract, and the only way to make them enforceable against successors is by a chain of contracts between the original parties and their successors.6 11. In Uganda, restrictive covenants apply when there is a dominant land and a servient land (Bakibinga 2018: 49–51). 12. Covenants in Malaysia are considered contracts (Maidin and Ali 2014: 94), which presumably do not run. 13. India and Pakistan (TPA) §44 stipulates that successors in ­co-­ownership interests are subject to existing conditions and liabilities, but no authorities elaborate on the conditions and liabilities. Nor can I find discussions regarding notice or registration requirements.

B Administration Administrative decision-making rules in 112 jurisdictions were identified and summarized below (see also Figure 8.2). Notably, several Latin American countries have no explicit rule on co-ownership in general, and administration of it in particular – for instance, Chile, Colombia, Ecuador, El Salvador, and Uruguay (Barua 2018: 61). Many jurisdictions prescribe that any co-owner can take necessary means (such as interrupting prescription) to preserve co-owned things; for example, Eritrea §1093 and Iraq §1066. This kind of rule is not included in the summary below. Many jurisdictions prescribe a (super-)majority rule, and the natural question is how majority is assessed: Is it one-person-one-vote or voting by shares? Among the 70 jurisdictions that use majority or super-majority voting, 26 use the percentage of shares, whereas 36 use the value of shares to determine majority. Given that, in a co-owned tangible thing, the percentage of shares perfectly correlates with the value – because, unlike 6

I thank Prof. Alvin See for this observation.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

212

part ii: immovable property

Administration of co-ownership A fiduciary decides Always unanimity Sometimes < unanimity Always < unanimity No information or no rule

Figure 8.2  Administration of co-owned land

in corporate law, there is no “dual class” of shares in tangible things – these 62 jurisdictions adopt essentially the same rule. In Liechtenstein (SR) §26b, Mexico §947, Quebec §1026, Senegal (CF) §452, Switzerland §§647a–647e,7 Taiwan §820, Thailand §1358, and Turkey §688, both the number of co-owners and the percentage/value of shares have to reach a (super-)majority to make administrative decisions.

1  Always Less than Unanimity Type P1. Simple majority rule. Afghanistan §1938, Albania §203, Angola, Argentina §1994, Bolivia §164, Brazil §1323, Bulgaria (OA) §32, Cambodia §208, Cape Verde, Cuba §164, Equatorial Guinea §398, Estonia (TLPA) §72, Georgia §957, Guinea-Bissau, Honduras §2219, Iraq §1064, Japan §252, Liechtenstein (SR) §26b, Mexico §946, Mongolia §488.1, Mozambique, Nicaragua §1699, Panama §406, Paraguay §2098, Philippines §492, Poland §201, Portugal §1407, Puerto Rico §332, Quebec §1026, Romania (Stanescu 2017: 68), Sao Tome and Principe, Scotland (Bell and Guthrie 1899: 427), Senegal (CF) §452, Slovakia §139, South Korea §265, Spain §398, Switzerland §§647a–647e, Taiwan §820, Thailand §1358, TimorLeste §1325, and Venezuela §764 adopt this. Type P2. Super-majority rule. China §301 and Guatemala §488 require 67% of co-owners to agree and Tunisia §68 requires 75%. Type P3. Some administrative decisions are subject to a simple majority vote, while others are subject to a super-majority vote. Several 7

In some instances, majority is ascertained in terms of number of co-owners; in other instances, majority is ascertained in terms of the number of co-owners and shares.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

213

Middle-East countries belong to this type: Algeria §§716–717, Bahrain §§779–780, Egypt §§828–829, Jordan §§1034–1035, Kuwait §821 & §824, Libya §§836­–838, Qatar §855 & §858, Syria §§782–784, United Arab Emirates §§1155–1156. So do Czech §§1128–1129 and Macau §1304.

2  Sometimes Less than Unanimity Type Q1. Unanimity is, in principle, required, but in some scenarios, a simple majority vote suffices: Austria §833, Azerbaijan §215.3, Germany §§744–745, Greece §§788–790, and Peru §971. Type Q2. Some administrative decisions are subject to a simple majority vote, while others require consensus. Types Q1 and Q2 are different largely in terms of styles. Type Q2 includes Croatia (AOORR) §40, Ethiopia §§1165–1166, Hungary §V:78, Israel (LL) §30, Macedonia (AOORR) §36, Norway (CA) §4, Serbia (LBOPR) §15, Slovenia (Škerl and Vlahek 2010: 88), and Turkey §§689–692. Type Q3. Some administrative decisions are subject to a two-thirds super-majority vote, while others require consensus. France §815-3 belongs to this type.8 Type Q4. In Italy §1105 & §1108, some administrative decisions are subject to a two-thirds super-majority vote; some, a simple majority vote; and others, unanimity. 3  Always Unanimity Type R. Armenia §191, Australia (Edgeworth 2017: 264), Belarus §250.1, Belgium §577-2, Burundi §§II:I:31–32, California and New York (Dagan and Heller 2001: 614; Singer 2010: 360), Eritrea §1093, Finland (CA) §5, Hong Kong,9 India (Krishnaswami 2002: 353), Iran §576, Ireland,10 Kazakhstan §213, Kyrgyzstan §269, Laos (PL) §25, Latvia §1068, Lithuania §4.75, Louisiana §803, Luxembourg §815-3, Malawi (RLA) §§96–97, Malaysia (NLC) §343, Malta §493, Moldova §347, Morocco (Gignoux 2015: 13), the Netherlands §III:170, New Zealand (Brown 2017: 538), North Korea §42, Ontario (La Forest 2017a: 14-32–14-33), Russia §247, Rwanda §II:32, Singapore (See et al. 2018: 78), South Africa (Van der Merwe et al. 2002: 319), Suriname §748, Sweden (CA) §2, 8

Dyson (2003: 147–154) discusses the complicated and interesting evolution of the ­co-­ownership administration rules in France. Note that Dyson calls the stipulations in §815-3 and following the “old statutory rules” that co-owners are able to sidestep and opt for the “new statutory rules” in §1873-1 and following. 9 I thank Prof. Say Goo for this observation. 10 I thank Prof. Una Woods for this observation.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

214

part ii: immovable property

Tajikistan §295, Turkmenistan §194, Ukraine §358, Uzbekistan §219, and Vietnam §217.

4  A Fiduciary Decides Type S1. Seychelles §818: “If the property subject to co-ownership is immovable, the rights of the co-owners shall be held on their behalf by a fiduciary through whom only they may act.” Type S2. In England and Wales, the 1925 reform imposes a trust in all tenancy in common; in other words, co-owners will hold beneficial interests in trust as tenants in common (Sparkes 2019: 64–69). Trustees will make the administrative decisions. C Sale A total of 103 jurisdictions adopt an unanimity rule for selling co-owned land. Seventy-five of these explicitly require unanimity. Twenty-eight only implicitly so; as 21 of the 28 adopt a less-than-unanimity rule for administration, it is possible as a statutory construction to require only a (super-) majority to sell the land. In the United States, the general rule is that “[a] sale or encumbrance by one tenant in common of more than his or her share does not bind the others” (Wolf 2021: Volume 7, §50.06) – see also Buller v. Buller, 62 Cal. App. 2d 694, 145 P.2d 653 (1944). Hence, everyone’s consent is required to bind all co-tenants. The unanimity rule should not be surprising. Below, this chapter lists only the several countries that explicitly adopt a rule of sale that does not require unanimity: Type T1. China §301 allows a 2/3 super-majority in shares to sell the thing in its entirety. Type T2. Co-owners with a 3/4 super-majority in shares (number of co-owners must be larger than one) can propose to sell things in their entirety, but if other co-owners dissent, the court may stop sales prejudicial to the interests of minority co-owners: Algeria §720, Bahrain §786, Egypt §827 & §832, Kuwait §820 & §827, Libya §836 & §838, Qatar §861, and Syria §787. Type U. Under law in Seychelles §§818–821, the rights of co-owners are held on their behalf by a fiduciary. Co-owned immovable things will be sold via private sales or public auctions unless all co-owners agree to postpone selling. In public auctions, courts may partition the thing in kind instead.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

215

Type V. In England, trustees decide, but if trustees cannot agree on whether to sell, any interested person may apply to courts for an order to sell (Dixon 2016: 138). Note that, by way of comparison, in New South Wales, Australia (CA) §66G, courts have discretion, on the application of one or more ­co-­owners, to appoint trustees to sell or partition; co-owners holding at least 50% if a plot is entitled “almost as of right” to a court order for partition or sale (Edgeworth 2017: 286–288). Presumably, co-owners in Australia can by consensus sell the entire co-owned land to other parties without judicial intervention or an appointment of trustees. Rules in Ireland (de Londras 2011: 170) are similar to those in Australia.

II  Tragedy of the Common-Ownership? The two prototypes of inefficient equilibrium of resource uses are the tragedy of the commons (Hardin 1968) and the tragedy of the anticommons (Heller 1998).11 Open-access commons arise when no one has the legal right to exclude, whereas anticommons emerge when many parties have the legal right to exclude. Because in rem exclusion is the sine qua non of property (cf. Merrill 1998; Chang and Smith 2012; Smith 2012b; Merrill 2014), ­open-access commons are not subject to any property rights.12 The discussion of commons can thus be linked to the two most influential law and economics articles (Coase 1960; Calabresi and Melamed 1972) – two pivotal studies on (in)complete property rights and assignment of entitlements. Since these articles were published, the property literature has tended to hold the view that if transaction costs are low or zero, no matter how entitlements are assigned, voluntary transactions will increase allocative benefits (Chapter 2). Put differently, property rights are considered a good thing, a pre-requisite for efficient transactions, as long as the costs of establishing property rights justify the benefits (Demsetz 1967). It was in this context that Heller (1998) suggested that too many property

11

12

The idea of anticommons can be traced back to at least Augustin Cournot’s work in the nineteenth century (Kominers and Weyl 2011). In the legal academia, Michelman (1982) and Ellickson (1993: 1322 fn.22) discuss anticommons issues before Heller (1998). Now there are also rich discussions of “semicommons” (Smith 2000; 2005; 2008; Fennell 2011a; 2011b), referring to properties that are partly privately owned and partly collectively owned. Limited-access commons, by contrast, are “property on the outside” (Rose 1998: 144).

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

216

part ii: immovable property

rights (anticommons) – in the same, contiguous, or complementary resource – might hamper efficiency as well.13 The theoretical necessity of re-evaluating the efficiency of co-ownership in light of the discussion of the two tragedies is apparent. On the one hand, some scholars assert that tenancy in common produces the tragedy of the commons (Craig-Taylor 2000; Dagan and Heller 2001: 551–552; Mitchell 2001; Rivers 2007: 50–52),14 which is commonly characterized as a resource being overused by commoners. On the other hand, Heller (2008: 108) considers tenancy in common as creating the tragedy of the anticommons, suggesting that a co-owned resource is underused. This part analyzes what tragedy, if any, co-ownership creates. More specifically, this part inquires: Does co-­ownership often lead to under-use? Does co-ownership often lead to under-investment? Are these results inevitable and unsolvable?

A  Underuse or Overuse? Co-ownership creates overuse if co-owners internalize all the benefits of use but externalize part of the associated costs. Each co-owner has an incentive to possess the co-owned resource and to overuse it, as she would obtain the full use value while bearing only part of the costs. The first subsection first examines rules in common-law jurisdictions, and the second subsection moves on to civil-code jurisdictions. The third subsection ­discusses whether there is an ideal decision-making rule.

1  Common-Law Jurisdictions Three conditions have to be met for co-owned resources to be overused. The first condition is that any co-owner can possess and use the entire co-owned property (or disproportionately). In common-law jurisdictions that requires “unity of possession” in tenancy in common, such as Australia (Hepburn 2002: 144–147) and New Zealand (Williams 2011: 93–95), a co-owner is entitled to possess and use the whole land. The second condition is that a possessing co-owner does not have to share benefits (such as natural fruits or rent) with fellow co-owners. In the 13

14

Heller (2008) further provides a myriad of examples of anticommons across various legal fields in many countries. Heller (2008) has become influential, though his theory has its share of critics (Claeys 2011; Epstein 2011). While Dagan and Heller (2001: 611) emphasize that the common-law doctrine of tenancy in common “facilitates a race to overuse,” Dagan and Heller (2001: 614) also point out that “absent partition,” tenancy in common will lead to underuse.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

217

United States, the law is divergent. In some US states, when a p ­ ossessory co-owner benefits from her sole occupation of the co-owned property, she has to account to other non-possessory co-owners for the reasonable rental value of the property (Stoebuck and Whitman 2000:  205; Dagan and Heller 2001: 611).15 This rule does not give co-owners an incentive to overuse the co-owned resource, and it may even induce co-owners to underuse, as some may anticipate free-riding on other co-owners’ efforts. By contrast, in Canada (La Forest 2017a: 14-34) and in a majority of states in the United States (Stoebuck and Whitman 2000: 205; Dagan and Heller 2001: 611; Dukeminier et al. 2010: 357), possessing co-owners do not have to account to other non-possessory co-owners. Only possessory coowners who go too far as to oust other co-owners would have to pay damages (Stoebuck and Whitman 2000: 203–204). Australia and New Zealand have a functionally equivalent rule that requires possessing co-owners to pay occupation rent, if they exclude other co-owners. If we reasonably assume that most non-possessory co-owners who have not struck deals with possessory co-owners would protest the latter’s unilateral use, in all the aforementioned common-law jurisdictions, possessory co-owners have to compensate non-possessory co-owners. The second condition thus can hardly be met. The third condition is that possessory co-owners can demand nonpossessory co-owners to contribute to general upkeep costs, for example, sharing improvement expenses. Dukeminier et al. (2010: 357) point out that in the United States, possessory co-owners cannot ask for contributions to the payment of taxes if they are in sole possession of the property and the enjoyed use value surpasses the cost. Similarly, in Canada, possessory co-owners cannot ask for sharing in repair or improvement expenses unless accounting to other co-owners the received profits and sharing their occupational rent (La Forest 2017a: 14-34). Possessory coowners thus internalize all the costs. Hence, the take-away is that, in these common-law jurisdictions, a co-owner either internalizes the entire costs and benefits or internalizes part of the costs and part of the benefits. This is not a scheme that leads to overusing. If, eventually, in voluntary or judicial partition actions, the 15

If possessory co-owners receive fruits or yields from co-owned land or collect rents from third parties, she must “account to [non-possessory] cotenants for the amount received” (Dukeminier et al. 2010: 357). “In the final accounting incident to every partition action, each cotenant may be charged with rents and profits actually received by him in excess of his pro rata share” (Stoebuck and Whitman 2000: 220).

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

218

part ii: immovable property

possessory co-owner’s time and expenses for managing the revenues are not fully credited or reimbursed, internalized costs along with externalized benefits would lead to underuse of the property.

2  Civil-Code Jurisdictions Many civil-code jurisdictions have clear rules regarding administrative plans, which make overusing co-owned resources less likely. Many countries, at the same time, stipulate rules regarding whether any given co-owner may possess and/or use the entire land unilaterally. These stipulations, however, often use broad terms and it is hard to ascertain what they entail and whether they conflict with privately contracted administrative plans. By my best reading, in 32 jurisdictions,16 a co-owner may use the entire land – unless an administrative plan stipulates otherwise – but cannot exclude other co-owners or adversely affect their interests. Several other jurisdictions require possessory co-owners who exclude fellow co-owners to compensate them.17 Their rules are similar to the common-law rule and are unlikely to result in overuse. In 17 jurisdictions,18 a co-owner is entitled to demand a proportionate part of the property for possession and use. Under this set of rules, coowned land should be slightly underused instead of overused. If using part of the co-owned land is economically feasible for most or all co-owners, the use of the co-owned land should be close to optimality, though still suboptimal. When proportional use is not feasible, some or all parts of the land will be left barren. In, for example, Paraguay §2084, South Africa (Van der Merwe et al. 2002: 320), and Taiwan §818 & §820, a co-owner cannot legally possess any part of the land, however tiny, without authorization from an existing administrative agreement. It goes without saying that this type of rule does not lead to overuse. Rather, underuse is the problem. 16

17 18

For example, Argentina §1986, Bolivia §160, Equatorial Guinea §394, Estonia (TLPA) §72, Georgia §955, Germany §818 (Dannemann and Schulze 2020: 1478), Greece §787, Guatemala §487, Honduras §2215, Hungary §V:74, Israel (LL) §31, Italy §1102 (explicitly stipulating that a co-owner who uses the land bears the improvement cost himself), Louisiana §802, Macau §1302, Mexico §943, Nicaragua §1695, Panama §402, Philippines §486, Poland §206, Portugal §1406, Rwanda §II:31bis, Spain §394, Timor-Leste §1326, Turkey §693, and Venezuela §761. For example, Peru §975 and Romania §636. They are Armenia §191, Azerbaijan §215, Bahrain §778, Belarus §250.2, Croatia (AOORR) §38, Kazakhstan §213, Kyrgyzstan §269, Kuwait §819, Lithuania §4.75, Moldova §347, Norway (CA) §3, Russia §247, Serbia (LBOPR) §14, Slovenia (Škerl and Vlahek 2020: 121), Tajikistan §295, Ukraine §358, and Uzbekistan §219.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

219

Finally, no jurisdiction requires a non-possessory co-owner to share all operating and improvement costs. There are rules, like Panama §403, that require all co-owners to share the cost of conservation, but that does not include the improvement costs made by possessory co-owners. Nicaragua §1697, for instance, explicitly prescribes that co-owners who have not consented to a use plan are not obliged to share the costs arising from the property. That lawmakers do not combine a unilateral right to possess and a universal duty to share all costs is not surprising. This kind of rule would create a bad incentive for co-owners to overuse.

3  The Calculus of Consent Co-ownership calls for a governance rule, and 112 jurisdictions stipulate explicit rules for administration of co-owned land. Which decision rule for administrative plans is the most efficient? One of the many insights in Buchanan and Tullock (1965) is that a simple majority rule is not a magic threshold in a constitutional democracy. An unanimity rule ensures that everyone’s preference is satisfied and no externality is imposed, but is the costliest for reaching a decision. The polar opposite rule, that one dictator makes all the decisions, minimizes decision-making costs but is likely to satisfy the preferences of only a small number of citizens. Contrary to some people’s intuition and prevalent practice, the simple majority rule does not necessarily strike the best balance because the increase in ­decision-making costs may not be linear and the minority may or may not have more strongly held preferences. If collective decision-making costs vary by region, country, and/or culture, the optimal decision-making rule will vary. While this book does not generally inquire what explains jurisdictional differences in property laws, it is worth asking here what may drive the variations in co-ownership rules. Following Licht et al. (2005), who explore how laws on the books reflect cultural values, this chapter uses the Hofstede (1991; 2001) data on individualism (versus collectivism) to test a potential relationship between law and culture. In the Hofstede data, individualism “legitimizes the vigorous pursuit of personal interests rather than deference to others’ decisions and interests” (Licht et al. 2005: 236). High individualism creates two opposite forces relevant to Buchanan and Tullock (1965). On the one hand, higher individualism is associated with higher decision-making cost. Thus, to avoid impasse, lawmakers may decrease the threshold for co-owners’ making collective decisions. On the other hand, higher individualism means strongly held preferences. Thus, lawmakers may instead increase the threshold for co-owners’ making

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

220

part ii: immovable property

collective decisions. Accordingly, assuming lawmakers pay attention to the existing culture and assuming each nation’s culture has not changed between the time a country enacted its co-owners’ voting rule and the time Hofstede measured its national culture, one can posit two opposite hypotheses. Hypothesis 1(a): The level of individualism is positively correlated with the percentage of required votes. This theory states that more individualistic co-owners have their own (different) agendas and thus a simple majority rule may frustrate a large minority’s plans. An unanimity rule would then push co-owners to make compromises that accommodate co-owners’ preferences as much as possible, so that intensive preferences held by the minority co-owners will not be easily sacrificed. Hypothesis 1(b): The level of individualism is negatively correlated with the percentage of required notes. The theory here is that more individualistic co-owners are more likely to be “hold-ins” (Parchomovsky and Siegelman 2004) or hold-outs. An unanimity rule would lead to deadlocks. Figure 8.3 is consistent with Hypothesis 1(a). The more individualistic a country is, the higher the required percentage of votes is. A related empirical exercise would be to compare successful deal rates among coowners in dyads of countries with very similar individualism scores but different co-owner voting rules. This empirical exercise is nearly impossible to conduct because it is hard to track successful deal rates among co-owners.19 Another issue is the unit of voting. For modern democratic elections, “one person, one vote” has become a universal deliberative standard. However, as summarized in Part I.B, no jurisdictions use the number of co-owners alone in determining majorities. Co-ownership governance regimes, like corporate governance regimes, emphasize the import of percentage of shares. Given that the law generally (for other reasons) allows co-owners to transfer their own shares at will, a voting rule that counts heads would easily fall prey to strategic behavior. A co-owner with, say, 10% of the shares could divide it into 100 portions and transfer them to 100 escrows to become the majority in terms of number of co-owners. The predicted consequence of voting by headcount is fragmentation of share votes – some of the escrows could decide to go their own way, potentially leading to fragmentary land problem (Chapter 7). 19

Chang (2012d) is an attempt to measure successful deal rates among co-owners when it comes to partition.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

221

80 70 50

60

% of votes required

90

100

managing co-ownership

0

20

40

60

80

100

Individualism

Figure 8.3  Voting rule versus individualism Notes: N = 58. One circle, one jurisdiction. If the linear fitted line shown in the figure is replaced by a quadratic fitted line or a LOWESS (Locally Weighted Scatterplot Smoothing) line, they will still show an upward trend. Pearson correlation coefficient = 0.451 (p-value < 0.001).

B Underinvestment In a similar vein, co-owned resources are underinvested because coowners who unilaterally improve co-owned things take significant risks that they will not be fully reimbursed. In the United States, Stoebuck and Whitman (2000: 208, 220) observe that during the partition procedure, improvements (to the extent they have increased the value of the property) will be credited to the co-owners who undertook them. This qualification (must increase property value) and the delayed reimbursement for incurred costs reduce co-owners’ incentives to invest in the property, not to mention the fact that when improvements cost more than they yield in terms of market value, the improver will not be compensated (Dagan and Heller 2001: 612–613; Dukeminier et al. 2010: 358). In other countries, any unilateral investment not contemplated in an administrative plan does not bind other co-owners. Depending on the unjust enrichment law, the investing co-owner may demand other

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

222

part ii: immovable property

co-owners to disgorge their enrichment, but many hurdles have to be overcome (first and foremost, the investment has to benefit all the coowners). Moreover, the threat of other co-owners unilaterally exercising their right to partition will give co-owners pause in making any investment (Dagan and Heller 2001: 606). Even the existence of a non-partition agreement (see below) would be unlikely to provide sufficient incentives for long-term investment. Therefore, there is no scenario where ­co-­ownership would lead to overinvestment, while underinvestment is likely a serious problem. In sum, theoretically, underuse and underinvestment of co-owned property is likely. Underuse and underinvestment are generally regarded as the trademark of anticommons (but compare Fennell 2011a: 42–43).20 Hence, if one has to choose between commons and anticommons to describe coownership, anticommons appear to be a better fit than (­limited-access) commons (Holderness 2003: 76).21

C  Not Necessarily Tragic Co-owned resources tend to be inefficiently underused and underinvested in.22 The more critical point is that this inefficient anticommons is not necessarily tragic. Commentators have argued that anticommons and commons are not always tragic because some resources are better unused or widely used (Heller 1998: 673–675; Fennell 2004a: 912–913; 2011a: 35–36). Further, Ostrom (1990: 91–102) has demonstrated that commons can be successfully managed to avoid tragedy. In the co-ownership context, anticommons is not tragic when the parties involved can easily get out of it. For co-ownership to be tragic, co-owners have to be trapped in the low-level equilibrium of underuse

20

21

22

I agree with Fennell (2011a) that the anticommons (underuse) versus commons (overuse) dichotomy is over-simplifying. However, as it is still the common way to label, and this chapter responds to the literature that uses this conventional dichotomy, this chapter still consider co-ownership as anticommons. The more important point should be that resources held in tenancy in common tend to be underused and underinvested, rather than how we label this phenomenon. In addition, in commons, “each group member has the unilateral ability to transfer commons resources to herself” (Fennell 2004a: 916), but a co-owner can only do so under limited circumstances in some jurisdictions. Not all over- or underuse of co-owned resources are inefficient, though. For example, when co-owners reach an agreement to preserve the co-owned land – intentionally “underusing” it – it is not inefficient. Inefficient underuse arises when co-owned properties would have

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

223

and underinvestment. In the typical tragedy of the anticommons, customs, legal rules, and regulations that can resolve or greatly ameliorate the extent of underuse are absent or malfunctioning. Co-ownership, however, should not be considered tragic because each co-owner’s unilateral right to petition for partition (see Chapter 7) can end any suboptimal use.23 Hence, ordinarily, co-ownership does not create a tragedy of the anticommons.24 Are transaction and information costs among co-owners too high for them to reach agreement? Apparently this is an empirical question and the answers may vary greatly by context and jurisdiction. Chang (2012d) offers the first (and so far perhaps the only) quantitative study on institution costs in the context of co-ownership partition. The finding is that “partition by consensual agreement” makes up about 83% of all partition actions between 2005 and 2010 in Taiwan – this is the narrowly and conservatively defined cooperation rate among co-owners. Moreover, because “partition following mediation” and “partition following settlement” can be reached only if every co-owner agrees to cooperate, they can also be counted as cooperation among co-owners. So, broadly defined, the estimated cooperation rate is around 93%. Cooperation among co-owners does not fail often, perhaps for the following reasons. First, co-owners are often relatives (particularly in the case of heir property) or business partners, not strangers. This could make bargaining more amicable.25 In addition, strategic noncooperation is not

23

24

25

been used in their most valuable way had it not been for the high transaction costs that hinder agreements of all co-owners. Co-ownership will not lead to tragedies if co-owners can cooperate. Scholars disagree as to whether the threat of partition may facilitate (Merrill and Smith 2007b: 643) or hamper (Heller 2008: 124) cooperation in using and investing among co-owners. New York adopted the UPHPA in 2020. There are different governance rules when the property interest is inherited versus not inherited. When the property is co-owned by heirs, a minority co-owner cannot force a sale through partition but must instead allow for buy-out and follow the other UPHPA rules. When the property is not “heir property,” a co-owner in NY still has an absolute right to partition and force a sale. Chang (2012d) argues that the decline of Black farms in the American South, a famous example, is neither an example for the tragedy of the commons (Craig-Taylor 2000: 772; Dagan and Heller 2001: 551; Mitchell 2001) nor one for the tragedy of the anticommons (Heller 2008: 121–125). The sale (or, for that matter, loss) of Black farms actually avoids underuse and underinvestment, rather than creating them — though it does aggravate racial injustice. Claeys (2011: 34–35) also doubts whether co-ownership and the partition law are to blame for the loss of Black farms. This point is related to Ellickson (1991: 161–183; 1993: 1320), who study close-knit groups’ developing norms that maximize aggregate welfare.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

224

part ii: immovable property

prevalent because the hold-out threat is not credible when any co-owner can petition the court for judicial partition. While not every co-owner would prefer the default judicial partition approach, it takes only one coowner to file for judicial partition. Thus, strategic noncooperation by a few co-owners will not block the partition process. In addition, because co-owners’ shares are generally clear-cut, the court is unlikely to bestow disproportionally large plots or proceeds to any co-owner. Hence, strategic co-owners would not expect that their requests for a lion’s share would be granted by the court. Bargaining under the shadow of law (Mnookin and Kornhauser 1979), other co-owners would not yield to strategic co-owners’ disproportionate request. What makes cooperation fail? When it comes to co-owners, perhaps the more the messier. In theory, bargaining costs are positively correlated with the number of co-owners. Thus, it is reasonable to posit that co-owned properties with more owners tend to end up being partitioned by the court. Chang (2012d) compares the distribution of the number of co-owners in partition litigation with that of co-owners in general (before any partition actions are completed). The former, as compared to the latter, is a little skewed to the left, showing that properties co-owned by more parties are indeed more prone to noncooperation. Nevertheless, they generally resemble each other. Therefore, the number of co-owners should not be the only driving force for noncooperation. Maybe higher-valued co-owned property would make co-owners’ bargaining more contentious. Chang and Hubbard (2021), using more than 185,000 civil litigation cases in Taiwan from 2010 to 2015, find that settlement rates decrease (litigation rates increase) when stakes increase.

D  Partial Solutions to Underuse and Underinvestment Even though co-ownership does not necessarily produce tragedies, underuse and underinvestment of co-owned properties are still inefficient. Two scenarios will cause underuse and underinvestment: Co-owners cannot reach a deal, or co-owners can reach a deal with each other but cannot bind later transferees. This section explores a few partial solutions that might ameliorate the underuse and underinvestment due to the second scenario.26 26

A more radical approach would be to impose a trust structure on co-ownership (Lehavi 2020), like England and Seychelles have done, or to transform co-ownership to legal entities in which parties own shares.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

225

Although partition can resolve the underuse and underinvestment, the threat of partition may impede optimal use and investment. No-partition covenants, summarized in Part I.A, could neutralize the threat. Its relatively wide use suggests that lawmakers are concerned about the ex ante investment-depressing effect imposed by unilateral partition requests. Still, even for co-owners who have managed to enter an administrative plan of co-owned resources, in addition to a nopartition agreement, their expectations could be disrupted when one co-owner (strategically or not) sells her share to a third party who,27 not bound by the prior agreement among the co-owners, requests a different administrative plan for the property, or requests partition outright. Allowing such covenants to run with shares and bind successors prevents strategic behavior – a displeased co-owner transferring a tiny share to a strawman who would then file for partition – and encourages shortto mid-term investment. To reduce third-party information costs and prevent original co-owners’ strategic behaviors (forging agreements after the fact, e.g.), notice should be a prerequisite for co-owner agreements to run. If a country’s registry is reasonably functioning, notice should be provided mainly through registries. Especially in countries where co-owners can take unilateral actions in using all or part of co-owned resources, it is often difficult for third-parties to figure out whether an administrative plan exists – not to mention that such a plan could stipulate conservation (not use!) which is even harder to observe. “Governance” regimes (Smith 2002) like co-ownership administrative plans and no-partition agreements, however, may also create underuse and underinvestment if they extend beyond their usefulness.28 The original co-owners who make plans may not be able to envision the most valuable use, say, 50 years later. The once highest and best use could, over time, become an inefficient choice.29 If the law allows such covenants to 27

28 29

The parties may include an agreement not to sell, or co-owners have a statutory right of first refusal in share sales – at least 36 jurisdictions have such a rule; for example, China §305, Latvia §1073, and Turkmenistan §194. But if one co-owner does sell her interest to a third party despite the agreement or the right of first refusal, other co-owners may not be entitled to ask for specific performance (revoking the sale), and a contractual obligation not to sell does not automatically run to the third party (Part I.A). Smith (2005: 310) points out that flexibility is needed when a governance regime has outlived its usefulness. In the case of racially restrictive covenants, the covenants may become unconstitutional and yet still have de facto binding effects (Brooks and Rose 2013).

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

226

part ii: immovable property

remain in force indefinitely, this would create a serious “assembly problem” – which Fennell (2011a: 41; 2019b: 29–39) argues is the nature of the anticommons problem – as later-generation, more numerous co-owners could adjust the suboptimal use only through unanimous agreements to reverse the course of the original plan. To prevent this type of anticommons, some countries set a cap on the maximal duration of administrative plans and no-partition agreements, while others don’t. Some countries give courts equity power to terminate no-partition covenants before expiry, while others do not (Table 8.1). An overly simplistic design may not be ideal. A country with a rule that recognizes renewable30 five-year no-partition covenants allows co-owners to re-consider the most valuable use of their properties at least every five years. On the other hand, mandatory reviews also reduce the value of administrative plans because even if a plan lays out the most efficient use for years to come, it is valid for at most five years. Strategic bargaining at renewal negotiations may negate any efficiency gain that administrative plans produce. As the number of co-owners usually increases over time, there is no guarantee that the expanding roster can always manage to reach an efficient joint administrative decision. By contrast, a covenant that can last indefinitely (or 99 years) could perpetuate an efficient administrative plan, but also make it nearly impossible for later co-owners to remove that plan, even if outdated, unless courts have equity power to terminate it. Another type of design, adopted by 33 jurisdictions (Figure 8.1), is to give courts discretion to halt partition. This is a vintage example of ­opportunism-reducing meta-law (Smith 2021a): the “law” is that any coowner has unilateral power to partition, but “meta-law” dictates that a coowner cannot demand partition at a time that frustrates the plans of other co-owners. This meta-law bars co-owners from acting s­ trategically – but for the judicial discretionary power, under the shadow of law, any coowner can threaten to demand partition, and by doing so solicit side payments from other co-owners who have more at stake. If courts could be efficiency-minded and well-informed about the potential best use of co-owned resources, the best design would be a 30

As said, not all jurisdictions explicitly allow renewal of no-partition covenants and almost no jurisdictions make clear whether consensus is required. There seems to be no efficiency reason not to allow renewal of no-partition agreement if all co-owners at the time of renewal agree.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

227

rule-standard combination: a clear rule regarding the cap on maximal duration that is relatively long, say 30–50 years, in combination with a standard that allows courts to terminate covenants when circumstances change. On top of that, a judicial power to postpone partition would be efficient as well. A clear and long maximal period enables co-owners who are able to enter into a no-partition agreement to also reach an administrative agreement that makes efficient use and investment plans (this includes delegating the decision-making power to an agent) during the term.31 As co-owners who form these plans cannot anticipate all future events and bind themselves (or their successors) to all the useful exit plans ex ante, the equity power that courts have is a safety valve for co-owners. Co-owners can entrust courts to terminate the agreement on their behalf when circumstances change down the road. Holding the nature of the use plan constant, the shorter the maximal duration for no-partition agreements, the more judicial discretion to postpone partition can be justified. It is not ideal to use the power to postpone alone. Aside from the fact that many countries only allow courts to postpone partition for a few years only (not enough for co-owners to carry out a wealth-enhancing plan), it is more difficult for co-owners ex ante to anticipate for how long, and whether, the court would postpone partition requests. Put differently, the judicial power to postpone is also a safety valve, but only a safety valve as it does not enable co-owners to enter efficient plans but only prevents coowners from inefficiently breaking efficient plans. Selling co-owned resources to a single person could, of course, ameliorate the underuse and underinvestment problem. In jurisdictions like Taiwan where co-owners seem to be cooperative (the individualism score of Taiwan is 17 (Hofstede 1991; 2001), making it more collectivistic than most jurisdictions32), there is little need to dial down the requirement to sell co-owned resources. China’s individualism score is slightly greater at 20 but includes a two-thirds super-majority rule. This makes 31

32

Interestingly, of the 77 civil-code jurisdictions with covenants not to partition, only two countries – France §1873-2 and Macedonia (AOORR) §50 – explicitly require consensus among all co-owners. Other countries’ codes simply refer to the (co-owners’) agreement, and it is not crystal clear that consensus is required to reach a no-partition covenant, especially so in countries where (super-)majority votes can determine an administrative plan. Put differently, the votes needed to enter a no-partition covenant are higher than or equal to those needed to reach an administrative plan. Hence, it should not be that difficult for the co-owners who are able to unanimously agree not to partition for a while to also enter into an administrative plan. Among the 72 jurisdictions with individualism score, the median is 47, the 25 percentile is 26, and the 10 percentile is 15.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

228

part ii: immovable property

Common interest community Commonhold Coop & condo Condo Neither No information

Figure 8.4  Condominium, cooperative, and commonhold

sense because, in China, land is state-owned or collectively owned, so the super-majority rule applies to buildings and chattels only, but not land. It is more difficult for co-owners to share the use of buildings and chattels than land, so a sale-facilitative regime is efficient. In jurisdictions where co-owners are more individualistic, dialing down the required percentage of votes could decrease the probabilities of hold-ins and hold-outs, at the expense of sacrificing legitimate idiosyncratic values held by outliers. As summarized in Part I.B.4, other than China, only several Middle-East countries allow a mechanism for less than unanimous agreement to sell co-owned resource (their Hofstede individualism scores are missing). As this rule was not transplanted from Western countries, its adoption may reflect a local wish to facilitate co-ownership dissolution. Finally, a more radical approach is to transform co-ownership into trusts (the English rule) or holding companies. This would be desirable when lawmakers expect or even wish co-ownership to sustain, as in ­common-interest communities. Three-fourths of the studied jurisdictions and all states/provinces in the United States and Canada recognize condominium, whereas residential co-operatives are explicitly adopted in only a few – mainly common-law jurisdictions and the Nordic countries. The English commonhold and the Irish counterpart (Van der Merwe 2015: 43–44) are an interesting middle ground between condominium and cooperative (Figure 8.4). Both cooperative and commonhold necessarily embody a legal person, while co-owners under the condominium regime do not necessarily form any legal entity. Mandatory rules in the condominium regime push this co-ownership form toward an essential legal entity. Limited by space, this chapter is not able to explore whether it is efficient to transform co-ownership into legal entities.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

managing co-ownership

229

III Conclusion The moral of this chapter is that before we dub a property regime anticommons or call it tragic, careful theoretical and empirical analyses are required. Anticommons may be prevalent, but not every form of property inclines toward the creation of the tragedy of the anticommons. Some authors worry that tenancy in common as a co-ownership form creates the tragedy of the commons, or of the anticommons. This chapter shows that, as long as co-owners keep their right to petition for partition, coownership produces neither type of tragedy, even though co-owned resources indeed tend to be underused and underinvested, as compared to solely owned resources. How serious underuse and underinvestment are depends on whether transaction and information costs among coowners are high. Therefore, there is no globally efficient voting rule for the administration and sale of co-owned resources. As for covenants not to partition, the more efficient design is to combine a rule of long maximal period with a standard of judicial early termination, and perhaps also with the discretionary judicial power to postpone partition for a few years. No jurisdiction has enacted this design, however. The jurisdictions coming closest are Argentina and Italy, which cap the duration at ten years and allow courts to terminate no-partition covenants before their expiry, and to postpone partition for at least five years.

https://doi.org/10.1017/9781009236553.011 Published online by Cambridge University Press

9 Access to Landlocked Land Hybrid Entitlement Protection Few things are as certain as death, taxes, and the legal entanglement that follows a sale of landlocked real estate. Bob Daniels and Sons v. Weaver, 681 P.2d 1010, 1013 (Idaho Ct. App. 1984)

Phil owns Blackacre, which had access to a public road until the state exercised eminent domain to build a fire station, rendering Blackacre landlocked.1 How can Phil and other landowners who involuntarily lost their access to public roads (“landlocked owners”) regain their connection to the outer world in order to effectively utilize their properties? Access to landlocked land is a universal legal problem.2 Civil-code jurisdictions often group the related access doctrines under the heading of “legal servitude of passage,”3 while most states in the United States deal with the same access problems by two legal schemes: the easement of necessity in the common law and the statutory easement prescribed by state statutes. The basic structures of legal solutions to access to landlocked land are similar (yet not the same). For statutory easements, the common requirements are: (1) necessity of the passage; (2) compensation for the owner of the servient land (“neighbors” hereinafter); and (3) no 1 2

3

This fact pattern is taken from In the Matter of Opening a Private Road for the Benefit of O’Reilly, 5 A.3d 246 (2010). The concept can be readily extended to a broad swath of legal issues such as landlocked states’ access to seas, landlocked cities’ access to interstate highways, building habitat (or wildlife) corridors to preserve biodiversity, and even constructing cross-continent oil pipelines (Chang 2016b: 221). Chang (2022b: 156) reports that in Taiwan in 2006–2010, there were more than 1,000 first-instance court decisions on access to landlocked land. I do not know how often this doctrine has been invoked elsewhere. Legal servitudes, according to Louisiana §659, are “limitations on ownership established by law for the benefit of the general public or for the benefit of particular persons” (Wilmore 1986; Sentell 1994; Yiannopoulos 1996; Merwe 1999; Huffstetler 2002). This chapter uses legal servitude of passage as the umbrella term for easement of necessity and statutory easement.

230

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

231

better alternative (least damage to the neighboring servient land; hereinafter “least damage”). For easement of necessity, a right of way, which may or may not be gratuitous, is generally granted to a landlocked owner if the land becomes landlocked due to severance. This chapter will use the two American terms and discuss easement of necessity and statutory easement separately, as they present different economic problems. Legal servitude of passage is used as the umbrella term for the two schemes. Note that in civil-code jurisdictions, both types of easements are of course prescribed by statutes. This chapter is structured as follows: Part I compares the legal servitude of passage doctrine around the world. The following parts economically interpret the doctrine. Part II describes the fundamental economic rationale for statutory easements. Parts III, IV, and V provide economic analysis of the three major requirements for statutory easements: compensation to the neighbors, necessity of the passage, and least damage to the neighboring land, respectively. Part VI economically justifies ­easement of necessity. Part VII concludes.

I  Comparative Overview The easement of necessity doctrine (also called easement by necessity or ways of necessity) implies an easement right for parcels that become landlocked after partition or partial transfer. That is, the context in which easement of necessity arises is limited. Nonetheless, the common law, for instance, did not develop a general right of access to landlocked land.4 In the United States, 22 states, through statutes (thus the name statutory easement), permit landlocked owners to petition local governments to condemn easements over neighboring land (Hernandez 2005: 112; Bruce and Ely 2011: 4:14). Statutory easements are considered “private takings” (Bell 2009)5 and administrative agencies or courts are often involved in the process of establishing a statutory easement (Merrill and Smith 2010: 204; Bruce and Ely 2011: 4:14). By contrast, 97 civil-code jurisdictions 4 5

Restatement (Third) of Property: Servitudes §2.15 comment a (2000) (addressing the American common law). The US Supreme Court, in Leo Sheep Co. v. United States, 440 U.S. 668, 679–80 (1979), has recognized that easements of necessity and eminent domain are alternative ways to reach the same results for the state. The supreme courts in some US states, such as Michigan (Tolksdorf v. Griffith, 626 N.W. 2d 163) and Texas (Waggoner v. Gleghorn, 378 S.W. 2d 47; 354 S.W.2d 923), went as far as to declare statutory easement an unconstitutional taking of the neighbor’s property for private use. See also Fambrough (2010: 3).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

232

Part II: Immovable Property

Easement of necessity Type C Type B2 Type B1 Types A1 & A2 No information or no rule

Figure 9.1  Typology of easement of necessity

recognize both easements of necessity and statutory easements. Although civil lawyers may debate the nature of such easements, analogies to private condemnation are rarely, if ever, made. This part summarizes the legal servitude of passage doctrine around the world. Section A reviews easements of necessity. Section B describes statutory easements.

A  Easement of Necessity Other than the compensation issue, jurisdictions that recognize the easement of necessity doctrine have similar rules. In American common law, for instance, an easement of necessity is granted when the following four requirements are met: (1) prior common ownership of the dominant and the servient land; (2) transfer of part of the land (severance); (3) (strict or reasonable) necessity for an easement at severance; and (4) continuing (strict or reasonable) necessity for an easement (Bruce and Ely 2011: 4:6). (3) and (4) are required in statutory easement, too. The distinctions between easement of necessity and statutory easement are that the former has (1) and (2). Studied jurisdictions with easement of necessity can be classified into several groups based on different compensation requirements. Figure 9.1 shows the detailed typology. In countries that prescribe rules on statutory easement in general, but do not stipulate a separate rule for easement of necessity, the rule for statutory easement is usually broad enough to include easement of necessity. As the former rule often (but not always6) 6

The statutory easement rules in Burundi §II:II:125, China §291 & §296, Hungary §V:160, and South Pacific countries (Farran 2013: 53) do not explicitly require compensation. In Malaysia (Maidin and Ali 2014: 92–93), neighbors are not entitled to compensation.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

233

gives rise to compensation duty by landlocked owners, landlocked owners in the easement of necessity context may also be obliged to compensate their neighbors. Here, only countries with explicit rules for easement of necessity are included. Type A1. Countries prescribe gratuitous passage: Australia and New Zealand (Bradbrook 1983), Bolivia §264, Brazil §1285 (the “toleration” requirement is interpreted to mean no compensation), Cambodia §144, Canada (Ziff 2014: 386), Chile §850, Colombia §908, Costa Rica §400, Czech §1033, Ecuador §886, El Salvador §852, England and Wales (McFarlane 2008: 855–863), Equatorial Guinea §567, Guatemala §792, Hong Kong (Mau 2010: 68),7 Honduras §829, Ireland (de Londras 2011: 289), Italy §1054, Japan §213, Louisiana §694, Malta §448, Nicaragua §1630, Panama §549, Paraguay §2211, Peru §1053, Philippines §652, Puerto Rico §503, Quebec §999, South Africa (Joubert and Faris 2010: 475), South Korea §220, Spain §567, Taiwan §789, Thailand §1350, Uruguay §587, Venezuela §663, and Vietnam §254. Type A2. The easement of necessity doctrine does not address whether compensation is required. Following the common-law tradition, these jurisdictions are likely to not require compensation: California (28 Cal. Jur. 3d Easements and Licenses §§34–38), India & Pakistan (EA) §28, New York (49 N.Y. Jur. 2d Easements §84), Scotland (Gretton et al. 2017: 182), and Singapore (See et al. 2018: 194–195). Type B1. The law is silent on whether neighbors in cases of easement of necessity should be compensated, but compensation is explicitly required in statutory easement (the two easements are stipulated separately). Afghanistan §1920, Algeria §697, Cuba §171, Egypt §812, Eritrea §1036, Estonia (TLPA) §156, Ethiopia §1223, Germany §918,8 Greece §1015, Iraq §1059, Jordan §1291, Libya §821, Mongolia §138.4, the Netherlands §5:57, Qatar §842, Romania §618, and United Arab Emirates §1381. Type B2. Belgium, Burkina Faso, Comoros, Dominican Republic, France, Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, Senegal, Seychelles, and Togo, all in §682, prescribe a general legal servitude of passage with compensation duty, and in §684 prescribe that if landlocking is attributable to severance, partition, or partial transfer, passage should be set through land owned by a co-partitioner, heir or contracting party – compensation is not explicitly required. Nonetheless, if such a passage is insufficient, an easement should be set according to §682. Tunisia (CDR) 7 8

Prof. Say Goo confirmed that no compensation is required. Roth (2017: §918 Nos. 7–9) interpret the German Civil Code to prescribe compensation.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

234

Part II: Immovable Property

§§177–179 and Morocco (LRRP) §§142–144 have similar stipulations. Type B2 differs from Type B1 because it spells out that if easement of necessity is insufficient, statutory easement can be established. Type C. The duty to compensate is explicitly stipulated. Liechtenstein (SR) §102, Switzerland §694, and Turkey §747 have essentially the same rule: “This right is in the first place exercised against the neighbor who, in the light of existing ownership and access circumstances, may most reasonably be expected to grant such right of way….” It is unclear, however, whether the quoted stipulation considers requirements (1) and (2).

B  Statutory Easements In general, the key elements for statutory easements are necessary passage,9 compensation,10 and involuntary landlocking11 (e.g., Bruce and Ely 2011: §4:14). The rule regarding the location of the right-of-way is the most diverse, as shown in Table 9.1 and Figure 9.2. Forty-one jurisdictions stipulate that the right-of-way shall be set at the place that causes the least damage to the neighboring land: Afghanistan §1919, Algeria §696, Argentina §2181, Azerbaijan §175.1, Belgium §683, Bhutan (LA) §272, China §296, Croatia (AOORR) §224, Cuba §171, Egypt §812, Eritrea §1036, Ethiopia §1224, Hungary §V:162, Iraq §1059, Japan §211, Jordan §1290, Libya §821, Liechtenstein (SR) §102, Macau §1444, Macedonia (AOORR) §197, Malta §447, Moldova §392, Poland §145, Portugal §1553 and its former colonies, Qatar §842, Romania §617, Serbia (LBOPR) §50 & §53, Slovenia (Škerl 9

10

11

For example, Colombia §905, Ecuador §883, Georgia §180.1, Germany §917, Kazakhstan (LC) §69, Louisiana §690, Taiwan §787, and Turkmenistan §202. France §682 seems to give landlocked owners more leeway, as the right of way shall be “sufficient for the complete servicing” of the landlocked land. The statutory easement doctrine in 107 civil-code jurisdictions explicitly requires compensation to neighbors: for example, France §682, Italy §1053, Louisiana §689, Macau §1445, the Netherlands §V:57, Portugal §1554, Russia §274, Switzerland §694, Taiwan §787, and Timor-Leste §1444. Eight of them explicitly allow landlocked owners to compensate periodically, instead of paying lump-sum: Algeria §701, Germany §917, Iraq §1059, Japan §212, Malta §449, Quebec §1001, Thailand §1349, and Venezuela §664. The voluntary act exception is emphasized in some civil codes; for example, Germany §918, Louisiana §693, Mongolia §138.3, Taiwan §787, and Turkmenistan §202. That is, an owner who intentionally makes a parcel landlocked is not eligible for a statutory easement. Portugal §1552 adopts a unique rule, under which such owners can still acquire statutory easements through paying heightened damages, which cannot surpass twice the amount of the ordinary damages and are assessed according to the culpability of the landlocked owners.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

235

Table 9.1  Location for statutory easement Criterion

Frequency

Percent

Least damage Least damage and shortest Shortest Neighbor decides Courts and/or experts decide Governments decide Other rules No rule for location

44 29 6 1 15 1 4 13

39 26 5 1 13 1 4 12

Total

113

100

Notes: This table includes only the core 156 studied jurisdictions.

Statutory Easement No rules regarding location Discretion & other rules Width Neighbor decides Least damage & shortest Shortest Least damage No information

Figure 9.2  Typology of statutory easement

and Vlahek 2020: 154), South Korea §219, Switzerland §694, Taiwan §787, Thailand §1349, Timor-Leste §1443, Turkey §747, United Arab Emirates §1380, and Vietnam §254. Thirty-three jurisdictions add an additional requirement that the passage should be the shortest possible: Albania §277, Bolivia §262, Cambodia §144, Equatorial Guinea §565, France §683, Guinea §620, Haiti §550, Indonesia §668, Italy §1051, Louisiana §692, Monaco §§568–569, Morocco (LRRP) §143, the Netherlands §V:57, Nicaragua §1628, Panama §547, the Philippines §650, Puerto Rico §501, South Africa (Van der Merwe et al. 2002: 432), Spain §565, Tunisia (CDR) §178, Uruguay §582, and Venezuela §661, as well as §§683–684 in Burkina Faso, Comoros, Dominican Republic, Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, Senegal, Seychelles, and Togo.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

236

Part II: Immovable Property

Burundi §II:II:125, Guatemala §790, Laos (PL) §49, and Paraguay §2209 simply require the shortest route. Chile §848, Colombia §906, Ecuador §884, El Salvador §850, and Honduras §827 delegate the task to experts. Cyprus (Neocleous 2000: 597; Synodinou 2020: 80) (director of the registry decides), Kyrgyzstan ­§233-13, and Norway (LA) §16 leave it to government agencies to decide. Courts in Armenia §214, Australia (Hepburn 2002: 122), Belarus §268, Costa Rica §§397–398 (focusing on the width of the passage), Estonia (TLPA) §156, Germany §917, Greece §1013, Lithuania §4.43, New York (Bruce and Ely 2011: §4:14), New Zealand (Williams 2011: 78), Russia §274, Singapore (See et al. 2018: 195), Slovakia §151o, Tanzania (LA) §148, Uganda (ARA), and Uzbekistan §173 have discretion in the location of the easement to be established. Mexico §1099 adopts the “landlocked-owner chooses” rule.12 Other rules are adopted in Quebec §998 (“the most natural way out, taking into consideration the condition of the place, the benefit to the enclosed land and the inconvenience caused by the right of way to the land on which it is exercised”), Brazil §1285 (“more naturally and easily lends itself to the passage”) (Romañach Jr. 2015b: 237), Czech §1033 (“the most natural access”), Hong Kong (Mau 2010: 66), and Peru §1052. Georgia §180, Kazakhstan (LC), Malaysia, Mongolia §138, and Turkmenistan §202 do not stipulate how locations should be set. Interestingly, few US state statutes use similar language.13 The American state legislatures usually choose to specify the maximum and minimum width of the easement, but some states also prescribe that the route has to be the shortest (Bruce and Ely 2011: §4:14). The length of the passage affects the economic loss of the neighbor but not the way the landlocked plot can be used. By contrast, the width of the passage could affect both. For instance, commercial timber operations may not be feasible with a narrow, prescribed easement width. A few civil-code jurisdictions emphasize that landlocked owners may have the right to construct a road (instead of merely passing by) if 12

13

Mexico §§1100–1102 further stipulates that if the landlocked owner chooses a location that is impracticable or very burdensome to the neighbor, courts can demand landlocked owners pick another location. Courts which are still not satisfied with the new location have to choose the location that is “most convenient” while reconciling the interests of the two parties. Moreover, when several neighbors’ plots could provide access to public roads, the shortest distance rule applies, as long as the shortest passage is not costly and uncomfortable. For example, Arkansas uses “the least inconvenience” (Ark. Code Ann. §§27-66-401 (a)(1) (Michie 1994)).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

237

necessary; for example, Japan §211, Taiwan §788, and Louisiana §691. An interesting variation to this doctrine is Russia §274, which stipulates that the legal servitude of passage guarantees passage across the neighboring land plot both on foot and by motor vehicle.

II  A Bargaining Power Theory The starting point for my economic analysis is the reasonable assumption that landlocked owners generally value the land used for passage more than do the owners of the passage (Sterk 1987: 76), at least for the first several units of the extent of passage (0–Q3 in Figure 9.3).14 Therefore, the legal servitude of passage doctrine increases allocative benefits because it assigns the land used for passage – the resource in question – to the user who values it more. The more difficult question is what concrete designs best promote allocative benefits while not incurring significant institution cost (see Chapter 2). In the Calabresi and Melamed (1972) framework, there are two ways to deal with this issue.15 First, the property rule may apply. Under this paradigm, Phil has to bargain with his neighbor Dora to cross her land. If no deal is reached, Phil’s land will be rendered useless and valueless. By contrast, if the liability rule reigns, Phil can cross Dora’s land, in any manner, as long as he compensates for her losses in market value, possibly making “over-optimal” (like beyond Q4 in Figure 9.3) crossing in some instances due to incomplete compensation – as market value may be lower than a neighbor’s economic value. This chapter will argue that neither the property rule nor the liability rule is the most efficient solution to the problem at hand. This chapter proposes a “hybrid rule,” which can be conceptualized as a liability rule of limited extent, plus a residual property rule. The hybrid rule postulates the following: instead of the dire social choice between leaving landlocked land barren due to excessive fees and unconsummated transactions (pure property rule), and allowing landlocked owners to pass through the servient land in over-optimally (pure liability rule), the better 14

15

In other words, if the landlocked land and the adjacent land are owned by a single owner, she will create a passage over the adjacent land for the landlocked land. For discussions of the conceptual framework of a single owner, see Coase (1960); Baxter and Altree (1972: 2–4); Epstein (1993: 556–557); Schäfer and Ott (2022: 538–539). Most literature implicitly or explicitly characterizes the structure of entitlement protections as an either-or choice (Ayres and Talley 1995; Ayres and Balkin 1996; Kaplow and Shavell 1996; Ayres 1998; 2005).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

238

Part II: Immovable Property

Figure 9.3  Marginal costs and marginal benefits of passage Note: MCEV represents the marginal cost of a passage if the economic value of the neighboring land is considered. MCMV represents the marginal cost of a passage if only the market value of the neighboring land is considered. MR shows the marginal revenue (benefit) of a passage.

design is limited liability backstopped by the property rule. In short, landlocked owners cannot set the extent of passage as they like; rather, the extent of passage must be “necessary.” Indeed, the legal servitude of passage laws adopted in most studied jurisdictions already exhibit the spirit of the hybrid rule. The hybrid rule is more efficient than pure property or liability rules at solving the access issue because the hybrid rule better facilitates Coasean bargaining between the landlocked owner and her neighbor.16 Under the property rule, the bargaining power of the neighbor17 is 16

17

Chang (2016b: 225–227) discusses how the hybrid rule proposed here belongs to a family that might be called “split-the-difference rule” and its application to Pile v. Pedrick and how the hybrid rule differs from Polinsky (1980: 1086–1088)’s “intermediate entitlements.” For simplicity’s sake, I assume now there is only one neighbor. This assumption will be lifted later on in this chapter.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

239

much larger than that of the landlocked owner, and there is asymmetric information as to the reservation value of the landlocked owner. These two factors may doom voluntary transactions, rendering landlocked parcels worthless; at the very least, the transaction would be costly due to the wide bargaining range. Under the (unconstrained) liability rule, landlocked owners can acquire their desirable extent of passage rights and thus do not have incentive to further bargain with the neighbor. Nevertheless, because the compensation assessed by the court under the liability rule generally fails to consider the neighbor’s subjective value, and the court does not always accurately assess market value (Chang 2011; 2013; Chang et al. 2016), landlocked owners often do not fully internalize the total social cost of their easement. Therefore, landlocked owners often over-use. By contrast, the hybrid rule only awards limited passage right to landlocked owners, and most, but not all, of the value of the landlocked land is preserved, thus balancing the bargaining power between the two parties. Once the playing field is leveled, voluntary transactions (which will fully compensate any further loss of the neighbor) that ensue will attain an optimal extent of passage. In other words, beyond the limited extent of passage (attainable under the liability rule) that the neighbor is forced to accept, the neighbor’s entitlement is protected by the property rule. It is theoretically meaningful to distinguish between the prototypical liability rule and the hybrid rule. The prior literature – for example, Fennell (2007: 1427) and Parchomovsky and Stein (2009: 1825) – appears to assume, explicitly or implicitly, that under the prototypical liability rules the holder of call option can infringe upon others’ property interests to the extent the former prefers, as long as the holder compensates the latter for the loss. By contrast, the extent of trespass or infringement is legally constrained under the hybrid rule. As demonstrated below, due to informational constraint, courts and landlocked owners are unlikely to identify the socially optimal extent of passage. In other words, the prototypical liability rule would lead to too much passage. By contrast, the liability rule part of the hybrid rule would lead to too little passage. With the presence of the sunk cost fallacy, neighbors are less likely to bargain with landlocked owners in the scenario of too much passage. Starting with the hybrid rule is more likely to lead to a passage of optimal extent. Below, I elaborate on why the property rule should be replaced with the hybrid rule. Part IV will demonstrate why the hybrid rule is more efficient than the liability rule.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

240

Part II: Immovable Property

A  Overcoming Asymmetrical Bargaining Power and Information Under the property rule, without statutory easements, the bargaining powers of the two parties are asymmetrical, and as such bargaining may fail. Without a right-of-way, a landlocked parcel may be worthless (Dana and Merrill 2002: 181), whereas a passage is just one of the many possible ways to use neighboring land. Thus, landlocked owners have much more to lose should negotiations fail. Both parties know this. With their superior bargaining power, the neighbor is likely to ask for a price that is beyond the opportunity cost of using their land as passage, or, simply put, ask for the lion’s share of the economic value of the landlocked land. Under neoclassical economic theory, rational parties are likely to strike a bargain with the neighbor taking the largest slice of the pie. Nevertheless, there is also experimental evidence that people who consider the deal unfair might be willing to sacrifice their own economic benefit in order to punish a greedy opposite party (Güth et al. 1982; Kahneman et al. 1986; Kahneman 2011: 305–08). That is, landlocked owners may rather leave the land barren than give most of the property value to their neighbors. Asymmetric information aggravates the bargaining problem.18 Neighbors who try to split the surplus evenly face the problem of ascertaining the total size of the pie. The economic value (i.e., market value plus subjective value) of the landlocked plot is its owner’s private i­ nformation – neighbors can only guess. If neighbors over-estimate the size of the pie, they may propose (and insist upon) deals that landlocked owners deem unfair. Greedy neighbors may even request a price beyond the difference in economic value of the landlocked parcel with and without a right-of-way (this difference in value is the most that the landlocked owner is willing to pay). Hence, an efficient transaction may fail due to imperfect information. 18

One may contend that asymmetric information and bargaining power are common in many transaction settings, and challenge whether the hybrid rule or the liability rule would then be justified in all of these settings. My response is that a non-property rule such as the hybrid rule is justified in the access problem, but probably not in other settings, because here it is pretty clear that the landlocked owner generally values the entitlement more, while in most other settings it is unclear whether the party without the entitlement is a higher valuer. For a theoretical framework of adopting property rule or non-property rule, see Chang (2015c) and Chapter 2. Note that although the landlocked owner values the passage right highly, it does not lead to the conclusion that the owner should be allowed to purchase the neighboring land under the hybrid rule. Ownership does not give the landlocked owner more passage benefits than an easement that runs with the land, but depriving the neighbor of ownership (with only market value compensation) imposes more losses.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

241

Bargaining does not always fail, but a successful access deal may come with a high price. The conventional wisdom19 is that a wide bargaining range yields high transaction costs (Epstein 1997: 2094; Ayres and Madison 1999: 107; Posner 2011: 89). Because nearly the total value of the landlocked parcel is at stake, the bargaining range is wide. The social welfare gain of this trade, after a prolonged bargaining process, may be largely siphoned away by high transaction costs. The hybrid rule embodied in the legal servitude of passage law ameliorates, if not solves, the above bargaining problem. A prescribed (albeit limited) right-of-way ensures the preservation of the value of the landlocked land. To attain an optimal level of access, landlocked owners have to further bargain with the neighbors. With a narrower bargaining range and a more balanced bargaining power (for the landlocked owner has less to lose now), transactions are more likely to be consummated20 at a much lower cost.21 Asymmetric information is still a concern, but the hybrid rule does not exacerbate it. The normative Coase theorem prescribes that the law should be structured to remove the impediments to private agreements (Cooter and Ulen 2012: 91–93). This is what the legal servitude of passage does. One may contend that it would not be inefficient if the transaction fails and the neighbor buys the landlocked land from the landlocked owners (at a very low price) and becomes sole owner with an incentive to maximize the joint value of the two land parcels. Indeed, Portugal §1551 allows certain urban neighbors to exercise a call option to buy landlocked land at a reasonable price. There could be several responses to this argument. First, neighbors who acquire landlocked parcels at a low price may not value them more than landlocked owners did and even if neighbors, as a single 19 20

21

Judge Posner expresses this view in his judicial opinion, Walgreen Co. v. Sara Creek Property Co., 966 F. 2d 273, 276 (7th Cir. 1992). Empirical evidence on whether parties always bargain rationally to enhance their mutual interests is still scant. Farnsworth (1999), a notable exception, finds that animosity from litigation deters further transactions. More empirical studies on Coasean bargaining, however, are still required to enable us to better understand the nature and interaction of emotion, transaction costs, and bargaining. For empirical analyses on Farnsworth (1999)’s idea, see Chang et al. (2019) and Chang and Lin (2021). The transaction cost comparison will not be complete without considering the litigation cost under the hybrid rule. It is difficult to ascertain in the abstract whether the litigation cost under the hybrid rule, or the bargaining cost under the property rule, is lower. As argued below, the hybrid rule is transaction-facilitating. I conjecture that, all things considered, the total transaction costs under the hybrid rule would not be higher than those under the property rule in the access problem.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

242

Part II: Immovable Property

owner, aimed to maximize the total value of the two parcels, the two parcels could have been more valuable in the hands of their two original owners (given a certain extent of passage for the landlocked parcel).22 Second, in the context of statutory easements, landowners become landlocked involuntarily and usually unexpectedly. If landowners would have no prescribed passage when their land suddenly became landlocked, some (particularly those who are risk-averse) would have cared enough to bargain for a passage option or right (which was useless at the time of the transaction and often never needed), to prevent a sudden drop of property value. Total transaction costs from such bargaining are presumably higher than what would arise from bargaining between the few landowners who find themselves landlocked and their neighbors. Finally, insurance might not always be available, as most insurance policies do not cover losses resulting from governmental actions, earthquakes, landslides, tsunami, etc., events which can cause sudden landlocking. The adverse selection problem might also dissuade insurance companies from insuring against such events. Under my proposed hybrid rule, the prescribed right-of-way will not maximize landlocked owners’ net benefits. As such, they would have incentives to enter serious negotiations with neighbors. The lower the neighbor’s subjective value, the more likely that the disputing parties could avoid litigation and reach an agreement that set an optimal extent of passage, as the neighbor’s willingness to accept would not be far from the expected court-adjudicated compensation. In sum, asymmetrical bargaining power and information warrants the use of the hybrid rule, rather than the property rule. Below, I turn to the issue of the different probabilities of successful bargaining under the liability and hybrid rules.

B  Bilateral Monopoly Theory Reconsidered My bargaining power theory differs from the oft-raised bilateral monopoly theory (Sterk 1987: 57–58; Kelly 2009: 231; 2011: 349), though they involve similar issues. Bilateral monopoly theory characterizes the access to landlocked land issue as having only one buyer (a landlocked owner, the monopsonist23) and one seller (a neighbor, the monopolist). Nevertheless, this theory should be reconsidered on both fronts. Landlocked owners 22 23

If landlocked owners value pre-landlocking parcels less than neighbors, the former would have sold them to the latter. “A monopsonist is a buyer who faces an upward-sloping supply curve” (Landsburg 2008: 488).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

243

are not monopsonists if neighbors can use their land in ways other than selling a right of passage, like growing flowers. It is reasonable to assume that neighboring parcels usually have many possible uses. Depicting landlocked owners as monopsonists may overestimate their bargaining power. Moreover, neighbors may not be monopolists. A landlocked parcel could have only one adjacent plot or many. In the latter scenario, there is no monopoly of passages; rather, there could be a competitive market of passage rights. Such parcels are (technically speaking) landlocked, but their owners do not necessarily stand in a weaker bargaining position when dealing with neighbors who may compete for selling a passage right.24 Hence, characterizing neighbors as monopolists may sometimes over-estimate their bargaining power and obscure the nature of the problem. Nonetheless, neighbors do not have to be monopolists to have strong bargaining power. They can be oligopolists or form a cartel.25 Thus, the bilateral monopoly theory may mislead unsophisticated courts or other third-party adjudicators to underestimate the bargaining power of neighbors when a few adjacent parcels could serve as the servient land. Ultimately, this chapter prefers using the term asymmetric bargaining power to describe the nature of the problem here.

III  The Compensation Element Requiring landlocked owners to compensate neighbors is justifiable from an economic perspective. Compensation forces landlocked owners to internalize the costs their passage imposes on neighbors. The necessity doctrine and the least damage rule (discussed below) should, in most cases, ensure that the benefits of passage are greater than the costs. Nevertheless, compensation requirements still have the following efficiency advantages: 24

25

In this scenario, the court (or the state administrative agency that handles private condemnation in the United States) should refuse to grant a statutory easement to the landlocked owner and leave it to voluntary transactions to solve the passage issues, for the following reasons: First, property rule protection of the neighbors’ entitlements preserves autonomy and subjective value. Second, as elaborated below, the court (or the administrative agency) tends to undercompensate the owner of servient land, while voluntary transactions do not. In short, asymmetric bargaining power is the foundation to statutory easement; competition between neighbors to offer passage erodes the foundation. This cartel could be more powerful than cartels in other contexts. First, most antitrust laws would probably not define the coalition of neighbors as a cartel. Second, an act of passage (the right of which is the product controlled and sold by the cartel) is easily observable, making it easier for the cartel to discipline defecting neighbors. I am not aware of any evidence of the existence of such a cartel.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

244

Part II: Immovable Property

(1) The compensation requirement ensures that landlocked owners value the passage more than the assessed compensation. Without a compensation requirement, landlocked owners may insist on passing through adjacent plots in a not-wealth-maximizing fashion, and prevail, as landlocked owners do not internalize the costs. (2) Under a no-compensation rule, landlocked owners would prefer litigating in court to bargaining with neighbors who require positive compensation, increasing judicial burden. Neighbors would litigate in court if they thought they had a fair chance of convincing the court that another neighbor’s land as passage would incur less damage. In sum, when the parties are unlikely to arrange access before landlocking occurs, the compensation rule should be more efficient than the no-compensation rule.26 In reality, full compensation is not always feasible. The court measures the amount of compensation by computing the difference between the market value of the property before and after an easement is prescribed. Neighbors/landowners, however, measure their losses from forced passage by the changes in their economic value, which is subjective and not accounted for by a third party (such as courts). Therefore, if neighbors lose subjective value due to prescribed easement, it will not be compensated. In other words, the market value standard might undercompensate the neighbors because their subjective value is not considered (Posner 2011: 87).27 As a result, the cost of statutory easements is not always fully internalized by landlocked owners. The following analyses will be based on this claim.

IV  The Necessity Requirement This part explains why the hybrid rule is more efficient than the liability rule in the access to landlocked land problem. A limited legal right-of-way 26

27

One may contend that compensation for legal servitude, just like compensation for eminent domain, would induce property owners to over-rely and ignore the probability of the forced transaction, unless the compensation is lump-sum (Blume and Rubinfeld 1984). As I have elaborated elsewhere in the eminent domain context, market value of real properties is often assessed in a lump-sum fashion, and thus over-reliance (also framed as moral hazard) is more of a theoretical concern that will not realize in practice if appraisal methods do not change (Chang 2012a; 2013). Compensation for statutory easement is also assessed based on the differences in market value; as a result, over-reliance would not be a realistic problem that would push for a reform to reduce compensation. Granted, systemic under-compensation is less frequent in partial takings such as statutory easement than in complete physical takings. The prescribed passage is usually set at the least used part of a plot – certainly not through buildings – and landowners may not attach much, if any, subjective value to the vacant part of their properties.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

245

is more efficient than a more expansive one. An expansive easement sets the “extent” of passage (captured by the X-axis in Figure 9.3) at the point where its perceived marginal social cost equals the perceived marginal social benefits (Q4), whereas a limited easement is set at where the perceived marginal social benefit of passage is about to decline sharply – and the marginal social benefit is higher than the marginal social cost (Q1). (Note that the social cost is the next best use of the passage, whereas the social benefit is the preserved value of the landlocked land.) A limited right-of-way is able to preserve much of the value of landlocked land and balance the bargaining powers of the two parties, reducing transaction costs for future bargaining. At the same time, neighbors’ uncompensated loss is minimized. Finally, the necessity requirement should be interpreted to prescribe such a limited statutory easement. In what follows, I elaborate the above arguments.

A  MR = MC Leads to Too Much Passage The intuition for some law-and-economists may be to set the solution to problems at the point where MR = MC (marginal revenue equals marginal cost). A pure liability rule implies this formula because it does not impose any internal constraint on the extent of losses that the infringers can impose on others (Fennell 2007: 1427). However, I argue that in the context of statutory easements, this formula, MR = MC, will produce inefficiency. Assuming that a third-party adjudicator accurately assesses the market value of the neighboring land and the social benefit of passage, and following the mandate of MR = MC, the adjudicator will set the extent of passage at Q4 in Figure 9.3, because the MCMV curve that represents observable marginal social costs (marginal losses in market value) crosses the MR curve that denotes the marginal social benefits at Q4. By contrast, an omniscient social planner would have set the extent of passage at Q3 in Figure 9.3, because that is where the actual marginal social costs (marginal losses in economic value, represented by MCEV) equal the marginal social benefits. In other words, Q3 represents the optimal extent of passage, but a third-party adjudicator will award Q4 (more than Q3) to the landlocked owners – an inefficient decision, as the extent of passage is over-optimal. Even in cases where neighbors lose no subjective value from the prescribed easement, and thus MCEV = MCMV, the MR = MC formula does not necessarily lead to the socially optimal result. My prior empirical study shows that courts do not always assess market value accurately (Chang 2011). The court’s perceived social costs of passage may be higher

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

246

Part II: Immovable Property

than, or lower than, the actual social costs. Thus, the adjudicated extent of passage will deviate from Q3. Inaccurate assessment of property value is another reason why property law should leave as much decision-making to the disputing parties as possible – a precept that this book has continuously espoused.

B  “Too Little” Is More Efficient than “Too Much” Given that market value is the most practicable compensation standard and can hardly be amended, it is the MR = MC formula that has to be discarded as a judicial standard to avoid the inefficiency described above. This chapter argues that the extent of statutory easements should be set at a point where the slope of the marginal benefit curve (MR) drastically changes, as in Q1 of Figure 9.3. The hybrid rule in this context should be interpreted as such. Many economic models assume that marginal social benefits decrease smoothly. In the context of statutory easements, I believe that it is more realistic to assume that there is a kink point, at which the marginal social benefit abruptly decreases.28 Consider this example: the extent of passage increases when the landlocked owners change the manner in which they pass through the neighboring land from walking on foot, to riding a bicycle, driving a car on a single-lane road, driving on a six-lane road, etc. In many cases, the marginal social benefit sharply decreases after landlocked owners can drive a car on a single-lane road.29 Setting the extent of passage in this way also ensures that the marginal social benefit of the statutory easement is high, and most likely higher than the marginal social cost (measured in economic value) – that is, MR > MC. Of course, in cases where the observable MCMV curve crosses the MR curve before the latter reaches the turning point, it is more efficient for the court to set the extent of passage at MCMV = MR. Nevertheless, this should happen only when the landlocked parcel is not very valuable in the first place, or when the passage across the servient land has other high-yielding usages. Given that economic value is not observable to anyone other than the owner herself, neither MR = MC nor MR > MC or other rules can guarantee efficient allocation in all access cases. Nevertheless, a limited extent of statutory easement like Q1 (determined by MR > MC and the change of 28 29

When the MR curve is indeed smooth, the court should determine an extent of passage where a large portion of the value of the landlocked parcel has already been preserved. The single lane road is just an illustration of what Q1 may be like. It is certainly not a general case. When the building on the landlocked land is a high-rise rather than a single-family house, Q1 may be better represented by a two-lane or four-lane road.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

247

slope of MR), embodying the hybrid rule, should be more efficient than a “too much” statutory easement like Q4 (determined by MCMV = MR), embodying the liability rule, for the following reasons: First, the accurate marginal analysis required by MR = MC is difficult for a third-party adjudicator to implement. Under my proposal, accurate marginal analysis is not needed. The adjudicator only has to know that the marginal benefit is about to change sharply, on the reasonable assumption that in most cases, MC is lower than MR before the turning point at the MR curve (such as Q1 in Figure 9.3). Based on my reading of two hundred judicial cases in Taiwan on the access issue, judges had no difficulty setting passage extent on a case-by-case basis, according to the evidence of costs and benefits submitted by both parties. Put differently, the informational requirement of my proposal should be manageable: the landlocked owner would present a schedule of marginal and total benefits at several focalpoint passage extents (such as riding a bicycle and driving on a one-lane road), whereas the neighbor would try to discredit the schedule (demonstrating, e.g., that the marginal benefits sharply decline at an earlier point than that for which the landlocked owner petitions) and persuade the judges that the decrease in the value of her land due to the prescribed passage is greater than the benefit claimed by the landlocked owner. By contrast, the accurate marginal analysis as required by MR = MC calls for much more evidence, such as a detailed schedule of the marginal costs and marginal revenues. The take-away point is that even when subjective value is zero and market value is accurately assessed, the adjudication cost under the hybrid rule remains lower than under the liability rule. Second, too little passage (Q1) is more likely to move to the optimal passage (Q3) than is too much passage (Q4). A limited passage right balances bargaining powers and lower transaction costs because the value of the landlocked land is largely preserved, shrinking the bargaining range. Under the conventional law and economics framework, transaction costs in the Q4 passage context would not be larger than those in the Q1 passage context. Nonetheless, if neighbors suffer from sunk cost fallacy (Kahneman 2011: 342–346), they may resist paying twice for the same resource (the right to use between Q3 and Q4). Neighbors’ feeling that the liability rule is unfair might also prevent a deal from going through.

C  Necessity as MR > MC In terms of statutory interpretations, it should not be difficult to interpret the ubiquitous “necessity” requirement as stipulating the extent of

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

248

Part II: Immovable Property

statutory easements at the point where the slope of the marginal benefit curve (MR) drastically changes. In addition, no matter whether or not the MR curve is smooth, the necessity requirement prescribes that the marginal revenue of the passage has to be larger than (or equal to) its marginal cost. Given that the marginal revenue of access is decreasing and the marginal cost is increasing, the MR ≥ MC interpretation ensures that the total revenue of the access right is higher than its total cost. Hence, if a parcel of landlocked land is small and low-value, the more valuable neighboring parcel does not necessarily have to endure a statutory easement, because MC may be higher than MR. The necessity doctrine is interpreted to ensure that only higher-valuing landlocked owners will be granted a right-of-way. Czech §1032 explicitly prescribes this rule. Only a few jurisdictions, including Japan, Taiwan, and states in the United States, explicitly allow landlocked owners to build a road over adjacent land, if necessary, instead of just passing through. The necessity requirement in these statutes should be interpreted in the same way as the necessity requirement discussed above. Namely, if the marginal social benefit has sharply declined before a road is built, the road should not be considered necessary.

V  Passage Location: The Least Damage and Other Rules The necessity requirement should be interpreted to ensure that social benefits of passage are sufficiently high, and the compensation requirement internalizes costs. This part discusses the third prong of the statutory easement doctrine – the location of the passage, which concerns both the costs and benefits of passage. The doctrines regarding passage location are diverse. Seventy-three jurisdictions prescribe that the statutory passage has to be set so as to cause the least damage to the neighboring land, while six stipulate an even more specific rule – the passage must be the shortest possible route. Many US states are different, emphasizing the allowable width of the passage. Although the least damage rule has intuitive appeal, it is not necessarily the most efficient rule. Part A will discuss this counter-intuitive point in more detail. Part B discusses the length and width stipulations.

A  Four Prototypes: “Least Damage” Probably Most Efficient There are four prototypical methods to determine the passage location. As shown in Table 9.2, some approaches are pro-neighbor while others favor landlocked owners. In addition, there are objective and subjective

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

249

Table 9.2  Choosers’ utility function under four prototypical methods of determining passage location

Distributional Pro-neighbor effects Pro-landlocked owner

Objective (court adjudicates)

Subjective (one party decides)

(1) Least damage Minimize CMV (3) Most benefits Maximize BSOCIAL

(2) Neighbor chooses Minimize CEV–CMV (4) Landlocked owner chooses Maximize BSOCIAL–CMV

approaches. Under the objective approach, a dispute over the passage location often ends up in court. Under the subjective approach, the location is determined at the will of one of the parties. Thus, the two parties would usually settle. This two-by-two matrix constitutes four methods of determining passage locations. The social welfare function is BSOCIAL–CEV, but as Table 9.2 shows, neither method induces the decision-makers to make the socially efficient choice. (The utility function takes as given the requirement to compensate the neighbor’s loss in market value.) One lesson of Table 9.2 is that if the difference between CEV and CMV is negligible, method (4), which embodies the liability rule, maximizes social welfare,30 as well as reduces litigation. The problems are, first, subjective value is not always negligible. Moreover, [Maximize BSOCIAL–CMV] may not always fully delineate the landlocked owners’ utility function. Whether future transactions between the two parties are feasible or not, landlocked owners may have an incentive to behave strategically under method (4). (Thus, Mexico adopts this method but adds a safety valve under which courts can intervene.) For instance, if the transaction is not likely, vengeful landlocked owners have the power to choose the location that maximizes (or increases) neighbors’ losses and pain. If further transactions are feasible, landlocked owners may choose a passage location that imposes the maximal cost on neighbors, as long as the cost > [BSOCIAL–CMV], because landlocked owners could gain more than [BSOCIAL–CMV] from neighbors in future deals. In sum, the landlocked-owner-chooses rule has pros and cons and is not a clear winner in terms of efficiency. 30

The intuition is that landlocked owners who enjoy the benefit of accessing public roads have to internalize the costs.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

250

Part II: Immovable Property

Nor do the three other methods always dictate an efficient choice. Neighbors under method (2) have similar strategic incentives as landlocked owners under method (4). Or they may focus on minimizing their own losses. Courts bound under method (3) that are not efficiency-minded might entirely ignore the social costs of passage. The least damage rule, method (1), appears to lead courts to the other extreme – considering only costs. The least damage rule, however, is only one of the requirements of statutory easement. The necessity doctrine, another element, ensures that the benefit of passage (preserved land value) is large. Given that the extent of passage has to be set at where the marginal benefit of passing is about to sharply decline, the least damage rule only works to determine which of the potential locations (all producing similar marginal benefits) imposes the least cost on the neighbor. This is probably why 73 jurisdictions gravitate toward method (1).31 Put differently, the least damage rule exhibits the spirit of the hybrid rule. The hybrid rule recognizes that courts are unlikely to make the optimal decision all the time, but an entirely handsoff approach would not work well, either. Thus, legal intervention should be kept to a minimum (as long as the land value has been largely preserved), to keep down losses to the neighbor and to clear hurdles to future transactions. Only the least damage rule fits the hybrid rule.

B  Length and Width Given that the least damage doctrine is, overall, the most efficient method to determine passage locations, we can further evaluate the statutes that prescribe the length and width of statutory easements. Length is certainly one of the factors that influence the damage caused to neighbors, and probably the most objective factor. Prescribing the shortest passage can also save litigation costs. Nevertheless, using just length as a proxy for damage may do neighbors a disservice, as the shortest passage may cross a valuable part of the neighboring land. Twenty-nine jurisdictions use the shortest route rule and the least damage rule together. It is unclear how courts in those jurisdictions interpret these rules jointly. If they are interpreted to mean that the court has to choose a route of passage that causes the least damage from several possible shortest routes, the neighbors’ losses are not 31

Method (3), combined with the necessity requirement, may also be a plausible solution. Nonetheless, in real-world litigation, costs are often easily demonstrated and proved than benefits (here, changes in market value are at best a proxy for benefits and may not be easily appraised). This may be the reason for the popularity of method (1) over method (3).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

251

necessarily minimized and litigation costs may not be reduced by much. The court still has to assess the damage caused by each candidate route, and the route that causes the least damage may not be the shortest one. Moreover, seven jurisdictions consider only the length, which is less efficient than the rules in the 73 jurisdictions that account for the least damage caused to neighbors. I do not have a good theory as to why some states in the United States choose to focus solely on the width of the passage. Perhaps it has to do with the fact that land in Europe and Asia is generally more fragmented than that in the United States. While width could certainly affect the loss to neighboring land and the benefit of passage for landlocked owners, it can be covered by the necessity requirement.32 Without specifications on other dimensions of the passage, the state courts in these jurisdictions lack the statutory support to facilitate efficient transactions. Hence, emphasis on width alone is not a wise policy choice. Think of it this way: in a state where a width limitation replaces the least damage rule, theoretically, a landlocked owner could demand the passage be located anywhere on the neighboring plot – creating disproportionate damage to neighbors. Judges may draw on certain equity principle to limit landlocked owners’ choice set, but such a principle would still embody something like the least damage rule. Hence, to increase legal certainty and reduce litigation costs, even when state legislatures would prefer to prescribe the width of passages, they should do so while also stipulating the least damage rule.

VI  Ex Ante Viewpoint and Easements of Necessity The major differences in doctrines between easements of necessity and statutory easements are two-fold: First, in the case of statutory easements, the legal right-of-way could be imposed on any adjacent land parcel as long as the plot suffers the least damage among neighboring parcels (or the passage over the plot is the shortest). By contrast, pursuant to the easement of necessity doctrine, the landlocked owners can only gain access over land “held by the grantor at the time of the conveyance.”33 Second, in 37 jurisdictions, easements of necessity are gratuitous. This part explains why these two differences make economic sense from an ex ante 32 33

Kansas law (Kan. Stat. Ann. §68-117 (2002)) stipulates a 100-feet maximum width and a 40-feet minimum width. 40 feet seems pretty wide for a statutory easement, let alone 100 feet. See Restatement (Third) of Property: Servitudes §2.15 comment a (2000).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

252

Part II: Immovable Property

perspective (see Chapter 2).34 As for the necessity and passage location requirements for easements of necessity, they should be interpreted in the same way as those for statutory easements, elaborated above.

A  Restriction on the Choice of Servient Land Land becomes landlocked for several possible reasons: voluntary acts of landowners themselves (such as severance), third-party actions (such as building a highway), abolition of public roads by the government, and acts of God (such as an earthquake or tsunami). The statutory easement law deals with land becoming landlocked due to the last three reasons, whereas the easement of necessity law handles land that loses its access for the first reason. More precisely, there are two types of voluntary acts: one is a landowner’s decision to block an existing passage; the other is (co-)owners’ decision to partition or partially transfer land and leave part of the land parcel landlocked. Several civil codes explicitly stipulate that no easements will be prescribed following the first type of voluntary act,35 because otherwise landowners may have an incentive to extinguish their existing routes through their own land and seek legal servitude of passage over their neighbors’ plots instead. This would externalize some of the costs of passage (due to under-compensation) and reduce the value of the neighboring land. Put differently, if voluntary landlocking is not prohibited, the easement of necessity doctrine essentially gives the landlocked owner an option to switch from the property rule to the hybrid rule. When the land becomes landlocked due to partition, easement of necessity doctrines around the world also put restrictions on landlocked owners’ choice of servient land. Namely, landlocked owners can only pass through 34 35

The traditional justifications are the inferred intent of the parties and public policy “favoring the full and productive use of land” (Bruce and Ely 2011: §4:5). Portugal §1552 and Timor-Leste §1442 adopt a unique rule, under which such owners can still acquire statutory easements through paying heightened damages, which cannot surpass twice the amount of the ordinary damages, and are assessed according to the culpability of the landlocked owners. Subjective value, one would guess, usually is not as large as market value; thus, doubling the amount of compensation could presumably fully compensate neighbors and deter some landlocked owners. Nevertheless, scaling the damages according to the culpability of the landlocked owners is paying attention to the wrong factor. The critical criterion for determining the amount of damages should be the loss by neighbors. But, again, the court cannot accurately assess the subjective value of neighbors. I doubt that adopting this unique rule has made land use in Portugal more efficient. Macau §1443 has a similar rule but caps at triple damages. Romania §618 has a similar rule requiring double compensation.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

P

A

B

C

253

Public road

Figure 9.4  Hypothetical scenario for easements of necessity

land parcels owned by a former co-owner (or her transferee) or land parcels that have been part of the same plots with the landlocked parcels. This rule bars externalization of passage costs and induces owners of potential landlocked land and servient land to arrange for passage before any land becomes landlocked. Co-owners partitioning the co-owned land parcel or an owner transferring part of her land are able to avoid creating landlocked parcels by planning for the right of passage when partitioning or transferring. For example, co-owners may maintain co-ownership of a lengthy plot and use it as a passage, or transferors can structure a voluntary right-ofway between potentially landlocked parcels and adjacent parcels. Without the easement of necessity doctrine, but with the statutory easement law, (co-)owners do not always have incentives to arrange for voluntary passage because doing so reduces the value of their own land. Take Figure 9.4 as an example. Assume that plots A, B, and C are co-owned by Allen, Bob, and Carol (alternatively, Carol owns plots A, B, and C but is planning to transfer A to Allen and B to Bob). Plot P is owned by Peter. Plot A will become landlocked after partition or transfer. Allen, Bob, and Carol may agree to continue co-owning a lengthy plot along the boundary line between Plots B and C for Allen’s passage after partition. This plan, however, reduces the total size of the land parcel available for partition, thus decreasing the value of the properties that they each receive. Without the easement of necessity doctrine, Allen, Bob, and Carol may have incentives to do nothing about passage and let Plot A become landlocked. Allen would then petition the court for a statutory easement over Plot P. This plan becomes particularly likely when Allen, Bob, and Carol could persuade the court that passing through Plot P causes the least damage.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

254

Part II: Immovable Property

Failing to arrange for passage rights when partitioning or severing produces inefficient results. A statutory easement over Plot P will leave some losses by Peter uncompensated. That is, Allen, Bob, and Carol externalize some of the passage costs to Peter and, if the court follows the MR = MC formula and adopts the market value compensation criterion, Allen will excessively exploit Plot P. By contrast, if Allen, Bob, and Carol agree to establish a passage somewhere in Plots B or C, there will be no uncompensated loss, and their easement covenant will ensure that Allen and his successors are unable to use the passage at an inefficient level. Therefore, the easement of necessity doctrine that restricts landlocked owners’ choice of servient land is efficient. In jurisdictions that adopt the registration-of-right system (instead of the recording system), the registrars generally substantively review the content of the registration. One way to further ensure that easements of necessity will be pre-arranged is to authorize the registrar to deny registrations before the co-owners work out a voluntary easement. That is, coowners have to file for registration of severance or partition of properties and registration of an easement at the same time. In this sense, registration systems can complement, or even substitute, the easement of necessity doctrine. I am not aware of any jurisdiction with a registration system that has adopted this mechanism. This mechanism should now be easier than ever to implement.

B  The Gratuity Rule In general, whether or not a voluntary right-of-way is gratuitous does not influence the joint interest of grantors and grantees. The transaction prices will reflect whether the owner of the servient land would receive cash flow in the future. Yet the gratuity stipulation for statutory easement can reduce the ex ante transaction cost and the amount of litigation. A gratuitous r­ ight-of-way is a passage with a price tag of zero, while a right-of-way with compensation requirement is a passage with a price tag of market value, an objective but uncertain amount of money, subject to judicial or administrative assessment ex post. Zero dollar is a clear price that landowners Allen, Bob, and Carol can easily contract around – they can adjust the plot size or the amount of compensation to reflect the gratuitous nature of the passage, while market value is more uncertain and thus costlier to contract around. Moreover, if Allen, Bob, and Carol somehow ignore the passage issue when partitioning or severing, they will not litigate over the amount

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

access to landlocked land

255

of compensation when the law adopts the clear gratuity requirement – but they may if the law prescribes any kind of positive compensation. The number of lawsuits thus decreases due to the gratuity requirement. While in 37 jurisdictions the easement of necessity is gratuitous, at least 39 jurisdictions require landlocked owners to compensate the owners of servient land.36 The compensation requirement is not groundless. A requirement to pay market value may better ensure the efficiency of the passage of landlocked owners. Nevertheless, because the gratuitous passage still has to meet the requirement of necessity and least damage, it is likely that the c­ ourt-adjudicated right-of-way would be efficient – in this case, compensation of market value would be merely a transfer of wealth.37 One may also contend that the prospect of paying compensation may prod the potential compensator to take actions ex ante, but this bargaining stimulus is about as strong as that provided by the prospect of not receiving compensation for passage. After these two effects cancel out, the no-compensation rule is still more efficient than the requirement to pay market value, because it is a clearer benchmark to be contracted around and it reduces litigation. The conclusion reached here differs from that in the context of statutory easement, mainly because the parties in the context of easement of necessity could have bargained ex ante to avoid the landlocking problem. To further induce the parties to make ex ante passage arrangements, a penalty default rule could be designed. A penalty default rule puts the party with more information at a disadvantage so as to nudge that party to bargain with the other party ex ante, bringing the issue to the latter’s attention (Ayres and Gertner 1989: 91). In the property context, grantors are generally more informative regarding the property. The law can prescribe that if plots retained by grantors become landlocked, grantors have to compensate grantees for the passage. By contrast, if grantors’ plot will be the servient land, the passage is gratuitous. Consequently, grantors are more likely to bargain with grantees regarding the passage and compensation issues, thus revealing useful information for grantees. 36

37

Additionally, 30 plus jurisdictions like Germany do not clearly require compensation for easement of necessity, but compensation duty may be inferred (for Germany, see Roth 2017: §918 Rn. 7–9). This point applies to statutory easement as well. Nonetheless, a no-compensation statutory easement doctrine could not induce optimal ex ante planning, as becoming the neighbor of a landlocked parcel usually comes as a surprise. Moreover, a no-compensation rule may induce strategic behavior of the landlocked owner, who prefers to cross other’s land for free (than passing her own land).

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

256

Part II: Immovable Property

VII Conclusion This chapter demonstrates that a hybrid rule combining limited liability and residual property rules, applied to the problem of statutory easements, is more efficient than either a pure property rule or a pure liability rule. In this context, a hybrid rule requires that passages be set at the place where it causes the least damage to neighbors; the extent of the passage be set at the point when the social cost of passage is about to sharply decline; and the landlocked owners compensate the owners of servient land. The easement of necessity doctrine, on the other hand, induces ex ante Coasean bargaining by setting a default rule (no compensation) that is easy to contract around and by limiting the providers of passage to those who could have planned for the access ex ante. In sum, the relevant doctrines, when properly interpreted, mostly make economic sense. Jurisdictions that have both the easement of necessity and statutory easement and enact the most efficient rules include Bolivia, Cuba, Czech, Equatorial Guinea, Japan, Macau, Malta, Nicaragua, Romania, South Korea, Taiwan, Thailand, Venezuela, and Vietnam.

https://doi.org/10.1017/9781009236553.012 Published online by Cambridge University Press

PART III Movable Property

https://doi.org/10.1017/9781009236553.013 Published online by Cambridge University Press

https://doi.org/10.1017/9781009236553.013 Published online by Cambridge University Press

10 Good-Faith Purchase Proposing Fractional Ownership and Internal Auction

Here we meet the Eternal Triangle of the Law: an honest man (A), a rascal (B), and another honest man (C). Typically, the rascal imposes upon both of them … and leaves to the law the problem of deciding which of them shall bear the loss. Casner and Leac (1950: 179), cited from Mautner (1991: 95)

In a good-faith purchase dispute, the purchaser, the original owner, and the intermediary have been called “the eternal triangle of law.” An intermediary steals or embezzles movables from the original owner, then sells it to a purchaser who is not aware of the lack of authority behind the transfer. By the time the original owner finds the purchaser, the intermediary may be long gone or judgment-proof. Thus, two usually innocent parties, the original owner and the good-faith purchaser, are left to argue their cases in court, and lawmakers have to decide whom to favor. The good-faith purchase doctrine is a headache for anyone who believes that a private law doctrine should converge toward one universal, efficient solution. Levmore (1987: 11) more than three decades ago, observed that the good-faith purchase doctrine is divergent.1 As yet, there is no sign of imminent convergence. Using the largest,2 hand-coded data set on the 1

2

Klerman (2015: 279) observes that art auction houses forum-shop to auction off paintings in jurisdictions that are more pro-purchaser – but compare the observation of Klerman and Shortland (2022) that the dominant position of the U.S. art market has made the very ­pro-original owner good-faith purchase law disproportionately influential. For a critical review of the case law regarding stolen art in the United States, see Hawkins et al. (1995). Stolen art is also regulated by the 1995 UNIDROIT Convention on Stolen or Illegally Exported Cultural Objects as well as Directive 2014/60/EU of May 15, 2014. This chapter discusses only the national laws. Prior works such as Schwartz and Scott (2011) include a few dozen jurisdictions. DariMattiacci and Guerriero (2015), Dari-Mattiacci et al. (2016), and Dari-Mattiacci and Guerriero (2019) include 126 jurisdictions.

259

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

260

part iii: movable property

good-faith purchase doctrine from 246 jurisdictions around the world,3 this chapter finds that 21 variants of this doctrine exist. Perhaps this is not surprising, as the good-faith purchase doctrine is arguably the most difficult property issue. In many property disputes, there are only two parties (trespassers versus owners, for instance). At least for lawmakers who have not heard of the reciprocity theory in Coase (1960), the wrongdoer is often easy to find: people who trespass, create nuisance, fail to pay rent, etc. True, property is in rem (Merrill and Smith 2001b; 2011; Smith 2012b), and property doctrines typically affect third parties – but they are faceless and unspecified in most property disputes. The good-faith purchase doctrine is different. Two parties, not one, are innocent, and there is an additional, culpable party, usually out of the picture. There is actually a fourth actor – a final seller, who may be a merchant dealer, or one-shot dealer, who is neither the embezzler nor the thief (Schwartz and Scott 2011: 1338). Sellers’ interests are deeply influenced by the good-faith purchase doctrine. Optimizing the incentive schemes of four actors with different interests is difficult,4 and efficiency-minded lawmakers may emphasize different aspects of the incentive scheme. The doctrine, thus, diverges. The good-faith purchase doctrine has traditionally covered all kinds of movables. For illustration, readers can keep in mind looted paintings during World War II, embezzled Rolex watches, and stolen cars as prime examples. As a stylized description, the movable thing in question is either stolen from original owner O, by thief T, or sold by embezzler E without O’s authorization (e.g., O entrusts E with his watch for repair, and E “­embezzles” the watch by selling it). T (or E) then sells the item to either one-shot dealer D (who occasionally trades in used watches) or merchant dealer M (a professional, second-hand, high-end watch shop). D or M finally sells the item to purchaser P. The good-faith purchase doctrine has to decide whether to favor P or O as the rightful owner, but jurisdictions have come to different conclusions when T or E who initiates the

3

4

My property law data set contains comprehensive coding of property doctrines in 156 jurisdictions. For this chapter, I also coded 46 other U.S. states (New York, California, and Louisiana originally included; Hawaii still not included) plus Washington, DC, 11 Canadian provinces (Ontario and Quebec originally included), and 32 Mexican provinces (the Federal Civil Code of Mexico originally included). 156 + 46 + 1 + 11 + 32 = 246. I have not seen a mathematical law-and-economics model that tries to solve the optimal solutions for four parties at the same time.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

261

unauthorized transfer, and when D or M sells to P. The good-faith purchase problem may better be described as an eternal rectangle.

I  Comparative Overview This part explains variations in the good-faith purchase doctrine.5 The taxonomy is based on whether good faith is explicitly required, whether the fact that goods were stolen affects the level of purchaser protection, whether the original owner can reimburse the purchaser to re-gain ownership, and other factors. Figure 10.1 shows the taxonomy. Sections A, B, and C briefly recount laws in countries that have only bare traces of this doctrine, while Sections D, E, and F focus on the 201 jurisdictions that engage with some form of this doctrine.6 Jurisdictions in Section D adopt categorical approaches, under which stolen goods always revert to their original owners. Those in Section E adopt binary approaches, under which a distinction is drawn between stolen and nonstolen goods, but in either situation good-faith purchasers will not necessarily lose possession of the movables without compensation. That is, good-faith purchasers may, under certain conditions, receive compensation or be able to keep the movables in question. Those in Section F adopt unitary approaches, under which the issue of whether or not goods are stolen is immaterial. Each can be further classified into subgroups, as explained below.

5 6

Due to word limitation, Part I, which is based on Chang (2020a), has omitted some details. Several flow charts and world maps can be found there, too. Note the differences in the definition of good faith: 29 jurisdictions explicitly require purchasers to be in good faith rather than merely innocent – innocent purchasers should have known of title defects: for example, Armenia §275, Azerbaijan §182.1, Belarus §283.1, Croatia (AOORR) §118, Eritrea §987, Estonia (TLPA) §95, Georgia §187, Lithuania §4.96, Louisiana §523, Macedonia (AOORR) §147, Malta §531, Moldova §331, the Netherlands §III:11, Philippines §1127, Romania §938, Russia §302, Serbia (LBOPR) §72, Slovenia (Škerl and Vlahek 2010: 136), and Switzerland §933. Some jurisdictions, using more tort-like terminology, exclude the presumption of good faith if the buyer is grossly negligent or negligent. The former term is used in Germany §932, Greece §1037, Taiwan §948, Qatar §935, and Italy §1147, while the latter term is used in South Korea §249, Cambodia §193, and Japan §192. Dari-Mattiacci and Guerriero (2015: 553) call “did not know and should not have known” “objective good-faith.” Landes and Posner (1996) and Sterk (2012) describe purchasers who did not know but should have known as “innocent” and keep the term “good faith” for purchasers who did not know and should not have known from an economic standpoint. For the evolution of the definition of good faith in American law, see Balganesh (2016: 280–282).

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

262

part iii: movable property

Good−faith purchaser classification Unitary approach Binary approach Categorical approach Possession certifies title No GFP Not coded

Figure 10.1  Taxonomy of good-faith purchase doctrines

A  The Nemo Dat Doctrine Type A. Under the nemo dat doctrine (Krier 2006: 67–68; Singer 2010: 809), which prescribes that no one may transfer more than she has, even the most careful good-faith purchaser cannot acquire ownership, unless through adverse possession of movables. The 12 Type-A jurisdictions, for example, Denmark,7 Iran,8 and North Korea, may not explicitly stipulate the nemo dat doctrine, but they afford good-faith purchasers limited, if any, protection against original owners. As such, in many of these countries, no rule related to good-faith purchasers can be identified. Nine of the twelve jurisdictions, however, allow good-faith possessors to acquire ownership of the goods through acquisitive prescription (see Chapter 4). Thus, while these good-faith purchasers cannot immediately acquire ownership via the good-faith purchase doctrine, after the statute of limitations runs, they may still acquire ownership. This is substantively equivalent to giving acquirers good-faith purchaser protection, albeit subject to return duties within a few years. Put differently, good-faith purchase is akin to adverse possession in good faith with a statute of limitation of zero days. Finally, note that my coding and grouping do not take account of the statute of limitations as it is not always easy to identify whether the ordinary statute of limitations applies to the claims of original owners. 7 8

Faerstad and Lilja (2010: 271–274) note that only original owners’ carelessness and passivity would tip the case in favor of a good-faith purchaser. Iran §§324–325 has the French doctrine of “possession is indicative of ownership,” but when proven otherwise, a possessor qua purchaser does not have the good-faith purchase doctrine to protect himself. That is, Iran does not have the equivalent of France §2276II (the stolen goods exception, see below), but only France §2276I.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

263

Countries in this group, such as North Korea §62, may allow original owners to recover the thing from a good-faith purchaser no matter how the latter acquires it, but nonetheless subjects the claim to a one-year statute of limitation (§260).9 That is, after one year, good-faith purchasers essentially retain de facto, if not de jure, ownership and possession. These countries are still classified as adopting the nemo dat doctrine. In addition to the application of the acquisitive prescription doctrine and shorter statute of limitations, there are other ways that the law could favor good-faith purchasers. For example, in O’Keefe v. Snyder, 416 A.2d 862 (N.J., 1980), a famous U.S. case, the court ruled that the statute of limitations would begin to run if original owners failed to use due diligence to locate stolen goods (in that case, paintings by Georgia O’Keeffe).

B  Good-Faith Purchase Doctrine Not Found Type B. Some jurisdictions do not have a civil code and no good-faith purchase doctrine has been identified. They may have no such doctrine at all, or the stipulation may be too obscure to find. Among these, Zimbabwe (PA) §4 allows for prescriptive acquisition of movables after 30 years of possession, in good and bad faith alike.

C  Good or Bad Faith Not Distinguished Type C. The French group, led by France, stipulates the doctrine of “possession equivalent to ownership certificate.”10 Pursuant to the textual interpretation of, for example, France §2276 Para. 1, good faith is not important for current possessor to keep movables, as possession itself certifies the possessor’s ownership. That is, at least the law on the books enables bad-faith purchasers to keep goods that were not stolen.11 Under France §2276 Para. 2 and in countries that emulate France by carving out an exception, original owners can claim, within three years, stolen or lost goods from their current possessors – that is, the current possessor’s “ownership certificate” can be proven ungrounded! Adopters of the 9 10

11

Colombia and South Africa, similarly, have no good-faith purchase doctrine, but the statute of limitation applies to the original owner’s claim. Alternatively, “possession is indicative of ownership.” This stipulation essentially creates a presumption of ownership, which can be proven otherwise by a nonpossessor (Aynès 2008: 156). French courts and scholars, however, have contended that possessors have to be in good faith to acquire ownership instantaneously (Aynès 2008: 157; Ritaine 2011: 113).

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

264

part iii: movable property

French doctrine include, for example, Belgium, Burkina Faso, Burundi §III:658, Comoros, Democratic Republic of Congo §658, Dominican Republic, Haiti §2044, Ivory Coast, Luxembourg, Madagascar, Mauritius §2282, Niger, Togo (§2279 if unnoted above), and four Mexican states (Guanajuato, Puebla, Tlaxcala, and Yucatán). Seychelles §2279, however, stipulates that only possession in good faith is equivalent to title certificate. Guinea §788 is similar to France §2276, but extends the prescription period to 30 years when possessors are bad-faith.

D  Categorical Approach: Stolen Goods Always Revert 1  Non-Stolen Goods Not Always Protected Type D1. Stolen goods always revert to their original owners. Non-stolen goods, by contrast, will revert to original owners unless otherwise stipulated. Most commonwealth countries belonging to Type D1 have a Sale of Goods Act, which usually introduces the good-faith purchase doctrine with words such as the following: “where goods are sold by a person who is not the owner thereof, and who does not sell them under the authority or with the consent of the owner, the buyer acquires no better title to the goods than the seller had, unless…” and then spells out the exceptions. Interestingly, as Table 10.1 shows, the exceptions are not the same. While all jurisdictions in this group adopt the “voidable title rule”12 and the “entrustment rule,”13 the “market overt rule”14 is adopted in only a few of them. American common law adopts the nemo dat doctrine as the starting point and the Uniform Commercial Code stipulates only two exceptions: the voidable title rule and the entrustment rule.15 12

13

14

15

Using Ireland (SGA) §23 as an example, the voidable title rule means that “[w]hen the seller of goods has a voidable title thereto, but his title has not been voided at the time of the sale, the buyer acquires a good title to the goods, provided he buys them in good faith and without notice of the seller’s defect of title.” Using Singapore (FA) §2 stipulates the entrustment rule: “Where a mercantile agent is, with the consent of the owner, in possession of goods … any sale … made by him when acting in the ordinary course of business of a mercantile agent, shall … be as valid as if he were expressly authorised by the owner of the goods to make the same: Provided that the person taking under the disposition acts in good faith….” Cf. Ontario (SGA) §25. Using Nigeria (SGA) §22 as an example: “Where goods are sold in market overt, according to the usage of the market, the buyer acquires a good title to the goods, provided he buys them in good faith and without notice of any defect or want of title on the part of the seller.” As Smith (2012a: 2121–2221) points out, to economize on concepts, any property system had better start with the nemo dat principle and then add exceptions, rather than having merely a number of good-faith purchase rules that sometimes do not apply.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

265

Table 10.1  Good-faith purchase rules in the English group

Voidable rule Australia Canada, excluding Quebec Cyprus England Hong Kong India Ireland Malawi Malaysia New Zealand Nigeria Pakistan Scotland Singapore Tanzania 49 U.S. states plus D.C.; Louisiana not included Uganda *

Entrustment Market rule overt rule

✓ ✓

✓ ✓

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓* ✓ ✓ ✓ ✓ ✓ ✓* ✓ ✓ ✓ ✓



Sale of Goods Act Year / section numbers 1954 / 27 & 29 1990 / 24 & 25 **

✓ ✓ ✓







1959 / 27 & 29 1979 / 23 & 25 1893 / 24 & 25 & 27 1930 / 27 & 29 1893 / 22 & 23 & 25 1967 / 24 & 26 1957 / 27 & 29 (CCLA) §149 & §151 1893 / 21 & 22 & 23 1930 / 27 & 29 1979 / 23 & 25 1999 / 23 & 24 1963 / 24 & 25 & 27 (UCC) §2-403

1932 / 23 & 25

The entrustment rule in India and Pakistan has a mercantile agent twist. Each Canadian province has its own SGA. Here, Ontario is used as an example.

**

Type D1 is categorical regarding stolen goods. As the entrustment rule and the voidable title rule, by definition, do not apply to stolen goods, good-faith purchasers cannot keep stolen movables. Stolen goods may be sold on the open market, but countries that adopt the market overt rule, such as Nigeria (SGA) §24, Ireland (SGA) §24, and Tanzania (SGA) §26, also stipulate that the market overt rule is not applicable to stolen goods.

2  Purchasers Always Keep Non-Stolen Goods Type D2. Original owners can always recover stolen goods, whereas good-faith purchasers can always keep the purchased non-stolen goods:

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

266

part iii: movable property

Albania §166, Armenia §275, Belarus §283.1, Estonia (TLPA) §95, Guinea §788, Iraq §1164, Kyrgyzstan §291, Lithuania §4.96, Peru §948, Russia §302, and Vietnam §167. Norway’s doctrine is not crystal clear, but should be close to this group (Faerstad and Lilja 2010: 258–267).

3  Purchasers May Be Reimbursed for Non-Stolen Goods Type D3. This chapter follows Schwartz and Scott (2011) and uses the term “reimbursement rule” to describe a rule that allows original owners to regain ownership as long as they pay good-faith purchasers compensation or reimburse what the latter have paid. In Sweden and Finland, a reimbursement rule applies to non-stolen goods (Kuusinen 2010: 374; Lilja 2010: 130–131). E  Binary Approach: Stolen or Not Matters 1  Good-Faith Enough to Keep Non-Stolen Goods Type E1. Good-faith purchasers acquire ownership of non-stolen movables. Nonetheless, Type E1 differs in how they treat buyers of stolen goods, elaborated below.16 a. Market Overt Rule for Stolen Goods  Type E1a. Good-faith purchasers can keep the stolen things in question if they were bought on the open market: Bulgaria (OA) §78, Croatia (AOORR) §118, Germany §932 & §935, Greece §§1038–1039, Moldova §331, Poland §169, and Turkmenistan §211. b. Reimbursement Version of the Market Overt Rule for Stolen Goods  Type E1b. Original owners cannot recover the stolen things without reimbursing good-faith purchasers in the amount that the latter paid, if the latter acquired the stolen things on the open market: Equatorial Guinea §464, Eritrea §§990–991, Spain §464, and Tunisia (CDR) §54. c. Market Overt Rule and Merchant Dealer Rule for Stolen Goods  Type E1c. Good-faith purchasers can keep stolen movables in question if they were bought on the open market or from merchant 16

In some Type-E1 jurisdictions, such as Germany and Thailand, auction is also a type of sale venue that elevates the protection of the good-faith purchaser. However, in order not to further complicate the taxonomy, the classifications below ignore whether or not auction is singled out for heightened protection.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

267

dealers who sell similar items as their trade: Austria §367 and the Netherlands §III:86. d. Reimbursement Version of the Market Overt Rule and Merchant Dealer Rule for Stolen Goods  Type E1d. Original owners cannot recover stolen things without reimbursing good-faith purchasers what the latter have paid, if the latter acquired the stolen movables on the open market or from merchant dealers: Algeria §836, Argentina §1895 & §2259, Bahrain §908, Bolivia §102, Cambodia §194, Costa Rica §481, Egypt §977, Ethiopia §§1165–1166, Japan §§192–194, Jordan §§1189–1190, Kuwait §939, Libya §§980–981, Liechtenstein (SR) §513, Louisiana §§522–524, Mexico §§798–799, Monaco §§2098–2100, Paraguay §2412, Philippines §559, Qatar §972, Romania §937, Seychelles §§2279–2280, South Korea §§250– 251, Switzerland §934, Syria §928, Taiwan §950, Tajikistan §323, Thailand §1332, United Arab Emirates §§1325–1326, Uruguay §1213, and Venezuela §795 & §1986. e. Reimbursement Rule  Type E1e. Original owners cannot recover stolen things without reimbursing good-faith purchasers: Afghanistan §2289, Latvia §§1065–1066, and Malta §§558–559 & §2155. f. Auction Rule or Judicial Proceeding  Type E1f. Azerbaijan §182 affords more protection only to purchasers of stolen goods at auction – these purchasers can keep the goods. Georgia §187 allows purchasers of stolen goods on the open market, or at auctions, to keep the thing in question. Kazakhstan §261, Ukraine §388, and Uzbekistan §229 give heightened protection to those who purchase in judicial auctions. g. Reimbursement Version of Public Auction and Annual Market Rule for Stolen Goods  Type E1g. Suriname §637 & §1176a is the only example. For stolen goods, reimbursement is required only when purchasers have bought things in public auctions or “annual markets.”

2  Good-Faith Not Enough to Keep Non-Stolen Goods with Certainty Type E2. Macedonia (AOORR) §147 has a complicated rule. First, regarding non-stolen goods, good-faith purchasers have to buy the movables at auction or from merchant dealers to be able to keep them. However, original owners can reimburse good-faith purchasers to recover the things if they are of special import to original owners. Second, regarding stolen

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

268

part iii: movable property

goods, original owners can recover the things unless good-faith ­purchasers buy them in an auction.

F  Unitary Approach: Stolen or Not Does Not Matter 1  Strongest Good-Faith Purchase Protection Type F1. Italy §1153, Hungary §V:39,17 Guatemala §1794, and Mongolia §114 allow good-faith purchasers to keep movables without almost any condition other than being in good faith. 2  Reimbursement Rule Type F2. Laos (PL) §58 adopts the reimbursement rule, whereas Serbia (LBOPR) §31 adopts the reimbursement rule if the thing in question has special meaning to original owners (recall the Macedonia rule above). In Slovenia, good-faith purchasers have to purchase on the open market, at auction, or from merchant dealers to keep the goods. Nonetheless, even meeting the requirement, good-faith purchasers may have to surrender the thing to original owners if the thing is of special import to original owners and original owners compensate good-faith purchasers (Škerl and Vlahek 2010: 128). 3  Market Overt and/or Merchant Dealer Rules Type F3. Original owners can recover movables if buyers do not purchase them from merchant dealers or on the open market. Nonetheless, if sales took place on the open market or in a merchant’s store, there are some further variations: a. Can Keep if Buying on the Open Market  Type F3a. Israel (SL) §34 and Brazil §1268 adopt the market overt rule – buying in the right place, good-faith purchasers can keep the goods. For simplicity’s sake, this chapter takes some liberty in classifying Czech in this group as well. Czech allows good-faith purchasers to keep the goods, if the sellers and buyers are in “commercial relations” (Tichy 2010: 325–326). b. Reimbursement Version or Either or Both Rules  Type F3b. Chile §890, Ecuador §§933–934, Indonesia §582, Macau §885, Quebec §§1713– 1714, Timor-Leste §1221, and Turkey §§989–991 adopt the reimbursement 17

The sale has to be “in the course of trade,” but it is unclear how strictly it is interpreted in Hungary.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

269

version of the market overt rule and/or the reimbursement version of the merchant dealer rule. c. Reimbursement Version of the Merchant Dealer Rule  Type F3c. Portugal and its former colonies adopt the reimbursement version of the merchant dealer rule. Portugal §1301 is a good example of how the doctrine of acquisitive prescription in movables mixes with the goodfaith purchase doctrine. The good-faith purchase doctrine is classified as a special rule under the Section of Acquisitive Prescription in Movables in the Portuguese civil codes. Nonetheless, Portugal §1301, unlike §§1298–1300 which deal with typical acquisitive prescription issues and contain stipulations regarding prescription periods, does not stipulate a prescription period, which makes the good-faith purchase doctrine an apparent outlier, as it is the final article in the Chapter of Prescription Period. d. Blending Two Doctrines  Type F3d. Nicaragua §§899–901 is unique and also blends the doctrine regarding acquisitive prescription of movables and that regarding good-faith purchase. In Nicaragua, prescription periods for stolen goods (five years) and non-stolen goods (two years) are different, but if buyers purchase goods on the open market or from merchant dealers, the reimbursement version applies when original owners demand recovery.

4  Unique Chinese Rule Type F4. China §§311–312 is idiosyncratic. The reimbursement version of the merchant dealer rule applies to lost goods, but not stolen goods. II  A Revised Optimization Theory This chapter proposes a new design that promotes efficiency better than competing designs.18 As Table 10.2 shows, when an original owner fails to spend optimally on prevention, but a purchaser spends optimally on verification, the latter should be legally protected from challenge; in the opposite scenario, the former should be protected. When both fail 18

Guerriero (2016) and Dari-Mattiacci and Guerriero (2015) argue that there is no single correct rule, as country- and region-specific market characteristics and cultures may lead to locally efficient regimes that diverge. Levmore (1987) argues that reasonable persons can disagree on what the correct design for the good-faith purchase doctrine is.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

270

part iii: movable property

Table 10.2  Summary of design Purchaser Spent ≧ optimal verification Original owner

Spent ≧ optimal prevention Spent < optimal prevention

Spent < optimal verification

Fractional Protect original ownership and owner internal auction Protect purchaser Protect original owner

to spend optimally, original owners are protected, as their prevention efforts are likely to be independent of the content of the good-faith purchase doctrine. When both spend optimally, the court can assign each party a 50% share of the thing in question and host an internal auction between an original owner and a good-faith purchaser. The winner in the auction is the higher-valuer and thus takes home full ownership, whereas the low bidder in the auction will be compensated by the winning bidder. Efficiency is thus achieved. Giving each of the two innocent parties half ownership is an incredibly simple solution. Yet, no jurisdiction has adopted it!19 The starting point of my economic analysis is based on the optimization theory of Schwartz and Scott (2011: 1347–1353). Their basic idea is that original owners should be induced to exert optimal preventive measures, whereas purchasers should be incentivized to undertake optimal verification. Socially optimal prevention occurs when original owners’ marginal prevention costs equal the decrease in the probability of losing the goods (as a function of the prevention effort) times the value of the goods. Socially optimal verification is where purchasers’ (or dealers’) marginal verification costs equal the increased probability of purchasing full ownership to the goods (as a function of the verification effort) times the value of the goods. My revised theory differs from Schwartz and Scott (2011) in the following ways: (1) search costs can be ignored; (2) original owners’ prevention 19

One might challenge: if the proposal is really good, why haven’t any jurisdictions adopted it? As said, part of the reason is that lawmakers do not necessarily care about efficiency. Part of the reason is path dependence and the fact that the good-faith purchase doctrine in all jurisdictions was enacted well before the theoretical foundation of my proposal was advanced.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

271

level is largely invariant to the content of the good-faith purchase doctrine and close to the optimal level; (3) merchant dealers play a key role; and (4) allocative benefits should be explicitly included in the social goal.20 At best, under the optimization theory by Schwartz and Scott (2011), both original owners and purchasers spend at the exact optimal points. This is better than a regime in which original owners are careless and externalize the social cost of theft and embezzlement, or a regime in which purchasers shy away from transactions because the law demands inefficiently high verification. In other words, a regime designed by the optimization theory produces more transactions and less thefts and embezzlements at the lowest cost. Nonetheless, this theory alone does not explicitly consider which party should gain ownership of the goods in question. Unlike in torts, where, after both parties have behaved optimally, who bears the loss is only of distributive concern, in property, the (valuable) goods remain. If market friction (Coase 1960) or animosity (Farnsworth 1999) bars transactions after litigation, legal rules should endeavor to increase allocative benefits. My revised optimization theory addresses this.

A  Maximize Efficiency in Property Law The first and foremost economic question is what the good-faith purchase doctrine should maximize. In Schwartz and Scott (2011), the goal is to induce original owners to spend optimally on preventing involuntary transfer of movables and searching for them after conversion, and to induce potential purchasers to verify ownership optimally. This is, as Schwartz and Scott recognize, very tort-like thinking.21 Schäfer and Ott (2004: 421)22 and Cooter and Ulen (2012: 152–153) take a position to 20

21

22

A brief note on social versus private costs and benefits. My (uncontroversial) view is that all private costs are also social, but not all social costs will fall on actors in question. In this context, original owners’ search costs are social costs, but often largely externalized to the police. By contrast, not all private benefits are social benefits. Once stolen or lost, things recovered by original owners are their private benefits. Yet, the social benefit of recovery is lower than the private benefit – the social benefit may even be negative, if the good-faith purchaser values the thing more. Smith (2017: 810) also observes that the law and economics literature tends to take a tortlike perspective toward sequential exchanges, using the good-faith purchase problem as an example. Arruñada (2017: 756) proposes a theory of property as sequential exchanges, and characterizes Schwartz and Scott (2011) and Dari-Mattiacci et al. (2016) as adopting the single-exchange model. Schäfer and Ott (2022: 507–510), however, offer new analysis and discuss Schwartz and Scott (2011).

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

272

part iii: movable property

minimize total costs; thus, they assign responsibility to the party who is the least cost avoider. Medina (2003: 348) pursues “highest expected value of the ownership right.” The value function is T = (1 – α)R + αB. R is the ­owner’s reservation value and B is the buyer’s willingness to pay. The weight, α, can vary by context. One would pause to ponder why a pure tort-law approach advocated by Schwartz and Scott is a good fit. While economic analysis of law often champions a unified means of thinking about law, and efficiency is indeed the common goal, differences across legal fields make the relevant concerns distinct (Rose 1997; Chang 2015c). In property law and economics, the primary goal has been called allocative efficiency (Epstein 1994: 31; Merrill 2009: 488), which is achieved when resources go to the party who most values them. Chapter 2 describes this in terms of increasing allocative benefits, a key component in efficiency analysis in property law. This allocative goal cannot be achieved simply by inducing optimal prevention, search, and verification. Increasing allocative benefits requires that the thing in question go to original owners or purchasers, depending on whose economic value is higher. Medina’s approach, by contrast, maximizes a weighted average of R and B, often yielding a different result. Shavell (2004: 39) maximizes “the expected value of property minus the costs of efforts” when analyzing the good-faith purchase doctrine. While Shavell does not elaborate this function, it appears to be consistent with the efficiency criterion proposed in Chapter 2. Much of the prior literature on the good-faith purchase doctrine either over-emphasizes the institution cost or pays attention to only the allocation benefits.23 Both should be considered. Given the goal of efficiency, it should not be difficult to justify the exclusion of bad-faith purchasers.24 Indeed, sometimes bad-faith purchasers might be the higher valuer. Nonetheless, the costs associated with efforts by bad-faith parties to refrain from purchasing things with only defective rights are low, and the social benefits of, for instance, reducing property 23

24

Dari-Mattiacci and Guerriero (2015: 555–556) use “incentive vs value” to characterize concerns that largely overlap with the cost and benefit ledgers of my efficiency standard. In observations by Dari-Mattiacci and Guerriero (2015: 555–556), the two goals appear to be taken as alternatives in the literature, not goals that should be jointly maximized. ­Dari-Mattiacci et al. (2016: 42) further argue that value allocation is more relevant than inducing original owners and purchasers to behave efficiently. That is, the law should focus on allocative efficiency. Bad-faith purchasers never acquire de jure ownership, but they keep de facto ownership if not caught. Bell (2015) points out that the adverse possession doctrine is used to solve the gap between de jure and de facto property rights.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

273

crimes are high. Usually close substitutes exist, so bad-faith parties can buy elsewhere. If the things in question are unique, it should not be difficult for bad-faith parties to locate the real owners and bargain with them. The more challenging question is: to maximize efficiency, under what other conditions should good-faith purchasers be protected? As stated above, much of the literature on good-faith purchase focuses on lowering institution cost. Allocative benefits are increased only as a by-product (if mentioned at all) of keeping things in the hands of their original owners (assuming that they are higher-valuers) or inducing purchasers to be informed of unauthorized sales (as reducing theft and embezzlement keeps things in the hands of their owners). A large part of the difficulty is that a general case for original owners or purchasers being higher-valuers under certain conditions is hardly convincing. A view perhaps implicitly held may be that voluntary transactions will redress most, if not all, of the misallocation; thus, lawmakers of a good-faith purchase rule need not be concerned about the misallocation. The following analysis will address the aforementioned questions and explore what rules enhance ex ante efficiency. For the same reason, institution costs will be the main focus, whereas allocative benefits can only be considered indirectly. The analytical result will be combined with a radical proposal that ensures ex post allocative efficiency, discussed in Part III.

B  Original Owners’ Prevention Costs In the design of Schwartz and Scott (2011: 1340), original owners who are not negligent in prevention can always recover involuntarily transferred goods, so as to give original owners the desired incentives.25 Buyers’ strict liability increases their level of verification, but still not to the optimal level. That is, even with the tort approach of Schwartz and Scott (2011), not all parties can be induced to behave efficiently at the same time. Optimal prevention is indeed a social goal, but perhaps lawmakers need not focus on facilitating optimal prevention. Schwartz and Scott posit that the original owners’ prevention level is a function of the probability of recovering stolen and embezzled goods, and thus the prospect of recovery leads to a suboptimal level of prevention. Nevertheless, if original owners 25

No jurisdictions except Denmark appear to require the original owners to be nonnegligent. An optimal prevention rule can be doctrinally implemented with a rule that allows owners to recover only when they are not (grossly) negligent in prevention.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

274

part iii: movable property

do not expect to recover their things once stolen or embezzled – this chapter conjectures that many people, ex ante, perceive it this way, and in many jurisdictions the actual probabilities of recovery may indeed be low – their prevention level will not be far from the optimal level, no matter whether under the nemo dat doctrine or under the most generous good-faith purchase doctrine.26 It follows that original owners’ prevention level is nearly independent of the doctrinal substance, and it is usually close to the optimal level.27 If this is the case, lawmakers should pay attention to regulating behaviors of dealers and consumers, but not those of original owners. Therefore, unlike the design in Schwartz and Scott (2011), my design focuses on the incentives of purchasers, not original owners (more on this below). A theory can only be conditionally correct. Whether my proposed design or Schwartz and Scott (2011)’s works better critically depends on the assumption that many people perceive the probability of recovery to be low. There may be cross-jurisdictional variations. In countries where law enforcement is effective, the perceived probability of recovery is higher, and the doctrinal design is more likely to affect owners’ level of precaution. There may also be intra-jurisdictional variations. Movables that will be kept intact (e.g., paintings) and used publicly (e.g., high-end cars) are more likely to be recovered, while movables that will be dismantled (low-end cars) or used in production are less likely to be recovered. When the perceived probability of recovery is low, my model may create lower institution costs; when it is high, Schwartz and Scott’s model may create lower institution costs. The idea of optimal prevention is more useful in criticizing a prevalent form of good-faith purchase doctrine adopted in jurisdictions such as Germany and Taiwan.28 There, original owners are favored when it comes 26

27

28

But compare Bell and Parchomovsky (2012a), who argue that owners do not always have incentives to protect their own things, especially when they can externalize the search costs to the police if the things are stolen or lost. If this assumption is correct, the potential problem of one party’s optimal verification (prevention) level being contingent on the other party’s prevention (verification) level goes away. This is because prevention takes place first and is close to the optimal level. Thus, purchasers who verify later can safely assume that the prevention level is close to optimal and set their verification level accordingly. The following jurisdictions protect original owners’ lost and stolen goods at the same level: Afghanistan §2289, Algeria §836, Armenia §275, Azerbaijan §182.2, Bahrain §908, Belarus §283.1, Belgium §2279, Bolivia §102, Bulgaria (OA) §78, Cambodia §194, Costa Rica §482, Croatia (AOORR) §118, Egypt §977, Equatorial Guinea §464, Georgia §187.2, Germany §935, Greece §1038, Haiti §2044, Japan §193, Kazakhstan §261, Kuwait §939, Kyrgyzstan §291, Libya §981, Liechtenstein (SR) §513, Lithuania §4.96, Louisiana §524, Macedonia (AOORR) §147,

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

275

to not only stolen goods but also lost goods and other involuntarily dispossessed movables. If anything, the loss of valuable goods (those worth searching out and litigating over) is often a result of reckless owners’ suboptimal prevention level. If social litigation costs are low enough, lawmakers could welcome the investment of time and resources by original owners of lost goods in going to court to prove that they have exerted optimal prevention. As Schwartz and Scott demonstrate, real-world courts have to rely on proxies to ascertain whether optimal prevention costs have been invested. Under this realistic constraint, not allowing owners of lost goods to recover at all expends fewer social resources, while not necessarily compromising resource allocation. Hence, efficiency-minded lawmakers should not create the same rule for stolen goods and lost goods. Most movables owners nonetheless will spend optimal costs to prevent losing their things because they are not entitled to their recovery. Many prior studies on the doctrine focus on stolen goods and how to set the right incentives for original owners, purchasers, and thieves (BenShahar 1997; Salomons 2009; 2011; 2013). Positive laws often share this focus. Among the 201 jurisdictions with good-faith purchase doctrines, 176 (88%) distinguish stolen from non-stolen goods (Figure 10.2).29 Under my framework, stolen goods and embezzled goods should not be distinguished. Original owners and purchasers have to spend optimally to be protected against both theft and embezzlement. Given this framework, the only reason to differentiate between stolen and embezzled goods is that (1) original owners systematically fail to spend optimally when it comes to embezzled goods or (2) purchasers systematically fail to spend optimally when it comes to stolen goods; or both. Hence, the law can categorically choose to protect one party once the nature of the goods (stolen or embezzled) is known. In the jurisdictions surveyed in Part I, stolen goods are often singled out and the legal protection goes to original owners. This makes economic sense only if condition (2) above holds. As purchasers, especially ordinary consumers, ex ante have no way to distinguish stolen goods from embezzled goods, it is doubtful that purchasers can spend optimally on verifying ownership when it comes to embezzled goods but fail to

29

Malta §559, Mexico §799, Moldova §331, Paraguay §2412, Poland §169, Qatar §972, Romania §937, Russia §302, South Korea §250, Switzerland §934, Syria §928, Taiwan §949, Tajikistan §323, Turkmenistan §211, Ukraine §388, Uzbekistan §229, Venezuela §795, Vietnam §167, and common-law jurisdictions that generally protect original owners. Also, according to DCFR VIII-3:101, good-faith purchasers of stolen goods do not immediately acquire ownership, unless they do so in the ordinary course of business.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

part iii: movable property

276

Market overt rule and theft rule Both Market overt rule Stolen goods distinguished Neither No GFP Not coded

Figure 10.2  Stolen goods and market overt rule

do so in terms of stolen goods. Hence, the categorical distinction prevalent in 88% of the surveyed jurisdictions has a dubious efficiency foundation.

C  Original Owners’ Search Costs Schwartz and Scott (2011: 1344–1345), unlike the prior literature, emphasize that search costs should be optimal. In their framework, searching is a means to deter theft and embezzlement, to recover stolen and embezzled goods, and to enable original owners to meet purchasers for further bargaining. Granted, reducing property crimes and recovering stolen and embezzled goods are important social goals (Hasen and McAdams 1997). Nonetheless, the cost of searching in modern society has largely been externalized to the police. The design of the good-faith purchase doctrine is highly unlikely to induce original owners and the police to search jointly at the socially optimal level. Even when the search costs by the police are treated as exogenous and ignored, it is unclear whether the actual search level under any design of the good-faith purchase doctrine is higher or lower than the socially optimal level because the benefits of searching are often beyond the information held by original owners, and some of the benefits of searching are externalized. (E.g., what is the marginal effect of theft reduction by an owner’s ad hoc searching?) Therefore, it is not necessarily more or less efficient when a new design of the doctrine increases original owners’ incentives to search. As for searching’s facilitating negotiation between original owners and purchasers, Schwartz and Scott follow the Coasean paradigm in leaving the two parties to bargain with each other, implicitly assuming that the

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

277

transaction costs between them are not prohibitively high. While this chapter is generally sympathetic to this approach, in this context, voluntary bargaining will not always maximize allocative benefits. Part III champions a new approach to deal with this issue. In light of the preceding reasons, the following analysis will not consider the optimization of social searching costs.

D  Purchasers’ Verification Costs Purchasers should be induced to conduct optimal verification. Schwartz and Scott (2011: 1356) point out that because purchasers who buy goods with flawed title will not always be found, their verification level will be suboptimal. Schwartz and Scott (2011: 1338) also treat merchant dealers as purchasers for their analytical purposes. However, in terms of optimal verification, it is worth distinguishing between merchant dealers, oneshot dealers (who are not merchant dealers), and consumers. Under the optimization paradigm, Ben-Shahar and Porat (2016; 2021: 62) propose personalizing the level of care. While their main example is accident law, the logic largely applies to the good-faith purchase context. As Ben-Shahar and Porat (2016) point out, the level of care can be personalized along two dimensions: skill and risk. In short, their argument is that, holding other factors constant, more skillful actors should be held to a higher level of care than the less skilled, as the former can prevent bad outcomes more effectively. In addition, riskier actors should be held to a higher standard of care than the safer, as the formers’ effort is more likely to reduce the probability of bad outcomes. Unfortunately, the two dimensions often counteract each other. That is, skillful actors are often also safe, and unskilled actors are often also risky. For instance, in my context, merchant dealers are more skillful in identifying title defects than one-shot dealers. Nonetheless, merchant dealers are also safer than oneshot dealers because thieves and embezzlers are prone to sell to one-shot dealers lest their property crimes be exposed by professional middlemen. Therefore, the baseline probability of selling movables with title defects by one-shot dealers is higher than that by merchant dealers. Which dimension, skill or risk, should policy-makers choose to determine levels of care? Like Ben-Shahar and Porat (2016: 668), this chapter tends to favor the risk dimension. First, as Schwartz (1989) also points out, skill-based personalization gives rise to an ill ex ante effect of inducing dealers to refrain from acquiring expertise. That is, dealers are inclined to remain one-shot dealers, rather than expanding their business or

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

278

part iii: movable property

cultivating their professional knowledge. By contrast, imposing higher levels of care on risky actors induces them to become safer types. In this context, one-shot dealers will seek to upgrade to merchant dealers, so as to reduce their duty of care. Based on the assumption that thieves and embezzlers are more likely to sell to one-shot dealers, reducing the ratio of one-shot dealers to merchant dealers may even deter property crimes. Hence, risk-based personalization is more efficient. Second, from the skill perspective, consumers are simply unskilled, and the law can do nothing to induce them to become better at distinguishing goods sold with and without authorization. No personalization can be done. However, consumers can easily distinguish merchant dealers and one-shot dealers using common sense. Ordinary people also know that in general merchant dealers are more “reliable.”30 From the risk perspective, consumers who buy from merchant dealers can be considered the safe type, whereas those who buy from one-shot dealers can be considered the risky type.31 Personalization law should then impose the requirement of a higher level of care on consumers buying from oneshot dealers. In short, merchant dealers and consumers who buy from merchant dealers should be held to a lower level of care standard, whereas one-shot dealers and consumers who buy from them should be subject to a higher level of care requirement. If the private law in a jurisdiction stipulates that once dealers are good-faith purchasers, consumers are protected no matter whether the latter should have known, consumers who buy from merchant dealers are more likely to acquire perfect ownership. The reason is that either merchant dealers have verified optimally and are considered legitimate sellers, or consumers have met the lower threshold of optimal verification. If the private law in a jurisdiction requires consumers to be good-faith no matter whether dealers are negligent or not, buying from merchant dealers still makes it easier for consumers to be considered acting in good faith. Consumers will be attracted to merchant dealers. 30

31

Thus, under my proposed regime, consumers do not have to know the law; rather, they only need to know that merchant dealers are more reliable, and less trouble would follow if purchasing from merchant dealers. Merchant dealers are in a better position to know the law and merchant dealers have incentives to inform consumers that they are better protected if purchasing from the former. This argument assumes that even if a lower level of care were required of merchant dealers, after all the verification is done, the ex post probability for consumers to acquire goods with title defects from merchant dealers is still lower than that from one-shot dealers.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

279

In the real world, it would be costly to ascertain on a case-by-case basis whether customers who buy from merchant dealers and one-shot dealers have exerted optimal verification. A shortcut is to adopt the market overt plus merchant dealer rule, under which a customer who buys from a merchant dealer who displays goods openly for sale will be categorically protected, whereas a customer who buys in other venues will not be protected.32 Merchant dealers under this short-cut rule still have the incentive to screen goods offered for sale, as they have reputational concerns and will be liable to original owners for unjust enrichment if it is deemed that they should have known title defects and thus will not be protected.

E  Double Negligence Who should be protected when both original owners and purchasers are negligent? Schwartz and Scott (2011: 1365) favor the purchasers, in order to minimize the disruption of involuntary dispossession. The better economic defense for their position might be that an opposite rule would not give original owners incentives to maintain optimal prevention. In my theory, even lacking legal requirements, original owners will not deviate far from optimal prevention levels so protecting them when both parties are negligent will not induce suboptimal prevention levels but rather increase verification levels. In other words, purchasers will never be protected if they fail to verify optimally.33 Purchasers’ (expected) verification level will still be suboptimal because not all goods with title defects will be found; thus, a proposal by Schwartz and Scott (2011: 1342) to impose a civil fine on bad-faith dealers is worth considering to increase their level of verification to the range of optimality.

F  Double Non-Negligence In assuming away certain constraints, the preceding analysis suggests that both original owners and purchasers would not behave optimally. 32

33

Some forms of the market overt rule are adopted in 98 of the 201 (49%) jurisdictions that have a good-faith purchase doctrine (Figure 10.2). Here, I count in the reimbursement version of these rules. Note that even when merchant dealers have sold goods with defective title to consumers, if not having spent optimally on verification, they are liable in conversion, torts, or unjust enrichment (depending on the private law in the jurisdiction under discussion) to original owners. This is the case no matter whether consumers are acting in good faith or not.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

280

part iii: movable property

Nonetheless, this does not imply that all owners and purchasers will in fact behave suboptimally. First, some purchasers may over-estimate the real probability of identifying stolen and embezzled goods through verification, and thus spend more on verification than those with perfect information. Second, verification efforts are not always continuous; rather, they may be conducted in lumpy chunks and as a batch (Fennell 2012; 2019b). (Otherwise, double non-negligence would be a null set and not worth discussing.) While in reality, courts can only rely on proxies to examine whether they have behaved optimally, there should be many cases in which it is clear that both parties have done their due diligence. When both parties have spent optimally on prevention or verification, it appears that lawmakers can choose to protect either one. Lawmakers around the world have painstakingly designed rules to assign entitlements to either party in full. The implicit assumption is that the ownership of movables should not be divided. This might seem reasonable until we pause to ponder why lawmakers in other property issues, such as accessio (see Chapter 13), often adopt equal-share co-ownership as the solution.34 As described in Part I, some jurisdictions, while beginning with the fullownership assignment rule, also adopt the reimbursement version of the market overt rule under which an original owner can reimburse a good-faith purchaser for the price paid and recover the thing in question. The reimbursement rule could promote resource allocation when the thing in question has no close substitutes for its original owner for idiosyncratic reasons and she is the higher-valuer. It is not always first-best, however, as a purchaser may be the higher-valuer. Subjective value is in general non-verifiable by third parties (compare, e.g., Levmore 1982; Bell and Parchomovsky 2007; Chang 2012c; 2013; Chang and Fennell 2014b). To elicit private valuation, the reimbursement rule could be tinkered with to become a higher-order liability rule (Ayres and Balkin 1996). That is, original owners should be required to pay purchasers, say, 110% of the latter’s payment. Still, secondorder liability is insufficient to maximize allocative benefits, as purchasers’ willingness to pay may be 120% of the first actual payment. Part III proposes a radical change that solves this difficult informational problem. Adopting that design in the double non-­negligence scenario can maximize allocative benefits while keeping the institution costs of my proposed regime low. 34

The parallel issue in accident law is residual loss sharing when both the injurer and the victim are nonnegligent (see, e.g., Carbonara et al. 2016). The accident-law framework is not readily applicable to the good-faith purchase issue discussed here.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

281

III  Fractional Ownership and Internal Auction Part II concludes with a dilemma: whom should the law protect when both parties have behaved optimally? This is a question of allocative benefits, which is not a goal advanced in most prior studies. This part escapes from the all-or-nothing presumption of the current paradigm and entertains a radical proposal, which should apply when there is no strong normative reason to protect either party in full – in my case, when both parties have behaved optimally in reducing institution costs. This part proposes a mechanism that can ensure ex post allocative efficiency.35 Here, “ensure” is used in a strict sense, meaning that under a legal rule, movables in question will be allocated to, say, purchasers if and only if purchasers value them more than original owners – often called “ex post efficiency” in the mechanism design literature (Myerson and Satterthwaite 1983; Myerson 2008). Whom between original owners and good-faith purchasers values movables in question more is unobservable to judges, so a standard that simply gives judges discretion in assigning entitlements cannot ensure allocative efficiency, either. This part advocates a regime with fractional ownership and internal auctions, based on the ideas developed in the bilateral trading literature (one of the key questions in mechanism design in economics) in general and in the writings on internal auctions (Ayres and Balkin 1996: 707–717) and fractional property rights (Ayres and Talley 1995: 1072) in particular. This line of literature would propose a mechanism under which when an original owner and a good-faith purchaser co-own things initially, the party who places the highest bid in internal auctions will become the sole owner after paying the other party the difference between the two bids. The mechanism design literature in economics has proved that as long as neither party owns more than a 79% share, the parties will bid with their reservation values, and allocative efficiency will always be attained. Deviating from the 50–50 allocation of ownership increases courts’ decision-making costs. As a starter, this part argues for the application of 50–50 fractional ownership on top of the rules championed in Part II. 35

One may contend that original owners are systematically the higher valuers, as purchasers could have bought the goods from original owners but did not, and question why an auction is needed to figure this out. Nonetheless, the same goods (say, a used BMW) sold by original owners and by merchant dealers may very well be different products for purchasers. More specifically, purchasers may be willing to pay more to a BMW dealer for a second-hand car than to an ordinary car owner (the original owner), because the latter has no reputation concern and no expertise, and her warranty is less valuable. Therefore, the fact that purchasers did not buy the goods from original owners does not necessarily show that original owners are higher valuers.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

282

part iii: movable property

Internal auctions under 50–50 fractional ownership can be administered at low cost by court clerks. This proposed regime can ensure ex post allocative efficiency. I elaborate these points below.

A  Myerson and Satterthwaite Impossibility Theorem It is impossible for unitary ownership (full ownership going to either original owners or purchasers), which is adopted in all the surveyed 246 jurisdictions, to ensure allocative efficiency. Nobel laureate Roger Myerson and his co-author Mark Satterthwaite demonstrate that the following four requirements cannot be satisfied at the same time: (1)  The two parties do not in net lose money from the trade (“ex ante individual rationality”); (2)  Courts, states, or any institution that organize the trade or auction do not have to subsidize the two parties in bidding; (3)  Two parties both honestly reveal their true values in the trade/auction – “(Bayesian) incentive compatible” (Myerson 1979: 63; 2008)36; and (4)  Ex post efficiency (Holmström and Myerson 1983: 1805). This is sometimes called the Myerson-Satterthwaite Impossibility Theorem. This does not mean that no form of efficiency will ever be attained. Instead, a “double auction”37 with the “split-the-difference rule” is “incentive efficient” (Myerson 1979: 67; Klemperer 1999), meaning that allocative efficiency can be achieved when the buyer’s value exceeds the seller’s value by at least one quarter, on the condition that both parties know the distribution of value of the other party (Myerson and Satterthwaite 1983: 276–277). Still, as the value distribution does not always meet this requirement, ex post efficiency cannot always be attained. Moving beyond the bilateral trading context, efficiency is more likely to be attained. Rustichini et al. (1994) demonstrate that when multiple parties can participate in the transaction – like in an ordinary market – the four conditions can be achieved together. Wilson (1985) demonstrates that double auctions can achieve incentive efficiency if the numbers of buyers and sellers are sufficiently large. Nonetheless, if property law for other reasons is not inclined to involve parties other than original owners and purchasers in the allocation of entitlements, desirable results in Rustichini et al. (1994) are no condolence. 36 37

In other words, incentive compatibility guarantees that telling the truth is a Bayesian Nash equilibrium (Myerson 1979: 63). For plain explanations of double auctions to lawyers, see Ayres and Talley (1995: 1075).

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

283

B  Equal-Share Co-Ownership in Bidding Games Ensures Allocative Efficiency Importantly, Cramton et al. (1987) show that if ownership is shared by the two or more parties, and no party owns a predominant share, ex post efficiency can be ensured while meeting the Myerson and Satterthwaite requirements above.38 In particular, in a two-party setting, if each owns half, ex post efficiency can be ensured.39 If lawmakers believe, as I do, that when both parties have met the requirements to verify and prevent optimally, assigning ownership of the movables in question in full to one of them seems unfair, then giving each party a 50% share not only appears to be more fair, but it also ensures that in subsequent bilateral trading (hereinafter called an internal auction) the party who values movables more can acquire full ownership.40 More generally, the model of Cramton et al. (1987) shows that, as long as neither of the two parties owns more than a 79% share, ex post efficiency41 can be ensured.42 Once ownership shares have been assigned, the court can host a bidding game (i.e., an internal auction). In a typical auction, there is an 38 39

40

41

42

For extension of Cramton et al. (1987), see Schweizer (2006) and Preston McAfee (1992). Posner and Weyl (2017; 2018: 30–79) apply Cramton et al. (1987) in the legal context. The conclusion of Cramton et al. (1987) depends on the conditions that the distributions of value of the parties are the same, and the parties’ valuations are independent of one another (“private values”). Moldovanu (2002), Fieseler et al. (2003) and Jehiel and Pauzner (2006) show that if their valuations depend on their own private information as well as private information of others, efficiency is not always possible. In the good-faith purchase setting, original owners and purchasers should be able to come to their own valuations with just their own private information. Thus, we need not be concerned about the result based on positively interdependent valuations. Parchomovsky et al. (2007) argue for the split-the-difference rule (i.e., original owners and purchasers each receive half of the things in question) in the good-faith purchase setting and some other private law contexts. Parchomovsky et al. (2007) justify the rule on three grounds: efficiency (risk-averse persons prefer it), fairness, and doctrine integrity. Parchomovsky et al. (2007) do not use allocative efficiency and the accompanying institution costs to justify fractional ownership. In addition, the internal auction mechanism also induces more ex ante investment from purchasers. Under the current regimes anywhere, purchasers may lose the entire ownership to original owners without being compensated. Under the internal auction mechanism, purchasers can safely invest. If purchasers are the highest bidder in the internal auction, they can keep the movables and only have to pay monetary compensation to the other bidder. Even if purchasers lose in the auction, the value of the investment is reflected in their own bids and thus will be compensated. The setup in Cramton et al. (1987) is to dissolve partnership efficiently. In the real world, a partner with the lion’s share may exist. Thus, not all partnerships can be dissolved efficiently. Chang and Fennell (2014b) discuss partitions of co-owned land, where co-tenants

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

284

part iii: movable property

unquestioned owner, and multiple parties seek to win the bid to acquire ownership from owners. By contrast, in an internal auction like this one, ownership assignment is the issue itself, and in this context, there are only two parties in the bidding process. In the Cramton et al. (1987) design, before auctions start, parties have to give side payments, the amount of which depends on the shares and the lowest possible value of the movables in question. In the following double auction, the one who places the higher bid gets ownership, and this party pays the other party the difference of the two bids.43 The idea of fractional ownership, while endorsed by leading scholars (see also Rose 1997; Fennell 2005; 2009), has not really taken off. The conservative mindset of lawmakers may be part of the reason. It may also be attributed to the lack of a proper setting to start experimenting. Once fractional ownership has been proved in the real world to be useful and to facilitate efficiency, other reforms may follow. The good-faith purchase problem is a proper setting to start with, as this is a scenario in which no party is an unquestionable full owner. Fractional ownership in this context, rather than carving out shares from a full ownership and awarding them to nonowners, is created out of deadlock – there is no clear reason who should be credited as the full owner.

C  Determining and Assigning Fractional Ownership How should the fraction be determined? Put differently, how should shares of ownership be assigned between original owners and good-faith purchasers? In the two-party setting, as long as one party owns between 50% and 79%, the ownership distribution does not hinder parties’ revelation of reservation prices. This opens up an even more innovative way to design the good-faith purchase doctrine. The internal auction regime achieves allocative efficiency when parties have met the optimality requirement

43

with large shares also may exist. Chang and Fennell (2014b) thus propose an alternative mechanism instead. Brooks et al. (2010) discuss a related application in law, called Texas Shootout. See also Chapter 7. An ardent Coasean may contend that such an auction process is not necessary, since in a low-transaction-cost setting, Coasean bargaining following clear entitlement allocations reaches the same result. Chang and Lin (2021) review the theoretical literature on Coasean bargaining after litigation, and conduct the first large-scale, systematic empirical analysis of real-world post-property-litigation bargaining between parties. They find that parties do bargain over inefficient allocation of entitlements. Mechanism design scholars would contend that without the help of a double auction and other revelation mechanisms, parties will not always reveal their true value.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

285

for precaution and verification. Shares can be assigned in ways to achieve non-efficiency normative goals. For instance, if there is a huge wealth gap between the two parties (perhaps unlikely when both have paid for the same movables), the ownership distribution can be assigned progressively. There are, however, reasons to stick to the 50–50 distribution of ownership between the original owners and purchasers. First, under the bidding game design in Cramton et al. (1987), no side payment has to be paid if the two sides have the exact same shares. Calculation of the side payment depends on a certain distribution (F in their article) that the court does not always know easily. In other words, when the distribution of ownership deviates from 50–50, courts have to incur additional costs in assessing the side payments, which makes my proposed mechanism less likely to be implemented in real life. Second, Cramton et al. (1987: 624–626) also show that when neither side owns more than a 57.7% share, other forms of auctions, such as the wellknown first-price auction and second-price auction, can be used instead of the bidding game which Cramton et al. (1987: 620–621) design, and ex post efficiency is still ensured. Varying ownership shares between 50% and 57.7% creates only minimal achievement of other goals. Using the better known first- or second-price auction in court proceedings also decreases potential missteps in applying more complicated or less known auctions. Firstprice auctions are already used in courts in many jurisdictions (Chapter 7). Therefore, the administrative cost of such auctions should be fairly low. Finally, this reduces judges’ decision-making costs regarding share assignments and increases predictability, which further encourages parties to settle,44 as parties can hold internal auctions with equal shares by themselves. Therefore, a conservative version of my radical proposal would initially implement a 50–50 split of ownership.

D  Incorporation with the Doctrine The precondition to apply fractional ownership and internal auctions is that both parties have done their “efficiency duties.” That is, original owners have spent optimal prevention costs, and purchasers have spent optimal verification costs. If lawmakers adopt the market overt plus merchant dealer rule as a proxy for the optimal verification design, an internal auction with fractional ownership is required when consumers purchase

44

For empirical works on what drives settlement, see, for example, Chang and Klerman (2022) and Chang and Hubbard (2021).

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

286

part iii: movable property

on the open market from merchant dealers, and original owners are not obviously negligent. The probability of having to bid to retain ownership or losing in a bid could reduce original owners’ and consumers’ ex ante incentive to prevent and verify. Still, an internal auction will be held only if, ex post, courts find that both parties have verified and prevented optimally. Both original owners and purchasers are thus unlikely to be wrongly incentivized by the internal auction design. The fractional ownership and internal auction design can be combined with any existing rule laid out in Part I. For instance, a country with the theft rule can stipulate it when things are stolen. A country with strong protection of consumers can stipulate it when things are bought from one-shot dealers – a reason for reducing the protection of consumers. Therefore, even if the analysis in Part II does not convince ­lawmakers, they can still take the idea in this part and incorporate it in their current law. Some may question whether the fractional ownership and internal auction design would be too costly to administer. This chapter disagrees. Internal auction is simpler than, or at least as simple as, the court auctions used in many jurisdictions. Awarding co-ownership of movables to the two parties is not judicially costly. Even when the goods in question are not very valuable, the marginal cost of adopting internal auction, as compared to assigning full ownership to one party outright, is very low and may be worthwhile for the sake of allocative efficiency. What could be costly in applying this design is for courts to ascertain whether both parties have done their due diligence. It is apparently costly to do so with mathematical precision, as Part II.F has discussed. Even for courts to use a rule of thumb to ascertain whether the prevention and verification are roughly optimal is not costless, and it may very well be costlier than the regime under which courts only have to figure out whether purchasers are good-faith. Again, the fractional ownership and internal auction design can be combined with the latter regime and any other regime to increase allocative efficiency. Finally, lawmakers enacting the proposed scheme should pay attention to the design of the related contractual and tort claims as well. If original owners win a bid and compensate purchasers, original owners have a claim against embezzlers or thieves, though the latter is often long gone. If purchasers win a bid and compensate original owners, purchasers have a contractual claim against dealers, who are often identifiable and solvent.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

good-faith purchase

287

Use the contractual claim as an example, if dealers have to fully reimburse purchasers’ winning bid, purchasers have an incentive to exaggerate their economic value – purchasers win, dealers pay the bill; purchasers lose, exaggeration does not harm under first-price auction and increases the compensation under second-price auction. Hence, for efficiency’s sake, a restitution approach under which dealers refund what purchasers have paid is better than an expectation damages approach under which purchasers are likely reimbursed, because under the former, what dealers will pay purchasers is independent from how much purchasers bid. Under the latter, the “no subsidy” requirement in the Impossibility Theorem is essentially violated.

IV Conclusion This chapter argues that merchant dealers selling on the open market should be held to a lower standard of care, while other dealers should be held to a higher standard of care. Consumers who buy from the former should also be held to a lower standard of care. If courts prefer to further reduce their decision-making costs, the market overt plus merchant dealer rule may be adopted instead. That is, good-faith persons who buy from merchant dealers on the open market will be protected absolutely, but not others. Countries in Types F3b and F3c, such as Brazil, Chile, Ecuador, Indonesia, Israel, Macau, Portugal, Quebec, Timor-Leste, and Turkey have already enacted a good-faith purchase doctrine that is close to the market overt rule, though none has stipulated the doctrine exactly the way this chapter advocates. Other jurisdictions fail first and foremost because they distinguish stolen and non-stolen goods. Although the aforementioned jurisdictions come close to adopting a design that is ex ante efficient, no studied jurisdiction has implemented internal auctions among original owners and good-faith purchasers. All jurisdictions, thus, fail to achieve ex post allocative efficiency. An internal auction with fractional ownership ensures that both parties will bid with their reservation value. The structure of internal auctions avoids the oft-erratic task – required under the current paradigm – of assigning the ownership of the movables in question either completely to original owners or completely to good-faith purchasers, when both may not have behaved entirely optimally and it is unclear which party values it more. When an original owner and a purchaser are assigned equal shares, the internal auction is easy to implement, and it can be combined with any existing good-faith purchase design.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

288

part iii: movable property

While internal auctions with equal shares could achieve ex post allocative efficiency, neither the personalized optimal prevention and verification design nor the market overt plus merchant dealer rule could ensure ex ante efficiency. There is no perfect design, and policymakers in the 246 jurisdictions that the author has painstakingly studied have to make trade-offs. Nonetheless, the analysis in this chapter suggests that if policymakers care mostly about efficiency, there are improvements to be made. In a world with 21 variants of the good-faith purchase doctrine, championing one solution that fits all will inevitably invite push-back. My proposed framework, however, should be general enough to fit various countries, even though their property crime rates, reliability of movables registries, information costs in markets, etc. differ. First, to the extent that variances in these background conditions do not lead us to believe that original owners or purchasers are always the higher valuer, a mechanism that can help tease out higher valuers is still needed. Fractional ownership and internal auction should be universally applicable to do just that.45 Second, optimality is context-dependent. When property crime rates are higher, both original owners and purchasers have to spend more to be counted as fulfilling the optimality requirement. When movables like cars are subject to reliable registration, purchasers’ checking the car registry would suffice for the optimal verification requirement. Therefore, the one framework proposed here would, in practice, lead to different equilibria across countries. 45

Still, admittedly, if there are huge wealth gaps between the two parties, and liquidity is a problem, auctions may not reveal higher valuers. But again, an alternative rule that, say, always favors the poorer party or original owner is not a more allocatively efficient rule.

https://doi.org/10.1017/9781009236553.014 Published online by Cambridge University Press

11 Finders, Keepers A Minority Rule Finders, keepers; losers, weepers.

Despite being one of the most popular legal maxims, “finders, ­keepers” is the rule in a minority jurisdiction. That said, in English-speaking ­common-law jurisdictions, finders do become keepers, as they are not required to submit found things to the state. Losers, however, are not actually weepers, because losers qua owners can recover lost items from finders within the statute of limitations. Thus, “finders, keepers; losers, weepers” maybe both the most popular, and misleading, legal maxim (Heller and Salzman 2021: 50). This chapter summarizes two finder doctrines: one regards finders of ordinary lost things, and the other concerns finders of troves of buried treasure. After the comparative overview, this chapter reflects on the merits of various legal schemes.

I  Comparative Overview This chapter concerns only lost things, not things mislaid.1 In addition, it discusses mandatory rewards for finders, not those voluntary promises of prizes owners announce in an effort to recover lost goods. Promises of prizes are enforced in most, if not all, jurisdictions – see, for example, China §317. The mandatory rewards discussed here must exceed mere reimbursement of necessary expenses incurred in the finding. Section A gives an overview of the finder doctrine. The movable things found are lost by the original owners and later recovered by finders, most likely by chance. Most jurisdictions stipulate that a certain period of time after 1

Merrill and Smith (2010: 39) explain that mislaid things are deliberately placed somewhere by their owners and then forgotten. The analysis of this chapter may or may not apply to mislaid things and I did not code rules regarding mislaid things. This topic will be left for another day.

289

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

290

part iii: movable property

Finders of ordinary lost things Against the world except owners May acquire ownership Never acquire ownership Not coded or no found rule

Figure 11.1  Typology of the finder law

a person reports finding an item to the relevant authority (often the police department), ownership is forfeited to either the finders or the state (including local municipalities). Many jurisdictions mandate that the original owners offer a reward to the finder – usually a certain fraction of the value of the item found.2 Section B graphically presents the required fraction of value, and the length of time. Section C summarizes the doctrine regarding the finding of treasure troves. A number of countries enacted additional rules regarding cultural artifacts, which in general belong to the state. Such additional rules are not discussed in this chapter. Treasure troves differ from the ordinary found things as they are concealed in immovable or movable things. The prime example for the rules coded and presented in this chapter regards valuable movables, such as a box of gold coins, buried underground. Some countries have additional rules for shipwreck (e.g., Iran §178) or animals (e.g., Vietnam §§231–233). These are not covered in this chapter.

A  Finders of Lost Things Figure 11.1 provides an overview of the finder law around the world. The following subsections further differentiate the rules within each group.

2

The chapter also does not deal with the issue of who the finder in law is when, for instance, a person finds a lost thing in another’s apartment or in a public space. Briefly, on the one hand, if a store rather than a patron is considered the finder, the store has reputational concern and is more likely to keep lost things for the original owners. On the other hand, if patrons are not considered the finder, they will not receive any reward, increasing their incentives to misappropriate lost things.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

291

1  Finders Never Acquire Ownership Type A1. Original owners always retain ownership and do not have a duty to offer a reward. India (ICA) §§168–169,3 Pakistan,4 and North Korea §62 are in this group. In India (LA) §68 and Pakistan (LA) §48, however, owners’ recovery claims are subject to a three-year limitation period. Type A2. Finders are entitled to neither ownership nor rewards; found things either return to original owners or go to the state. In China §§314–318, finders must notify owners or deliver lost things to the police. After one year, the lost things belong to the state. In Cuba §193, finders must deliver lost things to the police. Owners can reclaim them within three months; if they do not, the state should put the found thing to its best use from a socioeconomic standpoint. Type A3. Finders have claims for rewards against owners (if they announce themselves) or (if not) the state which acquires ownership of found things. Bulgaria (OA) §§88–89 and Slovakia §135 adopt this type. Type A4. If owners appear, finders have claims for rewards against owners. If not, the state auctions off the found things and shares the proceeds with finders: Argentina §1956, Bolivia §144 (25% of the net proceeds), Brazil §§1233–1237 (judges decide the amount; ≥5%), Chile §§629–633 (50% of the net proceeds), Ecuador §645 (50% of the net proceeds), Norway (LPA) §§6–9 (Faerstad and Lilja 2010: 281) (33% of the net proceeds); Denmark (LPA) §§3–7 (Faerstad and Lilja 2010: 282) (33% of the net proceeds), Honduras §§683–686 (25% of the net proceeds), El Salvador §§610–613 (25% of the net proceeds), Paraguay §§2035–2038 (judges decide the amount), Peru §§932–933 (50% of the net proceeds), and Uruguay §726 (50% of the net proceeds). Type A5. Original owners have no duty to reward finders. If the original owners do not show up, the state auctions off the found things and shares the proceeds with the finders. Only Mexico §§777–781 (25% of the net proceeds) belongs to this type. Type A6. Original owners have duties to reward finders. If original owners do not show up, the things found belong to the state, which does not appear to have reward duties. Colombia §708, Guatemala §§596–598, and Panama §§361–362 fall in this type. Type A7. Original owners have duties to reward finders. If original owners do not show up, the law is silent. Ethiopia §§1157–1158 is the sole example. 3 4

Finders can sell the found things only under limited circumstances. www.islam21c.com/islamic-law/the-fiqh-of-finders-keepers/

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

292

part iii: movable property

2  Finders May Acquire Ownership Type B1. Finders acquire ownership after some time, if original owners do not show up. Original owners appearing in time do not have to reward finders. California §2080.3, Finland (Hollo 2019: 158), Iran §163 (more complicated than other B1 jurisdictions), Israel (RLPL) §4, Louisiana §3419, Monaco §601, New York (PPL) §257, Slovenia (Škerl and Vlahek 2020: 199), South Africa (Van der Merwe et al. 2002: 196),5 South Korea §253, and Sweden (Lilja 2010: 145) belong to this group. Type B2. Finders acquire ownership after some time, if owners do not show up. Finders have claims for rewards against original owners who show up before the prescription period runs. Albania §183, Armenia §§180–182, Austria §395, Azerbaijan §186, Belarus §§228–230, Cambodia §191, Costa Rica §§501–502, Croatia (AOORR) §137, Czech §§1056–1057, Equatorial Guinea §§615–616, Eritrea §§1002–1003, Estonia (TLPA) §§100–101, Georgia §191, Germany §§971–973, Greece §§1086–1088, Hungary §§V:62–V:63, Italy §§929–930, Japan §240, Kazakhstan §§245–246, Kyrgyzstan §§259–260, Laos (PL) §38, Latvia §§948–949, Liechtenstein (SR) §191, Lithuania §4.62, Macau §1247, Macedonia (AOORR) §§138– 139, Malta §564, Moldova §§325–326, Mongolia §116, the Netherlands §§V:6–V:10, Nicaragua §§698–699, Philippines §§719–720, Poland §§186– 187, Portugal §1323, Puerto Rico §§555–556, Quebec §§942–946 (owners must give finders the cost of administration and “where applicable, the value of the work done”), Romania §§942–945, Russia §§228–229, Spain §§615–616, Switzerland §722, Taiwan §§805–807,6 Tajikistan §§252–253, Thailand §§1324–1325, Timor-Leste §1243, Turkmenistan §214, Ukraine §§338–339, Uzbekistan §§193–194, and Venezuela §§803–804. Type B3. After some time, finders may acquire ownership or receive rewards from the state. One example is Vietnam §230: A year after the relevant authority has issued a public notice regarding the lost thing, either finders acquire ownership of the lost item (when its value is worth ten months of statutory minimum wages), or finders are entitled to compensation that equals ten months of statutory minimum wages, plus 50% of the value beyond the preceding threshold. If original owners make a claim, the law is silent – presumably, finders will not receive a reward. The other example is Scotland (CGSA) §§68–70: After two months, the chief constable may give the property to a finder or pay her a reward. 5 6

The prescription period is 30 years (Lee 1953: 131). Chang (2016a: 226) observes that the most frequently amended stipulations in the Book of Things in the Taiwan Civil Code are those regarding the finder-keeper rule.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

293

Owners who show up in time must pay finders a reward, in the amount determined by the chief constable.

3  Finders Have Property Rights against all the World but Owner Type B4. This is the common-law rule: finders have property rights against everyone else in the world but the owner. Functionally, this rule is equivalent to Type B1. Like those in Type B1, the common law jurisdictions do not prescribe rewards. In addition, if owners do not show up in time, finders become de jure owners. In common-law jurisdictions, owners’ claims against finders could be time-barred. Australia (Hepburn 2002: 174) (6 years; LAA §5), Canada (Ziff 2014: 153) (2 years; LA §5), England and Wales (Sparkes 2019: 165–167) (6 years; LA §2), Hong Kong (Mau 2010: 5) (6 years; LO §4), Ireland (cf. Woods 2011: 120) (6 years; SL §11), Malaysia (Maidin and Ali 2014: 148) (6 years; LA §6), New Zealand (Williams 2011: 109) (6 years; SL § 11), Nigeria (Smith 2018: 126) (6 years; LLLS §8), Singapore (See et al. 2018: 298) (6 years; LA §6), and Uganda (Bakibinga 2018: 83) (6 years; LA §3) have this rule. Cyprus (Synodinou 2020: 153) also belongs to this group. 4  No Tailor-Made Rule Type Z1. France §717 and the former colonies of France opt to enact the finder doctrine in other statutes.7 Notably, France §2276 (in the old code, §2279) and its former colonies adopt the famous doctrine of “possession equivalent to ownership certificate” (see Chapter 10). The same provision carves out an exception for lost and stolen things. This doctrine thus becomes the foundation for issues regarding finders-keepers, acquisitive prescription of movable ownership, good-faith purchase, and first possession. Among jurisdictions that do not have the finder doctrine, countries in the former Ottoman Empire, which have been heavily influenced by the French Civil Code, are notable for not enacting such a doctrine in their civil codes.8 The omission might be because the French Civil Code has no clear rule. Notably, the French Civil Code, as described below, does include the treasure trove doctrine, as do many Middle East countries. Type Z2. Without a finder doctrine, but with a doctrine on acquisitive prescription of movable things, finders who keep found things to themselves 7 8

In the case of France, other statutes only deal with finding in specific contexts (Terré and Philippe 2006: para. 424). United Arab Emirates §1207 refers to special statutes that govern lost properties.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

294

part iii: movable property

but fulfill the possession requirement could nonetheless acquire ownership after some time. A number of countries for which I cannot locate a finder doctrine have such an acquisitive prescription doctrine for even bad-faith possessors qua finders. These include Afghanistan §2279, Algeria §827, Egypt §968, Jordan §1181, Kuwait §935, Libya §972, Qatar §966, Syria §917, United Arab Emirates §1317, and Zimbabwe (PA) §4. It is worth emphasizing that even in jurisdictions with a finder doctrine, as long as the bad-faith acquisitive prescription of movables is recognized, finders may still be able acquire ownership through acquisitive prescription. These jurisdictions include, for example, Albania §168, Bolivia §149, Brazil §1260, and South Korea §246.

B  Reward Fraction and Waiting Period This section shows the distributions of reward by fractional value and waiting period in the finder law. I define reward fraction as the percentage of the monetary value of the found item that must be paid by the owner to the finder under a mandatory rule. I define waiting period as the number of days after which an owner loses ownership rights over their lost item. Figure 11.2 shows the reward fractions and waiting period in the 100 studied jurisdictions. The linear fitted line shown in Figure 11.2 only considers specific fractions (at, or below, 50%) and specific waiting periods (at, or below, 2 years). Table 11.1 shows that a number of countries adopt a fixed fraction, some impose a cap, while some allows flexible rewards. Fifteen studied jurisdictions exempt finders’ reporting duties if the value of the found things is below a certain threshold. The threshold ranges from 0.12 USD to 250 USD. Brazil §1237 Sole Paragraph does not exempt reporting duties but a municipality may abandon low-value things in favor of finders. In Norway, the police may return movables of little value to finders (Faerstad and Lilja 2010: 281).

C  Finders of Treasure Trove Many jurisdictions explicitly limit the application of the treasure trove doctrine9 things found “by chance” (e.g., Chile §626, Iran §173, Libya §876, Spain §351, Puerto Rico §554, and Venezuela §800) and to finders who do not trespass on others’ land (e.g., Chile §626, Moldova §327, 9

Izuel (1992: 1665) summarizes the definition of treasure trove in American case law.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

295

Table 11.1  Fraction type Types

Frequency

Minimal Fixed Fixed but regressive Flexible Maximal Maximal but regressive N/A

2 14 15 14 23 3 30

Total

101

50% 33% None 10% 20%

Reward fraction

Certain %

Notes: “but regressive” means that the law prescribes a fixed or maximal baseline reward fraction, but if the value of the found things surpassed a certain threshold, the reward fraction reduces for the value beyond the threshold.

3 months

6 months

1 year

2 years

Certain days

Waiting period (in log10 of days) % of value (graphing)

Fitted values

Figure 11.2  Reward fraction and waiting period Notes: N = 100. Jitter effects apply. Jurisdictions that give certain unspecified rewards are placed at the top of the Y-axis, whereas those that extinguish original ownership after certain unspecified days are placed on the right of the X-axis — these jurisdictions are not considered in the fitted line.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

296

part iii: movable property

Finders of treasure troves Locus owners acquire ownership Finders & locus owners share ownership May acquire ownership Never acquire ownership Not coded or no found rule

Figure 11.3  Typology of the treasure trove law

New York (PPL) §256, and Russia §233). This section summarizes laws that apply when these conditions are met. If the rule regarding finders of treasure troves is the same as that regarding finders of ordinary things, I use the same labels (Type A1 and Type A2, for instance). In many jurisdictions, the treasure trove doctrine is silent about the original owners’ rights, or implies that original owners may claim the treasure but that they are unlikely to be identified. Hence, while some jurisdictions request finders to report the finding to a government agency, the reporting duties are not as prevalent as those for lost things. Moreover, the government agency’s task is not to locate the original owners, but to assess whether the found treasure is of sufficient cultural value to be housed in a museum. In general, this doctrine focuses on the distribution of the treasure among several parties: finders, locus owners (beneath whose land was the treasure found), and the state. Figure 11.3 provides an overview of the treasure trove law.

1  Finders Never Acquire Ownership Type A0. Finders get neither ownership nor a reward; found treasures belong to locus owners or the state (if public land). Iraq §1101 is the sole example.10 Type A2. Finders get neither ownership nor rewards; found treasures either return to their original owners or belong to the state. China §319 is the only example. 10

A majority of American states adopt the rule that treasure troves belong to locus owners (Heller and Salzman 2021: 128). In Corliss v. Wenner, rendered by Court of Appeals of Idaho (2001), the finders are employed by the locus owner, so finders do not receive rewards.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

297

Type A3. Owners who show up must reward finders; if treasures are unclaimed, finders receive rewards from the state. Slovakia §135, giving a 10% reward, is the sole example. Type A8. Found treasures belong to the state – for some countries in this group, a condition “if original owners cannot be found” may have to be satisfied – and the state gives rewards to finders. In the case of England and Wales, Treasure Act 1996 prescribes that treasure belongs to the Crown (Sparkes 2019: 167), which can give rewards (§10), no matter whether original owners can be identified. Bulgaria (OA) §91 (25%), Cuba §195 (25%), Hungary §V:64 (reasonable rewards), Laos (PL) §41 (15%), Poland §189 (reasonable rewards), and Thailand §1328 (33%), on the other hand, leave room for original owners to claim the treasure. When original owners do so, the law is unclear as to whether original owners must reward finders. Hence, Type A8 differs from A3 because A8 does not explicitly prescribe original owners a duty to reward finders. Type A9. Finders cannot acquire ownership, which always goes to locus owners, who give rewards. Albania §186, Czech §1064, Eritrea §1004, Estonia (TLPA) §104, Ethiopia §1159, Switzerland §723, and Turkey §772 belong to this type. These countries all give rewards equivalent to 50% of the value of the treasure. Type A10. Finders cannot acquire ownership, which always go to the state, which gives rewards to finders and locus owners. Croatia (AOORR) §140 (5% to finders and 5% to landowners), Macedonia (AOORR) §143 (7.5% to finders and 7.5% to landowners), Malaysia (Maidin and Ali 2014: 126–127) (TTA) §§29–30 (amount being what the state thinks fit), and Tajikistan §257 (finders and locus owners sharing 50% of the value) constitute this type. Type A11. Finders cannot acquire ownership, which always goes to the state, which does not give rewards. Australia (Mann and Blunden 2010),11 California,12 Ireland (de Londras 2011: 10–11), Israel,13 and Maldives (LA) §38 are examples. 11 12

13

Compare Liu (2014)’s comment on an Australian case regarding a briefcase found in an apartment by the police during criminal investigation. This is not a typical treasure trove. California Civil Code does not contain a provision on treasure troves, and there appear to be no leading cases on this issue. 61 A.L.R. 4th 1180 (2021) does not contain any Californian case. Izuel (1992: 1669–1670) observes several different approaches adopted by the American states, but does not cite any California case or statute. California is classified as this type, because California (PRC) §6313 stipulates that “abandoned shipwrecks, archeological sites, and historic resources in submerged lands vested in state.” The law of Israel regarding treasure trove is quite complicated. Even leaving aside the law related to cultural antiquity, finders of ordinary treasure trove would still not acquire ownership under any scenario.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

298

part iii: movable property

2  Finders May Acquire Ownership Type B2. Finders acquire ownership after some time, if owners do not make a claim. Original owners who announce themselves before the prescription period expires must reward finders. Liechtenstein (SR) §192, which gives reasonable rewards, is the sole example. Type B3. After some time, finders may obtain ownership or rewards from the state. In Scotland, a finder of treasure troves sometimes acquires ownership, and sometimes receives rewards from the state, depending on whether a museum, through the state, “claims” the found treasure trove.14 Vietnam §229 follows the same rule. 3  Finders and Locus Owners Share Ownership Type C1, the 50:50 rule: Finders and locus owners are co-owners, each receiving a 50% share. Type C1 is similar to Type A9, which gives full ownership to locus owners, but finders are entitled to a reward worth 50% of the value of the treasure. Included in this type are: Argentina §1953, Armenia §186, Austria §399, Azerbaijan §187, Belarus §234, Belgium §716, Bolivia §146, Brazil §1264, Burkina Faso §716, Cambodia §192, Chile §626, Colombia §701, Comoros §716, Costa Rica §498, Dominican Republic §716, Ecuador §642, El Salvador §607, Equatorial Guinea §351, France §716, Georgia §192, Germany §984, Greece §1093, Guatemala §§594–595, Haiti §576, Honduras §680, India (ITTA) §10, Indonesia §587, Italy §932, Ivory Coast §716, Japan §241, Kazakhstan §247, Kyrgyzstan §261, Latvia §953, Libya §876, Louisiana §3420, Luxembourg §716, Macau §1248, Malta §563, Mauritius §716, Mexico §877, Moldova §327, Monaco §600, Mongolia §118.1, the Netherlands §V:13, Nicaragua §619, Niger §716, Norway (LPA) §1, Pakistan (ITTA) §10, Panama §340, Paraguay §2040, Peru §935, Philippines §438, Portugal §1324, Puerto Rico §285 & §554, Quebec §938, Romania §946, Russia §233, Seychelles §716, Slovenia (Škerl and Vlahek 2020: 199–200), South Africa (Van der Merwe et al. 2002: 198), South Korea §254, Spain §351, Suriname §642, Taiwan §808, TimorLeste §1244, Togo §716, Turkmenistan §215, Ukraine §343, Uruguay §721, Uzbekistan §196, and Venezuela §800. Type C2, the 25:75 rule: Finders and locus owners are co-owners, receiving 25% and 75% shares, respectively. Lithuania §4.65 is the only example. 14

See Treasure Trove in Scotland: A Code of Practice, https://treasuretrovescotland.co.uk/ sites/default/assets/File/tt-code-jan-2016.pdf.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

299

Type C3, the 20:80 rule: Finders and locus owners are co-owners, receiving 20% and 80% shares, respectively. Bahrain §846, Kuwait §877, and Qatar §906 are in this group. Type C4, the 20:60:20 rule, under which finders get 20%, locus owners get 60%, and the state gets 20% of the value of the treasure. Syria §830 and Tunisia (CDR) §25 belong to this type.

4  Locus Owners Acquire Ownership Type D1. Things below the surface of land belong to locus owner. Included in D1 are Canada (Ziff 2014: 154), Hong Kong (Mau 2010: 5), New York (PPL) §251 (presuming treasure trove to be lost property), New Zealand (HNZPT 2019: 10), and Singapore (Lee 2004: 264–265). Type D2. Finders receive neither ownership nor rewards. Locus owners own the treasure trove found underground, but they must pay 20% of the value to the state. Jordan §1078 and United Arab Emirates §1205 fall into this type. Type D3. A treasure trove found on someone else’s land belongs to locus owners. A treasure trove found on unclaimed land, however, belongs to finders. Iran §§174–176. Note that Type D3 is the equivalent to Type D1 when there is no more unclaimed land or when treasure troves are not located beneath such land. II  Economic Analysis Sections A and B engage with existing economic theories regarding finders but argue that finders’ searching costs (generally zero) need not be considered and original owners’ prevention costs are unlikely to be affected by the finder law. Instead, Sections A and B advance a simple economic theory of my own and propose a “schedule approach” in designing rewards. Section C finds that cultural variations are not correlated with differences in reward conditions. Section D argues that the state should get involved when it comes to treasure troves, but leave most valuations (by, e.g., giving rewards) to finders and locus owners.

A  Rewards Increase Incentives to Turn in Found Movables The foundation of my analysis in Sections A and B is that original owners generally value lost things more than do finders, and thus finders’ turning in found things to the police is the best outcome. The police station serves as the focal point for original owners to turn to for assistance. The key

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

300

part iii: movable property

economic problem is how to induce finders to neither misappropriate nor to leave things alone. Shavell (2004: 40–45) distinguishes between two situations: (1) when original owners have had a good opportunity to recover; (2) original owners have not had good opportunity to recover. In (1), Shavell argues that the original owner rule (i.e., original owners keep ownership) is more efficient, as otherwise original owners would spend excessively in prevention. In (2), Shavell argues that the “finders-keepers rule”15 is superior to the “original owner rule,” because there is a greater recovery of lost property. An even more superior rule is the combination of “original owner rule” and a mandatory reward paid by original owners to finders because it provides incentives for finders to help return lost things and mitigates the problem of excessive loss prevention of owners (see also Posner 2011: 103). In addition, awarding ownership to finders when original owners do not show up provides finders with an incentive to search for lost things. No jurisdictions’ finder laws distinguish these two situations. When original owners can easily recover lost items, a different doctrine regarding mislaid things may apply, though. The original owner rule for easily recoverable things would work only if there were sufficiently strong social norms (not discussed here) or a legal norm that induces finders not to misappropriate found things. The “stick” in this context is often criminal penalties. Civil codes in some studied jurisdictions refer to criminal penalties for finders who do not follow the prescribed procedure (e.g., Guatemala §598). Some jurisdictions (e.g., Taiwan) do not contain such references but do have such criminal penalties in place. To the extent that the deterrent effect of criminal penalties is sufficiently strong, the original owner rule incentivizes finders to leave the thing alone or to report their findings to the police, so that original owners can quickly recover their lost property. Nonetheless, the expected criminal punishment maybe insufficiently strong if, for example, the probability of being caught and prosecuted is very low. Finders, realizing that turning in found things brings no personal benefit, but is instead a hassle,16 may decide instead to 15 16

Shavell (2004: 34) characterizes the finders-keepers rule as “under which the finder is deemed to be the owner of anything that he has found.” When I was 13, I found a lost wallet and sent it to a police station. I ended up spending three hours filling out paperwork. Also, a finder may get into trouble. Imagine a finder who turns in a wallet containing 5,000 dollars to the police, and later the original owner shows up claiming that there were 10,000 dollars in the wallet. When the lost thing is an antique, original owners may blame finders for damaging it. In a society lacking social trust, finders may decide to leave found things alone in order not to get into trouble. The theoretical take-away point is that while finders may generally have zero searching costs, they have positive reporting costs.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

301

keep them. Some of these opportunists, however, may turn in the found things if they can receive a reward (the “carrot”). Therefore, the original owner rule – that is, no reward for finders – is not always efficient because lost things are kept by finders. Under (2), notably, only three studied jurisdictions adopt a pure original owner rule and none adopts a pure finders-keepers rule. By contrast, in 54 jurisdictions, finders are entitled to rewards and have the chance of becoming new owners after some time (Types B2 and B3). Shavell (2004)’s economic analysis is based on the premise that finders make investments in searching, and the level of investment is partly driven by the prospect of getting ownership or rewards. I suspect that lawmakers did not focus on searching investment. I also doubt that many people, without a promise of prizes, are out and about searching for ordinary lost things. Moreover, in the treasure trove context at least, many jurisdictions demand the opposite – finders must encounter a treasure trove by chance. An intentional effort (such as using metal detectors) is prohibited. When the lost things are everyday items, what lawmakers had in mind also appears to be a finder’s inadvertently bumping into a lost thing, without making any findingspecific investment, though the by-chance requirement is rare. My simple theory, instead, is that the mandatory reward and the prospect of gaining ownership is to increase finders’ (opportunity) cost of misappropriation or doing nothing; in other words, rewards give finders incentives to turn in found things. The higher the reward fraction and the shorter the waiting period before gaining ownership, the stronger the incentives for finders to turn in found property to the police because the expected payoff is higher. Shavell (2004)’s analysis may better apply to scenarios where a person loses, say, her wedding ring or pet, and promises for a prize to a finder. Motivated by the prize, finders embark on the detective work. However, as Shavell (2004: 36) points out, when there are multiple searchers, the incentives to search could be socially excessive (see also Merrill 2009).

B  Schedule Approach to Reward Design The finder doctrine should be designed with only the incentives of finders in mind. Shavell (2004: 43) points out that “the higher the reward fraction, the greater the incentive for recovery… but the more excessive is the incentive of original owners to prevent loss.” Nonetheless, as Chapter 10 argues, original owners are unlikely to adjust their investment in precautions according to the reward fraction and the waiting period. Owners’ incentives are to economically reduce the probability of their things being

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

302

part iii: movable property

exposed to theft, robbery, loss, unwanted improvement or some combination thereof. Any single, small-scale adjustment in one movable property doctrine is unlikely to change owners’ behaviors. By contrast, the choice of finders with legal knowledge between turning the property in to the police and misappropriate it depends on the reward fraction and waiting period. Holding the deterrence effect of the criminal punishment constant, the slimmer the chance of receiving rewards or acquiring ownership of found things, or the longer the waiting period, the more likely that finders will misappropriate the found things. Therefore, finders’ incentives are the key. If my theory of increasing finders’ opportunity cost of misappropriation and idling were also the main concern of lawmakers, one would expect laws to provide generous rewards and short waiting periods. Shown in Figure 11.2, 10% and one year are the most popular choices, as 17 jurisdictions pick these two together. Admittedly, on its face, 10% does not appear to be particularly attractive, and one year not particularly short. Either lawmakers believe that this reward fraction is sufficiently generous, or they have been unconcerned with finders’ incentives. A challenge to my theory is: if original owners would not spend excessively on loss prevention, even under high reward fractions and short waiting periods, why not stipulate the most generous possible reward conditions, such as 95% of the value, going to finders? This would maximize the opportunity cost of misappropriation and idling. In terms of cash, for instance, surely losers of cash would prefer to get 5% or 10% back than getting nothing back. In addition, whether original owners or finders get cash is a matter of distribution, not efficiency. That said, a legal rule that conveys the social meaning that ones who lose cash only “deserve” to get 5% back is inconsistent with the sense of fairness in most societies. For other kinds of lost movables, under the assumption of original owners’ being the higher valuer and not suffering from the sunk cost fallacy, between finders’ misappropriation and original owners’ recovery and rewarding finders, a high reward fraction is efficient, as it facilitates recovery. Original owners can always opt not to claim found things, leaving them to finders. (The two parties may even negotiate to reduce the reward fraction.) Again, the society-wide sense of fairness constitutes a constraint on any efficiency-oriented-only design. This explains why, as shown in Figure 11.2, except for one 50% country, no jurisdiction adopts a >33% reward fraction. The fairness concern may also justify not adopting an otherwise efficient scheme. The next question is, under the aforementioned constraint, what type of reward fraction design shown in Table 11.1 is better? Shavell (2004: 43) argues that “[i]f the reward depends on the optimal recovery effort, it can

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

303

be shown to lead to optimal behavior for both [original owners and finders], and likewise if the reward is paid by the state to the finder.” That is, Shavell proposes a flexible reward fraction determined by ex post judicial examination of whether the recovery effort is optimal. Indeed, it makes more sense to reward finders more when finders’ cooperation (i.e., not misappropriating) is critical and/or costly. But consider this scenario: if the lost things are found in, say, a supermarket, any given finder’s time and effort is not important, as the lost things will sooner or later be found by the owners or other patrons. Fully informed finders who expect to get a low reward (as their efforts are tiny) are inclined to misappropriate. My proposal of a schedule approach is to set rewards based on thing types, which approximate finders’ opportunities to misappropriate. If the lost things are cash, the reward fraction should be highest, to increase finders’ opportunity cost of misappropriation.17 If the lost things are goods that finders can enjoy if misappropriated, the reward fraction should be high. If the lost things are, for example, bank checks which are valuable to losers but valueless to finders (as finders cannot cash in the checks), the reward fraction should be low. Alternatively, a fixed reward (say, 100 USD) can be used. If the lost things are, for example, passports which have no market value and are a big hassle to lose control of, a fixed reward should be used. These examples are illustrative, not exhaustive. Whether, for example, credit cards and smart phones are more like cash or checks – or a stand-alone category – depends on the existing technology. A point worth noting is that many countries spend a lot of ink detailing the rules for found things that are perishable, but none of them set different rewards based on thing types.

C  Cultural Variation as Explanation for Variation? Figure 11.2 shows substantial variations in reward fractions and waiting periods across studied jurisdictions. One plausible explanation for this variation may be lawmakers’ responses to different levels of local opportunistic behaviors. Countries where ordinary people are more likely to misappropriate found things are more likely to give higher rewards and/or impose shorter waiting periods. More specifically, I propose two indices to measure the amount of expected payoffs for finders. The first index is simply the reward fraction. The higher the fraction, the higher the expected payoff. The second index is the product of the reward fraction 17

In Israel, for instance, waiting period for found cash is two months whereas that for other found goods are four months. I thank Roy Shapira for this example.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

part iii: movable property

304 Power Distance

Masculinity

0

20

40

60

80

100

30 20 10 0

0

0

10

10

20

20

30

30

Individualism

0

Pearson Correlation Coefficient=-0.164

20

40

60

80

100

30 20

20

10

10

0

0 100

100

Indulgence Versus Restraint

30

30 20 10 0

50

0

Pearson Correlation Coefficient=0.172

50

Pearson Correlation Coefficient=-0.354

Longterm Orientation

Uncertainty Avoidance

0

0

Pearson Correlation Coefficient=0.168

20

40

60

80

100

0

Pearson Correlation Coefficient=0.087

20

40

60

80

100

Pearson Correlation Coefficient=-0.009

Figure 11.4  Reward % by Hofstede data

Harmony

Hierarchy

3.5

4

4.5

0 10 20 30

0 10 20 30

0 10 20 30

Embedded

3

Pearson Correlation Coefficient=0.070

3.5

4

4.5

Mastery

4

4.2

4.4

3

3.5

4

4.5

0 10 20 30

4.4

4.6

4.8

5

3

3.5

0 10 20 30 2.5

Pearson Correlation Coefficient=-0.247

Egalitarianism

4.2

2.5

Intellectual Autonomy

0 10 20 30 3.8

Pearson Correlation Coefficient=-0.325

2

Pearson Correlation Coefficient=-0.178

Affective Autonomy

0 10 20 30 3.6

1.5

Pearson Correlation Coefficient=0.153

5.2

Pearson Correlation Coefficient=-0.106

Figure 11.5  Reward % by Schwartz data

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

3.5

4

4.5

5

5.5

Pearson Correlation Coefficient=-0.009

finders, keepers: a minority rule Power Distance

Masculinity

20

40

60

80

100

140 105 70 0

0

35

35

35

70

70

105

105

140

140

Individualism

0 0

0

Pearson Correlation Coefficient=-0.107

20

40

60

80

100

100

Indulgence Versus Restraint

0

0

35

35

70

70

105

105

140

140

140 105 70 35

100

0

Pearson Correlation Coefficient=0.026

50

Pearson Correlation Coefficient=-0.304

Longterm Orientation

0

50

0

Pearson Correlation Coefficient=0.011

Uncertainty Avoidance

0

305

20

40

60

80

100

0

Pearson Correlation Coefficient=-0.104

20

40

60

80

100

Pearson Correlation Coefficient=0.076

Figure 11.6  Expected payoff by Hofstede data

Harmony

Hierarchy

3.5

4

4.5

0 35 70 105140

0 35 70 105140

0 35 70 105140

Embedded

3

Pearson Correlation Coefficient=-0.191

3.5

4

4.5

Mastery

4

4.2

4.4

3

3.5

4

4.5

0 35 70 105140

4.4

4.6

4.8

5

3

3.5

0 35 70 105140

2.5

Pearson Correlation Coefficient=-0.299

Egalitarianism

4.2

2.5

Intellectual Autonomy

0 35 70 105140

3.8

Pearson Correlation Coefficient=-0.035

2

Pearson Correlation Coefficient=0.012

Affective Autonomy

0 35 70 105140

3.6

1.5

Pearson Correlation Coefficient=0.186

5.2

Pearson Correlation Coefficient=0.066

Figure 11.7  Expected payoff by Schwartz data

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

3.5

4

4.5

5

5.5

Pearson Correlation Coefficient=-0.090

306

part iii: movable property

and the inverse of waiting period (in years). For instance, a country with a 20% reward fraction and a three-month waiting period would have an index value of 20 × (1/0.25) = 80. The higher the index value, the higher the expected payoff for turning in found things. The question arises: are there data that consistently capture the tendency to misappropriate found things across countries? Schwartz (1999) and Hofstede (1991; 2001) each coded several dimensions of cultures across dozens of countries, and their data have been used in legal studies (e.g., Licht et al. 2005). As Figures 11.4–11.7 show, none of the 13 cultural dimensions has a strong correlation with either finder index I created. The highest correlation coefficient in terms of absolute value is about −0.35 with masculinity, suggesting that lawmakers in a more masculine society are more prone to provide lower expected rewards. This result is not only weak in a statistical sense but also counter-intuitive. The lack of strong results suggests that (1) perhaps none of the 13 cultural dimensions in the two datasets alone captures the opportunism I focus on; (2) even if all lawmakers in the world intend to set reward conditions in response to local conditions, they may reasonably (for informational constraint) set different conditions to the same level of opportunism; (3) perhaps my theory is wrong, as lawmakers do not care about reducing incentives for misappropriation; or (4) perhaps the expected payoff is only one important factor, and the statistical significance of which would only be revealed when controlling for other relevant factors, such as the availability and punishment level of criminal sanctions. This is a topic worth exploring in future studies.

D  Should Treasure Trove Law Be Different? To put the treasure trove doctrine in context, note that the first possession doctrine (i.e., unowned movables belong to a person who exerts actual control first) is explicitly recognized in 101 jurisdictions (65%).18 Given that cultural artifacts are excluded from the treasure trove doctrine, and given that original owners are hard to identify, why should the law not simply give finders of treasure troves the status of first possessors? Put 18

In addition, Mongolia §104.1 and Uzbekistan §191 instead apply the acquisitive prescription doctrine to unowned (often abandoned) movables. Mexico §§774–775 and Slovakia §135 apply the finder doctrine to unowned movables. In Australia (Moore et al. 2016: 167–168), Cyprus (Shaelou et al. 2010: 573), and New Zealand (Williams 2011: 129), first possessors only have relative title. Iran §28 prescribes that unowned things should be used for the need of the poor. In Francophone countries, possession is certificate of ownership.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

finders, keepers: a minority rule

307

differently, why does allocating treasure troves to locus owners or the state increase economic efficiency? Perhaps it does not matter, at least as a first-order matter. Once the treasure has been uncovered, who is entitled to it is a matter of distribution, not efficiency. The doctrine applies only when finders encounter treasure troves by chance. In all jurisdictions, finders who search for treasures intentionally will not be entitled to the treasure. From an ex ante standpoint, this rule on the one hand discourages searching behaviors19 and on the other hand induces intentional searchers to keep the secret to themselves. Providing finders with nothing but a “thank you” creates similar dynamics even if finders encounter treasure trove by chance but are aware of the legal rule. That is, finders may decide to keep it secret. Especially when locus owners also receive nothing (e.g., Types A0, A2, and A8), locus owners, who are likely to become aware of the treasure, and finders may collude and share the treasures, instead of reporting to the state. The actual efficiency loss is the cost incurred by finders to keep it secret, though the cost may not be high. Sometimes, a treasure trove has cultural value, but finders are not in a position to evaluate that value. To nudge finders to present found treasures to the state, it is more effective for the state to keep cultural artifacts while rewarding finders.20 When it comes to ordinary treasures, the state should take, if anything, at most a small fraction, and leave the rest to finders and locus owners. As finders encounter treasure troves by luck and locus owners do not own underground treasure, neither party deserves to be rewarded more than the other. Locus owners should perhaps take less. Finders surely know what they have found. If revealing the fact to locus owners leads to a huge loss, finders would try to hide the treasure – the cost of hiding is a social waste. Giving locus owners a small share in exchange for their cooperation, by contrast, may be a deal that many finders would strike. 19 20

Whether searching is efficient depends on what the next best things searchers would do. Thus, discouraging searching may or may not be a good thing. Some may believe that the non-pecuniary reward such as the glow that finders will bask in for being a good citizen is what motivates people to take the time to report to the state, and providing rewards would dim the glow. A counter argument is that the rewards are claims held by finders, who can choose not to exercise them to send the signal of being a good citizen. If the law does not reward finders in any case, finders cannot signal their altruism, because others may suspect that they have reported lost goods due to ignorance of the law. If the goal of the law is to maximize the frequency of recovering lost things, the take-away point is that using non-legal rewards (by not giving monetary rewards) is at least a doubleedged sword, if not the inferior approach.

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

308

part iii: movable property

III Conclusion Regarding the treasure trove doctrine, Tajikistan, Syria, and Tunisia come closest to the more efficient rule that I propose. In Syria and Tunisia, finders and locus owners combined are entitled to 80% of the value, but the latter’s share is three times the former’s, while a higher share for finders reduce socially wasteful expenditure in hiding the treasure troves. In Tajikistan, the two parties combined are entitled to 50% of the value. Laws in these jurisdictions are favored mainly because the state is involved, so that cultural artifacts may be singled out, and the state does not take the lion’s share. A number of countries allow finders and locus owners to divide the treasure troves between themselves. When the treasure troves have no cultural value, this is fine. However, the state should have an opportunity to examine the treasure trove. Ordinary lost things are not cultural artifacts, so the justification for the state to give rewards is weaker. A legal scheme like Types A4 and B2 in which finders may acquire ownership and at the very least receive rewards from original owners is preferable. When it comes to reward designs, all factors considered, no country is perfect, but Bolivia (Type A4) and Kyrgyzstan (Type B2) come close by adopting fixed reward fractions (20%) and short waiting periods (three months).

https://doi.org/10.1017/9781009236553.015 Published online by Cambridge University Press

12 Specificatio Do What the Romans Did This chapter discusses the specificatio doctrine – alternatively called the processing, manufacturing, mistaken improvement, or alteration doctrine (Ziff 2014: 132). Ecuador §678, provides illustrative examples of specificatio: a person brews wine from another’s grapes, makes a cup from another’s silver, or builds a ship from another’s wood. The doctrine, which can be traced to Roman law (Merrill 2009: 463–464; Newman 2011: 274–276), appears in 111 jurisdictions (71%), and the modern doctrine is not much different from the Roman law.1 Under the specificatio doctrine an improver may become a co-owner, or outright owner, of an improved movable, after mixing another’s movables with their own labor. The two parties involved are referred to as improvers and material owners. According to the accession principle, which is defined as ownership “established by assigning resources to the owner of some other thing that is already owned” (Merrill 2009: 460),2 the general principle of property law is that material owners continue to be owners of new things. Thus, the core issue in the specificatio doctrine is under what conditions improvers ought to become owners of the products. Early English commentators confusingly put specificatio and accessio under one accession doctrine (Note 2020: 2383). This chapter only deals with specificatio, and leaves accessio and confusio to the next chapter. Additionally, as in other chapters, this chapter only deals with the specificatio doctrine as applied to corporeal things, but not its application in intellectual property law (Lemley 1997: 1070; Koh 1998; Merrill and Smith 2007b: 171; Lee 2011; Newman 2011; Varadarajan 2014; Note 2020). To illustrate the legal issue and economic justification for specificatio, consider the following hypothetical examples: 1 2

Notably, none of the Middle East countries stipulate this doctrine. “Accession is grounded in a conception of original ownership as an attachment to existing ownership rights” (Merrill 2009: 463). Accordingly, a newborn kitten belongs to the mother cat’s owner, and trees belong to the owner of the soil.

309

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

310

part iii: movable property

The log example: Orly, the owner, an environmentalist, buys a large tract of land to preserve its forest. She believes that fallen trees should stay where they are to nurture the soil. Iman, the improver, comes across the area. Mistakenly believing that the logs are unowned, Iman tows them away and makes them into hoops.3 Orly values the logs at $20 apiece and values the hoops at their market value of $100, whereas Iman, a proud carpenter, values them at $125. The wine example: Orly owns a small vineyard. The fruits in her garden could sell for $350 on the market, but Orly never sells them because she has an idiosyncratic taste for grapes she grows herself; put differently, the value of the grapes for Orly is $900, and no one has bothered to offer more than $350 before. Orly’s new neighbor, Iman, makes a mistake regarding their property boundary and thinks that the grapes grow on his land. Iman makes these grapes into a bottle of wine, which has a going market price of $800, while his reservation price is $850. Orly, who holds a religious belief against drinking, thinks the wine is valueless. The car example: Orly has a 15-year-old car.4 She sends it to a repair shop to change its tires. The shop owner’s son, Iman, a gifted car mechanic, renovates the old model, thinking it is a lemon his father has bought for him. Iman values the car at its market value in the second-hand market ($400), plus $50 of subjective value from a few days of hard work. Orly appreciates the upgrade and her subjective value for the renovated car is $500, based on her emotional attachment for the car resulting from her 15-year relationship with it.

I  Comparative Overview Many jurisdictions explicitly specify that the specificatio doctrine is a default rule (“unless otherwise specified by contract …”), for example, Russia §220 and Tajikistan §244. By contrast, countries like Germany §950 and Taiwan §814 do not explicitly spell out the default rule nature of this doctrine. 3

4

Part of the facts are borrowed from the American leading case Wetherbee v. Green, 22 Mich. 311 (1871). Two other cases involve the processing of timber. See Beede v. Lamprey, 64 N.H. 510 (1888); E.E. Bolles Wooden Ware Co. v. United States, 106 U.S. 432 (1882). Levmore (1985: 86) notes several modern American specificatio cases involving mistaken improvement of cars, including Walden v. Vera’s Auto Body Serv., 94 Misc. 2d 792 (1978); Greenwood v. Bennet, 3 W.L.R. 691 (Can. C.A. 1972).

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

311

Most jurisdictions adopt one or two tests to determine whether ownership of improved things should pass to the improvers. One is the transformation test, under which an improver has so transformed the object physically that retrieving or tracing the original object is difficult; the other is the disparity-of-value test, under which the value of the improved thing relative to that of the unimproved thing is sufficiently great as to pass some threshold (Koh 1998: 331; Merrill and Smith 2007a: 1878–1879; Smith 2007: 1769). The transformation test and the disparity-of-value test are collectively referred to in this chapter as the ownership transfer rule. All studied jurisdictions with the specificatio doctrine explicitly require improvers to compensate material owners should the former acquire ownership of the processed things. This is intuitive and its efficiency should not be controversial as a no-compensation rule would induce more people to improve nonconsensually and pretend that they have acted as good-faith improvers. Also, the compensation requirement ensures that improvers value the thing at least at market value (which is the compensation criterion). Conversely, some jurisdictions require material owners to compensate improvers for their labor. Below I refer to this as the compensation rule and focus on this issue.

A  Restoring to Status Quo Ante The first step in the application of the specificatio doctrine is to ascertain whether the improvements in question are reversible. If the status quo ante may be restored, material owners can take back their materials without being enriched by the improvement. The studied jurisdictions respond to this issue in three different ways. First, Argentina §1957, Austria §415, South Africa,5 Timor-Leste §1256, among others, emphasize that the processing has to be irreversible. That is, if processing is reversible, improvers will not acquire ownership of improved things. Rather, material owners take back materials in their original form. Below, I summarize only rules regarding irreversible improvement. Second, a number of countries, including France §570, Malta §577, Paraguay §2047, Seychelles 5

Van der Merwe et al. (2002: 223) offer several interesting examples: “The requirement of non-reducibility is dependent on contemporary technological expertise and knowledge. The requirement implies literally that whenever an artistic creation, like a silver goblet, can physically be reduced to its former state, for example by melting it down, the specificans is not entitled to his handiwork … if a statue is made of stolen bronze, the artist will not become the owner; if, however, he uses marble, he becomes the owner. The issue of irreducibility should serve only as one criterion to determine the allocation of ownership.”

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

312

part iii: movable property

§570, Thailand §1317, Uruguay §740, and Venezuela §576, disregard the feasibility of restoration. Third, Taiwan §814 and South Korea §259, to name but a couple, are agnostic about the possibility of restoration.

B  Co-Ownership versus Sole Ownership Countries with the specificatio doctrine can be first categorized by whether the first-order result is to assign ownership of processed things to a single owner, or to both parties as co-owners. Co-ownership is the rule adopted by a minority of seven jurisdictions. Georgia, Mongolia §120.1, and Turkmenistan have an obscure transformation test, no disparity-of-value test, and do not distinguish between good- and bad-faith improvers. There, improvers and material owners become co-owners. In addition, in Croatia (Josipovic 2013: 187–188) – which adopts only the transformation test – both good- and bad-faith improvers may become co-owners with the ­original owners, but good- and bad-faith improvers are treated differently in the partition process. In Austria §415, both good- and bad-faith improvers may become co-owners with the original owners, but the party without fault has an option to either (1) become a single owner and compensate the other or (2) sell his share to the other. If no party is at fault, the party with the more valuable share has the option to choose between options (1) and (2). In Ireland, recently there has been a trend toward co-ownership between material owners and improvers (de Londras 2011: 235–236). Israel (MPL) §6 applies the rules for accessio and confusio mutatis mutandis. That is, in principle, the two parties are co-owners, but if the difference in value between the improvement and the property is too vast, the improved thing belongs to the higher-valuing owner alone (with a duty to compensate).

C  Two Tests Most jurisdictions assign ownership of processed things to a single owner.6 They can be classified into five types, based on whether they adopt the transformation test and the disparity-of-value test, as shown in Table 12.1. The lower right cell of Table 12.1 is divided into two types because most countries require both tests to be met, while American 6

Macedonia (AOORR) §114 and Serbia (LBOPR) §22 use sole ownership as the first-order rule but create co-ownership if the values of labor and materials are the same and improvers are good-faith.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

313

Table 12.1  Two tests Disparity-of-value test

Transformation test

Total

No Yes

No

Yes

Total

3 21

19 (both) 68 (either) U.S.*

22 89

24

87

111

Notes: * The American common law is not coded by me and thus not included in the summary statistics. See footnote 7 and accompanying text.

common law requires only one of the tests to be met. The surprising result of the comparative exercise is that no countries have come up with a third test, and the formulation of the two tests is largely the same. As Part II elaborates, this result is remarkable because the application of either test is hardly inevitable. First of all, property law can stick to the baseline rule that material owners retain ownership, as long as they do not consent to improvement, and then leave it to unjust enrichment law to deal with the remaining wrinkles. In addition, the disparity-of-value test is neither apparently just nor efficient, and yet countries converge to the same understanding of this test. Type A contains only three countries which stipulate neither tests. China §322 prescribes that ownership of improved things should be determined based on how best to increase the utility of things and to protect the party without fault. Peru §937 stipulates no test at all, and ownership always go to improvers if they are good-faith. Israel (MPL) §6 (Greenfield 2010: 118) applies the accessio and confusio doctrine to the specificatio doctrine mutatis mutandis; that is, co-ownership is the principle and sole ownership the exception. The value of the processing is treated like the value of movable things in the application. Type B requires only the disparity-of-value test. Notably, while countries formulate this test in a few (slightly different) ways, the core of the test is always a comparison between the value of the original materials and that of the improvement. Cambodia §199 and South Korea §259 adopt a twist: If the processing party provides part of the materials, such a party shall acquire ownership of the processed thing so long as the increase in the value of the thing attributable to the processing or reworking plus the

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

314

part iii: movable property

specificatio Co-ownership Type A (neither) Type B (value) Type C (transform) Type D (both) Type E (either) Not coded or no doctrine found

Figure 12.1  Use of the two tests

value of the materials provided by the processing party exceeds the value of the materials provided by the other person. Type C includes the transformation test but not the disparity-of-value test. My coding of the transformation test may be over-inclusive, as the test is counted loosely – legal texts referring to the fact that improvers make something “new” are counted as adopting the transformation test. Indeed, 50 of the 89 jurisdictions (56%) coded as enacting the transformation test belong to this ambiguous type. One hard case is Bulgaria (OA) §94 which prescribes that “a person who has made a new item out of another’s material ….” Bulgaria and 49 other countries with similar ambiguous formulations were coded as having the transformation test. By contrast, 39 jurisdictions are more explicit about the condition of transformation. Easy cases include, for example, Latvia §985 which stipulates that “if … something new has been created, such that the materials used in the composition thereof have lost their former and acquired a new form ….” A number of countries, such as Moldova §330, provide concrete examples of transformation: “Processing shall include writing, drawing, painting, stamping, engraving or any other transformation of surface.” Type D has the highest frequency, requiring both tests to be met. A typical legal structure used by Type D countries is: if improvers make new things, it belongs to the owner of the raw materials; but if the value of labor surpasses that of the material, improvers acquire ownership. This makes the transformation test the first step whereas the disparity-of-value test the second step. No statute is explicit about if the disparity-of-value test is met but processed things are essentially the same, whether improvers may acquire ownership. The doctrinal answer should be no.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

315

U.S. common law, exemplifying Type E, by contrast, adopts an eitheror approach (Koh 1998: 321–334).7 If either test is met, an improver acquires ownership. If an improver neither transforms the property nor significantly increases the value of the property, the original owner retains ownership, but scholars disagree as to whether the owner has to compensate the improver for her labor and other inputs – compare Smith (2007: 1769) with Merrill (2009: 466) and Levmore (1985: 84–85).

D  Good versus Bad Faith From another angle, the studied jurisdictions can be divided into nine types based on how bad-faith improvers are treated. Section C can be thought of as how good-faith improvers fare. Only jurisdictions that adopt the sole ownership rule are included below. Types 0–6 are summarized visually in Figure 12.2. Type 0 does not distinguish between good- and bad-faith improvers. They include Bolivia §148, Cambodia §199, England,8 Germany §950, Greece §1061, Indonesia §606, Italy §940, Ireland,9 Japan §246, Malaysia,10 Monaco §458, Mongolia §120, the Netherlands §V:16, Quebec §972, Romania §598, Scotland,11 Singapore,12 Slovenia (Škerl and Vlahek 2010: 142), South Korea §259, Suriname §660, Taiwan §814, Thailand §1317, and Uganda.13 Improvers are compensated if they fail to meet the conditions for the ownership transfer rule. Type 1 does not distinguish between good- and bad-faith improvers, but good-faith material owners are treated differently. Pursuant to the “French §576 rule” (see also Chapter 13), when good-faith material owners may 7

Note that the New York Jurisprudence, Second Edition, written by Thomas M. Fleming, available on WestLaw as NYJUR ACCESSION §3, suggests that New York does not recognize the disparity-of-value test. In addition, California §1028 does not appear to adopt a transformation test. 8 Hayton and Matthews (2007: 50) and McFarlane (2008: 161) do not discuss whether good and bad faith make a difference. 9 de Londras (2011: 235–236) does not discuss whether good and bad faith make a difference. 10 Maidin and Ali (2014: 148–149, 173) do not discuss whether good and bad faith make a difference. 11 Gretton et al. (2017: 143) contend that “the better view is that state of mind should be relevant only to the issue of compensation.” Therefore, once the transformation test is passed, even bad-faith improvers acquire ownership of the processed things. Bad-faith improvers are punished when they process others’ materials and fail to transform them. Material owners do not have to compensate improvers. 12 Yee and Tjio (2000: 304–305) do not discuss whether good and bad faith make a difference. 13 Bakibinga (2018: 85) does not discuss whether good and bad faith make a difference.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

316

part iii: movable property

Figure 12.2  Structural relationship in Types 0–6

claim ownership of the improved things by compensating improvers,14 they can instead demand either restitution of materials in kind, or compensation in the amount of the value of their materials. Type 1 is adopted in §576 in Belgium, Burkina Faso, Comoros, Dominican Republic, France, Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, Senegal, Seychelles, and Togo, as well as Costa Rica §517, Haiti §476, Louisiana §515, Malta §583, Monaco §462, Uruguay §744, and Venezuela §580. 14

France §576 does not stipulate whether improvers have to be bad-faith. §576 appears to imply the inapplicability of §571, the disparity of value test of which would assign improved things to improvers.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

317

Type 2. Courts may find that the improved thing belongs to the owners of raw materials, if improvers are bad-faith, on the condition that material owners compensate the improvers: Albania §177, Burundi §II:I:29, Eritrea §1021, Ethiopia §1182, Liechtenstein (SR) §194, Louisiana §512, Rwanda §II:29, Slovakia §135b, Switzerland §726, and Turkey §775. Type 3 is similar to Type 2, but is a rule rather than a standard. Put differently, if improvers are bad-faith, raw material owners will have ownership on the condition of compensation; no discretion is explicitly given to courts. South Africa is an example, though its law is under debate (Van der Merwe et al. 2002: 223–224). In Portugal §1337 and its former colonies (e.g., Timor-Leste §1257), compensation is required only if the increased value surpasses one-third of the original value. Type 4 is similar to Type 3, without the explicit condition of compensation. In other words, if improvers are bad-faith, raw material owners will gain ownership and there is no requirement to compensate improvers: American common law (Koh 1998: 326; Smith 2007: 1769), Brazil §1269–1271, Bulgaria (OA) §94, California §1031, Cuba §181, Denmark,15 Estonia (TLPA) §106, Lithuania §4.55, Macedonia (AOORR) §114, Moldova §330, Peru §937 (implying this rule), Poland §192, Serbia (LBOPR) §22, Ukraine §332, and Vietnam §227. Moreover, Chile §667, Colombia §737, Ecuador §683, El Salvador §648, Honduras §657 share the same rule, summarized below16: bad-faith improvers who used materials of good-faith owners “without just cause for error” shall be liable to forfeit their rights, no matter their labor value significantly exceeds the value of materials or not. Type 5. If improvers are bad-faith, raw material owners can choose between being compensated for losses and becoming owners on the condition of compensating improvers: Argentina §1957, Czech §1075, Hungary §5:65, Macau §1256,17 and Paraguay §2047. Type 6 is similar to Type 5, but the option of becoming owners is not matched with the duty to compensate improvers. Hence, if improvers are bad-faith, raw material owners can choose between being compensated

15 16 17

Danish law is more complicated than the typical Type 4 (Faerstad and Lilja 2010: 256–258). Chile §665, Colombia §735, Ecuador §681, El Salvador §646, and Honduras §656 are the French 576 rule. Macau, like Portugal, also stipulates that compensation is required only if the increased value surpasses one-third of the original value. Macau, unlike Portugal, gives material owners options to be compensated. Thus, Macau is not classified in the same group as Portugal.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

318

part iii: movable property

specificatio Type 8 Type 7 Type 6 Type 5 Type 4 Type 3 Type 2 Type 1 Type 0 Not coded or no doctrine found

Figure 12.3  Different ways to treat bad-faith party

for losses and becoming owners: Armenia §174, Azerbaijan §188, Belarus §221, Equatorial Guinea §383, Guatemala §698–700, Kazakhstan §237, Kyrgyzstan §253, Latvia §982,18 Mexico §931, Nicaragua §654, Panama §395, the Philippines §474,19 Puerto Rico §318, Russia §220, Spain §383, and Tajikistan §244. Type 7. China §322 again employs an idiosyncratic standard: courts should assign ownership of processed things according to parties’ fault and utilities of things. A party who is at fault has to compensate the other party. A bad-faith improver should not be able to receive compensation under Chinese law, and may even have to compensate the original owner, even if the latter retains ownership. Type 8. Treaties on property law in Australia (Chambers 2013: 392–395), Canada (Ziff 2014: 132), Finland (Hollo 2019: 155), and Sweden (Faerstad and Lilja 2010: 119–121) appear to take into account of bad faith as well, but the effect of bad faith has not been settled.

E Summary Figure 12.1 shows which countries adopt the transformation test and/ or the disparity-of-value test. Figure 12.2 demonstrates the differences among Types 0–6. Figure 12.3 shows how bad faith makes a difference. Figures 12.4 and 12.5 show the overall distribution of the doctrinal design

18 19

Latvia §982 spells out a Type-6 rule most clearly, explicitly stipulating that there is no duty to compensate. Philippines §474 adds a limit on material owners’ options: if the labor value is “considerably” higher than the material value, material owners cannot become owners. The disparity-of-value test in the Philippines does not include the “considerably” condition.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

319

Figure 12.4  Overall typology: co-ownership part

of the specificatio doctrine. Unreported statistics would show that Type D2 (18 jurisdictions) and Type D7 (13 jurisdictions) are the most popular choices.

II  Ownership Transfer Rule The specificatio doctrine in most countries, as applied to the three examples presented at the beginning of this chapter, does not always produce the most efficient results. Pursuant to the specificatio doctrine in most jurisdictions, Iman the improver in the above examples will get the ownership of the properties (hoops, the bottle of wine, and the car) after he compensates Orly, the original owner, for the pre-improvement market value of the properties. Nonetheless, the wine example shows that the improvement is unnecessary as the grapes are most valuable as fruits for Orly. Nonetheless, once the transformation is completed, Iman values the new product more and should gain ownership. In the best-case scenario,

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

320

part iii: movable property

Figure 12.5  Overall typology: sole ownership part

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

321

Iman would have verified the land boundary at low cost and refrained from making wine from Orly’s grapes. In the car example, renovation is value-enhancing for both parties, but the original owner still values it more and should be able to retain ownership, if the goal is to maximize allocative benefits. In the log example, the specificatio doctrine rightly re-assigns ownership to Iman, and compensation to Orly should enable her to continue her efforts in protecting the environment in other ways. Although the specificatio doctrine does not ensure efficiency in every case, it could be more efficient than the prototypical property rule and liability rule if rightly construed. In what follows, I will elaborate on the interpretation and reformation of the specificatio doctrine that is most likely to be efficiency-enhancing. The ownership transfer rule is a type of pliability rule – it adds a condition to Rule 2. Bell and Parchomovsky (2002: 5) define the pliability rule as “contingent rules that provide an entitlement owner with property rule or liability rule protection as long as some specified condition obtains; however, once the relevant condition changes, a different rule protects the entitlement – either liability or property, as the circumstances dictate.” If Rule 2 (a prototypical liability rule) replaces the ownership transfer rule, a mistaken improver can gain ownership so long as she is willing to compensate the original owner, no matter how little she has changed the movable. Under this kind of the specificatio doctrine, anyone who is interested in someone else’s movables can force the issue by minimally tinkering with them. This chapter will argue that the ownership transfer rule is more efficient than the liability rule, and perhaps more efficient than the property rule, as well. Few countries use co-ownership to deal with specificatio. This makes sense. Specificatio is the combination of A’s materials and B’s labor (and perhaps B’s materials and IP rights, too). It is hard enough to evaluate B’s labor, and no easy task to evaluate the relative contribution of A and B, which is essential to determine the shares of A and B if co-ownership is the rule. Once A and B become co-owners, their unwillingness to engage in bargaining and litigation may hinder efficient use of the co-owned thing, at least before partition (see Chapter 7). Also, under my theory, labor need not be evaluated, as the improver either acquires the ownership of the processed things or is not compensated for the improvement.20 20

That improvers need not be compensated is a key argument in this chapter. If improvers have to be compensated, another type of specificatio doctrine may be warranted. Consider the scenario where the materials are fungible and the price of labor is difficult to appraise,

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

322

part iii: movable property

A  High Institution Cost? Chapter 2 and Chang (2015c) demonstrate that in the field of property, when institution cost is low, the property rule is indeed more efficient than the liability rule. Thus, a liability rule (or a pliability rule) can only be justified given high institution costs. This section rejects the hold-out rationale in the specificatio context, but argues that when the cost of verifying ownership of a certain thing is higher than the benefit, investing in discovering the identity of the correct transacting partner is not socially preferable.

1  Hold-Out Problem? One or both parties with hold-out power may create a situation in which high transaction costs are incurred.21 Nonetheless, few, if any, specificatio problems arise in the bilateral monopoly context – that is, there is only one buyer and one seller (see Chapter 5).22 Original owners may not want to sell at all, or could easily find other trading partners. Improvers usually have many alternative things to work on. Moreover, in countries where improvers must be acting in good-faith to be compensated, they would not have negotiated with the original owners and proceeded against the owners’ wishes in good faith. Original owners also cannot be said to holdout against potential improvers if they never refuse to deal. Hence, holdout is not a source of high institution cost in the specificatio context. 2  Verification Cost and Benefit A source of high institution cost worthy of consideration is the improvers’ verification cost. Verifying ownership, in most movables, can be costly due to a deficient system of record keeping. Regardless of whether the ownership transfer rule or the property rule applies, those who firmly believe that they are the true owners will not incur verification expenses. Most of us are confident about what we own and what we do not own,

21 22

in terms of judicial decision-making cost, it may be easier to give improved things to improvers and ask improvers to either compensate material owners or provide the same of fungible goods. In the case of Orly, this proposed rule leads to the wrong result. This proposal assumes that people generally do not have high subjective value on fungible goods, which admittedly is often true. I thank Richard Epstein for pushing me on this point. Epstein (1997: 2092) discusses using liability rules to overcome hold-out problems. Granted, ex post (after the improvement), the improver and the original owner may fall into the bilateral monopoly context. Nevertheless, economic analysis is mainly an ex ante analysis, rather than an ex post analysis (Chapter 2). Ex post, the two parties in most lawsuits are entrenched in the bilateral monopoly scenario. Ex ante, one-on-one bargaining is not coterminous with bilateral monopoly, as explained in the text.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

323

and we are typically correct in our judgment – therefore, the legislative (or judicial) choice of the property rule over the ownership transfer rule will not motivate many people to verify ownership. Only people who are unsure of their ownership have incentives to double-check. Sterk (2008)’s core insight is that not all verifications are worth their cost.23 Of course, Sterk (2008) does not dismiss information as useless. When the benefit of verification exceeds its cost, it is efficient for a potential improver unsure of her ownership to accumulate information. Verification brings social benefit for the following reason: the original owner may attach “subjective value”24 to the pre-improved thing, and the owner’s ex ante (pre-improvement) total economic value may be higher than the potential improver’s ex post economic value. That is, an “improvement” in terms of market value could actually reduce efficiency (recall the wine example above). Verification may expose ownership information and facilitate bargaining between the potential improver and the owner. As argued above, generally there should be no serious transactional hurdles between these two parties ex ante. Hence, no matter the results of bargaining, the thing in question is likely to be used in its most valuable way. That is, if any improvement contract follows, the improvement can be viewed as efficiency-enhancing and the consensual nature of the improvement bars litigation. No deals suggest that the original owner’s idiosyncratic value dwarfs the increase in market value. A verification that does not lead to improvement action prevents a decline in efficiency.

3  Good Faith versus Bad Faith The ownership transfer rule should be applied only when the verification cost is higher than the verification benefit. That is, an improver whose verification cost is lower than the expected verification benefit should not receive the ownership. Sterk (2012) proposes to import the negligence principle from tort law to property law. In the good-faith purchase context, Landes and Posner (1996: 190–192) distinguish between good-faith and innocent (Chapter 10). That is, an innocent improver fails to expend a reasonable amount in verifying ownership – that is, an improver could be innocent by refusing to make any inquiry about ownership. Mackaay (2021: 291) also defines good faith as having taken all justified precautions. 23 24

See also Demsetz (2011a: 9): “[T]here exists an efficient amount of ignorance in an economic system if the cost of acquiring information is positive,” and also Demsetz (2011b: S12). Subjective value is the difference between economic value and market value. See the following footnotes for detailed explanations.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

324

part iii: movable property

Following this line of argument, improvers who either know or “should have known” (the innocent) should be classified as bad-faith. The distinction between good faith and bad faith, from an efficiency standpoint, is whether the verification cost is lower than the expected verification benefit. Compared to the specificatio doctrine, which does not distinguish between the foundations of good faith, this broader interpretation of bad faith and the prevention of bad-faith improvers from acquiring ownership can induce potential improvers to conduct cost-justified information investigations. Such efforts may reveal that improvers are not owners. In this scenario, improvers have to bargain with the true owners for permission to improve because the property rule will be enforced to protect owners. If cost-justified searches have been done, and no evidence against improvers as owners is produced, the specificatio doctrine would apply. Improvers still will not necessarily gain ownership, as that depends on whether the two tests are met. Taking account of improvers’ “negligence” is more than academic fantasizing. Koh (1998: 237) has observed that “courts have been reluctant to apply the law of specificatio to provide negligent improvers of property with compensation for their efforts” (emphasis added), suggesting that at least American courts have found ways to distinguish between improvers who had low verification costs and those with high.25 Therefore, good faith in the specificatio context should, and can, be narrowly construed as an additional verification cost surpassing the verification benefit. Implementation of the specificatio doctrine with the “negligence” element is realistic, but can hardly be mathematically precise.26 While the verification cost is known ex ante, verification benefit can only be 25

26

Sterk (2008: 1313), however, argues that it is difficult for courts to differentiate good faith from bad faith, and the distinction will create perverse incentives to be good-faith, even when the verification cost is low. Nonetheless, if bad-faith improvers could gain ownership, they will not verify even if the cost is low. There are two types of bad faith: one, a party who knows who the real owner is; the other, a party who knows she does not own the item, but does not know who owns it. There should be a very strong case against giving the first type of bad-faith improvers ownership, as their verification cost is zero. The cost for courts to distinguish the two types of bad faith or distinguish between good faith and bad faith is a serious concern. When the background evidentiary rule or the inability of fact-finders (judges or juries) makes such a distinction difficult, the case for the different treatments are weakened. Keep in mind that courts are often required to distinguish good faith and bad faith in many other contexts, such as the good-faith purchase rule. Questioning courts’ ability to single out bad-faith improvers also casts doubts on their ability to sort out badfaith agents in other contexts. That is, my specificatio regime (and perhaps all regimes) is unlikely to be efficiencymaximizing, but it should be efficiency-increasing.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

325

ascertained by bargaining with the real owner ex post. In other words, the verification benefit is probabilistic ex ante. The problem is that before verification, potential improvers do not know the identity of the real owner. Thus, without the presence of an omniscient adjudicator at all times, what the specificatio doctrine can do is require improvers to conduct a basic search, something like checking ownership as a reasonable person,27 and when either the probability of improvers not being owners or the expected stakes increase, acquiring more ownership information. When are the stakes higher? As a general matter, when the thing in question is not a commodity, but something more personal (perhaps a refurbished used car), and likely to destroy the owner’s subjective value, and the likelihood of decreasing social wealth (and thus the verification benefit) is high.28 Bad-faith improvers should not be eligible for favorable treatment based on the specificatio doctrine.29 A bad-faith improver already knows that she has no right over the thing in question, although she may not know the identity of the real owner.30 Verification cost can be considered zero for knowing improvers, as they can refrain from working on another’s thing without any additional searching. At the very least, compared to the verification cost of good-faith improvers, that of bad-faith improvers is generally lower. A categorical rule against bad-faith improvements would save some adjudication cost, although such a rule does not always ensure maximization of allocative benefits ex post. My intuition is that such a categorical rule is, in the long run, more efficient than a balancing test, though admittedly it is an empirical question. The categorical rule against acting in bad faith is easier to enforce due to its moral salience (Merrill and Smith 2007a). In this respect, the 26 jurisdictions that adopt 27

28

29

30

A textbook example: what should Peter do when he finds and likes a log at his camping site? It is reasonable and low-cost for Peter to check with the camp administrator or the ranger or cowboy who swings by, before he takes the log as his own and starts working on it. One may worry that the accumulated verification cost across the whole society would be huge. As argued above, only people who are not sure of the ownership of thing would actually incur verification expenses. Thus, the accumulated cost should not be exaggerated. There are disputes in the Netherlands and other countries as to whether million-dollar, leased aircraft engines would become owned by the aircraft frame owners through the accession doctrine (Gomez 2010). These aircraft engines are leased through complicated financing contracts and all parties are well aware of the nature of the deal. There is no policy reason that the specificatio doctrine would apply in this context, where everyone knows the deal. Moreover, the specificatio doctrine should clearly be a default rule (DCFR: 4267). Contracts should have specified who owns what. Bad-faith improvers might be more inclined to steal the things in question because they have no chance of gaining ownership through the specificatio doctrine. But the criminal punishment should keep the stealing incentives in check.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

326

part iii: movable property

Type 0 (no distinction between good faith and bad faith) have the worst rule. 55 of the 111 jurisdictions (51%) that assign ownership of improved things to good-faith original owners and do not require them to compensate bad-faith improvers have the more efficient rule (Types 1, 4, and 6).

B  Allocative Benefits? When institution cost is high, the efficient allocation of resources depends on how property rights are assigned (e.g., Miceli 2004: 175). Even when institution cost is low, a good entitlement allocation rule saves on institution cost. Hence, the normative goals of property law include assigning entitlements of properties to the party who values them more (Epstein 1994: 31; Krier and Schwab 1995: 446; Merrill 2009: 488).31 Usually, current property owners are those ones who value their things most, as they have already out-bid others in a market economy to acquire ownership and have held on to the things. To ensure maximization of allocative benefits, the property rule is designated the default rule under property law, under which ownership is transferred only through voluntary transactions, which occur only when a buyer’s value (willingness to pay) is higher than a seller’s value (willingness to accept). Thus, resources flow to the parties who value them more. The ownership transfer rule, however, calls for an alternative mechanism. The main issue of this section is then whether the ownership transfer rule allocates resources more efficiently than the property rule. Epstein (1995: 118) argues that specificatio “tends to award the things to the party who values it most.” This section elaborates the reasons underlying Epstein’s claim. The starting point is that the original owner generally values the thing more ex ante. The problem is under what conditions improvers will become the higher-valuer ex post. No rule is likely to judge the high-valuer accurately in every case. The property rule might favor the owner when the resource is more valuable in the improver’s hand, whereas the liability rule might transfer ownership when the owner can better use the resource. Is the ownership transfer rule as a pliability rule better able to identify the high-valuer than the two prototypical rules? My argument is that when both the transformation test and the disparity-of-value test are satisfied, the ownership transfer rule is more likely to lead to higher allocative benefits than the two prototypical rules. I examine the two tests in turn. 31

Cooter and Ulen (2012: 91–93) call it the “normative Hobbes theorem.”

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

327

The transformation test could increase allocative benefits because transformation (such as from logs to hoops, and from grapes to wines) is likely to create subjective value for improvers and eliminate the subjective value of the original owners. In the wine example, Iman may enjoy the wine he has made more than do other connoisseurs; that is, the transformation creates subjective value for Iman. Orly is a fruit lover who does not drink wine. Her affection for the home-grown grapes disappears after she can no longer consume them. It is generally difficult to determine whether the original owner’s ex ante or ex post value or the improver’s ex post value is higher. My intuition is that it would be better to adhere to the property rule and allow original owners to retain ownership when improved things are not transformed. By contrast, if they have been transformed, the possible creation of subjective value for improvers, and the possible elimination of subjective value for original owners, will likely leave improvers as the high-valuer. Nonetheless, transformation does not always work like a zero-sum game in terms of subjective value. Thus, applying the transformation test and re-assigning ownership to improvers does not always maximize allocative benefits. That improvers increase market value more, in and of itself, is not important for allocative benefits. To maximize efficiency, what matters is who values the thing more, not who contributes more to its market value. The former value is economic value (market value plus subjective value), which is hard for third parties to observe. Throughout this book, the economic analysis is concerned with economic value, but troubled by the difficulty of measuring it objectively. Contribution to market value could be (and has been; see Part I) solved through unjust enrichment or tailoredmade compensation rules. The law does not have to award ownership to improvers simply because improvers significantly increase market value. The disparity-of-value test could increase allocative benefits when (1) improvers greatly increase market value of the things and (2) the economic value of original owners remains more or less the same and is exceeded by the post-improvement market value – even before consideration of the improver’s ex post economic value, which is higher than their ex post market value.32 In this case, ownership transfers are preferred because after the fact improvers value the thing more. Original owners’ economic value, however, does not always stay put. In the car example, if Orly adjusts her economic value to reflect the new market 32

For further discussion of how subjective value may change in the face of vastly rising market value, see Chang and Fennell (2014b) and Fennell (2013a).

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

328

part iii: movable property

value (with her subjective value remaining intact), she could still value the thing more than Iman; thus, the property rule should be used. Hence, the disparity-of-value test does not ensure maximization of allocative benefits, either. Moreover, holding other conditions equal, it is hard to think of a real-world example where a non-transformative improvement that increases market value by a lot would not update material owners’ economic value. That is, the second-hand car example may arguably be a typical case. A more modest claim would be more convincing: the disparity-of-value33 test, in itself, does not justify the ownership transfer rule; nonetheless, between a transformation that does not significantly increase market value and one that simultaneously, significantly, increases market value, in the latter case it is more likely that improvers’ ex post economic value will be greater than the owners’ ex post economic value, if not their ex ante economic value as well. Thus, the strongest case for applying the ownership transfer rule is when both the transformation test and disparity-of-value test are met. Recall that it is Type D in Part I. I prefer Type D because of my premise that the property rule is generally more efficient in property law, and a deviation from it requires strong justifications. Type D is more likely to transfer ownership involuntarily when the transfer actually increases allocative benefits than Types A, B, C, and E. Thus, Type D should be favored.34

C  Other Justifications? Chapter 2 points out that five aspects should be explored to determine whether deviation from the property rule would be efficient, and above I have explored two aspects (high institution cost in Section A and allocative benefits in Section B). This section discusses the other three concerns: (negligible) ex ante investment incentive, ex post feasibility, and (low) decision-making cost by courts. I examine them in turn. First, not all movables can be significantly invested ex ante. Generally, ex ante under-investment is not a serious problem in the specificatio context. Thus, the property rule does not gain an edge as compared to the ownership transfer rule. 33 34

I do not yet have a good theory of how to define “disparity” of value. Perhaps it is better left to the court to determine on a case-by-case basis. Note, however, that, as Smith (2007: 1770) points out, the disparity-of-value test is better than the transformation test in reducing judicial decision-making costs. This is a reason for favoring Type B over Type D.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

329

Second, in certain contexts, the property rule is inapplicable after the fact (Rose 1997: 2181). Not here. The original owners of the improved things can simply retain ownership. Third, the judicial (error) cost under the ownership transfer rule is higher than that under the property rule. If the property rule is the norm, original owners retain ownership as long as they meet their burden of proving ownership. The compensation question can be avoided altogether (more on this below). Under the specificatio doctrine, by contrast, courts have to estimate the property value before and after the improvement. While assessing the value of commodities is not difficult, movables that can be significantly improved are generally not commodities, but things of which the market value before and after improvement can vary widely. This poses a risk of inaccurate judicial assessment of property value. Moreover, courts have to ascertain whether improvers are acting in good faith. Hence, the judicial decision-making cost is obviously higher under the specificatio doctrine than under the property rule.35 This is a good reason for not expanding the scope of the specificatio doctrine,36 or even to stick to the property rule. Original owners’ prevention cost is another factor worth discussing in comparing the relative efficiency of the property rule and the ownership transfer rule, though I will eventually dismiss this concern. Owners with positive subjective value will presumably spend more on prevention costs under the ownership transfer rule than under the property rule because under the former regime, owners are at risk of losing their subjective value. The existence of prevention costs under the ownership transfer rule, in this respect, gives the property rule an edge. Sterk (2008: 1315), however, has persuasively argued that when both verification cost and prevention cost are positive, the original owners will still often be the cheapest cost avoider. Original owners can often mark off their property right effectively at low cost, and the marking is useful to all potential improvers, whereas verification produces useful information for only the verifying 35

36

This is yet another example of how the property rule reduces the “measurement cost” of producing information about assets and activities (Smith 2004b: 1723). Note, however, that Epstein (2014) points out that the specificatio doctrine avoids the ticklish issues of valuing the labor of the improver, which reduces the judicial decision-making cost. This argument is based on the stance that improver’s labor has to be compensated. See Part III. As Epstein (1995: 66) points out, “assigning the property right to the party to whom it is more valuable is not panacea. It ignores the cost of administering the property-rights system, which might be lower under a simpler criterion for assigning rights.”

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

330

part iii: movable property

improver. Put differently, an owner spending a small amount in prevention saves potential improvers much in aggregate. Furthermore, owners spend prevention cost only to the extent of the expected loss in subjective value. Thus, when subjective value is small, prevention cost is negligible. Finally, for owners whose subjective value is reduced to zero with any alteration to their properties, the prevention cost does not vary with entitled legal protection because their goal is to have their properties remain untouched. More generally, very few people make prevention decisions based on one particular doctrine. Law-conscious property owners will also consider the good-faith purchase doctrine, for example. Given that goods that are exposed to unconsented improvement are also in danger of the more serious problem of theft, owners of movables probably respond to the various risks of losses by locking their houses and cars and put their most precious things in safety boxes – regardless of how the specificatio doctrine is stipulated. In sum, although prevention cost does exist, it is most likely invariable to the content of specificatio. Therefore, owners’ prevention cost can be ignored when the specificatio doctrine is designed. *** In summary, the ownership transfer rule could enhance efficiency if narrowly construed. First, only good-faith improvers who have conducted cost-justified verification can enjoy the protection of the ownership transfer rule. This strict construction better avoids improvements that decrease economic value. Second, both the transformation test and the disparityof-value test should be met, in order to limit the application of the ownership transfer rule to cases where improvers are very likely to value the resource more. The limited application also avoids the judicial error cost accompanying the pliability rule. When the ownership transfer rule is inapplicable, the property rule applies. The original owner retains ownership of the processed property. The contribution of the improver’s labor and materials (like water, in the wine example), even her intellectual property rights, will be assigned to the original owner through the accession principle. Whether it is efficient to require the original owner to compensate the improver for her contribution is explored in the next part.

III  The Compensation Rule In addition to the ownership transfer rule, the specificatio doctrine contains another non-property rule. When the ownership transfer rule is

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

331

inapplicable, owners (may) have to compensate improvers for their labor. I call this the compensation rule. This less noticed part of the specificatio doctrine could be characterized as a Rule 5, a type of put-option rule (Ayres 1998; 2005), particularly when improvers are bad-faith,37 because improvers force owners to purchase their labor. As explicated above, the compensation rule is controversial in the United States, but it is contained in 51 studied jurisdictions, even when improvers are bad-faith. Dickinson (1985: 75) has claimed that “[a]s an abstract principle, a mistaken improver’s claim for restitution is unassailable: conscience and equity are on his or her side.” While that might be the case, and probably the major reason that this compensation requirement is prevalent, this part argues that the compensation rule does not appear to be economically efficient. Compensation is economically justified if original owners or improvers change their ex ante behaviors to be more efficient because of the compensation rule. Original owners, as discussed above, are unlikely to set prevention efforts according to the content of the specificatio doctrine. Improvers’ incentives to invest in improvements may be reduced to reflect the risk of receiving nothing for improving. Nonetheless, this is not necessarily a bad thing. The issue of compensating improvers arises only when original owners are highly likely to value the resource more. To reiterate, non-transformative and marginally market-value-increasing improvements are unlikely to destroy owners’ subjective value or create a large subjective value for improvers. In addition, this kind of improvement is rarely so innovative that only a few high-end improvers are aware of the possibility of such changes.38 If original owners know of ways to change their things and have not contracted for it, they probably do not want such improvements.39 Hence, not improving could actually be more efficient, as 37

38

39

When improvers do not know but should have known their lack of ownership, they unintentionally process the thing. Thus, improvers do not consciously exercise a put option. Nevertheless, when improvers realize that they are not owners, they still demand actual owners pay them. In other words, improvers at least exercise their put option ex post. That is, I assume that minor improvements are common sense that both owners and improvers are aware of, while there is information asymmetry in major improvements – such that improvers, due to their expertise, are aware of the possibility as well as the cost and benefit of such renovation, whereas owners are not aware of them unless informed by experts. Such “improvements,” for the original owners, are probably personal property torts, because they reduce the economic value of the properties. Improvements do not lead to personal property tort actions, because improvements mean increases in market value. The court can only observe market value, but not economic value. Therefore, the court only finds personal property tort liabilities when there is a reduction in market value of

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

332

part iii: movable property

the higher-valuing owners prefer to use their money elsewhere. Moreover, original owners have little incentive to strategically expose their things to minor improvements that need not be compensated because this act will subject their things to the risk of theft or transformative improvement. By contrast, guaranteed compensation will induce improvers to process the things no matter what because at the very least they can recoup the expenses,40 leading to value-reducing improvements.41 Hence, the compensation rule does not appear to be efficient. More generally, put-option liability rules in property law tend to be inefficient (Chang 2014). The prior property literature is mostly conservative about implementing Rule 5 (Epstein 1998: 844),42 exemplified by the compensation rule here, as it obviously creates moral hazard (Levmore 1985: 87; Merrill 2009: 466).43 While excluding bad-faith improvers from acquiring ownership could keep the moral hazard at bay, some bad-faith improvers can pretend to be good-faith and force their work upon others. Since there is no economic reason to compensate improvers,44 even no-fault ones, the default position should be no compensation45 (see also Chapter 4).

40

41

42

43

44

45

the property at issue. In an ideal world, an owner who can prove that her subjective value is destroyed in the process of improvement is entitled to heightened (more than market value) damages – “propertized compensation” (Parchomovsky and Stein 2009: 1858). The compensation rule thus creates a problem similar to moral hazard problem under eminent domain law. Namely, when landowners are fully compensated by condemnors, landowners will over-invest because they can regard the probability of takings as zero (Blume et al. 1984: 78; Miceli and Segerson 2007: 27; Chang 2012a). Here, fully compensated improvers will over-improve. What if the improver turns out to be the actual owner but he does not improve due to the risk of not being compensated? Nonetheless, my hunch is that when improvers are actual owners, they tend to be certain of their ownership and do not refrain from improving. Porat (2009: 194) lists six cumulative conditions for compensating unrequested benefits that produce public goods. The compensation rule in the specificatio doctrine can hardly meet these requirements. In the 26 Type-0 jurisdictions that do not distinguish between good-faith and bad-faith improvers, the moral hazard problem is even more serious. Dari‐Mattiacci (2009) explains why restitution for “negative liability” should be limited to unintentional benefactor. In the adverse possession context, Fennell (2006: 1067) has persuasively argued that “[t]here is no reason to think that people who are making honest mistakes are necessarily also making efficient mistakes.” Thus, good-faith improvement is not coterminous with efficient improvement. American courts occasionally offer the original owner the choice of compensating or selling the thing to the improver at the price of pre-improvement value. Ditto in Types 5 and 6 countries. Levmore (1985: 87) points out that this approach does not work well in the specificatio context (other than automobile repairs) because of the difficult valuation problem and the fact that original owners can hardly buy a pre-improved, repair-needed thing with the selling price she receives.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

specificatio: do what the romans did

333

As briefed above, many jurisdictions clearly prescribe compensation to improvers who do not gain ownership, and no studied jurisdictions explicitly stipulate that original owners never have to compensate goodfaith improvers. Unlike the standard for distinguishing good and bad faith and the two tests accompanying the ownership transfer rule, the compensation rule leaves little room for courts’ discretion. Following Smith (2009a), compensation could be thought of as a “safety valve” for certain good-faith, efficiency-enhancing improvers.46 Before doctrinal changes, since compensation has to be awarded at least in some circumstances, I argue for a narrow interpretation of this requirement, under which compensation is only awarded to improvers who are good-faith without fault (for instance, under apparent title; see Chapter 447) or improvers who have expended significant resource in verification. The compensation, though, will not enhance efficiency.

IV Conclusion To be as efficiency-enhancing as possible, this chapter contends that modern specificatio should work in the following ways: improvements are irreversible. Bad faith improvers should never gain ownership through the specificatio doctrine. Good-faith improvers should be induced to conduct cost-justified verification, so that efficiency will not be reduced by unconsented alteration. The specificatio doctrine should exclude improvers who know, or should have known, from enjoying the benefit brought by the ownership transfer rule as they have failed to conduct the most basic ownership searching – creating a proper incentive scheme. In addition, when improvers transform movables and also greatly increase their market value, as compared to other scenarios, it is more likely that ex post improvers value them more than their original owner. Thus, to maximize allocative benefits, the ownership transfer rule is most warranted when both tests (transformation and disparity-of-value) are met. Finally, while the ownership-acquiring improver should compensate the original owner for market value, the ownership-retaining owner should not be required 46 47

Smith (2009a: 2127–2129) discusses the safety valve in the intellectual properties and building encroachment contexts. In the adverse possession context, Fennell (2006: 1083) argues that ownership insurance, improved ownership searching, etc. protect improvers better than the adverse possession doctrine. The alternatives that Fennell mentions are not available to movables in the specificatio context, so the justifications to protect good-faith improvers under apparent title should be stronger.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

334

part iii: movable property

to compensate the improver. The laborer’s reduced incentive to improve may actually be efficiency-enhancing. No specificatio doctrine stipulated in the studied jurisdictions exemplifies the most efficient version that this article lays out above. In the United States, where specificatio is a “venerable doctrine” (Merrill and Smith 2007a: 1878), only one test is required (Type E), and some states may have awarded an undeserving improver some compensation for her labor. Common-law courts, however, should be applauded for incorporating the “negligence” standard into the property doctrine. Type D rightly requires two tests for the ownership transfer rule to apply. Type 0 fails to distinguish between good-faith and bad-faith improvers, and Types 2, 3, and 5 request that bad-faith improvers be compensated by original owners, which could induce serious moral hazard problems. In these jurisdictions, bad-faith improvers still have a chance at getting ownership without the permission of original owners, and compensation to improvers is guaranteed. Overall, jurisdictions with Type D (and that are explicit about requiring transformation), plus Types 1, 4, or 6 have the most efficient specificatio doctrine. These are Malta, Moldova, the Philippines, Poland, Puerto Rico, Russia, and Venezuela.

https://doi.org/10.1017/9781009236553.016 Published online by Cambridge University Press

13 Accessio and Confusio No Sign of Convergence This chapter discusses accessio and confusio. DCFR VIII.–§5:203 defines accessio as when “goods owned by different persons are combined in the sense that separation would be impossible or economically unreasonable.” New York law offers a typical case of confusio: “A confusion of goods takes place when things of the same kind, belonging to different sources, are so intermingled that they can no longer be identified or distinguished” (1 N.Y. Jur. 2d Accession §9; emphasis added). The distinction between accessio and confusio may appear clear, but there are fuzzy edges. If my oils are mixed with your oils, the confusio doctrine kicks in. But what about my apple juices mixed with your orange juices? My milks mixed with your cereals? Are the latter two examples confusio or accessio? No statutes clarify the boundary between confusio and accessio. Some countries further distinguish confusio and commixture (commixtio). The former applies to mixture into one thing (such as mixture of liquids), while the latter applies to mixture into separate, multiple things (such as mixture of solids, like a pile of steel plates) (Erp and Akkermans 2012: 671; Gretton et al. 2017: 142). Most countries that make this distinction, such as Slovenia (Škerl and Vlahek 2010: 141), apply the same rule anyway. By contrast, South Africa offers a rare example of applying different rules to confusio and c­ ommixtio. The former leads to co-ownership, whereas the latter leads to separate ownerships of certain portions of the mixture (Van der Merwe et al. 2002: 224–225). This chapter summarizes law regarding inseparably merged or mixed things only. Some jurisdictions first spell out rules for separable merged or mixed things1 (usually, owners each take back their incorporated things), and then specify rules for inseparably merged or mixed things. This chapter coded and presented only the latter. A bit differently, France §566 1

Romania §600 adopts an idiosyncratic rule: owners of merged things may claim the separation of the property if the other owner does not suffer more than one-tenth of the property’s value (Stanescu 2017: 134–135).

335

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

336

part iii: movable property

Table 13.1  Comparison of terms under the accession principle Roman law Specificatio Confusio and commixtio Accessio

Restatement of the Law 4th, Property

Common law

DCFR

Combination Combination

Accession Confusion

Production Commingling

Combination

Accession

Combination

and 20 other jurisdictions stipulate that their accessio doctrine between movables applies even if the merged or mixed things are separable. A repeated example of a diamond merged with a gold ring appears in many civil codes, such as Ecuador §673, Honduras §648, Colombia §787, and El Salvador §638. A united diamond ring is separable; nonetheless, the ­accessio doctrine still applies. As the same rule applies to both separable and inseparable things, this is not a problem for Part I. Throughout the book, I describe accessio of two things as united, merged, or combined, while describing confusio of two things as mixed, mingled, or intermingled. I use these verbs interchangeably. In addition, I will stick to the Latin phrases accessio and confusio, rather than using its English counterpart accession and confusion because confusion is an everyday noun and accession in the English literature – confusingly – include specificatio and accessio but not confusio. Like Merrill (2009: 429), I will use the term accession when referring to the general principle of accession or referring to accessio, confusio, and specificatio combined. The hopeless term confusion is shown in Table 13.1. Some countries, for example, France §572 and Cuba §514, stipulate rules regarding a person’s improving of both things owned by themselves and others to form a new thing – that is, both accessio and specificatio take place. This chapter does not consider this type of rule.

I  Comparative Overview Part I is divided into four sections. The accessio doctrine includes three types of combinations: movable and movable, movable and immovable, and immovable and immovable. These are discussed in Section A, Section C (focusing on plants and seeds taking root in another’s land), and Section D (focusing on a building constructed on another’s land), respectively. Sections C and D are largely the same, with differences noted in Section D.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

337

Section B notes that the confusio rule is often the same as the accessio rule and summarizes the rules when they are not.

A  Accessio between Two Movables Figure 13.1 summarizes the typology of accessio between movables.

1  Rule: Sole Ownership a. Type A: Sole Ownership by Owner of Principal Thing  Type A. The owner of the “principal thing”2 always gets sole ownership, with compensation duty. Type A0 does not consider the parties’ good or bad faith. These include Afghanistan §2209, Bulgaria (Stoimenov 2011: 429), Eritrea §1020, Finland (Hollo 2019: 155), Iraq §1125, Slovakia (Petkov 2010: 419), and South Africa (Van der Merwe et al. 2002: 204–206). Also included in this group are Commonwealth countries such as Australia (Chambers 2013: 429–431), Cyprus (Synodinou 2020: 148), England (McFarlane 2008: 164; Sparkes 2019: 168), Ireland (Gardiner 2009: 233–234), New Zealand (Williams 2011: 130), Singapore (See et al. 2018: 301–304), and Uganda (Bakibinga 2018: 85), where the role of parties’ bad faith is either not explicit or explicitly disregarded. Other Type A countries either favor good-faith parties or disfavor badfaith parties. Several subtypes can be distinguished. Type A1. In France §565 and countries that closely follow French rules, the ground rule is that accessio, confusio, and specificatio regarding movables should be resolved according to “natural equity.” The ensuing stipulations are meant to provide guiding principles. As described below, several Middle East countries only transplanted the natural equity provision without importing the guiding principles. §§566–569 of the civil codes spell out the baseline rule of sole ownership. France §576, which has been transplanted widely beyond the former French colonies, stipulates that when materials of good-faith parties are used to make new things and good-faith parties are entitled to claim ownership of the new things, in 2

For the distinction between a principal thing/part/component and a subordinate/accessory thing/part/component, see DCFR Full (2009: 4305–4306). To avoid confusion with the accessory nature of mortgage and the accession principle, this book will not use the term accessory thing/part/component. In short, a subordinate thing serves a principal thing in its functioning. For instance, a TV screen is a principal thing whereas a remote control is a subordinate thing.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

338

part iii: movable property

Figure 13.1  Typology of accessio

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

339

lieu of acquiring ownership they may instead choose between restitution of materials and compensation for the value of the constituent m ­ aterials.3 This book calls this the “French §576 rule.” Note that the precondition for the aforementioned accessio rules to apply is the inapplicability of the famous French Civil Code §2276 (possession as certificate of ownership), which French commentators observe is infrequent (Cohet 2019: Paragraph 11; Reboul-Maupin 2020: §2A). France and its former colonies adopt Type A1. Type A2 has, in addition to the French §576 rule, rules under which parties “without just cause for error” (interpreted as bad faith under my framework) lose their materials (i.e., are not compensated): Chile §§665­ –667, Colombia §§735–737, Ecuador §§681–683, El Salvador §§646–648, and Honduras §§656–657.4 Type A3. All A3 jurisdictions have the following rule: when “subordinate parts” are more valuable, and their owners are good-faith about the merging, owners of valuable subordinate parts can demand that the united thing be separated and subordinate parts returned, even though the separation may cause some injury to the principal parts. Moldova §330 is the only example for Type A3a, which has only this baseline rule. Type A3b contains California §1031, which, in addition to this baseline rule, assigns ownership to the good-faith party if users of materials are bad-faith. Type A3c countries also have the French §576 rule: Costa Rica §§511–512 & §517, Haiti §§467–468 & §476, Louisiana §510 & §515, Uruguay §§737–738 & §744, and Monaco §454 & §462. Type A3d. Malta §575 & §583 and Venezuela §572 & §580 provide owners of subordinate parts an additional option that Type A3c does not award: acquire the whole and compensate the owner of the principal thing. That is, when subordinate parts are more valuable and their owners are good-faith about the union, owners of subordinate parts can either acquire united things by compensating owners of principal things, or demand that united things be separated and subordinate parts be returned even though separation may cause some injury to principal things. Additionally, both have the French §576 rule. 3 4

§576 applies to accessio, confusio, and specificatio of movables (Reboul-Maupin 2020: §2A3). The accessio doctrine in these countries applies to separable combined things. If one part of the material has no close substitute, its owner is good-faith, and separation does not damage the other part, the good-faith owner may demand separation: Chile §664, Colombia §734, Ecuador §680, El Salvador §645, and Honduras §655.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

340

part iii: movable property

Type-A3e contains the baseline rule of A3 (Spain §§375–378, Equatorial Guinea §§375–378, Nicaragua §§646–649, Panama §§387–390, Puerto Rico §§310–313, and the Philippines §§466–469) and an additional rule: bad-faith owners of subordinate parts who merge the two things lose their united components, whereas if it is bad-faith principal owners who merge the two things, owners of subordinate parts can choose between being compensated and re-taking subordinate components despite destroying principal things (Spain §379; Equatorial Guinea §379; Nicaragua §650; Panama §391; Puerto Rico §314; the Philippines §470). Type A4 is composed of two rules (not including the baseline rule of A3, though). First, A4 also has the additional rule that A3e has: Bad-faith owners of subordinate parts who merge the two things lose their united components, whereas if it is bad-faith principal owners who merge the two things, owners of subordinate parts can choose between being compensated and re-taking subordinate components despite destroying principal things (Mexico §§922–923; Guatemala §§691–692). Second, when unified things cannot be separated without decreasing the values of subordinate things, owners of principal things still have the right to demand separation, but are obliged to indemnify owners of subordinate things, provided that the latter is good-faith: Guatemala §690 and Mexico §921. Type A5. Scotland adopts the rule that if principal thing owners merge knowingly, they should compensate owners of subordinate things more (Gretton et al. 2017: 142). Type A6. Austria §415 achieves sole ownership via a winding path. Initially, combined things are co-owned. If one party is at fault while the other is not, the latter has the option either to acquire full ownership and compensate or to sell his share and be compensated. The former, depending on good or bad faith, has different responsibilities. If no party is at fault, the option belongs to the party with the more valuable component. b. Type B: Standard  Type B1. Canada (Ziff 2014: 130) and New York (1 N.Y. Jur. 2d Accession §7) courts consider, among others, whether a party is an intentional wrongdoer. c. Sole Ownership by Way of Auction  Type C1. Portugal §§1333–1334 and its former colonies (such as Timor-Leste §§1253–1254) adopt the following rule. If accessio arises from a good-faith action by an owner, and the owner of another component prefers to be compensated, the

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

341

former should compensate the latter and take sole ownership. Otherwise, sole ownership of merged things goes to the owner of the more valuable element, who must then compensate the other owner. If the elements are equivalently valuable, and the two parties cannot reach an agreement, the highest bidder in an internal auction acquires sole ownership and compensates the other bidder.5 If no parties are willing to bid, the merged things will be auctioned off in a public auction and proceeds will be divided pro rata. By contrast, if accessio arises from bad-faith actions, and merged things cannot be separated without damage, the inactive owner who did not trigger the accessio can choose between (1) acquiring the merged thing and compensating the active owner and (2) being compensated by the active owner.

2  Rule: Co-Ownership a. Simple Co-Ownership Rule  The common design in Type D is that when both parties are good-faith, they co-own merged things. Type-D countries can be divided into two subtypes. Type D0 does not distinguish between good and bad faith. Lithuania §4.54, Peru §937, and Poland §193 stipulate a simple co-ownership rule. Type D1 varies the legal effect by a party’s knowledge. For instance, Czech §1078 allows each co-owner to petition for partition in kind if it will not “interfere with the essence”; in addition, if a bad-faith party merges the things, the other, good-faith party has an option to sell its share to the ­former and be compensated. In Croatia (AOORR) §149, good-faith parties may choose between owning the merged thing by compensating the bad-faith party and being compensated. In Latvia §§980–982 and Macedonia (AOORR) §115, if the party who joins the things is badfaith, the good-faith party can choose between owning the merged thing without paying the bad-faith party, or being compensated. In Guinea §§555–556, the good-faith party can demand the bad-faith party return the merged materials or pay their monetary value. b. Co-Ownership in Principle  Type E. The baseline rule is when both parties are good-faith, they c­o-own a merged thing if neither

5

§1333’s formula to calculate the amount of compensation is ambiguous. It seems that if one bid is (b) and the other bid is (b + d), the former should be compensated with [b/ (b + b + d)] × (b + d).

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

342

part iii: movable property

component is principal; sole ownership is the exception, with a merged thing belonging to owners of principal components. Type E0 does not distinguish between good faith and bad faith of the parties, so Type E0 simply applies the baseline rule: Albania §176, Azerbaijan §191, Burundi §II:I:28, Cuba §182, DR Congo (LF) §28, Denmark and Norway (Faerstad and Lilja 2010: 254–255), Estonia (TLPA) §107, Ethiopia §1183, Georgia §194, Germany §947, Greece §1058, Israel (MPL) §4 (Greenfield 2010: 117), Liechtenstein (SR) §195, Mongolia §119.2 (the sole ownership exception kicks in if the parties have a dispute), Rwanda §II:28, Switzerland §727, Taiwan §812, Thailand §1316, Turkey §776, and Turkmenistan §217. Also classified as Type E0 but are somewhat different than the aforementioned countries are Indonesia §§607–608 and Suriname §§661–662. They stipulate co-­ownership when the combination is not attributable to human intervention and, when attributable to human intervention, ascribe sole ownership to inactive parties. Countries in Types E1–E3 stipulate different rules when one party is bad-faith. Type E1. Bad-faith owners of subordinate parts are compensated less: Bolivia §147, Hungary §V:66, and Italy §939. Type E2. In Brazil §1273 and Vietnam §225, if the party who joins the things is bad-faith, the good-faith party can choose between owning the merged thing by compensating the bad-faith party and being compensated. Type E3. In Serbia (LBOPR) §23, if the party who joins the things is bad-faith, the good-faith party can choose between owning the merged thing without compensating the bad-faith party and being compensated. The Serbian rule is similar to the Macedonian rule, but only the former adds an exception of sole ownership to the default co-­ownership rule. c. Co-Ownership as Exception  Type F0. Sole ownership is the principle (the party whose thing is more valuable acquires ownership) and co-ownership is the exception. As the conditions in Types E0 and F0 are quite similar, Types E0 and F0 may be just a matter of legislative drafting style rather than a matter of substantive differences. Type-F0 countries include Argentina §1958, Cambodia §198, Japan §244, the Netherlands §V:14, Paraguay §2048, Quebec §971 & §973, Romania §§598–601, Slovenia (Škerl and Vlahek 2010: 141), South Korea §257, and Sweden (Lilja 2010: 121). None of these countries consider knowledge of the parties.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

343

3 Standard Type G. Courts have judicial discretion about assigning ownership of merged things. Type G1. Middle East countries that borrow only the natural equity standard from France §565 falls in this category. These countries include Algeria §791, Bahrain §856, Egypt §931, Jordan §1145, Kuwait §887, Libya §935, Qatar §918, Syria §893, and United Arab Emirates §1274. These Middle East countries explicitly demand judges to consider the knowledge of the two parties, while the French natural equity standard itself does not explicitly account for this. Type G2. China §322 asks courts to assign ownership based on the parties’ fault and the utilities of the things in question. B  Confusio (Mixture) Confusio is often stipulated together with accessio. Indeed, 41 jurisdictions apply the same rules to accessio and confusio and nine jurisdictions apply the accessio rule to confusio mutatis mutandis. In 12 jurisdictions, however, the statutory texts on accessio do not seem to cover confusio, and there are no separate stipulations for confusio.6 Below, I focus on the 48 jurisdictions that adopt different confusio and accessio rules, and divide them into three groups based on how they adopt co-ownership as the solution to confusio: Some countries’ solution to confusio is not always co-ownership. They include Australia (Chambers 2013: 438–441), California §1030, Chile §663, Colombia §733, Costa Rica §§515–516, Ecuador §679, El Salvador §644, Eritrea §1022, France §573 and its former colonies, Haiti §§473–476, Honduras §654, Louisiana §514, Malta §§580–581, Monaco §§459–460, South Africa (Van der Merwe et al. 2002: 224–225), Uruguay §742, and Venezuela §574. In some other jurisdictions, the solution for confusio is always ­co-ownership if both parties are good-faith, but not always co-ownership if a party is bad-faith. Canada (Ziff 2014: 125–126) and New York (1 N.Y. Jur. 2d Accession §12) laws give the whole ownership to the good-faith party if mixing is intentional by the other party, a rule recommended 6

As many of the legal texts I have consulted are unofficial translations, and accessio and confusio may in those countries be consider as the same phenomenon, readers should take the null result with a grain of salt. That said, Stoimenov (2011: 429) explicitly notes that confusio is not stipulated in Bulgaria; same in Slovakia (Petkov 2010: 419).

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

344

part iii: movable property

by Blackstone (McFarlane 2008: 162). Also adopting the same rule are Equatorial Guinea §§381–382, Guatemala §§695–697, Mexico §§926–928, Moldova §330, Nicaragua §§652–653, Panama §§393–394, the Philippines §§472–473, Puerto Rico §§316–317, Spain §§381–382, and Vietnam §226. Finally, England (McFarlane 2008: 161–162), Ireland (Gardiner 2009: 234–235), Finland (Kuusinen 2010: 372), and New Zealand (Williams 2011: 130) explicitly treat bad-faith and good-faith confusio the same and always impose co-ownership on the two parties.

C  Accessio: Seeds or Plants to Land This section summarizes accessio rules from the viewpoint of a landowner. That is, what happens when, without her consent, plants and seeds are combined with her land by acts of owners of plants and seeds? Thus, when a civil code like France distinguishes outcomes based on whether combination is conducted by landowners or seeds owners, I only coded rules regarding acts by the latter.7 In the case of the French Civil Code, I coded the rule in §555 but not that in §554. Seventy-three jurisdictions have a rule like France §554. The typology below includes scenarios where landowners are goodfaith but seeds and plants owners may be either good- or bad-faith. That is, rules for other scenarios are omitted below.8 Type H. Landowners acquire ownership of the plants and seeds combined with their land, and, in general, landowners must compensate the good-faith owners of plants and seeds. Type H0. Legal effects do not vary by whether plants and seeds owners are good-faith or bad-faith: Australia (Chambers 2013: 133–135), Azerbaijan §135.10, Bulgaria (OA) §97, Cambodia §122, Colombia §739, Croatia (AOORR) §150, Cyprus (Synodinou 2020: 148), Denmark (von Bar 1999: 19), England (Sparkes 2019: 168), Georgia §193, Germany §946, Greece §1057, Honduras §659, Hong Kong (Goo and Lee 2019: 11), Ireland (de Londras 2011: 8–10) (the intention of the planters may be taken into 7

8

France §553 has an influential presumption: all constructions, plantings and works on or inside a piece of land are presumed made by the landowner, at his expenses and belonging to him, unless the contrary is proved. Libya §926, Egypt §922, Algeria §782, Venezuela §555, Iraq §1117, Syria §886, Spain §359, to name a few, have similar stipulations. Several countries, for example, Venezuela §557, enact rules for the scenario where both landowners and planters are bad-faith. In addition, several countries, for example, Egypt §925, give landowners put options to force planters to purchase land when compensating planters for the incremented value is too onerous for landowners.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

345

account), Japan §242, Malaysia (Maidin and Ali 2014: 149, 174), Mongolia §119.1, the Netherlands (Salomons and Haentjens 2017: 86), New York (1 N.Y. Jur. 2d Accession § 17), New Zealand (Williams 2011: 131), Ontario (La Forest 2017a: 20-2), Poland §191, Scotland (Gretton et al. 2017: 138– 139), Singapore (See et al. 2018: 67), Slovenia (Škerl and Vlahek 2020: 392), South Africa (Van der Merwe et al. 2002: 207), South Korea §256, Suriname §556, Sweden (von Bar 1999: 520), Taiwan §811, Turkmenistan §216, Uganda (Bakibinga 2018: 28), and Uruguay §751 (Barua 2018: 90). Type H1 distinguishes between whether owners of plants and seeds are good-faith or bad-faith, and their knowledge affects the amount of damages due: Austria §420, Chile §669, Ecuador §685, El Salvador §650, Finland (Kuusinen 2010: 371), Latvia §973 & §§978–979, Macau §1262, Malta §568, Serbia (LBOPR) §23, and Uruguay §751. Type H2. According to Puerto Rico §§297–298, landowners have a put option against owners of plants and seeds. That is, the former can force the latter to purchase the land. Type J0. Landownership extends to the plants, seeds, and buildings if they are not merged by trespassers. Landowners can require trespassers to remove buildings, though. Nigeria, which receives both English common law and Islamic law (Smith 2018: 61–63), is the only example. The key difference between Type J and Types H and K is that landowners do not have the option of compensating trespassers to acquire ownership of plants, seeds, and buildings that have been merged with the land. Type K. Landownership extends to plants and seeds. Landowners may request owners of plants and seeds remove their plants and seeds – the main difference between Type K and Type H is the power to demand removal. Type K0. Whether owners of plants and seeds are bad-faith does not matter, and landowners can always demand removal: Ethiopia §1175,9 Israel (LL) §21 (Greenfield 2010: 11),10 and Vietnam §225. Type K1. Landowners can always request removal, but planters’ goodand bad-faith matter in terms of compensation: Brazil §§1255–1257, California §1013 & §1013.5,11 Guinea §547, and Switzerland §§671–672. 9 10 11

Ethiopia §1180 distinguish bona fide and mala fide when it comes to accession of buildings to land. Israel (LL) §23 regarding unsettled land is not considered here. In California (and some other jurisdictions), a landowner can always request removal, but a good-faith planter “acting in good faith and erroneously believing because of a mistake either of law or fact that he has a right to do so, affixes improvements to the land of another, such person, or his successor in interest, shall have the right to remove such improvements upon payment, as their interests shall appear, to the owner of the land.”

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

346

part iii: movable property

Type K2. Landowners keep ownership that has extended to plants and seeds; when owners of plants and seeds are bad-faith, landowners can request plants and seeds be removed: §555 in Belgium, Burkina Faso, Comoros, Dominican Republic, France, Ivory Coast, Luxembourg, Madagascar, Mauritius, Niger, Senegal, Seychelles, and Togo, as well as Afghanistan §§2204–2205, Albania §175, Algeria §§784–785, Argentina §1962, Bahrain §852, Bolivia §129, Burundi §II:I:27, Costa Rica §508, DR Congo (Loi 1973) §23, Czech §1085, Egypt §§924–925, Equatorial Guinea §§363–364, Guatemala §§661–663, Haiti §461, Indonesia §604, Iraq §§1119–1120, Italy §§934–936, Jordan §§1140–1141, Kuwait §882, Libya §§928–929, Louisiana §496, Mexico §902, Moldova §329, Monaco §449, Morocco (LRRP) §18, Nicaragua §§630–633, Panama §§374–376, Paraguay §1984, the Philippines §§448–450, Quebec §§957–959, Rwanda §II:24, Spain §§361–363, Syria §§888–890, Thailand §§1310–1314, Tunisia (CDR) §36, United Arab Emirates §§1269–1270, and Venezuela §557. Type K3. Portugal §§1340–1341, Timor-Leste §§1260–1261, and other former Portugal colonies have a set of rules close to those in K2, but are not the same. When someone builds, sows, or plants on another’s land in good faith, outcomes differ based on whether the added value by construction, sowing, or planting surpasses the value of the pre-improvement land. If construction, sowing, or planting is more valuable, actors acquire landownership by compensation. If it is less valuable, landowners must compensate actors. If it is equally valuable as the land, the two parties enter an internal auction. (Macau §1262 does not use internal auction.) The comparison of values is not unique, as eight other jurisdictions (Albania, Brazil, Iraq, Jordan, Macau, Switzerland, Syria, and United Arab Emirates) have similar rules. The internal auction design, however, is unique to this group of countries. If planters are bad-faith, landowners can demand removal. Type K4. Romania §584 gives landowners a put option against planters. Otherwise, Type K4 is the same as Type K2. Type G2. China §322 asks courts to assign ownership based on the parties’ fault and the utilities of the things in question.

D  Accessio: Buildings to Land Ninety-seven jurisdictions apply the same rule no matter whether it is seeds and plants, or buildings, that are combined with land. Their rules will not be repeated in this section. Thirty-five jurisdictions have no rules in either context. In 11 jurisdictions, there are rules regarding accessio in the seeds and plants context but no rules regarding accessio in the building

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

347

context. Some of them, such as Germany §946, are coded as not having this doctrine because the law only regulates the combination of movables to immovables, and it is unclear whether a building-in-making would be considered as a movable in this context. Taiwan, on the other hand, recognizes buildings as separate immovable things, so a simple-minded accessio doctrine does not work. This section thus focuses on the nine jurisdictions that have no rules regarding accessio in the seeds and plants context but do have rules regarding accessio in the building context, as well as four jurisdictions that have different rules for both contexts. Below are the classifications of these jurisdictions. Type H0: Estonia (TLPA) §107, Slovakia §135C, and Turkey §722. Type H1: Latvia §§968–970 (though classified as H1, the same as its rule for accessio for plants and seeds, the rule for accessio for buildings to land is more complicated). Type K0: Croatia (AOORR) §152, Eritrea §1017, and Serbia (LBOPR) §25.12 Type K1: Liechtenstein (SR) §§53–54 and Macedonia (AOORR) §§118– 119. Landowners can always request removal in Macedonia, but builders’ being good-faith triggers a different set of other rules, including whether the value of constructions vis-à-vis land should be considered. Type K2: Cuba §180.1, Macau §§1259–1260, and Suriname §§657–658. Type K4: Hungary §§V:70–V:71 (Sandor 2017: 171–174) and Peru §941 & §943.

II  Economic Analysis A  Economic Formula for Separability The most puzzling stipulation summarized above is the rule that the accessio doctrine should apply even if merged things could be separated without much harm. As said, 21 jurisdictions, including France §566 and Haiti §467, adopt this position. Determining which component is principal and which is subordinate, assessing the amount of compensation due to the party who does not acquire ownership, etc. are not always easy. Why would lawmakers not prefer to order separation of the separable merged things even when two parties are good-faith? This does not appear to be economically sound. Outside of these 21 jurisdictions, at least in the 86 jurisdictions that ­explicitly limit the application of the accessio doctrine to inseparably merged 12

Croatia (AOORR) §153 & Serbia (LBOPR) §24 are about the context of good-faith builders and bad-faith landowners.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

348

part iii: movable property

things, the critical question is how to define separability. Ambiguous criteria such as “becoming an essential part” (e.g., Germany §946), “disproportionately high cost” (e.g., Taiwan §812), or “separation causing great harm” (e.g., Taiwan §812) are often used, but it only pushes the question a step back, as judges then must interpret “essential,” “disproportionate,” and “great.” In interpreting the confusio doctrine in Germany §948, Gaier (2013: §948 Rn. 3), in a famous commentary, proposes that if one of the two tests is met, it constitutes “disproportionately high cost”: expenses needed for separation > values of post-separation things (1) expenses needed for separation > values of intermingled thing (2) The problem is that the two tests make the threshold of “disproportionately high cost” too high. Suppose that the intermingled thing is valued at 200 USD, and the two post-separation things are valued at 95 USD each. If the expense needed is 189, it is not disproportionately high cost to separate the intermingled thing. Nonetheless, given the accessio or confusio, separation decreases social value by –199.13 This interpretation is inefficient. A more efficient interpretation is to adopt the following formula for accessio and confusio: if (the total values of the two post-separation things) minus (the value of the intermingled thing) > (expenses needed for separation), separation is economically warranted. If, according to the formula, separation is value-reducing, efficient accessio and confusio rule discussed below should apply. The values here are measured in terms of economic value, not market value (see Chapter 2); otherwise, separability would never be warranted. Economic values in this context, admittedly, would be difficult to measure accurately. Thus, the formula should be a guide that courts use pragmatically. Courts can ask parties to demonstrate whether the intermingling or merging eliminates the subjective value of one or both parties and whether separation would restore or preserve it (see related discussion regarding the transformation test in Chapter 12).

B  Bad-Faith Party Get Nothing This book has consistently argued that bad-faith parties should receive no compensation at all. No matter whether bad-faith parties take actions in merging and mingling or knowingly receive a good-faith merging and mingling, the better, “crystal” rule (Rose 1988) is that good-faith parties 13

95 × 2 − 200 − 189 = − 199.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

349

can choose one of the options: (1) acquire the merged or mingled thing without having to compensate the bad-faith party; (2) demand the badfaith party remove the addition and compensate any damages; or (3) require the bad-faith party to take the merged or mingled things and offer compensation. This crystal rule, however, may cause concern over whether owners of seeds may strategically put them on others’ land and try to mislead courts into believing that landowners are bad-faith. A more conservative approach is to stipulate that if the active party is bad-faith, the good-faith inactive party has one of the three options. Hence, inactive landowners will not stand to lose their land due to strategic accessio. Mexico §901, Panama §374, Peru §943, Puerto Rico §298, among others, have the correct rule that bad-faith builders and planters lose the building materials and seeds and are not titled to ask for compensation.14 Many countries’ laws, however, are orthogonal to even this conservative approach. Many K2 countries require landowners to compensate bad-faith improvers if good-faith landowners choose to keep the improvements: for example, Albania §175, Argentina §1962, Bahrain §852, France §555, Morocco (LRRP) §18, Moldova §329, Paraguay §1984, Portugal §1341, Qatar §911, Suriname §657, and Tunisia (CDR) §36.15

C  The Efficient Portugal Fractional Ownership and Internal Auction Design When both parties in confusio are good-faith, the simple efficient rule is co-ownership pro rata. As the nature of intermixing facilitates voluntary partitioning, the law can stop right here. When both parties in accessio are good-faith, the Portugal rule makes economic sense. In fact, Portugal §1333 & §1340 embody the fractional ownership and internal auction idea proposed in Chapter 10. (Any critique that suggests that the internal auction is too expensive or too impracticable should start with an empirical study of the Portugal law in action to underpin the argument.) Timor-Leste §1253 exhibits how to operationalize the whole idea very clearly: First, separate the merged things when it is economical to do so. Second, assign ownership of entire things to principal part owners who must compensate owners of subordinate parts, 14 15

Cf. Ethiopia §1175 & §1178: if landowners do not consent to improvements, builders and planters are not entitled to compensation. Quebec §958 is a compromise still unsatisfactory: bad-faith improvers can be compensated for collected fruits and revenues after deduction of the costs incurred.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

350

part iii: movable property

or present them with equivalent things. Third, if both things are of equal value, the two parties enter an internal auction. Fourth, if no one is willing to bid, merged things are sold in public auction and proceeds are divided pro rata. One caveat, however, is in order. In any auction (internal or public), liquidity constraint of bidders is a potential concern. Different local contexts may derail an ostensibly efficient design. Chapter  7 proposes a partition approach that attempts to ameliorate the liquidity problem. Here, limited access to credit may prevents higher valuers from bidding. Hence, while Portugal and other developed economies may be expected to have more robust credit markets, Portugal’s former colonies that follow the rule verbatim may not always achieve efficiency due to the limited and slow access to credits. Giving the entire ownership of merged things to principal part ­owners is a widely used rule. When the two components of a merged thing are indeed a principal thing and a subordinate thing, an internal auction is not necessarily more efficient than this rule. Internal auctions do not ensure efficiency when the two parties do not have roughly equal shares. A principal part, by definition, takes up the lion’s share. Thus, using internal auctions do not bring obvious efficiency upgrades over the current rule.

D  Simple (First-Order) Rule Type K3 makes it possible for plant and building owners to acquire landownership. Epstein (1995: 118) disagrees with this kind of rule and argues it is “better to give the good faith improver… the incentive to stay off ­[others’] property by limiting her rights.” As Chapter 5 has analyzed, with the advent of GPS and mapping technologies, it is less and less plausible that a person who makes major improvements is genuinely good-faith about the location of land boundaries. Making it strictly impossible for improvers to acquire landownership fends off strategic behaviors and incentivizes improvers to investigate land boundaries. The more difficult problem is if landowners (choose to) enjoy the benefits of the improvements, are they required to compensate planters and builders? An as-of-right compensation for good-faith planters and builders again may induce strategic behaviors. Courts can draw on equity (as meta-law) (Smith 2021a) to require landowners to pay when it is clearly unjust for planters and builders to come away empty-handed.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

accessio and confusio: no sign of convergence

351

III Conclusion For accessio between movables, Type C (Portugal, etc.) is the most efficient. It will be even more efficient if bad-faith actors will not be compensated. For confusio, the New York and Canada rule that punishes the bad-faith actor and make intermingled things co-owned pro rata when both parties are good faith is more efficient. As for accessio involving land, Type K2, followed in many countries, is the most efficient. A perhaps big puzzle is why the accessio doctrines for two movables and that regarding land have been so diverse. Chang and Smith (2019) conjecture that the accessio and confusio doctrines have not converged because the benefit is low, but this does not address the question of why the rules are so diverse in the first place. Given that these doctrines date back to Roman law – unless lawmakers misunderstand the substance of Roman law – one would expect that Roman law (or its modern incarnation) would be a focal point and followed, and yet it is not the case. Solving this puzzle must await future work.

https://doi.org/10.1017/9781009236553.017 Published online by Cambridge University Press

u Conclusion

I  Theoretical Implications In the Introduction, I list four theories with which this book engages. Now it is time to recapitulate the lessons learned. The four theories are discussed, below, in the same order in which they appeared.

A  Converged or Diverged? 1  Structural Aspects Structural aspects of property law show convergence. As several “new essentialists” (Wyman 2017), including myself, have argued (Merrill 1998; Smith 2002; 2004a; 2004b; Chang and Smith 2012; Smith 2012b; Merrill 2014), the inevitable baseline in property law is in rem exclusion. While my codebook does not explicitly code the civil-law rei vindicatio and the common-law trespass, conversion, replevin, from my impression of reading the sources, strong protection of property owners’ dominion is universal. More direct evidence is the often strong protection of possession.1 The right to roam (Alexander 2016; Klick and Parchomovsky 2016) is said to break the despotic in rem exclusion, but only 13 jurisdictions recognize the right to roam, and three of these only on state land. Additionally, former or current socialist countries like Russia, China, and North Korea §§40–41, recognize private property and provide owners with strong protection on the books. In rem exclusion responds to our world of positive institution cost. Jurisdictions thus develop similar strategies for reducing the expenditure of institution cost to reasonable levels. Acquisitive prescription of immovables (Chapter 5) and good-faith purchase (Chapter 10) are cases in point. While the detailed designs differ, almost all jurisdictions have 1

I will write two separate articles about possession, and the coding of possession laws is already included in the analysis in Chapter 1.

352

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

conclusion

353

these two doctrines that limit the amount of verification cost expended. Moreover, the numerus clausus principle and registration of ownership complement each other in reducing institution cost (Rose 2011). Very few countries adopt the opposite principle of numerus apertus (Chapter 3) and most require transfers of ownership to be registered to bind third parties (Chapter 4). Another way to control institution costs is to adopt only a small number of property forms. The recursive nature of property (when future interests are allowed) and the existence of the trust in common law jurisdictions make unnecessary a large number of forms. Even in civil-code jurisdictions where trust and recursiveness are weaker, basic forms, such as mortgage and usufruct, go a long way in increasing the value of property. Real easement as a module and interface can accommodate a wide variety of arrangements. Even when future interests are banned, a remainder could be set up using a strawman. Hence, it is not surprising that most jurisdictions do not have more than a dozen limited property forms (Chapter 3).

2  Interconnected and Divergent The second hypothesis tested in the book is whether interconnected aspect of the property system that started out differently would remain divergent. Doctrines in the areas of possession, sales, future interests, and trust support this hypothesis. Possession is the root of many property law doctrines, such as adverse possession, first possession, delivery of movables for sale purposes, etc. Having a doctrine of possession is structural: Possession is a commonsensical cue that ordinary persons in everyday life can understand (Merrill 2015). Using possession as a proxy for boundaries helps enforce the exclusion-based structure of property. Nonetheless, what kind of possession doctrine to implement in the law is stylistic and interconnected. That is, the specific definition of possession, while only different on the margin (such as whether a certain intention is required), is difficult to change. I have elaborated this point elsewhere (Chang 2015b; 2016c; Chang and Smith 2019) and will provide a full-length treatment in a future work; also, the point is clear from the possession-related variables used in Chapter 1. It is doubtful that, in practice, any such intention requirement will make a substantial and substantive difference. Still, the doctrine remains divergent, probably because possession is a fundamental concept and lawmakers worry that changes in definition may bring unintended consequences. Possessory notions are not only important for the protection of property rights through trespass and conversion, but they figure as well in such

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

354

part iii: movable property

doctrines as original acquisition and adverse possession – even in animal torts. Aspects of possession intersect with bailment and with agency, security interests, and the like. A change in the definition of possession might have implications throughout private law, thus giving legislators pause. Sales of movables and immovables are also a case in point (Chapter 4). Clearly, sales are connected with a complicated web of other property doctrines. For example, for a jurisdiction to change from an intention-based system to a delivery-based system, many foundational principles have to make way. The design of the register and the markets of middlemen and professionals are also affected by sales law. Many civil-code jurisdictions only embrace trust law reluctantly and half-heartedly, treating a trust as a two- or three-way contract. No civil codes have a section on future interests and many explicitly ban such a feudal idea, even though function-equivalent of future interests have been demanded and transacted. Both property-form trust and future interests touch the nerve of civil lawyers considering ownership as absolute and even indivisible.

3  Isolated and Convergent Isolated doctrines are more likely to be convergent if the benefit of adopting the focal rule is apparent. The best example is the concurrent ownership form for residential common-interest communities (Chapter 8). Theoretically, there are multiple ways to set up the form. Condominium (strata title), cooperative, and English commonhold are three real-world designs. This concurrent ownership form is a discrete device that does not interact with other doctrines in the property law system, as it is used almost exclusively in residential buildings, and by definition is not applicable to land, so recognizing this form would not interfere with the concurrent ownership regime in land. Late-coming lawmakers can choose any form to structure high-rise residential buildings. Lawmakers who want to adopt a form that stands the tests of time and different property systems may be inclined to opt for the most popular form. Lawmakers may also want to adopt the most popular form so that foreigners are willing to directly invest or to reduce the cognitive burden of citizens who are familiar with an existing form abroad. Thus, convergence across jurisdictions is expected. Indeed, the condominium form dominates. Tellingly, when Saudi Arabian lawmakers decided to recognize a form of residential common-interest communities for the first time in 2002, they chose condominium.2 2

I thank Prof. Faisal Alotaibi for this observation.

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

conclusion

355

B  The Use of Meta-Law While the ideas of law-as-principle and meta-law-as-exception may not sound ground-breaking, lawmakers designing the legal schemes discussed in this book often fail to recognize the utility of this structure. Some countries employ a complicated web of rules, trying to close every possible loophole. Some countries adopt a single standard, leaving the resulting headache to the courts. Put differently, the former has law but scarce meta-law, whereas the latter has meta-law but an obscure law. The former is not ideal because no set of rules can altogether eliminate opportunism. The latter is not ideal because without rules qua law providing safe harbors and sure shipwrecks (Chapter 6) – that is, behavioral guidance, law becomes unpredictable and the cost of legal compliance increases. The basic position of this book is that crystal-clear rules qua law that give bad-faith parties no property rights and no compensation should be favored. Sometimes, for example, in the case of acquisitive prescription (Chapter 5), even good-faith possessors should not acquire property rights as a rule. Good-faith parties may better be protected with tailored or untailored standards qua meta-law. Again, a property law system goes a long way toward increasing efficiency by stipulating clear rules that induce behaviors which maximize net social wealth and standards that are flexible in punishing opportunism.

C  Judge-Made Laws Are Not More Efficient The overall finding of this book also has implications for the debate on the efficiency of common law versus statutory law. Judge Richard Posner famously argues that judge-made common law tends to be efficient, while statutory law (in the United States) does not. This is one of the most debated questions in law-and-economics (Parisi 2004; Garoupa and Ligüerre 2011; Zywicki and Stringham 2013). Reading together the conclusions of each chapter (which points out which jurisdictions adopt the most efficient rule), I do not consider the judge-made common law (sometimes modified or supplemented by statutes) having a clear efficiency edge over the statutory civil law. Granted, equitable doctrines available in common-law jurisdictions provide a lot of room for courts to render efficient decisions, and this book cannot accommodate a full discussion of all the equitable doctrines. That is, this book compares laws (but not meta-laws, unless they are built in the studied doctrines) in common-law and civil-law jurisdictions. To the

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

356

part iii: movable property

extent that the common law has more active meta-law, my comparison would be unfair to common-law jurisdictions. Still, equity as meta-law has an exceptional character, and it would be unlikely for judges in every case to disregard the rules (law) and apply (correctly) the meta-law. Of course, in our age of causal revolution, to really test Judge Posner’s thesis, other factors have to be controlled for; to that end, it is best that whether a property doctrine is enacted or judge-made be exogenously determined or even randomly assigned. However, such scenarios are hard to come by. Still, under the Posner thesis, statutes may be inefficient because the shadow examples are regulations where either the beneficiaries or the regulated parties are discrete and they have incentives and political prowess to influence the law-making process. Private law, in general, and property law, in particular, often cast a wide net and affect almost everyone, but only in minor ways. I observed the reform dynamics of Taiwan’s property law and found that adoption of an expert-drafted amended code was long stalled as there was no political momentum to push for its enactment. During the three readings, other than one mortgage-related provision where the banks (whose interests are large and who are discrete) attempted to switch courses, other provisions were rubber-stamped by the legislature (Chang 2016a). Hence, my bold claim would be that generally, no matter whether private law doctrines are enacted by legislature or made by judges, their inclinations toward efficiency (or lack thereof) would be similar.

D  The Changing Familial Picture Chapter 1, from a macro perspective, and Chapters 3–13, from micro perspectives, demonstrate the heterogeneity within the common law and the civil law, especially the latter. While common-law jurisdictions still more or less follow English case law, this book demonstrates that a number of jurisdictions have made significant reforms – from the Australia-initiated Torrens registration system, to England’s reforms from 1925 onwards. By contrast, other than the Uniform Commercial Code, property laws in New York and California appear to lack milestone reforms like those in other anglophone jurisdictions. In the civil-law world, few features are shared other than the format of a civil code. Most jurisdictions categorized to the civil-law side of the world still stick to the civil code model, opting to put private law (sometimes private international law and intellectual property law as well) into one

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

conclusion

357

statute, a civil code. The most notable examples are the recent Chinese Civil Code of 2020 and the Laos Civil Code of 2020, each replacing a myriad of stand-alone statutes. Despite the same legislative style, civil codes contain very different substances. Some codes, like China’s, include public-law rules. Some codes, like North Korea’s, convey the national ideology. Roman law’s influence, at least in terms of terminologies, is apparent in some but scant in others. Figure 1.1 demonstrates how diverse the substantive doctrinal decisions are. Legal families are changing (Pargendler 2012b). While there is a socialist group, it should be noted that several former USSR republics, including Azerbaijan, Turkmenistan, Georgia, Estonia, and Latvia, have come to resemble German law more than Russian law. This is true also for many Eastern European countries. While Spain and many Latin American countries flock together, the recently reformed Argentina and Brazil codes appear to have embarked on a different path. Figure 1.1 also suggests that Peru, Paraguay, and Bolivia are also distant from the Spanish-speaking sub-cluster. While the former French colonies still more or less follow the original 1804 Napoleonic Code, France – and Belgium in 2021 – has moved on. Nonetheless, many academic articles (e.g., Anderson 2018) still rely on the overly simplistic trichotomy of French, German, and Scandinavian civil law (La Porta et al. 2008). This book shows that such a classification is incomplete and unduly ignores the heterogeneity within each modern legal family. Legal origin is not always a valid proxy for legal substance (Bradford et al. 2021).

II  More Topics The previous chapters lay out many topics that this book could not possibly explore in depth. There are, of course, many more comparisons to be done. One example is the right of first refusal, also called the pre-emptive right or the right of pre-emption. It could be voluntary or statutory. Its effect could be in personam or in rem. Statutory right of first refusal is sometimes awarded to a concurrent owner when a fellow concurrent owner sells her share to a third party3 and sometimes awarded to a holder of limited 3

See, for example, Afghanistan 2216, Albania §204, Algeria §721, Armenia §195, Azerbaijan §218, Bahrain §861, Belarus §253, Cambodia §564, China §305, Egypt §833, Ethiopia §1261, Georgia §173, Guatemala §498, Hungary §V:84, Iraq §1073, Jordan §1151, Kazakhstan §216, Kuwait §892, Kyrgyzstan §270, Libya §842, Luxembourg §815-14, Macau §1308, Mexico §973, Moldova §352, Mongolia §91, the Philippines §1620, Qatar §862, Russia §250, Seychelles §834, Switzerland §682, Tajikistan §298, Turkey §732, Ukraine §362, and Uzbekistan §224.

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

358

part iii: movable property

property rights when the landowner is selling. Perhaps an even bigger question is whether the right of first refusal is a “property” right at all. My data set cannot do justice to this issue, as the right of first refusal is often regulated outside of civil codes. Another example is statutory security rights (priorities). Researchers have to additionally canvas the law of obligations and various other regulatory and tax statutes to have a complete picture. The third example is concurrent ownership. This book devotes two chapters to the most common type of concurrent ownership, but obviously there are additional types (Chang 2023 forthcoming-a). The latecoming time-sharing has made its headway into at least the United States, Denmark (Olsen-Ring and Ring 2015: 112–113), France (Terré and Simler 2006: para. 552B) (in corporate form), Finland (Setten 2015: 271), Norway (Hegdal 2015: 912–913), Sweden (Schaeferdiek 2015: 1281), United Arab Emirates §1177, and a number of other European countries4 and commonlaw countries (Edgeworth 2017: 988). Residential concurrent ownership forms have been summarized in Figure 8.4. In addition, 65 jurisdictions have separate stipulations for a party wall, 13 jurisdictions (mostly Middle-East countries, plus notably China) recognize family concurrent ownership – that is, concurrent ownership by family members only, and concurrent owners are not limited to spouses. The most important form of concurrent ownership would be the one that applies to both movables and immovables, residential or otherwise. If spousal concurrent ownership form could be put aside first, not all countries appear to have a second general concurrent ownership form. One type of such a form is called “joint tenancy,” adopted in the common-law jurisdictions. Joint tenancy is featured by the right of survivorship (jus accresendi) and a concurrent owner’s unilateral right to ask for severance at any time. The second type may be called “joint and indivisible ownership,” which is adopted in, for example, China §297, South Korea §271, Switzerland §652, Taiwan §828, and Turkey §701. This type is featured by (1) implicit and often undetermined shares; (2) lack of unilateral right to ask for partition before the joint legal relationship (e.g., partnership) ends; and (3) individual shares and the entire co-owned things (by default) that are transferrable only with unanimous consent. The third type may be called “USSR joint ownership,” as it is adopted by many former USSR 4

Directive 2008/122/EC of the European Parliament and of the Council of 14 January 2009 on the protection of consumers in respect of certain aspects of timeshare, long-term holiday product, resale.

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

conclusion

359

Second Form of Co-ownership Simple form Idiosyncratic form USSR style Joint tenancy Joint and indivisible No such a form

Figure C.1  The second general concurrent ownership form

republics. The fourth type provides only a barebone definition of a second concurrent ownership form. Finally, some jurisdictions, like Scotland (Miller et al. 2009: 447), enact idiosyncratic rules. See Figure C.1. Obviously, the preceding analysis has only sketched the rules regarding several types of concurrent ownership and has put aside spousal concurrent ownership. Future works would have to consider rules stipulated in the Book of Contract/Obligation and the Book of Family.

III Extension Due to space limitations, I elaborate on some of the possible extensions in Chang (2023 forthcoming-b). That said, before I close the book, I would like to note a few research topics that have been suggested but which I have not been able to pursue. My codebook and data can be expanded. Scholars from, say, Iceland and Saudi Arabia could code the property laws of their countries and merge their coding with mine. Students of Roman law could code Roman law, and use the dyadic correlation coefficient approach used in Chapter 1 to measure the lasting influence of Roman law. Researchers can compare international treaties and cross-country scholarly efforts like Europe’s Draft Common Framework of Reference with the jurisdictions in my data set, measuring which jurisdiction’s jurisprudence has the upper hand in the regional or global integration of law. Researchers from the studied jurisdictions can extend the coding to additional topics (see, e.g., above) within and beyond property law. A panel data, perhaps a snapshot of property law in each jurisdiction every ten years, could shed more light on the convergence and divergence issue. A data set on property law in

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

360

part iii: movable property

action and how property law has been enforced (for the constitutional law in action data, see Gutmann et al. 2022) would greatly complement my law-on-the-books data. My ten-year effort of understanding property law around the world culminates in this book. There were countless hurdles, but the fruits are sweet and I am to share these fruits with you. Welcome to the party.

https://doi.org/10.1017/9781009236553.018 Published online by Cambridge University Press

DATA APPENDIX

The codebook and data in Stata’s dta format, the data in Excel format, and program codes to run all the analysis reported in this book in Stata’s do format are publicly available in my personal website: http://yunchien .lawschool.cornell.edu/data. The data set will be expanded and new versions will be released when new information is available. For citation to the data set, please use the following format: Yun-chien Chang, Comparative Property Law Data Set, in Data Appendix to Property Law: Comparative, Empirical, and Economic Analyses (Cambridge University Press 2023), available at http://yunchien.lawschool .cornell.edu/data. Figures used in this book are available upon request. Colleagues are welcome to use these figures for teaching purposes, as long as proper citations are provided. Additional, unreported figures are also available on-line.

361

Published online by Cambridge University Press

DATA APPENDIX

The codebook and data in Stata’s dta format, the data in Excel format, and program codes to run all the analysis reported in this book in Stata’s do format are publicly available in my personal website: http://yunchien .lawschool.cornell.edu/data. The data set will be expanded and new versions will be released when new information is available. For citation to the data set, please use the following format: Yun-chien Chang, Comparative Property Law Data Set, in Data Appendix to Property Law: Comparative, Empirical, and Economic Analyses (Cambridge University Press 2023), available at http://yunchien.lawschool .cornell.edu/data. Figures used in this book are available upon request. Colleagues are welcome to use these figures for teaching purposes, as long as proper citations are provided. Additional, unreported figures are also available on-line.

361

https://doi.org/10.1017/9781009236553.019 Published online by Cambridge University Press

METHOD APPENDIX

Formal sources of law are the foundation for coding. If a jurisdiction has a civil code, I coded its property law based on nearly exclusively its civil code. Sometimes, if I was aware of highly important complementary statutes such as California’s Commercial Code and Code of Civil Procedure, they were included in the coding. I faced a conundrum: limiting my attention to civil codes only would be consistent but sometimes too narrow, while expanding to include all accessible statutes increases accuracy in the coding but is inconsistent. In this book, I go down the latter route, but recognize that it is impossible to be comprehensive.1 I was able to acquire civil codes in 105 jurisdictions. Not all jurisdictions have published official English translations of their civil codes. I thus relied on unofficial English translations (those published in print are listed in the References section of this book2), unofficial Chinese translation,3 and original text in Chinese, Dutch, English, French, German, Portuguese, Spanish, and Turkish to understand the codes. For countries and jurisdictions without civil codes, I coded applicable property law based on any available source. I started with stand-alone statutes (such as the Sale of Goods Act in Commonwealth countries), and then consulted treaties, case books, monographs, journal articles, country reports, etc. In a few instances, multiple sources have different takes on what the case law is. I do not have a rigid hierarchy of sources, but endeavor to find more sources and ask local experts for helps (they are thanked in the Acknowledgment). 1

2 3

Indonesia offers an example of why other statutes should be considered, as rules regarding immovables in the Indonesia Civil Code may be largely replaced by the Basic Agrarian Law (Undang-Undang No. 5 Tahun 1960 Tentang Peraturan Dasar PokokPokok Agraria) and other laws. I was not aware of this until shortly before the book is going to production. (I thank Gary Bell for this observation.) As I am not capable of figuring out which provisions of the Civil Code are still in force, I leave the coding of Indonesia as is – based on the provisions in the Civil Code. This shows the limitation of what one scholar could do. Other English translations of civil codes and other statutes are listed below. Prof. XU Guodong and his protégés translated a dozen civil codes into Chinese. They are listed below.

362

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

363

Coding property law based on other sources of law, notably cases in civillaw countries, is a meaningful endeavor. However, the understanding of law in action in private law should start with law on the books, and the starting point for judicial interpretations of statutes is law on the books. As the first project of its kind, it is a reasonable choice to start with formal sources of law.4 I hope readers can judge tenderly of me, in light of the difficulty of grasping with statutes themselves. With the sources of law at hand, a decade ago I read about a dozen civil codes and designed the codebook (Data Appendix). I hired research assistants in and outside of Taiwan (they are thanked in the Acknowledgment). The codebook has been constantly revised and expanded to reflect more sophisticated understanding of laws around the world. Research assistants answered the questions about substance of property law in the codebook and provided citations to statutes or academic writings to support the coding. For every jurisdiction, at least three research assistants have double-checked the sources and coding. Then, I checked the coding “horizontally” to see if there is any internal coding inconsistency within a jurisdiction. Finally, I checked the coding “vertically” to examine, in every given question, whether there is any cross-jurisdictional coding inconsistency. For missing values (i.e., questions which RAs reported that no sources provide answers), I spent extra efforts in browsing through the sources and searching with keywords to make sure that the missingness is indeed due to the silence in the sources. When I put the book manuscript together in 2021 and 2022, I decided to provide a citation to every data point used in this book. Upon doing so, I personally checked all the cited sources one last time. Many citations are already contained in this book. Due to space limitation, some property issues, though considered in the analysis in Chapter 1, can only be elaborated elsewhere (with detailed citations). See my personal website for a constantly updated list of such works.

A List of Statutes Referenced in the Book CIVIL CODE Afghanistan www-cdn.law.stanford.edu/wp-content/uploads/2015/10/Civil-Code-ofAfghanistan-ALEP-Translation.pdf

4

In a comparable project on competition law, Bradford and Chilton (2018; 2019), Bradford et al. (2019b), and Bradford et al. (2019a) also only focus on statutes, but not court cases. In another comparable project on trust law, Hofri-Winogradow (2020) also codes only statutes.

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

364

method appendix

Albania www.crca.al/sites/default/files/publications/Albanian%20Civil%20%20Code .pdf Algeria See information regarding the Chinese translation below. Angola Follow Portugal. Argentina See Reference. Armenia www.ilo.org/dyn/natlex/docs/ELECTRONIC/83647/92533/F1027892254/ ARM83647.pdf Austria See information regarding the Chinese translation below. Azerbaijan https://justice.gov.az/senedler/46 (official website; no direct link) http://jafbase.fr/docAsie/Azerbajian/Civil_code_eng.pdf Bahrain www.moj.gov.bh/en/ (official website; no direct link) https://bahrainbusinesslaws.com/laws/Civil-Law (direct link) Belarus See Reference. Belgium www.droitbelge.be/codes.asp#civ Bolivia https://bolivia.infoleyes.com/norma/821/codigo-civil-cc Brazil See Reference. Burkina Faso www.mindbank.info/item/2515 Burundi https://resourceequity.org/record/830-burundi-civil-code/ California https://leginfo.legislature.ca.gov/faces/codesTOCSelected.xhtml?tocCode=CI V&tocTitle=+Civil+Code+-+CIV Cambodia www.ilo.org/dyn/natlex/natlex4.detail?p_lang=en&p_isn=93489&p_ country=KHM&p_count=183 Cape Verde Follow Portugal.

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

365

Chile www.bcn.cl/leychile/N?i=172986&f=2015-10-22&p See also Chinese translation below. China http://en.npc.gov.cn.cdurl.cn/pdf/civilcodeofthepeoplesrepublicofchina.pdf Colombia https://vlex.com.co/vid/codigo-civil-43010756 Comoros https://resourceequity.org/record/838-comoros-civil-code/ Costa Rica See Reference. Cuba https://resourceequity.org/record/635-cuba-civil-code/ Czech Republic Refworld | Czech Republic: Law 89/2012, the Civil Code of the Czech Republic Dominican Republic https://wipolex.wipo.int/en/text/468693 Ecuador https://wipolex.wipo.int/en/text/513604 Egypt See information regarding the Chinese translation below. El Salvador www.acnur.org/fileadmin/Documentos/BDL/2002/1844.pdf Equatorial Guinea www.mindbank.info/item/2560 Eritrea http://rodra.co.za/images/countries/eritrea/legislation/Eritrea-Civil Code2015.pdf France www.legifrance.gouv.fr/codes/id/LEGITEXT000006070721/. See also Chinese translation below. Georgia www.lexadin.nl/wlg/legis/nofr/oeur/arch/geo/CIVILCODE.pdf Germany www.gesetze-im-internet.de/englisch_bgb/ See also Chinese translation below. Greece See Reference.

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

366

method appendix

Guatemala https://wipolex.wipo.int/en/legislation/details/2022 Guinea https://resourceequity.org/record/1157-guinea-civil-code/ Guinea-Bissau https://data.unicef.org/wp-content/uploads/2017/12/codigo_civil_guine_ bissau_e_legisl_complementar.pdf Haiti https://babel.hathitrust.org/cgi/pt?id=nyp.33433008531075&view=1up& seq=11 Honduras https://wipolex.wipo.int/en/text/475701 Hungary See Reference. Indonesia www.refworld.org/pdfid/3ffbd0804.pdf Iran See Reference. Iraq https://resourceequity.org/record/750-iraq-civil-code/ Italy https://resourceequity.org/record/1068-italy-civil-code/ For Chinese translation, see below. Ivory Coast www.ivoire-juriste.com (official website; no direct link) https://loidici.biz/2018/08/19/le-code-civil/lois-article-par-article/codes/ Japan www.moj.go.jp/content/000056024.pdf For Chinese translation, see below. Jordan See Reference. Kazakhstan See Reference. Kuwait See Reference. Kyrgyzstan The 2013 version appears to be no longer available online. Various other websites have later amended versions. See, for example, https://ihl-databases.icrc.org/applic/ihl/ihl-nat.nsf/implementingLaws.xsp?do cumentId=B60B0283E6EFE4ABC1256E690032E7F6&action=openDocum

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

367

ent&xp_countrySelected=KG&xp_topicSelected=GVAL-992BUA&from= state&SessionID=DXZXD3QZJT or https://wipolex.wipo.int/en/legislation/details/18809. Latvia https://vvc.gov.lv/image/catalog/dokumenti/civillikums.pdf Libya See Reference. Liechtenstein www.gesetze.li/konso/1003.001 Lithuania https://e-seimas.lrs.lt/portal/legalActPrint/lt?jfwid=32ocqrv7x&documentId= TAIS.400592&category=TAD Louisiana https://lcco.law.lsu.edu/ Luxembourg https://legilux.public.lu/eli/etat/leg/code/civil/20220101 Only the latest code is available here. The previous version I downloaded is now not available. Macau https://bo.io.gov.mo/bo/i/99/31/codcivcn/indice.asp Madagascar www.droit-afrique.com/upload/doc/madagascar/Madagascar-Code-2000Civil.pdf Malta https://legislation.mt/eli/cap/16/eng/pdf Mauritius www.ilo.org/dyn/natlex/docs/ELECTRONIC/88152/114145/F-172904586/ MUS88152%20Fre.pdf Mexico See Reference. Moldova www.ebrd.com/downloads/legal/core/moldova.pdf Monaco www.legimonaco.mc/305//legismclois.nsf/ViewCode!OpenView&Start=1&C ount=300&RestrictToCategory=CODE%20CIVIL Mongolia https://resourceequity.org/record/778-mongolia-civil-code/ Mozambique Follow Portugal. The Netherlands www.dutchcivillaw.com/civilcodegeneral.htm See also the Chinese translation below.

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

368

method appendix

Nicaragua https://wipolex.wipo.int/en/legislation/details/9762 Niger https://resourceequity.org/record/1799-niger-civil-code/ North Korea See information regarding the Chinese translation below. Panama See Reference. Paraguay www.oas.org/juridico/spanish/mesicic3_pry_ley1183.pdf Peru See Reference. Philippines See information regarding the Chinese translation below. Poland See Reference. Portugal See information regarding the Chinese translation below. Puerto Rico https://bvirtualogp.pr.gov/ogp/Bvirtual/leyesreferencia/PDF/C%C3% B3digos/48-1930.pdf Qatar www.almeezan.qa/LawView.aspx?opt&LawID=2559&language=en Quebec www.legisquebec.gouv.qc.ca/en/document/cs/ccq-1991 Romania www.euroavocatura.ro/noul_cod_civil_comentat.php Russia See Reference. Rwanda https://resourceequity.org/record/888-rwanda-civil-code/ São Tomé e Principe Follow Portugal. Senegal https://drs-sfd.gouv.sn/sitedrs/index.php/2018/01/05/nouveau-codeobligations-civiles-commerciales/ Seychelles www.africanchildforum.org/clr/Legislation%20Per%20Country/Seychelles/ seychelles_civilcode_1976_en.pdf

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

369

Slovakia www.vyvlastnenie.sk/predpisy/obciansky-zakonnik/ South Korea https://elaw.klri.re.kr/eng_service/lawView.do?hseq=29453&lang=ENG See information regarding the Chinese translation below. Spain www.mjusticia.gob.es/es/AreaTematica/DocumentacionPublicaciones/ Documents/Spanish_Civil_Code_(Codigo_Civil_Espanol).PDF Suriname https://ilo.org/dyn/natlex/natlex4.detail?p_lang=en&p_isn=96939&p_ country=SUR&p_count=67&p_classification=01.03&p_classcount=1 Switzerland www.fedlex.admin.ch/eli/cc/24/233_245_233/en See information regarding the Chinese translation below. Syria Code Civil Syrien. 1949. Damas: Bureau des documentations syriennes et arabe. Taiwan https://mojlaw.moj.gov.tw/ENG/LawContentE.aspx?LSID=FL001351 Tajikistan https://cis-legislation.com/document.fwx?rgn=2142 Thailand See information regarding the Chinese translation below. Timor-Leste www.ilo.org/dyn/natlex/natlex4.detail?p_lang=&p_isn=89755& p_country=TMP&p_count=58 Togo www.mindbank.info/item/2519 Turkey www.mevzuat.gov.tr/mevzuatmetin/1.5.4721.pdf Turkmenistan See Reference. See also the Chinese translation below. Ukraine See Reference. United Arab Emirates See Reference. Also available online: https://lexemiratidotnet.files.wordpress .com/2011/07/uae-civil-code-_english-translation_.pdf Uruguay https://parlamento.gub.uy/sites/default/files/CodigoCivil2010-02 .pdf?width=800&height=600&hl=en_US1&iframe=true&rel=nofollow

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

370

method appendix

Uzbekistan See Reference. Venezuela https://wipolex.wipo.int/en/text.jsp?file_id=130145 Vietnam See information regarding the Chinese translation below.

Statutes in List of Abbreviations ADA: Assignment of Debts Act (Israel) [Informed by RAs] AOORR: Act on Ownership and Other Real Rights (Croatia and Macedonia) Croatia http://pak.hr/cke/propisi,%20zakoni/en/OwnershipandOtherRealRights/ EN.pdf Macedonia www.libertas-institut.com/de/MK/nationallaws/law_on_ownership_and_ other_real_rights.pdf APA: Acquisition of Property Act (Botswana) https://leap.unep.org/countries/bw/national-legislation/acquisitionproperty-act-chapter-3210 ARA: The Access to Road Act (Uganda) https://land.igad.int/index.php/documents-1/countries/uganda/legislationand-policies-7/legislation-7/1351-access-to-roads-act-1969-ch-350/file BG: Baurechtsgesetz (Austria) www.ris.bka.gv.at/GeltendeFassung.wxe?Abfrage=Bundesnormen&Gesetzesn ummer=10001732 CA: Co-ownership Act (Finland, Norway and Sweden) [Lag (1904:48 s.1) om samäganderätt in Swedish; Lov om sameige in Norwegian] [180/1985 in Finland] https://lagen.nu/1904:48_s.1 CA: Conveyancing Act of 1919 (New South Wales, Australia) https://legislation.nsw.gov.au/view/html/inforce/current/act-1919-006 CA: Company Act of 2013 (India) www.mca.gov.in/Ministry/pdf/CompaniesAct2013.pdf CC: Commercial Code (California) https://leginfo.legislature.ca.gov/faces/codesTOCSelected.xhtml?tocCode=C OM&tocTitle=+Commercial+Code+-+COM CCLA: Contract and Commercial Law Act (New Zealand) www.legislation.govt.nz/act/public/2017/0005/21.0/DLM6844033.html

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

371

CDR: Code des Droits Reels (Tunisia) www.droit-afrique.com/upload/doc/tunisie/Tunisie-Code-2011-droits-reels.pdf CF: Code de la Famille (Senegal) http://jafbase.fr/docAfrique/docAfrique/Senegal/SenegalFam1.pdf CL: Companies Law (Israel) www.icnl.org/israel_companieslaw/ CPC: Civil Procedure Code (California; Macau; Poland) California https://leginfo.legislature.ca.gov/faces/codes_displayexpandedbranch.xhtm l?tocCode=ccp&division=&title=&part=2.&chapter=&article=&nodetre epath=5 Macau https://bo.io.gov.mo/bo/i/99/40/codprocivcn/default.asp Poland www.global-regulation.com/translation/poland/7049655/act-of-17november-1964%252c-the-code-of-civil-procedure.html CPO: Conveyancing and Property Ordinance (Cap. 219) (Hong Kong) www.elegislation.gov.hk/hk/cap219 CGSA: Civic Government (Scotland) Act 1982 (Scotland) www.legislation.gov.uk/ukpga/1982/45/contents CRE: Code of Real Estate (Finland) www.finlex.fi/fi/laki/kaannokset/1995/en19950540_19980964.pdf DL: Decreto Legislativo Nº 1400 (Peru) https://busquedas.elperuano.pe/normaslegales/decreto-legislativo-queaprueba-el-regimen-de-garantia-mobil-decreto-legislativo-n-14001689445-5/ DRA: Deeds Registries Act (Zimbabwe) https://media.zimlii.org/files/legislation/akn-zw-act-1959-10-eng2016-12-31.pdf EA: Easement Act of 1882 (India and Pakistan) https://leap.unep.org/countries/pk/national-legislation/easements-act1882-act-no-v-1882 ELC: Estate and Land Code [Ordonnance N°00-027/P-Rm du 22 Mars 2000 Portant Code Domanial et Foncier] (Mali) www.informea.org/zh-hans/node/204543 EV: Erbbaurechtsverordnung (Germany) www.gesetze-im-internet.de/erbbauv/index.html FA: Factors Act of 1994 (Singapore) https://sso.agc.gov.sg/Act-Rev/386/Published?DocDate=19940520& ProvIds=pr2FCA: Floating Charge Act of 1958 (Japan) https://elaws.e-gov.go.jp/document?lawid=333AC0000000106

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

372

method appendix

FPA: Factory Pledge Act (South Korea) See information regarding the Chinese translation below. FPL: Framework Pledge Law (Bosnia and Herzegovina) https://advokat-prnjavorac.com/legislation/Framework-Pledge-Law-Bosniaand-Herzegovina.pdf GLC: General Law of Contracts and Other Obligations (DR Congo) www.leganet.cd/Legislation/Droit%20obligations-contrats/Decret .30.07.1988.obl.htm ICA: Indian Contract Act (India) https://legislative.gov.in/sites/default/files/A1872-09.pdf IRA: Indian Registration Act, 1908 (India and Pakistan) India www.indiacode.nic.in/bitstream/123456789/13236/1/the_registration_ act%2C_1908.pdf Pakistan https://wpc.org.pk/wp-content/uploads/2020/02/Registration-Act1908.pdf ITTA: Indian Treasure Trove Act, 1878 (India and Pakistan) India www.indiaculture.nic.in/sites/default/files/Legislations/9.pdf Pakistan www.punjab-zameen.gov.pk/Documents/LawsAndRules/TREASURETROVE%20ACT,%201878.pdf LA: Land Act (Bhutan, Maldives, Norway, Tanzania, and Zambia) Bhutan www.humanitarianlibrary.org/resource/land-act-bhutan-2007-0 Maldives https://leap.unep.org/countries/mv/national-legislation/maldivianland-act-act-no-1-2002 Norway https://app.uio.no/ub/ujur/oversatte-lover/data/lov-19950512-023-eng .html Tanzania https://landportal.org/library/resources/lex-faoc053051/landamendment-act-2004-no-2-2004 Zambia www.parliament.gov.zm/node/945 LA: Limitation Act (Malaysia, Ontario of Canada, India (1963), Pakistan, Singapore, Uganda) Malaysia www.zulrafique.com.my/ckfinder/userfiles/files/legislation%20update/ Act%20A1566%20-%20Limitation%20(Amendment)%20Act%20 2018%20(01777415xA335E).pdf

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

373

Ontario of Canada www.ontario.ca/laws/statute/02l24 India www.commonlii.org/in/legis/cen/num_act/la1963133/ Pakistan www.ma-law.org.pk/pdflaw/Limitation%20%20Act%20-%201908.pdf Singapore https://sso.agc.gov.sg/Act/LA1959 Uganda https://ulii.org/akn/ug/act/ord/1958/46/eng%402000-12-31 LAA: Limitation of Actions Act (Victoria Province of Australia) www.legislation.vic.gov.au/in-force/acts/limitation-actions-act-1958/107 LBOPR: Law on Basis of Ownership and Proprietary Relations (Serbia) www.osce.org/serbia/34272 LC: Land Code (Brunei, Kazakhstan (2003), and Sweden) Brunei www.agc.gov.bn/AGC%20Images/LOB/pdf/Chp.40.pdf Kazakhstan https://cis-legislation.com/document.fwx?rgn=3672 Sweden https://resourceequity.org/record/1138-sweden-land-code/ LCLRA: Land and Conveyancing Law Reform Act 2009 (Ireland) www.irishstatutebook.ie/eli/2009/act/27/enacted/en/html LCT: Liberian Code, Title 29 (Property Law) www.liberlii.org/lr/legis/codes/plt29lcolr449/ LDRA: Land and Deeds Registration Act (Zambia) www.parliament.gov.zm/sites/default/files/documents/acts/Lands%20 and%20Deeds%20Registry%20Act.pdf Ley 1676: Ley 1676 de 2013 (Colombia) https://vlex.com.co/vid/promueve-cra-dito-dictan-as-mobiliarias456012058 LF: Loi Foncière (Democratic Republic of Congo) www.conaref-rdc.org/wp-content/uploads/2018/12/20151125094842-36_loi_ fonci%C3%A8re.pdf LGM: Ley de Garantías Mobiliarias N° 9246 (Costa Rica) https://vlex.co.cr/vid/540106342 LH: Loi Hypothecaire of 1851 (Belgium) www.ejustice.just.fgov.be/cgi_loi/change_lg.pl?language=fr&la=F&cn=185112 1601&table_name=loi LL: Land Law, 5729-1969 (Israel) See Reference. LLLS: Limitation Law of Lagos State (Nigeria) https://lawnigeria.com/2019/10/limitation-law-of-lagos-state/

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

374

method appendix

LO: Limitation Ordinance (Hong Kong) www.elegislation.gov.hk/hk/cap347!en-zh-Hant-HK Loi 2013: Loi n° 2013-01 portant code foncier et domanial (Benin) www.droit-afrique.com/upload/doc/benin/Benin-Code-foncier-domanial2013.pdf LoP: Law on Pledge (Moldova) www.ebrd.com/downloads/legal/core/moldovals.pdf LoPA: Law on Property Act (Slovenia) www.ebrd.com/downloads/legal/core/slolom.pdf LoPA: Law of Property Act 1925 (England) www.legislation.gov.uk/ukpga/Geo5/15-16/20 LP: Law on Pledge (Kyrgyzstan) www.libertas-institut.com/de/Mittel-Osteuropa/Law%20on%20Pledge.pdf LPA: Lost Property Act (Denmark and Norway) Denmark (Hittegodsloven) www.retsinformation.dk/eli/lta/2014/879 Norway (Lov om hittegods [hittegodslova]) www.global-regulation.com/translation/norway/5961844/law-on-lostproperty-%255blost-act%255d.html LPA: Law of Property Act 1925 (England) www.legislation.gov.uk/ukpga/Geo5/15-16/20/contents LRA: Land Registration Act 2012 (Scotland) www.legislation.gov.uk/asp/2012/5/contents/enacted LRA: Land Registration Act (Norway) www.regjeringen.no/en/dokumenter/act-land-registration/id455461/ LRA: Land Registration Act, Chapter 334 (Tanzania) https://tanzania.eregulations.org/media/The%20Land%20Registration%20 Act.%20Cap%20334.pdf LRERS: Law No. 14 of 1964 on the Real Estate Registration System (Qatar) www.almeezan.qa/LawView.aspx?opt&LawID=2506&TYPE=PRINT&langua ge=en LRL: Land Reform Law Cap. 297 (Sri Lanka) www.commonlii.org/lk/legis/consol_act/lr297160.pdf LRO: Land Registration Ordinance (Hong Kong) www.elegislation.gov.hk/hk/cap128!en-zh-Hant-HK LRRP: Law on Registrable Real Property (Morocco) [la législation applicable aux immeubles immatriculés (B.O. 7 juin 1915)] https://bassamat-laraqui.com/code/dahir-19-rejeb-1333-fixant-la-legislationapplicable-aux-immeubles-immatricules-b-o-7-juin-1915/?pdf=4245 LTA: Land Transfer Act 1952 (New Zealand) www.legislation.govt.nz/act/public/2002/0011/latest/whole.html# DLM140144

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

375

LTA: Land Titles Act (Singapore; Ontario) Singapore https://sso.agc.gov.sg/Act/LTA1993 Ontario www.ontario.ca/laws/statute/90l05 MCL: Movable Collateral Law [Ley de Garantías Mobiliarias] (Honduras) http://felaban.s3-website-us-west-2.amazonaws.com/regulaciones/ archivo20140717152241PM.pdf MIPA: Moveable and Immovable Property Act (Bhutan) http://oag.gov.bt/wp-content/uploads/2010/05/Moveable-and-ImmovableProperty-Act-of-the-Kingdom-of-Bhutan-English-version.pdf MPL: Movable Property Law (Israel) See Reference. NLC: National Land Code (Malaysia) www.miea.com.my/sites/default/files/webmaster/National%20Land%20 Code%20Act%20828%20_Pewartaan__15Oct2020.pdf OA: Ownership Act (Bulgaria) https://leap.unep.org/countries/bg/national-legislation/ownership-act OLRS: Order relating to Land Registration System (Madagascar) [ORDONNANCE N°60-146 DU 3 OCTOBRE 1960 relative au régime foncier de l’immatriculation (J.O.R.M. n° 129 du 22.10.60, p.2205)] www.droit-afrique.com/upload/doc/madagascar/Madagascar-Ordonannce1960-146-regime-foncier.pdf PA: Prescription Act (Zimbabwe) https://old.zimlii.org/zw/legislation/act/1975/31 PL: Partition Law (No. 21 of 1977) (Sri Lanka) www.commonlii.org/lk/legis/num_act/pl21o1977192/ PL: Property Law of 1990 (Laos) Laos https://overseas.mofa.go.kr/la-en/brd/m_1889/view.do?seq=753494&s rchFr=&srchTo=&srchWord=&srchTp=&mu lti_itm_seq=0&itm_seq_1=0&itm_seq_2=0&company_ cd=&company_nm= PL: Prescription Law of 1958 (Israel) See Reference. PLA: Property Law Act (Australia 1974; New Zealand 2007) Australia www.austlii.edu.au/au/legis/qld/consol_act/pla1974179/ New Zealand www.legislation.govt.nz/act/public/2007/0091/latest/DLM968962 .html?search=qs_act%40bill%40regulation%40deemedreg_+Law+of+Pr operty+Act+1925+_resel_25_h&p=1&sr=1

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

376

method appendix

PLA: Prescription and Limitation Act 1973 (Scotland) www.legislation.gov.uk/ukpga/1973/52 PO: Partition Ordinance (Hong Kong) www.elegislation.gov.hk/hk/cap352 PPIP: The Punjab Partition of Immovable Property 2012 (Pakistan) http://punjablaws.gov.pk/laws/2528.html PPL: Personal Property Law (New York) https://newyork.public.law/laws/n.y._personal_property_law PPSTA: Personal Property Secured Transactions Act (Taiwan) https://law.moj.gov.tw/ENG/LawClass/LawAll.aspx?pcode=G0380024 PRC: Public Resource Code (California) https://leginfo.legislature.ca.gov/faces/codesTOCSelected.xhtml? tocCode=PRC PRL: Property Registration Act (Bahrain) https://bahrainbusinesslaws.com/laws/Property-Registration-Law PsL: Pledges Law 5727-1967 (Israel) See Reference. RA: Rentcharges Act (United Kingdom) www.legislation.gov.uk/ukpga/1977/30/contents RERL: Real Estate Registration Law (Dominican Republic) www.academia.edu/10092465/Normativa_de_la_Jurisdicci%C3%B3n_ Inmobiliaria RLA: Registered Land Act (Malawi) http://housingfinanceafrica.org/app/uploads/Malawi-Registered-LandAct-1967.pdf RLPL: Restoration of Lost Property Law 5733-1973 [Hok Hashavat Ave’da] (Israel) See Reference. RPA: Real Property Act (New South Wales Province of Australia) https://legislation.nsw.gov.au/view/html/inforce/current/act-1900-025 RPAPL: Real Property Actions and Proceedings Law (New York) https://newyork.public.law/laws/n.y._real_property_actions_and_ proceedings_law RPL: Real Property Law (New York) https://newyork.public.law/laws/n.y._real_property_law RTA: Registration of Titles Act (Uganda) https://ulii.org/akn/ug/act/ord/1922/22/eng%402011-09-02 SD: Sultani Decree 1980 (Oman) See Reference. SGA: Sale of Goods Act (all common-law jurisdictions) Australia www.legislation.act.gov.au/DownloadFile/a/1954-15/current/PDF/ 1954-15.PDF

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

377

Canada, excluding Quebec Ontario Sale of Goods Act www.ontario.ca/laws/statute/90s01#BK24 Nova Scotia Sale of Goods Act https://nslegislature.ca/sites/default/files/ legc/statutes/salegood.htm New Brunswick Sale of Goods Act http://laws.gnb.ca/en/showfulldoc/ cs/2016-c.110/#anchorga:l Manitoba Sale of Goods Act http://web2.gov.mb.ca/laws/statutes/ccsm/ s010e.php British Columbia Sale of Goods Act www.bclaws.ca/civix/document/id/ complete/statreg/96410_01 Prince Edward Island Sale of Goods Act www.princeedwardisland .ca/sites/default/files/legislation/S-01-Sale%20Of%20Goods%20Act.pdf Saskatchewan Sale of Goods Act www.qp.gov.sk.ca/documents/English/ Statutes/Statutes/S1.pdf Alberta Sale of Goods Act www.qp.alberta.ca/1266.cfm?page=S02 .cfm&leg_type=Acts&isbncln=9780779765874 Newfoundland and Labrador Sale of Goods Act http://assembly.nl.ca/ Legislation/sr/statutes/s06.htm Northwest Territories Sale of Goods Act www.justice.gov.nt.ca/en/files/ legislation/sale-of-goods/sale-of-goods.a.pdf Yukon Sale of Goods Act www.gov.yk.ca/legislation/acts/sago_c.pdf Nunavut Sale of Goods Act www.nunavutlegislation.ca/en/file-download/download/public/198 England www.legislation.gov.uk/ukpga/1979/54 Hong Kong www.elegislation.gov.hk/hk/cap26!en India https://indiankanoon.org/doc/651105/. Ireland www.irishstatutebook.ie/eli/1893/act/71/enacted/en/print Malawi www.malawilii.org/akn/mw/act/1967/14/eng@2014-12-31 Malaysia https://simplymalaysia.files.wordpress.com/2011/10/act-382-sale-of-goodsact-1957.pdf New Zealand www.legislation.govt.nz/act/public/1908/0168/latest/whole.html Nigeria The Sale of Goods Act 1893 has been re-enacted by states in Nigeria. See, for­ example, https://lawnigeria.com/2019/10/sale-of-goods-law-of-lagos-state-2/

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

378

method appendix

Pakistan www.ma-law.org.pk/pdflaw/Sale%20of%20Goods%20Act,%201930.pdf. Scotland www.legislation.gov.uk/ukpga/1979/54. Singapore https://sso.agc.gov.sg/Search/Content?Phrase=Sale%20of%20Goods%20 Act&PhraseType=AllTheseWords&In=InForce_Act_SL&Within=title Tanzania www.jamiiforums.com/attachments/the-sales-of-goods-act-cap-214-r-e2002-pdf.322303/ Uganda www.ulii.org/akn/ug/act/ord/1930/28/eng@2000-12-31 SGSSA: Sale of Goods and Supply of Services Act (Uganda)

https://ulii.org/akn/ug/act/2018/10/eng%402018-08-17 SGO: Sale of Goods Ordinance (Hong Kong) www.elegislation.gov.hk/hk/cap26 SL: Sale Law 1968 (Israel) www.israelinsurancelaw.com/sale-law-1968-2/ SL: Statute of Limitations (Ireland, New Zealand)

Ireland www.irishstatutebook.ie/eli/1957/act/6/enacted/en/html New Zealand www.legislation.govt.nz/act/public/2010/0110/latest/DLM2033120.html SR: Sachenrecht (Liechtenstein)

www.gesetze.li/konso/1923.004 TCA: Title Conditions Act 2003 (Scotland) www.legislation.gov.uk/asp/2003/9/contents TLPA: The Law of Property Act (Estonia) www.ebrd.com/downloads/legal/core/estlom.pdf TPA: Transfer of Property Act of 1882 (Myanmar and Pakistan)

Myanmar https://myanmar-law-library.org/topics/myanmar-property-law/thetransfer-of-property-act-1882.html Pakistan https://pja.gov.pk/transferofpropertyact TTA: Treasure Trove Act 1957 (Revised 1995) (Malaysia) www.commonlii.org/my/legis/consol_act/tta19571995270/ UCC: Uniform Commercial Code (states in the United States) www.law.cornell.edu/ucc UPHPA: Uniform Partition of Heirs Property Act (New York) www.nysenate.gov/legislation/laws/RPA/993

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

method appendix

379

A List of Chinese Translation of Civil Codes Used in This Book 于海涌、赵希璇:《瑞士民法典》,法律出版社2016年版。 尹田:《阿尔及利亚民法典》,中国法制出版社2002年版。 王书江:《日本民法典》,中国法制出版社2000年版。 王卫国、胡利玲、吴民许、陈龙江、喜佳:《荷兰民法典(第3、5、6 编)》,中国政法大学出版社2006年版。 伍光红、黄氏惠:《越南民法典》,商务印书馆2018年版。 李飞:《马耳他民法典》,厦门大学出版社2012年版。 周友军、杨垠红:《奥地利普通民法典》,清华大学出版社2013年版。 周喜梅:《泰王国民商法典》,中国法制出版社2013年版。 金玉珍:《韩国民法典、朝鲜民法》,北京大学出版社2009年版。 唐晓晴、曹锦俊、关冠雄、刘志强、艾林芝:《葡萄牙民法典》,北京 大学出版社2009年版。 徐涤宇:《智利共和国民法典》,金桥文化出版(香港)有限公司2002 年版。 徐涤宇:《智利共和国民法典》,北京大学出版社2014年版。 崔吉子:《韩国最新民法典》,北京大学出版社2010年版。 陈国柱:《意大利民法典》,中国人民大学出版社2010年版。 陈卫佐:《德国民法典》,法律出版社2010年版。 费安玲、丁玫、张宓:《意大利民法典》,中国政法大学出版社2004年 版。 黄文煌:《埃及民法典》,厦门大学出版社2008年版。 黄道秀:《俄罗斯联邦民法典》,北京大学出版社2007年版。 臺灣大學法律學院德國民法編譯委員會:《德國民法(上)—— 總則編、債編、物權編》,元照出版有限公司2016年版。 齐云:《巴西最新民法典》,中国法制出版社2009年版。 潘灯、马琴:《西班牙民法典》,中国政法大学出版社2013年版。 蒋军洲:《菲律宾民法典》,厦门大学出版社2011年版。 戴永盛:《奥地利普通民法典》,中国政法大学出版社2016年版。 戴永盛:《瑞士民法典》,中国政法大学出版社2016年版。 薛军:《埃塞俄比亚民法典》,中国法制出版社2002年版。 魏磊杰、朱渺、杨秋颜:《土库曼斯坦民法典》,厦门大学民法典2016 年版。 罗结珍:《法国民法典》,北京大学出版社2010年版。

https://doi.org/10.1017/9781009236553.020 Published online by Cambridge University Press

REFERENCES

Adal, Erhan. 2012. Fundamentals of Turkish Private Law. 10th ed. Istanbul: Legal Yaymcihk San. ve Tic. Ltd. St. Adler, Matthew D., and Eric A. Posner. 2006. New Foundations of Cost-Benefit Analysis. Cambridge, MA: Harvard University Press. Ai, Linzhi. 2013. Property Law of Macau. Macao SAR, China: Social Sciences Academic Press (China) [Written in Chinese]. Akkermans, Bram. 2008. The Principle of Numerus Clausus in European Property Law. Antwerp: Intersentia. 2017. The Numerus Clausus of Property Rights. In Comparative Property Law: Global Perspectives 100–120, edited by Michele Graziadei and Lionel Smith. Cheltenham, UK: Edward Elgar. Alessandri, Arturo, and Manuel Somarriva. 1974. Curso de Derecho Civil. 3rd ed. Santiago, Chile: Editorial Nascimento. Alexander, Gregory S. 2008. The Social-Obligation Norm in American Property Law. Cornell Law Review 94:745–819. 2016. The Sporting Life: Democratic Culture and the Historical Origins of the Scottish Right to Roam. University of Illinois Law Review 2016:321–370. Allen, Douglas W. 2000. Transaction Costs. In Encyclopedia of Law and Economics, Volume I. The History and Methodology of Law and Economics 893–926, edited by Boudewijn Bouckaert and Gerrit De Geest. Northampton, MA: Edward Elgar. Alston, Lee, and Bernardo Mueller. 2015. Towards a More Evolutionary Theory of Property Rights. Iowa Law Review 100:2255–2273. Alterini, Ignacio Ezequiel. 2018. Law of Property. In Introduction to the Law of Argentina 197–215, edited by Ursula Basset. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Anderson, Siwan. 2018. Legal Origin and Female HIV. American Economic Review 108 (6):1407–1439. Ansay, Tugrul. 2020. Law of Property. In Introduction to Turkish Law, edited by Tugrul Ansay, Don Wallace Jr. and Işık Önay. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Aristi, Rafael Sánchez, and Nieves Moralejo Imbernón. 2013. Property and Trust Law in Spain. Alphen aan den Rijn, the Netherlands: Wolters Kluwer.

380

Published online by Cambridge University Press

references

381

Armour, John, Simon Deakin, Priya Lele, and Mathias Siems. 2009. How Do Legal Rules Evolve? Evidence from Cross-National Comparison of Shareholder, Creditor and Worker Protection. American Journal of Comparative Law 57:579–630. Arruñada, Benito. 2012. Institutional Foundation of Impersonal Exchange: Theory and Policy of Contractual Registries. Chicago, IL: The University of Chicago Press. 2017. Property as Sequential Exchange: The Forgotten Limits of Private Contract. Journal of Institutional Economics 13 (4):753–783. Arruñada, Benito, and Nuno Garoupa. 2005. The Choice of Titling System in Land. Journal of Law and Economics 48 (2):709–727. Arruñada, Benito, Giorgio Zanarone, and Nuno Garoupa. 2019. Property Rights in Sequential Exchange. Journal of Law, Economics, and Organization 35 (1):127–153. Arruñada, Benito, Marco Fabbri, and Michael G. Faure. 2022. Land Titling and Litigation. Journal of Law & Economics 65 (1):131–156. Avraham, Ronen. 2004. Modular Liability Rules. International Review of Law and Economics 24 (3):269–297. Aynès, Laurent. 2008. Property Law. In Introduction to French Law 147–170, edited by George A. Bermann and Etienne Picard. Hague, the Netherlands: Kluwer Law. Ayres, Ian. 1998. Protecting Property with Puts. Valparaiso University Law Review 32 (3):793–832. 2005. Optional Law: The Structure of Legal Entitlements. Chicago, IL: The University of Chicago Press. 2006. Menus Matter. The University of Chicago Law Review 73:3–15. 2011. Regulating Opt-Out: An Economic Theory of Altering Rules. Yale Law Journal 121:2032–2116. Ayres, Ian, and J. M. Balkin. 1996. Legal Entitlements as Auctions: Property Rules, Liability Rules, and Beyond. Yale Law Journal 106:703–750. Ayres, Ian, and Robert Gertner. 1989. Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules. Yale Law Journal 99:87–130. 1999. Majoritarian vs. Minoritarian Defaults. Stanford Law Review 51:1591–1613. Ayres, Ian, and Paul M. Goldbart. 2001. Optimal Delegation and Decoupling in the Design of Liability Rules. Michigan Law Review 100:1–79. Ayres, Ian, and Kristin Madison. 1999. Threatening Inefficient Performance of Injunctions and Contracts. University of Pennsylvania Law Review 148 (1):45–108. Ayres, Ian, and Eric Talley. 1995. Solomonic Bargaining: Dividing a Legal Entitlement to Facilitate Coasean Trade. Yale Law Journal 104:1027–1117. Baker, Matthew, Thomas J. Miceli, C. F. Sirmans, and Geoffrey K. Turnbull. 2002. Optimal Title Search. The Journal of Legal Studies 31 (1):139–158.

Published online by Cambridge University Press

382

references

Bakibinga, David Justin. 2018. Property and Trust Law in Uganda. Alphen aan den Rijn: Wolters Kluwer. Balganesh, Shyam, Ted Sichelman, and Henry E. Smith, eds. 2022. Wesley Hohfeld a Century Later: Edited Work, Select Personal Papers, and Original Commentaries. Cambridge, UK: Cambridge University Press. Balganesh, Shyamkrishna. 2016. Copyright and Good Faith Purchasers. California Law Review 104:269–322. Bar-Gill, Oren, and Lucian Arye Bebchuk. 2010. Consent and Exchange. The Journal of Legal Studies 39 (2):375–397. Bar-Gill, Oren, and Nicola Persico. 2016. Exchange Efficiency with Weak Ownership Rights. American Economic Journal: Microeconomics 8 (4):230–267. 2019. Bounded Rationality and the Theory of Property. Notre Dame Law Review 94 (3):1019–1057. Barua, Mariana. 2018. Property and Trust Law in Uruguay. Alphen aan den Rijn the Netherlands: Wolters Kluwer. Barzel, Yoram. 1997. Economic Analysis of Property Rights. 2nd ed. Cambridge, UK: Cambridge University Press. Baucells, Manel, and Steven A. Lippmant. 2001. Justice Delayed Is Justice Denied: A Cooperative Game Theoretic Analysis of Hold-Up in Co-Ownership. Cardozo Law Review 22:1191–1250. Baxter, William F., and Lillian R. Altree. 1972. Legal Aspects of Airport Noise. Journal of Law & Economics 15:1–113. Bebchuk, Lucian Arye. 2001. Property Rights and Liability Rules: The Ex Ante View of the Cathedral. Michigan Law Review 100:601–639. Bell, Abraham. 2009. Private Takings. University of Chicago Law Review 76:517–585. 2015. Title in the Shadow of Possession. In Law and Economics of Possession 320– 335, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. Bell, Abraham, and Gideon Parchomovsky. 2002. Pliability Rules. Michigan Law Review 101:1–79. 2005. A Theory of Property. Cornell Law Review 90:531–615. 2007. Taking Compensation Private. Stanford Law Review 59:871–906. 2012a. The Case for Imperfect Enforcement of Property Rights. University of Pennsylvania Law Review:1927–1954. 2012b. The Privacy Interest in Property. University of Pennsylvania Law Review 167:869–920. 2014. Governing Communities by Auction. The University of Chicago Law Review 81:1–26. 2017. Partial Takings. Columbia Law Review 117:2043–2093. Bell, George Joseph, and William Guthrie. 1899. Principles of the Law of Scotland 10th ed. Edinburgh, UK: T. & T. Clark (available on Hein Online). Belli, Melvin M. 1959. Japanese Law. Hastings Law Journal 11:130–149.

Published online by Cambridge University Press

references

383

Ben-Shahar, Omri. 1997. Property Rights in Stolen Goods: An Economic Analysis. working paper (on file with author). Ben-Shahar, Omri, and Lisa Bernstein. 2000. The Secrecy Interest in Contract Law. The Yale Law Journal 109 (8):1885–1925. Ben-Shahar, Omri, and Ariel Porat. 2016. Personalizing Negligence Law. New York University Law Review 91:627–688. 2021. Personalized Law: Different Rules for Different People. Oxford, UK: Oxford University Press. Berkowitz, Daniel, Katharina Pistor, and Jean-Francois Richard. 2003a. Economic Development, Legality, and the Transplant Effect. European Economic Review 47 (1):165–195. 2003b. The Transplant Effect. American Journal of Comparative Law 51 (1):163–204. Blume, Lawrence, and Daniel L. Rubinfeld. 1984. Compensation for Takings: An Economic Analysis. California Law Review 72 (4):569–628. Blume, Lawrence, Daniel L. Rubinfeld, and Perry Shapiro. 1984. The Taking of Land: When Should Compensation Be Paid? The Quarterly Journal of Economics 99 (1):71–92. Bradbrook, A. J. 1983. Access to Landlocked Land: A Comparative Study of Legal Solutions. Sydney Law Review 10 (1):39–75. Bradford, Anu, Yun-chien Chang, Adam Chilton, and Nuno Garoupa. 2021. Do Legal Origins Predict Legal Substance? The Journal of Law and Economics 64 (2):207–231. Bradford, Anu, and Adam Chilton. 2018. Competition Law around the World from 1889 to 2010: The Competition Law Index. Journal of Competition Law & Economics 14 (3):393–432. 2019. Trade Openness and Antitrust Law. Journal of Law and Economics 62 (1):29–65. Bradford, Anu, Adam Chilton, Katerina Linos, and Alex Weaver. 2019a. The Global Dominance of European Competition Law Over American Antitrust Law. Journal of Empirical Legal Studies 16 (4):731–766. Bradford, Anu, Adam S. Chilton, Chris Megaw, and Nathaniel Sokol. 2019b. Competition Law Gone Global: Introducing the Comparative Competition Law and Enforcement Datasets. Journal of Empirical Legal Studies 16 (2):411–443. Brady, Maureen E. 2018. The Forgotten History of Metes and Bounds. Yale Law Journal 128:872–953. Brandsma, Frits. 2015. Actions in Roman and Civil Law for the Protection of Immovables. In Protection of Immovables in European Legal Systems 9–30, edited by Sonia Martin Santisteban and Peter Sparkes. Cambridge, UK: Cambridge University Press. Bridge, Michael, Louise Gullifer, Kevin F. K. Low, and McMeel Gerard. 2018. The Law of Personal Property. 2nd ed. London, UK: Thomson Reuters.

Published online by Cambridge University Press

384

references

Brooks, Richard R. W. 2002. The Relative Burden of Determining Property Rules and Liability Rules: Broken Elevators in the Cathedral. Northwestern University Law Review 97:267–317. Brooks, Richard R. W., Claudia M. Landeo, and Kathryn E. Spier. 2010. Trigger Happy or Gun Shy? Dissolving Common‐Value Partnerships with Texas Shootouts. The RAND Journal of Economics 41 (4):649–673. Brooks, Richard R. W., and Warren F. Schwartz. 2005. Legal Uncertainty, Economic Efficiency, and the Preliminary Injunction Doctrine. Stanford Law Review 58:381–410. Brooks, Richard R. W., and Carol Rose. 2013. Saving the Neighborhood: Racially Restrictive Covenants, Law, and Social Norms. Cambridge, MA: Harvard University Press. Brown, David. 2017. Co-Ownership. In New Zealand Land Law 503–570, edited by Elizabeth Toomey. Wellington, New Zealand: Thomson Reuters. Bruce, Jon W., and James W. Ely, Jr. 2011. The Law of Easements and Licenses in Land. New York, NY: West. Brzozowski, Adam. 2005. Property. In Introduction to Polish Law 54–69, edited by Stanislaw Frankowski. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Buchanan, James M., and Gordon Tullock. 1965. The Calculus of Consent: Logical Foundations of Constitutional Democracy. Ann Arbor, MI: University of Michigan Press. Burnham, William, Peter B. Maggs, and Gennady M. Danilenko. 2009. Law and Legal System in the Russian Federation. 4th ed. Huntington, NY: Juris Publishing. Bustillos, Espíndola, and Miguel Ángel. 2009. La Propiedad Inmobiliaria en México: Su Esencia Y Extensión. Revista Mexicana de Derecho 11:91–172. Butler, William E., ed. 1999. Turkmenistan Civil Code of Saparmurat Turkmenbashi. London, UK: Wildy, Simmonds & Hill Publishing. ed. 2000. Civil Code of the Republic Belarus. London, UK: Wildy, Simmonds & Hill Publishing. ed. 2002. Civil Code of the Russian Federation Parts One, Two, and Three: Edited, Compiled, and Translated from the Russian, with an Introduction. Oxford, UK: Oxford University Press. ed. 2007. Civil Code of the Republic Uzbekistan. London, UK: Wildy, Simmonds & Hill Publishing. ed. 2008. Civil Code of the Republic of Kazakhstan. London, UK: Wildy, Simmonds & Hill Publishing. ed. 2011. Civil Code of Ukraine and Law of Ukraine on Private International Law. London, UK: Wildy, Simmonds & Hill Publishing. Calabresi, Guido, and A. Douglas Melamed. 1972. Property Rules, Liability Rules, and Inalienability: One View of the Cathedral. Harvard Law Review 85:1089–1128.

Published online by Cambridge University Press

references

385

Carbonara, Emanuela, Alice Guerra, and Francesco Parisi. 2016. Sharing Residual Liability: The Cheapest Cost Avoider Revisited. The Journal of Legal Studies 45 (1):173–201. Casagrande, John G. 1985. Acquiring Property through Forced Partitioning Sales: Abuses and Remedies. Boston College Law Review 27:755–784. Casey, Anthony J., and Anthony Niblett. 2016. The Death of Rules and Standards. Indiana Law Journal 92:1401–1448. Casner, A. James, and W. Barton Leac. 1950. Cases and Text on Property. Boston, MA: Little, Brown. Cauffman, Caroline, and Vincent Sagaert. 2011. National Report on the Transfer of Movables in Belgium. In National Reports on the Transfer of Movables in Europe, Volume 4: France, Belgium, Bulgaria, Poland, Portugal 189–351, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Chambers, Robert. 2013. An Introduction to Property Law in Australia. 3rd ed. Sydney, Australia: Lawbook. Chang, Yun-chien. 2010. An Empirical Study of Compensation Paid in Eminent Domain Settlements: New York City 1990–2002. Journal of Legal Studies 39:201–244. 2011. An Empirical Study of Court-Adjudicated Takings Compensation in New York City: 1990–2003. Journal of Empirical Legal Studies 8 (2):384–412. 2012a. Economic Value or Fair Market Value: What Form of Takings Compensation Is Efficient? Supreme Court Economic Review 20:35–88. 2012b. Property Law with Chinese Characteristics: An Economic and Comparative Analysis. Brigham-Kanner Property Rights Conference Journal 1:345–372. 2012c. Self-Assessment of Takings Compensation: An Empirical Study. Journal of Law, Economics & Organization 28:265–285. 2012d. Tenancy in “Anticommons”? A Theoretical and Empirical Analysis of Co-ownership. Journal of Legal Analysis 4:515–553. 2013. Private Property and Takings Compensation: Theoretical Framework and Empirical Analysis. Northampton, MA: Edward Elgar. 2014. To Tear Down or Not to Tear Down? An Empirical Study of Boundary Encroachment Cases in Taiwan. In Empirical Legal Analysis: Assessing the Performance of Legal Institutions 144–158, edited by Yun-chien Chang. London, UK: Routledge. 2015a. An Economic and Comparative Analysis of Specificatio (the Accession Doctrine). European Journal of Law and Economics 39 (2):225–243. 2015b. The Economy of Concept and Possession. In Law and Economics of Possession 103–125, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. 2015c. Optional Law in Property: Theoretical Critiques. NYU Journal of Law and Liberty 9 (2):459–512.

Published online by Cambridge University Press

386

references

2016a. The Evolution of Property Law in Taiwan: An Unconventional Interest Group Story. In Private Law in China and Taiwan: Economic and Legal Analyses 212–244, edited by Yun-chien Chang, Wei Shen and Wen-yeu Wang. Cambridge, UK: Cambridge University Press. 2016b. Hybrid Rule: Hidden Entitlement Protection Rule in Access to Landlocked Land Doctrine. Tulane Law Review 88:217–257. 2016c. The Problematic Concept of Possession in DCFR: Lessons from Law and Economics of Possession. European Journal of Property Law 5 (1):4–25. 2017. Eminent Domain Law in Taiwan: New Law, Old Practice? In Eminent Domain: A Comparative Perspective 93–117, edited by Iljoong Kim, Hojun Lee and Ilya Somin. Cambridge, UK: Cambridge University Press. 2018a. Incorporated Organizations and Unincorporated Organizations according to the Theory of Asset Partitioning: A Missing Piece in General Provisions of Civil Code in China. Peking University Law Journal 30 (1):59–83 [in Chinese]. 2018b. Wealth Transfer Laws in 153 Jurisdictions: An Empirical Comparative Law Approach. Iowa Law Review 102 (5):1915–1944. 2020a. The Good-Faith Purchase Doctrine in 247 Jurisdictions. European Property Law Journal 9 (2–3):133–156. 2020b. Interpreting Private Law: A Social Scientific Approach. Taipei, Taiwan: New Sharing [in Chinese]. 2020c. Judges Avoid Ex Post but Not Ex Ante Inefficiency: Theory and Empirical Evidence from Taiwan. In Selection and Decision in Judicial Process around the World: Empirical Inquires 160–178, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. 2020d. The Property Structure of Rural Cultivated Land in China: Rethinking Membership Right and Tripartite Property Right. Nanjing University Law Journal 1 (1):81–100 [in Chinese]. 2021. Economic Analysis of Law: Methodology with Applications to Property Law. Taipei, Taiwan: Angle [in Chinese]. 2022a. Adverse Possession Laws in 203 Jurisdictions: Proposals for Reform. University of Pennsylvania Journal of International Law 43 (2):373–428. 2022b. Empirical Legal Studies: Principles, Methods, and Applications. 2nd ed. Taipei, Taiwan: New Sharing [in Chinese]. 2022c. Property Rights: (Probabilistically) Necessary or Sufficient for Economic Development in China and Beyond? Law & Social Inquiry 47 (2):727–733. 2023 forthcoming-a. An Empirical Comparative Analysis of Co-ownership Law. Renmin University Law Review [in Chinese]. 2023 forthcoming-b. Property and Comparative Empirical Legal Studies. In Research Agenda in Property Law, edited by Bram Akkermans. Northampton, MA: Edward Elgar. Chang, Yun-chien, Kong-pin Chen, and Yu-sheng Liu. 2016. The Unlawful Possession of Land and Unjust Enrichment Equivalent to Rent: Empirical

Published online by Cambridge University Press

references

387

Analysis and Policy Suggestions. Chengchi Law Review (144):81–153 [in Chinese]. Chang, Yun-chien, Weitseng Chen, and Ying Chieh Wu. 2022. Property and Trust Law in Taiwan. 2nd ed. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Chang, Yun-chien, and Lee Anne Fennell. 2014a. Appendices to Partition and Revelation. The University of Chicago Law Review Dialogue 81 (1):1–13. 2014b. Partition and Revelation. The University of Chicago Law Review 81 (1):27–51. Chang, Yun-chien, Nuno Garoupa, and Martin T. Wells. 2021. Drawing the Legal Family Tree: An Empirical Comparative Study of 170 Dimensions of Property Law in 129 Jurisdictions. Journal of Legal Analysis 13 (1):231–282. Chang, Yun-chien, and William H. J. Hubbard. 2021. New Empirical Tests for Classic Litigation Selection Models: Evidence from a Low Settlement Environment. American Law and Economics Review 23 (2):348–394. Chang, Yun-chien, David Ta-wei Hung, Chang-ching Lin, and Joseph Tao-yi Wang. 2019. Emotional Bargaining After Litigation: An Experimental Study of the Coase Theorem. Working paper, https://ssrn.com/abstract=3340930. Chang, Yun-chien, and Daniel Klerman. 2022. Settlement Around the World: Settlement Rates in the Largest Economies. Journal of Legal Analysis 14:80–175. Chang, Yun-chien, and Chang-ching Lin. 2021. Do Parties Negotiate after Trespass Litigation? An Empirical Study of Coasean Bargaining. Working paper, https://ssrn.com/abstract=2805063. Chang, Yun-chien, and Henry E. Smith. 2012. An Economic Analysis of Civil versus Common Law Property. Notre Dame Law Review 88 (1):1–55. 2015. The Numerus Clausus Principle, Property Custom, and the Emergence of New Property Forms. Iowa Law Review 100:2275–2308. 2016. Structure and Style in Comparative Property Law. In Research Handbook on Comparative Law and Economics 131–160, edited by Theodore Eisenberg and Giovanni B. Ramello. Northampton, MA: Edward Elgar. 2019. Convergence and Divergence in Systems of Property Law: Theoretical and Empirical Analyses. Southern California Law Review 92 (4):785–808. Chang, Yun-chien, and Ke Xu. 2018. Decentralized and Anomalous Interpretation of Chinese Private Law: Understanding a Bureaucratic and Political Judicial System. Minnesota Law Review 102 (4):1527–1562. Cheung, Steven N. S. 1998. The Transaction Costs Paradigm. Economic Inquiry 36:514–521. Chiang, Dominic Y. J. 2016. In Search of a Legal Solution for Boundary Encroachments in Hong Kong: Adverse Possession vs Statutory Encroachment. Hong Kong Journal of Legal Studies 10:59–82. Chilton, Adam, and Mila Versteeg. 2020. How Constitutional Rights Matter. Oxford, UK: Oxford University Press. Claeys, Eric. 2011. Exclusion and Exclusivity in Gridlock. Arizona Law Review 53:9–49.

Published online by Cambridge University Press

388

references

Coase, Ronald H. 1960. The Problem of Social Cost. Journal of Law & Economics 3:1–69. Coase, Ronald H., and Ning Wang. 2012. How China Became Capitalist. New York, NY: Palgrave Macmillan. Cohet, Frédérique. 2019. Chapitre 1 – Accession: Mode D’acquisition de la Propriété. In Répertoire de Droit Immobilier, https://jurisguide.univ-paris1.fr/fichesdocumentaires/encyclopedie-juridique-dalloz/. Paris, France: Dalloz. Cooter, Robert, and Thomas Ulen. 2012. Law and Economics. 6th ed. Boston, MA: Pearson/Addison Wesley. Craig-Taylor, Phyliss. 2000. Through A Colored Looking Glass: A View of Judicial Partition, Family Land Loss and Rule Setting. Washington University Law Quarterly 78:737–788. Cramton, Peter, Robert Gibbons, and Paul Klemperer. 1987. Dissolving a Partnership Efficiently. Econometrica 55 (3):615–632. Dagan, Hanoch, and Michael A. Heller. 2001. The Liberal Commons. Yale Law Journal 110:549–623. Dana, David, and Thomas W. Merrill. 2002. Property: Takings. New York, NY: Foundation Press. Dannemann, Gerhard, and Reiner Schulze, eds. 2020. German Civil Code Volume I: Books 1–3 Article-by-Article Commentary. Munich, Germany: C.H. Beck. Dari-Mattiacci, Giuseppe, and Carmine Guerriero. 2015. Law and Culture: A Theory of Comparative Variation in Bona Fide Purchase Rules. Oxford Journal of Legal Studies 35 (3):543–574. 2019. Divergent and Convergent at the Intersection of Property and Contract. Southern California Law Review 92 (4):809–850. Dari-Mattiacci, Giuseppe, Carmine Guerriero, and Zhenxing Huang. 2016. The Property-Contract Balance. Journal of Institutional and Theoretical Economics 172 (1):40–64. Dari‐Mattiacci, Giuseppe. 2009. Negative Liability. The Journal of Legal Studies 38 (1):21–59. De Geest, Gerrit, and Roger J. Van den Bergh, eds. 2004. Comparative Law and Economics. Cheltenham, UK: Edward Elgar. de Londras, Fiona. 2011. Principles of Irish Property Law. 2nd ed. Dublin, Ireland: Clarus Press. de Soto, Hernando. 2000. The Mystery of Capital New York, NY: Basic Books. Demsetz, Harold. 1967. Toward a Theory of Property Rights. American Economic Review 57:347–359. 2008. From Economic Man to Economic System: Essays on Human Behavior and the Institutions of Capitalism. Cambridge, UK: Cambridge University Press. 2011a. The Problem of Social Cost: What Problem? A Critique of the Reasoning of A.C. Pigou and R.H. Coase. Review of Law & Economics 7:1–13. 2011b. R.H. Coase and the Neoclassical Model of the Economic System. Journal of Law and Economics 54 (4):S7–S13.

Published online by Cambridge University Press

references

389

Depoorter, Ben. 2010. Adverse Possession. In Property Law and Economics 183–190, edited by Boudewijn Bouckaert. Northampton, MA: Edward Elgar. 2011. Fair Trespass. Columbia Law Review 111:1090–1135. Depoorter, Ben, and Paul Rubin. 2017. Judge-Made Law and the Common Law Process. In The Oxford Handbook of Law and Economics: Volume 3: Public Law and Legal Institutions 129–142, edited by Francesco Parisi. Oxford, UK: Oxford University Press. Deutch, Miguel. 1995. Property Law. In Introduction to the Law of Israel 81–93, edited by Amos Shapira and Keren C. DeWitt-Arar. Alphen aan den Rijn the Netherlands: Kluwer Law. Dickinson, Kevin H. 1985. Mistaken Improvers of Real Estate. North Carolina Law Review 64 (1):37–76. Dixon, Martin. 2016. Modern Land Law. 10th ed. New York, NY: Routledge. Djankov, Simeon, Andrei Shleifer, Caralee McLiesh, and Oliver Hart. 2008. Debt Enforcement around the World. Journal of Political Economy 116 (6):1105–1149. Domínguez Guillén, and María Candelaria. 2017. Curso de Derecho Civil III: Obligaciones. Caracas, Venezuela: Revista Venezolana de Legislación y Jurisprudencia. Drobnig, Ulrich. 1974. Security Over Corporeal Movables in Germany. In Security over Corporeal Movables 181–205, edited by J. G. Sauveplanne. Leiden, the Netherlands: A. W. Sijthoff. 2011a. Security Rights in Movables. In Towards a European Civil Code 1025–1042, edited by Arthur S. Hartkamp, Martijn W. Hesselink, Ewoud H. Hondius, Chantal Mak and C. Edgar du Perron. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. 2011b. Transfer of Property. In Towards a European Civil Code 1003–1023, edited by Arthur S. Hartkamp, Martijn W. Hesselink, Ewoud Hondius, C. Mak and Edgar Du Perron. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Druey, J. N. 2004. Law of Property. In Introduction to Swiss Law 93–106, edited by F. Dessemontet and T. Ansay. Alphen aan den Rijn, the Netherlands: Kluwer Law. Dukeminier, Jesse, James E. Krier, Gregory S. Alexander, and Michael H. Schill. 2010. Property. 7th ed. New York, NY: Aspen Publishers. Dyson, Henry. 2003. French Property and Inheritance Law: Principles and Practice. Oxford, UK: Oxford University Press. Easterbrook, Frank H. 1984. Foreword: The Court and the Economic System. Harvard Law Review 98:4–60. Edgeworth, Brendan. 2017. Butt’s Land Law. Sydney, Australia: Thomson Reuters. Eisenberg, Theodore, and Giovanni Battista Ramello, eds. 2016. Research Handbook on Comparative Law and Economics. Northampton, MA: Edward Elgar. Elkins, Zachary, Tom Ginsburg, and James Melton. 2009. The Endurance of National Constitutions. Cambridge, UK: Cambridge University Press.

Published online by Cambridge University Press

390

references

Ellickson, Robert C. 1986. Adverse Possession and Perpetuities Law: Two Dents in the Libertarian Model of Property Rights. Washington University Law Quarterly 64:723–737. 1991. Order without Law: How Neighbors Settle Disputes. Cambridge, MA: Harvard University Press. 1993. Property in Land. Yale Law Journal 102:1315–1400. 2012. The Costs of Complex Land Titles: Two Examples from China. BrighamKanner Property Rights Conference Journal 1:281–344. Emerich, Yaëll. 2018. Conceptualising Property Law: Integrating Common Law and Civil Law Traditions. Northampton, MA: Edward Elgar. Epstein, Richard A. 1986. Past and Future: The Temporal Dimension in the Law of Property. Washington University Law Quarterly 64:667–722. 1993. Holdouts, Externalities, and the Single Owner: One More Salute to Ronald Coase. Journal of Law & Economics 36:553–586. 1994. On the Optimal Mix of Private and Common Property. Social Philosophy and Policy 11:17–41. 1995. Simple Rules for A Complex World. Cambridge, MA: Harvard University Press. 1997. A Clear View of the Cathedral: The Dominance of Property Rules. Yale Law Journal 106:2091–2120. 1998. Protecting Property Rights with Legal Remedies: A Common Sense Reply to Professor Ayres. Valparaiso University Law Review 32:833–853. 2011. Heller’s Gridlock Economy in Perspective: Why There Is Too Little, Not Too Much Private Property. Arizona Law Review 53 (1):51–82. 2014. The Use and Limits of Self-Valuation Systems. The University of Chicago Law Review 81:109–129. Erp, Sjef van. 2002. A Comparative Analysis of Mortgage Law: Searching for Principles. In Land Law in Comparative Perspective, edited by Maria Elena Sanchez Jordan and Antonio Gambaro. The Hague, the Netherlands: Kluwer Law International. 2006. Comparative Property Law. In The Oxford Handbook of Comparative Law 1031–1057, edited by Mathias Reimann and Reinhard Zimmermann. Oxford, UK: Oxford University Press. Erp, Sjef van, and Bram Akkermans. 2010. Property Rights: A Comparative View. In Property Law and Economics 31–52, edited by Boudewijn Bouckaert. Northampton, MA: Edward Elgar. 2012. Cases, Materials and Text on Property Law. Oxford, UK: Hart. Faber, Wolfgang. 2009. National Report on the Transfer of Movables in Austria. In National Reports on the Transfer of Movables in Europe, Volume 1: Austria, Estonia, Italy, Slovenia 1–218, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Faerstad, Jan-Ove, and Martin Lilja. 2010. National Report on the Transfer of Movables in Norway and Denmark. In National Report on the Transfer of

Published online by Cambridge University Press

references

391

Movables in Europe, Volume 5: Sweden, Norway and Denmark, Finland, Spain 207–301, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Fambrough, Judon. 2010. Easements in Texas: Real Estate Center, Texas A & M University. Farnsworth, Ward. 1999. Do Parties to Nuisance Cases Bargain after Judgment? A Glimpse Inside the Cathedral. The University of Chicago Law Review 66 (2):373–436. 2016. The Empirical Accuracy and Judicial Use of the Coase Theorem (vel non). In The Elgar Companion to Ronald H. Coase 346–357, edited by Claude Menard and Elodie Bertrand. Northampton, MA: Edward Elgar. Farran, Sue. 2013. Property and Trust Law in South Pacific. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Faust, Florian. 2008. Comparative Law and Economic Analysis of Law. In The Oxford Handbook of Comparative Law 837–865, edited by Mathias Reimann and Reinhard Zimmermann. Oxford, UK: Oxford University Press. Fennell, Lee Anne. 2004a. Common Interest Tragedies. Northwestern University Law Review 98:907–990. 2004b. Taking Eminent Domain Apart. Michigan State Law Review 2004: 957–1004. 2005. Revealing Options. Harvard Law Review 118 (5):1399–1488. 2006. Efficient Trespass: The Case for “Bad Faith” Adverse Possession. Northwestern University Law Review 100:1037–1096. 2007. Property and Half-Torts. Yale Law Journal 116:1400–1471. 2009. The Unbounded Home: Property Values Beyond Property Lines. New Haven, CT: Yale University Press. 2011a. Commons, Anticommons, Semicommons. In Research Handbook on the Economics of Property Law 35–56, edited by Henry E. Smith. Northampton, MA: Edward Elgar. 2011b. Ostrom’s Law: Property Rights in the Commons. International Journal of the Commons 5 (1):9–27. 2011c. Property and Precaution. Journal of Tort Law 4 (2):art.3:1–63. 2012. Lumpy Property. University of Pennsylvania Law Review 160:1955–1993. 2013a. Just Enough. Columbia Law Review Sidebar 113:109–122. 2013b. The Problem of Resource Access. Harvard Law Review 126 (6):1471–1531. 2014. Forcings. Columbia Law Review 114:1297–1372. 2016. Fee Simple Obsolete. New York University Law Review 91 (6):1457–1516. 2019a. Property Attachments. The University of Chicago Law Review 87 (1):55–66. 2019b. Slices and Lumps: Division and Aggregation in Law and Life. Chicago, IL: Chicago University Press. Feys, Bert. 2006. A General Introduction to Norwegian Property Law. www.scribd .com/document/82917000/Norwegian-Property-Law.

Published online by Cambridge University Press

392

references

Field, Erica. 2004. Property Rights, Community Public Goods, and Household Time Allocation in Urban Squatter Communities: Evidence from Peru. William & Mary Law Review 45:837–887. Fieseler, Karsten, Thomas Kittsteiner, and Benny Moldovanu. 2003. Partnerships, Lemons, and Efficient Trade. Journal of Economic Theory 113 (2):223–234. Foëx, Bénédict, and Sylvain Marchand. 2010. National Report on the Transfer of Movables in Switzerland. In National Reports on the Transfer of Movables in Europe, Volume 6: The Netherlands, Switzerland, Czech Republic, Slovakia, Malta, Lavtia 167–253, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Frisby, Sandra, and Michael Jones. 2009. National Report on the Transfer of Movables in England and Wales. In National Reports on the Transfer of Movables in Europe, Volume 2: England and Wales, Ireland, Scotland, Cyprus 1–154, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter.Gaier, Reinhard. 2013. Münchener Kommentar zum Bürgerlichen Gesetzbuch. 6th ed. Munich, Germany: C.H. Beck. Galea, Patrick J. 2010. National Report on the Transfer of Movables in Malta. In National Reports on the Transfer of Movables in Europe, Volume 6: The Netherlands, Switzerland, Czech Republic, Slovakia, Malta, Lavtia 461–544, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Gambaro, Antonio, and Ugo Mattei. 2002. Property Law. In Introduction to Italian Law 283–316, edited by Jeffrey S. Lena and Ugo Mattei. The Hague, the Netherlands: Kluwer Law International. Gardiner, Caterina. 2009. National Report on the Transfer of Movables in Ireland. In National Reports on the Transfer of Movables in Europe, Volume 2: England and Wales, Ireland, Scotland, Cyprus 155–296, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Garoupa, Nuno, and Carlos Gómez Ligüerre. 2011. The Syndrome of the Efficiency of the Common Law. Boston University International Law Journal 29:287–335. Garrido, Pedro. 2010. National Report on the Real Property in Spain. In Real Property Law and Procedure in the European Union. Florence, Italy: European University Institute, www.eui.eu/Documents/DepartmentsCentres/Law/ ResearchTeaching/ResearchThemes/EuropeanPrivateLaw/RealProperty Project/Spain.pdf. Gaynanova, Irena, James. G. Booth, and Martin T. Wells. 2016. Simultaneous Sparse Estimation of Canonical Vectors in the p>>N Setting. Journal of the American Statistical Association 111 (514):696–706. Gignoux, Stéphane. 2015. Appréhender le droit marocain de l’immobilier: une approche transversale « droit privé – droit public ». www.umifre.fr/sites/ default/files/cjb_ee_30.pdf: Les Études et Essais du Centre Jacques-Berque. Gocza, Eva. 2013. Polgari Torvenykonyv – Civil Code Budapest, Hungary: Wolters Kluwer.

Published online by Cambridge University Press

references

393

Gold, Andrew S., John C. P. Goldberg, Daniel B. Kelly, Emily Sherwin, and Henry E. Smith. 2020. Introduction. In The Oxford Handbook of the New Private Law xv–xxviii, edited by Andrew S. Gold, John C. P. Goldberg, Daniel B. Kelly, Emily Sherwin and Henry E. Smith. Oxford, UK: Oxford University Press. Gomez, Lincoln D. 2010. Who Owns the (Aircraft) Engine(s)? The Doctrine of Accession Explained. http://lincolngomez.com/2010/02/11/aviation-enginesdoctrine-accession-gomez-bikker-arub/. González, Espinosa, and Jacinto Javier. 2015. Los Derechos Reales en el Ordenamiento Jurídico Panameño. Vol. 1. Panama City, Panama: Universidad de Panamá. Goo, Say, and Alice Lee. 2019. Land Law in Hong Kong. Hong Kong, China: LexisNexis. Gordley, James. 2006. Foundations of Private Law: Property, Tort, Contract, Unjust Enrichment. Oxford, UK: Oxford University Press. Graziadei, Michele. 2017. The Structure of Property Ownership and the Common Law/Civil Law Divide. In Comparative Property Law: Global Perspectives 71– 99, edited by Michele Graziadei and Lionel Smith. Cheltenham, UK: Edward Elgar. Greco, Alessio. 2009. National Report on the Transfer of Movables in Italy. In National Reports on the Transfer of Movables in Europe, Volume 1: Austria, Estonia, Italy, Slovenia 341–506, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Greenfield, Aryeh. 2010. Israel’s Real Estate and Movable Property Laws: Completely Revised and Updated English Translations, Incorporating All Amendments Legislated Until November 1, 2010. Translated by Aryeh Greenfield. 6th ed. Haifa, Israel: A.G. Publications. Gretton, George L., Andres J. M. Steven, and Alison E. C. Struthers. 2017. Property, Trusts and Succession. 3rd ed. London, UK: Bloomsbury. Grundmann, Stefan, Hans-W. Micklitz, and Moritz Renner. 2021. New Private Law Theory. Cambridge, UK: Cambridge University Press. Güth, Werner, Rolf Schmittberger, and Bernd Schwarze. 1982. An Experimental Analysis of Ultimatum Bargaining. Journal of Economic Behavior & Organization 3 (4):367–388. Guerriero, Carmine. 2016. Endogenous Property Rights. Journal of Law and Economics 59:313–358. Gutmann, Jerg, Bernd Hayo, and Stefan Voigt. 2014. Determinants of Constitutionally Safeguarded Judicial Review: Insights Based on a New Indicator. In Empirical Legal Analysis: Assessing the Performance of Legal Institutions 216–254, edited by Yun-chien Chang. London, UK: Routledge. Gutmann, Jerg, Katarzyna Metelska-Szaniawska, and Stefan Voigt. 2022. The Comparative Constitutional Compliance Database. www.google.com/ url?q=https%3A%2F%2Feconpapers.repec.org%2FRePEc%3Azbw%3Ailew ps%3A57&sa=D&sntz=1&usg=AOvVaw3KMIYP9Mqtbx_6MnIiQ35Q.

Published online by Cambridge University Press

394

references

Habdas, Magdalena. 2018. Property and Trust Law in Poland. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Hansmann, Henry, and Reinier Kraakman. 2000a. The End of History for Corporate Law. Georgia Law Journal 89:439–468. 2000b. The Essential Role of Organizational Law. Yale Law Journal 110:387–440. 2002. Property, Contract, and Verification: The Numerus Clausus Problem and the Divisibility of Rights. Journal of Legal Studies 31:S373–S420. Hardin, Garrett. 1968. The Tragedy of the Commons. Science 162:1243–1248. Hasen, Richard L., and Richard H. McAdams. 1997. The Surprisingly Complex Case against Theft. International Review of Law & Economics 3 (17):367–378. Hashem, Hisham R. 1990. The Jordan Civil Code of Moslem Jurisprudence. Amman, Jordan: al-Tawfiq Printing Press. Haug, Karoline Rakneberg. 2017. The Historical Development of the Scandinavian Functional Approach to Transfer of Ownership: A Tale of Change and Continuity. European Property Law Journal 6 (2):236–271. Hawkins, Ashton, Richard A. Rothman, and David B. Goldstein. 1995. A Tale of Two Innocents: Creating an Equitable Balance between the Rights of Former Owners and Good Faith Purchasers of Stolen Art. Fordham Law Review 64:49–96. Hayton, David, and Paul Matthews. 2007. Property and Trust Law in England and Wales. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Hegdal, Stein. 2015. Norwegen. In Handbuch Immobilienrecht in Europa : Zivil- und Steuerrechtliche Aspekte des Erwerbs, der Veräusserung und der Vererbung von Immobilien 901–973, edited by Susanne Frank and Thomas Wachter. Munich, Germany: C.F. Müller. Heller, Michael A. 1998. The Tragedy of the Anticommons: Property in the Transition from Marx to Markets. Harvard Law Review 111 (3):621–688. 2008. The Gridlock Economy: How Too Much Ownership Wrecks Markets, Stops Innovation, and Costs Lives. New York, NY: Basic Books. Heller, Michael, and James Salzman. 2021. Mine!: How the Hidden Rules of Ownership Control Our Lives. New York, NY: Atlantic Books. Helmholz, Richard H. 1983. Adverse Possession and Subjective Intent. Washington University Law Quarterly 61:331–358. Hennig, Christian. 2007. Cluster-Wise Assessment of Cluster Stability. Computational Statistics & Data Analysis 52 (1):258–271. Hepburn, Samantha J. 2002. Property and Trust Law in Australia. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Hernandez, Michael V. 2005. Restating Implied, Prescriptive, and Statutory Easements. Real Property Probate and Trust Journal 40:75–115. HNZPT. 2019. Guidelines for the Finding of Artefacts. www.heritage.org.nz/~/-/ media/b283ec5075024ca39c158a088a80af69.ashx: Heritage New Zealand Pouhere Taonga.

Published online by Cambridge University Press

references

395

Ho, Han-wei, Patrick Chung-Chia Huang, and Yun-chien Chang. 2023 forthcoming. Machine-Learning Methods. In Cambridge Handbook of Comparative Law, edited by Mathias Siems and Po Jen Yap. Cambridge, UK: Cambridge University Press. Ho, Lusina, and Rebecca Lee, eds. 2013. Trust Law in Asian Civil Law Jurisdictions: A Comparative Analysis. Cambridge, UK: Cambridge University Press. Hofri-Winogradow, Adam. 2020. The Statutory Liberalization of Trust Law across 152 Jurisdictions: Leaders, Laggards and the Market in Fiduciary Services. U.C. Davis Law Review 53:2313–2355. Hofstede, Geert. 1991. Cultures and Organizations: Software of the Mind. London, UK: McGraw-Hill. 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions and Organizations Across Nations. 2nd ed. London, UK: Sage. Hohfeld, Wesley Newcomb. 1913. Some Fundamental Legal Conceptions as Applied in Judicial Reasoning. The Yale Law Journal 23:16–59. 1917. Fundamental Legal Conceptions as Applied in Judicial Reasoning. The Yale Law Journal 26 (8):710–770. Holderness, Clifford G. 2003. Joint Ownership and Alienability. International Review of Law and Economics 23:75–100. Holguín, Juan Larrea. 2010. Derecho Civil Del Ecuador, Vol. V: Los Bienes Y La Posesión. Quito, Ecuador: Corporación de Estudios Y Publicaciones. Hollo, Erkki J. 2012. Property and Trust Law in Finland. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. 2019. Property and Trust Law in Finland. 2nd ed. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Holmström, Bengt, and Roger B. Myerson. 1983. Efficient and Durable Decision Rules with Incomplete Information. Econometrica 51 (6):1799–1819. Hovenkamp, Herbert, Sheldon F. Kurtz, and Thomas P. Gallanis. 2016. Principles of Property Law. 7th ed. St. Paul, MN: Thomson/West. Huffstetler, Scott D. 2002. Don’t Fence Me In: Louisiana’s Fourth Circuit Expands “Voluntariness” Under Louisiana Civil Code Article 693. Louisiana Law Review 63:111–126. Hylton, Keith N. 2006. Property Rules and Liability Rules, Once Again. Review of Law & Economics 2 (2):1–55. 2011. Property Rules and Defensive Conduct in Tort Law Theory. Journal of Tort Law 4 (1):1–31. Israel Law Resource Center. 2007. Laws of the State of Israel: Authorized Translation from the Hebrew. Vol. 12. Jerusalem, Israel: Government Printer. Izuel, Leeanna. 1992. Property Owners’ Constructive Possession of Treasure Trove: Rethinking the Finders Keepers Rule. UCLA Law Review 38:1659–1702. Jansen, J. E. 2012. Thieves and Squatters: Acquisitive and Extinctive Prescription in European Property Law. European Property Law Journal 1 (1):153–165.

Published online by Cambridge University Press

396

references

Jehiel, Philippe, and Ady Pauzner. 2006. Partnership Dissolution with Interdependent Values. The RAND Journal of Economics 37 (1):1–22. Jensen, Ulf. 2010. National Report on the Real Property in Sweden. In Real Property Law and Procedure in the European Union. Florence, Italy: European University Institute www.eui.eu/Documents/DepartmentsCentres/Law/ResearchTeaching/ ResearchThemes/EuropeanPrivateLaw/RealPropertyProject/Sweden.PDF. Josipovic, Tatjana. 2013. Property and Trust Law in Croatia. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Josipović, Tatjana. 2014. Property Law. In Introduction to the Law of Croatia 95–131, edi­ ted by Tatjana Josipović. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Joubert, Willem Adolf, and J. A. Faris. 2010. The Law of South Africa, Volume 24. Durban, South Africa: LexisNexis. Kahneman, Daniel. 2011. Thinking, Fast and Slow. New York, NY: Farrar, Straus and Giroux. Kahneman, Daniel, Jack L. Knetsch, and Richard Thaler. 1986. Fairness as a Constraint on Profit Seeking: Entitlements in the Market. American Economic Review 76 (4):728–741. Kaplow, Louis. 1992. Rules versus Standards: An Economic Analysis. Duke Law Journal 42:557–629. Kaplow, Louis, and Steven Shavell. 1996. Property Rules versus Liability Rules: An Economic Analysis. Harvard Law Review 109:713–790. 2002. Fairness versus Welfare. Cambridge, MA: Harvard University Press. Karam, Nicolas. 2011. The Civil Code of Kuwait: Decree Law No. 67 of 1980. Translated by Nicholas Karam. London, UK: Lexgulf Publishers Ltd. Karibali-Tsiptsiou, Yianna. 2012. Property and Trust Law in Greece. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Katz, Larissa. 2010. The Moral Paradox of Adverse Possession: Sovereignty and Revolution in Property Law. McGill Law Journal 55:47–80. Keesee, Allen P. K. 1982. Commercial Laws of the Middle East: Oman. London, UK: Oceana Publication. Kelly, Daniel B. 2009. Pretextual Takings: Of Private Developers, Local Governments, and Impermissible Favoritism. Supreme Court Economic Review 17:173–235. 2011. Acquiring Land through Eminent Domain: Justifications, Limitations, and Alternatives. In Research Handbook on the Economics of Property Law 344–371, edited by Kenneth Ayotte and Henry E. Smith. Cheltenham, UK; Northampton, MA: Edward Elgar. 2014. Restricting Testamentary Freedom: Ex Ante versus Ex Post Justifications. Fordham Law Review 82:1125–1185. Kenjebayeva, Aigoul. 2016. Property Law. In Introduction to the Law of Kazakhstan 109–125, edited by Zhenis Kembayev. Alphen aan den Rijn, the Netherlands: Wolters Kluwer.

Published online by Cambridge University Press

references

397

Kim, Jeong-Yoo. 2004. Good-faith Error and Intentional Trespassing in Adverse Possession. International Review of Law and Economics 24 (1):1–13. Kim, Kenise. 2010. Mixed Systems in Legal Origins Analysis. Southern California Law Review 83:693–730. Klaoudatou, Christina. 2011. National Report on the Transfer of Movables in Greece. In National Reports on the Transfer of Movables in Europe, Volume 3: Germany, Greece, Lithuania, Hungary 201–322, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Klass, Alexandra B. 2006. Adverse Possession and Conservation: Expanding Traditional Notions of Use and Possession. University of Colorado Law Review 77:283–333. Klauberg, Theis, and Julija Kolomijceva. 2010. National Report on the Transfer of Movables in Latvia. In National Reports on the Transfer of Movables in Europe, Volume 6: The Netherlands, Switzerland, Czech Republic, Slovakia, Malta, Lavtia 547–584, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyer. Kleijn, W. M., J. P. Jordaans, H. B. Krans, H. D. Ploeger, and F. A. Steketee (revised by J. M. J. Choros). 2006. Property Law. In Introduction to Dutch Law 103–134, edited by J. M. J. Chorus, Piet-Hein Gerver and Ewoud Hondius. The Hague, the Netherlands: Kluwer Law International. Klemperer, Paul. 1999. Auction Theory: A Guide to the Literature. Journal of Economic Surveys 13 (3):227–286. Klerman, Daniel. 2015. Jurisdiction, Choice of Law, and Property. In Law and Economics of Possession 266–289, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. Klerman, Daniel M., Paul G. Mahoney, Holger Spamann, and Mark I. Weinstein. 2011. Legal Origin or Colonial History? Journal of Legal Analysis 3 (2):379–409. Klerman, Daniel, and Anja Shortland. 2022. The Transformation of the Art World: Law, Norms, and Institutions. Theoretical Inquiries in Law 23 (1):219–242. Klick, Jonathan, and Gideon Parchomovsky. 2016. The Value of the Right to Exclude: An Empirical Assessment. University of Pennsylvania Law Review 165 (4):917–966. Koh, Jay L. 1998. From Hoops to Hard Drives: An Accession Law Approach to the Inevitable Misappropriation of Trade Secrets. American University Law Review 48:271–358. Kohler, Jürgen. 2005. Property Law (Sachenrecht). In Introduction to German Law 227–249, edited by Joachim Zekoll and Mathias Reimann. The Hague, the Netherlands: Kluwer Law International. Kominers, Scott D., and Eric G. Weyl. 2011. Holdout in the Assembly of Complements: A Problem for Market Design. American Economic Review 102 (3).

Published online by Cambridge University Press

398

references

Koo, Hui-wen. 2011. Property Rights, Land Prices, and Investment: A Study of the Taiwanese Land Registration System. Journal of Institutional and Theoretical Economics 167 (3):515–535. Kornhauser, Lewis A. 2022. The Economic Analysis of Law. In The Stanford Encyclopedia of Philosophy (Spring 2022 Edition), edited by Edward N. Zalta. https://plato.­ stanford.edu/entries/legal-econanalysis/. Metaphysics Research Lab, Stanford University. Koziol, Helmut, Peter Bydlinski, and Raimund Bollenberger. 2005. Kurzkommentar zum ABGB: Allgemeines bürgerliches Gesetzbuch samt Ehegesetz und Konsumentenschutzgesetz. Vienna, Austria: Springer. Krier, James E. 2006. Property. 17th ed. Chicago, IL: Gilbert Law Summaries. Krier, James E., and Stewart J. Schwab. 1995. Property Rules and Liability Rules: The Cathedral in Another Light. New York University Law Review 70:440–483. Krier, James E., and Christopher Serkin. 2004. Public Ruses. Michigan State Law Review 2004:859–876. Krishnaswami, M. 2002. Law of Adverse Possession. 13th ed. New Delhi, India: LexisNexis Butterworths. Kullerkupp, Kai. 2009. National Report on the Transfer of Movables in Estonia. In National Reports on the Transfer of Movables in Europe, Volume 1: Austria, Estonia, Italy, Slovenia 226–340, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Kuperman, Zachary D. 2011. Cutting the Baby in Half: An Economic Critique of Indivisible Resource Partition. Brooklyn Law Review 77:263–302. Kuusinen, Miki. 2010. National Report on the Transfer of Movables in Finland. In National Report on the Transfer of Movables in Europe, Volume 5: Sweden, Norway and Denmark, Finland, Spain 304–391, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Kwon, Youngjoon. 2013. Civil Law and Civil Procedure Law. In Introduction to Korean Law, edited by Korea Legislation Research Institute. Berlin, Germany: Springer. La Forest, Anne Warner. 2017a. Law of Real Property, Volume 2. 3rd ed. Toronto, Canada: Thomson Reuters. 2017b. Law of Real Property, Volume 3. 3rd ed. Toronto, Canada: Thomson Reuters. La Porta, Rafael, Florencio López-de-Silanes, and Andrei Shleifer. 2008. The Economic Consequences of Legal Origins. Journal of Economic Literature 46:285–332. Landeo, Claudia M., and Kathryn E. Spier. 2014. Irreconcilable Differences: Judicial Resolution of Business Deadlock. The University of Chicago Law Review 81:203–227. Landes, William M., and Richard A. Posner. 1996. The Economics of Legal Disputes over the Ownership of Works of Art and Other Collectibles. In Economics of the Arts: Selected Essays 177–219, edited by Victor Ginsburgh and PierreMichel Menger. Brussels, Belgium: Universite Libre de Bruxelles.

Published online by Cambridge University Press

references

399

Landsburg, Steven. 2008. Price Theory and Applications 7th ed. Mason, OH: Thomson South-Western. Langer, William. 1975. American Foods and Europe’s Population Growth. Journal of Social History 8:51–66. Law, David S. 2016. Constitutional Archetypes. Texas Law Review 95:153–243. Lee, Jack Tsen-Ta. 2004. Treaties, Time Limits and Treasure Trove: The Legal Protection of Cultural Objects in Singapore. Art Antiquity & Law 9:237–280. Lee, Peter. 2011. The Accession Insight and Patent Infringement Remedies. Michigan Law Review 110:175–233. Lee, Robert Warden. 1953. An Introduction to Roman Dutch Law. 5th ed. Oxford, UK: Clarendon Press. Lehavi, Amnon. 2020. The Law of Trusts and Collective Action: A New Approach to Property Deadlocks. University of Cincinnati Law Review 89:388–431. Leks, Eddy M. 2018. Property and Trust Law in Indonesia. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Lemley, Mark A. 1997. The Economics of Improvement in Intellectual Property Law. Texas Law Review 75:989–1084. Levinson, Daryl J. 2005. Empire-Building Government in Constitutional Law. Harvard Law Review 118:915–972. Levmore, Saul. 1982. Self-Assessed Valuation Systems for Tort and Other Law. Virginia Law Review 68 (4):771–861. 1985. Explaining Restitution. Virginia Law Review 71:65–124. 1987. Variety and Uniformity in the Treatment of the Good-faith Purchaser. Journal of Legal Studies 16 (1):43–65. 2019. Convergence and Downstream Divergence in Torts and Other Laws. Southern California Law Review 92:769–783. Lewinsohn-Zamir, Daphna. 2001. The Choice between Property Rules and Liability Rules Revisited: Critical Observations from Behavioral Studies. Texas Law Review 80:219–260. 2015. What Behavioral Studies Can Teach Jurists about Possession and Vice Versa. In Law and Economics of Possession 128–148, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. Libecap, Gary D., and Dean Lueck. 2011a. The Demarcation of Land and the Role of Coordinating Property Institutions. Journal of Political Economy 119 (3):426–467. 2011b. Land Demarcation Systems. In Research Handbook on the Economics of Property Law 257–295, edited by Kenneth Ayotte and Henry E. Smith. Northampton, MA: Edward Elgar. Libecap, Gary D., Dean Lueck, and Trevor O’Grady. 2011. Large-Scale Institutional Changes: Land Demarcation in the British Empire. Journal of Law and Economics 54 (4):S295–S327. Libya, Meredith O., and Ibrahim Massaud al-Arif Ansell. 1972. The Libyan Civil Code: An English Translation and a Comparison with the Egyptian Civil Code. Cambridge, UK: Oleander Press.

Published online by Cambridge University Press

400

references

Licht, Amir N., Chanan Goldschmidt, and Shalom H. Schwartz. 2005. Culture, Law, and Corporate Governance. International Review of Law and Economics 25:229–255. Lilja, Martin. 2010. National Report on the Transfer of Movables in Sweden. In National Report on the Transfer of Movables in Europe, Volume 5: Sweden, Norway and Denmark, Finland, Spain 1–204, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Lin, Yu-Hsin, and Yun-chien Chang. 2018. An Empirical Study of Corporate Default Rules and Menus in China, Hong Kong and Taiwan. Journal of Empirical Legal Studies 15 (4):875–915. Listokin, Yair. 2009. What Do Corporate Default Rules and Menus Do? An Empirical Examination. Journal of Empirical Legal Studies 6 (2):279–308. Liu, Michael. 2014. Possession in the Law of Finding: A Comparative of the Doctrine of Possession. Australian Property Law Journal 22:211–232. Locke, Natania. 2008. Security Granted by a Company over Its Movable Property: The Floating Charge and the General Notarial Bond Compared. The Comparative and International Law Journal of Southern Africa 41 (1):136–154. Lovett, John A. 2017a. Disseisin, Doubt, and Debate: Adverse Possession Scholarship in the United States (1881–1986). Texas A&M Law Review 5:1–63. 2017b. Precarious Possession. Louisiana Law Review 77:617–700. Mackaay, Ejan. 2021. Law and Economics for Civil Law Systems. 2nd ed. Northampton, MA: Edward Elgar. Maidin, Ainul Jaria Binti, and Zuraidah H. J. Ali. 2014. Property and Trust Law in Malaysia. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Mann, Trischa, and Audrey Blunden, eds. 2010. Australian Law Dictionary. Oxford, UK: Oxford University Press. Mattei, Ugo. 2000. Basic Principles of Property Law: A Comparative Legal and Economic Introduction. Westport, CT: Greenwood Press. Mattei, Ugo, Teemu Ruskola, and Antonio Gidi. 2009. Schlesinger’s Comparative Law. New York, NY: Foundation Press. Mattheou, Eugenia. 2010. Greek Report. In Real Property Law and Procedure in the European Union. Florence, Italy: European Law Institute. Mau, Stephen D. 2010. Property Law in Hong Kong: An Introductory Guide. Hong Kong, China: Hong Kong University Press. Mautner, Menachem. 1991. “The Eternal Triangles of the Law”: Toward a Theory of Priorities in Conflicts Involving Remote Parties. Michigan Law Review 90 (1):95–154. Maydanyk, Roman. 2017. Property and Trust Law in Ukraine. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Mayer, Christopher J. 1995. A Model of Negotiated Sales Applied to Real Estate Auctions. Journal of Urban Economics 38:1–22. 1998. Assessing the Performance of Real Estate Auctions. Real Estate Economics 26:41–66.

Published online by Cambridge University Press

references

401

McFarlane, Ben. 2008. The Structure of Property Law. Portland, OR: Hart. Medina, Barak. 2003. Augmenting the Value of Ownership by Protecting It Only Partially: The “Market-Overt” Rule Revisited. The Journal of Law, Economics, and Organization 19 (2):343–372. Meier, Luke. 2015. A Contextual Approach to Claim of Right in Adverse Possession Cases: On Van Valkenburgh v. Lutz, Bad Faith, and Mistaken Boundaries. Lewis & Clark Law Review 19:47–72. Merrill, Thomas W. 1985. Property Rules, Liability Rules, and Adverse Possession. Northwestern University Law Review 79:1122–1154. 1986. The Economics of Public Use. Cornell Law Review 72:61–116. 1998. Property and the Right to Exclude. Nebraska Law Review 77:730–755. 2009. Accession and Original Ownership. Journal of Legal Analysis 1 (2):459–510. 2012a. Property as Modularity. Harvard Law Review 125:151–163. 2012b. The Property Strategy. University of Pennsylvania Law Review 160: 2061–2095. 2014. Property and the Right to Exclude II. Brigham-Kanner Property Rights Conference Journal 3:1–25. 2015. Ownership and Possession. In Law and Economics of Possession 1–39, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. Merrill, Thomas W., and Henry E. Smith. 2000. Optimal Standardization in the Law of Property: The Numerus Clausus Principle. Yale Law Journal 110:1–70. 2001a. The Property/Contract Interface. Columbia Law Review 101:773–852. 2001b. What Happened to Property in Law and Economics. Yale Law Journal 111:357–398. 2007a. The Morality of Property. William and Mary Law Review 48:1849–1896. 2007b. Property: Principles and Policies. New York, NY: Foundation Press. 2010. The Oxford Introductions to U.S. Law: Property. New York, NY: Oxford University Press. 2011. Making Coasean Property More Coasean. Journal of Law and Economics 54 (4):S77–S104. 2012. Property: Principles and Policies. 2nd ed. New York, NY: Foundation Press. 2017. Property: Principles and Policies. 3rd ed. New York, NY: Foundation Press. 2020. The Architecture of Property. In Research Handbook on Private Law Theory 134–154, edited by Hanoch Dagan and Benjamin C. Zipursky. Northampton, MA: Edward Elgar Publishing. Merwe, C. G. Van der. 1999. The Louisiana Right to Forced Passage Compared to the South African Way of Necessity. Tulane Law Review 73:1363–1413. 2003. Interpretation of Common Law and Civil Law as Experienced in the South African and Scottish Law of Property. Tulane Law Review 78:257–290. 2004. The Law of Property. In Introduction to the Law of South Africa 201–242, edited by C. G. Van der Merwe and Jacques E. du Plessis. Alphen aan den Rijn, the Netherlands: Wolters Kluwer.

Published online by Cambridge University Press

402

references

Miceli, Thomas J. 2004. The Economic Approach to Law. Palo Alto, CA: Stanford Economics and Finance. Miceli, Thomas J., Henry J. Munneke, C. F. Sirmans, and Geoffrey K. Turnbull. 2002. Title Systems and Land Values. The Journal of Law and Economics 45 (2):565–582. Miceli, Thomas J., and Kathleen Segerson. 2007. The Economics of Eminent Domain: Private Property, Public Use, and Just Compensation. Boston, MA: Now Publishers. Miceli, Thomas J., and C. F. Sirmans. 1995a. An Economic Theory of Adverse Possession. International Review of Law and Economics 15:161–173. 1995b. The Economics of Land Transfer and Title Insurance. The Journal of Real Estate Finance and Economics 10 (1):81–88. 2000. Partition of Real Estate; or, Breaking Up Is (Not) Hard to Do. Journal of Legal Studies 29:783–796. Miceli, Thomas J., C. F. Sirmans, and Geoffrey K. Turnbull. 1998. Assurance and Incentives for Efficient Land Use. European Journal of Law and Economics 6 (3):305–323. 2000. The Dynamic Effects of Land Title Systems. Journal of Urban Economics 47 (3):370–389. 2001. The Property-Contract Boundary: An Economic Analysis of Leases. American Law and Economics Review 3:165–185. Michelman, Frank I. 1982. Ethics, Economics, and the Law of Property. In Nomos XXIV: Ethics, Economics, and the Law 3–40, edited by J. Roland Pennock and John W. Chapman. New York, NY: New York University Press. Mikelenas, Valentinas. 2011. National Report on the Transfer of Movables in Lithuania. In National Reports on the Transfer of Movables in Europe, Volume 3: Germany, Greece, Lithuania, Hungary 333–408, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Miller, David L. Carey, Malcolm M. Combe, Andrew J. M. Steven, and Scott Wortley. 2009. National Report on the Transfer of Movables in Scotland. In National Reports on the Transfer of Movables in Europe, Volume 2: England and Wales, Ireland, Scotland, Cyprus 311–470, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Millung-Christoffersen, Astrid. 2019. The Scandinavian “Functional” Approach to Movable Property from a Danish view: Including the Question of “Tradition.” European Property Law Journal 8 (1):4–22. Mitchell, Thomas W. 2001. From Reconstruction to Deconstruction: Undermining Black Landownership, Political Independence, and Community through Partition Sales of Tenancies in Common. Northwestern University Law Review 95:505–580. 2005. Destabilizing the Normalization of Rural Black Land Loss: A Critical Role for Legal Empiricism. Wisconsin Law Review 2005:557–616.

Published online by Cambridge University Press

references

403

Mitchell, Thomas W., Stephen Malpezzi, and Richard K. Green. 2010. Forced Sale Risk: Class, Race, and the “Double Discount.” Florida State University Law Review 37:589–658. Mitra, Arindam. 2007. Mitra’s Law of Possession and Ownership of Property. New Delhi, India: Sodhi Publications. Mnookin, Robert H., and Lewis Kornhauser. 1979. Bargaining in the Shadow of the Law: The Case of Divorce. Yale Law Journal 88 (5):950–997. Moldovanu, Benny. 2002. How to Dissolve a Partnership. Journal of Institutional and Theoretical Economics 158 (1):66–80. Moore, Anthony P., Scott Grattan, and Lynden Griggs. 2016. Australian Property Law: Cases and Materials. 5th ed. Sydney, Australia: Thomson Reuters. Morse, Susan C. 2016. Safe Harbors, Sure Shipwrecks. UC Davis Law Review 49:1385–1430. Mostert, Hanri, and Leon Verstappen. 2015. Practical Approaches to the Numerus Clausus of Land Rights: How Legal Professionals in South Africa and the Netherlands Deal with Certainty and Flexibility in Property Law. In Modern Studies in Property Law 351–372, edited by Warren Barr. London, UK: Hart. Myerson, Roger B. 1979. Incentive Compatibility and the Bargaining Problem. Econometrica 47 (1):61–73. 2008. Perspectives on Mechanism Design in Economic Theory. American Economic Review 98 (3):586–603. Myerson, Roger B., and Mark A. Satterthwaite. 1983. Efficient Mechanisms for Bilateral Trading. Journal of Economic Theory 29 (2):265–281. Nasarre-Aznar, Sergio. 2012. The Need for the Integration of the Mortgage Market in Europe. In The Future of European Property Law 79–122, edited by Sjef van Erp, Arthur Salomons and Bram Akkermans. Munich, Germany: Sellier European Law Publishers. Negbi, Z. 1975. The Polish Civil Code. Leiden, the Netherlands: A. W. Sijthoff. Neocleous, Andreas. 2000. Introduction to Cyprus Law. Albany, NY: Yorkhill Law. Netter, Jeffry M., Philip L. Hersch, and William D. Manson. 1986. An Economic Analysis of Adverse Possession Statutes. International Review of Law and Economics 6:217–228. Newman, Christopher M. 2011. Transformation in Property and Copyright. Villanova Law Review 56:251–325. Niblett, Anthony. 2017. On the Efficiency of the Common Law: An Application to the Recovery of Rewards. European Journal of Law and Economics 43 (3):393–417. 2020. How Lower Courts Respond to a Change in a Legal Rule. In Selection and Decision in Judicial Process around the World: Empirical Inquires 110–131, edited by Yun-chien Chang. Cambridge, UK: Cambridge University Press. Niblett, Anthony, Richard A. Posner, and Andrei Shleifer. 2010. The Evolution of a Legal Rule. The Journal of Legal Studies 39 (2):325–358.

Published online by Cambridge University Press

404

references

North, Douglass C. 1990. Institution, Institutional Change and Economic Performance. Cambridge, UK: Cambridge University Press. Note. 1906. The Japanese Civil Code. Canada Law Review 2:73–74. 2020. Accession on the Frontiers of Property. Harvard Law Review 133:2381–2402. O’Connor, Pamela. 2006. The Private Taking of Land: Adverse Possession, Encroachment by Buildings and Improvement under a Mistake. University of Western Australia Law Review 33 (1):31–62. Oda, Hiroshi. 2009. Japanese Law. 3rd ed. Oxford, UK: Oxford University Press. Olsen-Ring, Line, and Gerhard Ring. 2015. Dänemark. In Handbuch Immobilienrecht in Europa : Zivil- und Steuerrechtliche Aspekte des Erwerbs, der Veräusserung und der Vererbung von Immobilien 97–178, edited by Susanne Frank and Thomas Wachter. Munich, Germany: C.F. Müller. Ørebech, Peter. 2013. Western Scandinavia: Exit Bürgerliches Gesetzbuch – the Resurrection of Customary Laws. Texas International Law Journal 48:405–434. Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge, UK: Cambridge University Press. Oto-Peralías, Daniel, and Diego Romero-Ávila. 2014. The Distribution of Legal Traditions around the World: A Contribution to the Legal Origins Theory. Journal of Law and Economics 57:561–628. Pacanowska, Isabel V. González, and Carlos Manuel Díez Soto. 2010. National Report on the Transfer of Movables in Spain. In National Report on the Transfer of Movables in Europe, Volume 5: Sweden, Norway and Denmark, Finland, Spain 394–667, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Palmer, Vincent Valentine. 2001. A Descriptive and Comparative Overview. In Mixed Jurisdictions Worldwide: The Third Legal Family 19–92, edited by Vernon Valentine Palmer. Cambridge, UK: Cambridge University Press. Parchomovsky, Gideon, and Peter Siegelman. 2004. Selling Mayberry: Communities and Individuals in Law and Economics. California Law Review 92:75–146. Parchomovsky, Gideon, Peter Siegelman, and Steve Thel. 2007. Of Equal Wrongs and Half Rights. New York University Law Review 82:738–789. Parchomovsky, Gideon, and Alex Stein. 2009. Reconceptualizing Trespass. Northwestern University Law Review 103:1823–1862. Pargendler, Mariana. 2012a. Politics in the Origins: The Making of Corporate Law in Nineteenth-Century Brazil. The American Journal of Comparative Law 60 (3):805–850. 2012b. The Rise and Decline of Legal Families. American Journal of Comparative Law 60:1043–1074. Parisi, Francesco. 2004. The Efficiency of the Common Law Hypothesis. In The Encyclopedia of Public Choice 519–522, edited by Charles K. Rowley and Friedrich Schneider. New York, NY: Springer.

Published online by Cambridge University Press

references

405

Passinhas, Sandra. 2019. Property and Trust Law in Portugal. 2nd ed. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Peñalver, Eduardo M. 2009. Land Virtues. Cornell Law Review 94:821–888. Peñalver, Eduardo M., and Sonia Katyal. 2007. Property Outlaws. University of Pennsylvania Law Review 155:1095–1186. 2010. Property Outlaws: How Squatters, Pirates, and Protesters Improve the Law of Ownership. New Haven, CT: Yale University Press. Perez, Vanessa Casado, and Carlos Gomez Liguerre. 2019. From Nuisance to Environmental Protection in Continental Europe. South Carolina Law Review 92:1003–1030. Petkov, Ivan. 2010. National Report on the Transfer of Movables in Slovakia. In National Reports on the Transfer of Movables in Europe, Volume 6: The Netherlands, Switzerland, Czech Republic, Slovakia, Malta, Lavtia 351–449, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyer. Picker, Christian, and Sebastian Herrler. 2019. J. von Staudingers Kommentar zum Bürgerlichen Gesetzbuch: Staudinger BGB – Buch 3: Sachenrecht. Berlin, Germany: C.H. Beck. Picker, Randal C. 2009. Security Interests in Personal Property: Cases, Problems, and Materials. 4th ed. New York, NY: Foundation Press. Pilati, José Isaac. 2016. Property Law. In Introduction to Brazilian Law 73–86, edited by Fabiano Deffenti and Welber Barral. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Plassmann, Florenz, and T. Nicolaus Tideman. 2011. Marginal Cost Pricing and Eminent Domain. Foundations and Trends in Microeconomics 7 (1):1–110. Polinsky, Mitchell A. 1980. Resolving Nuisance Disputes: The Simple Economics of Injunctive and Damage Remedies. Stanford Law Review 32:1075–1112. Porat, Ariel. 2009. Private Production of Public Goods: Liability for Unrequested Benefits. Michigan Law Review 108:189–227. Porat, Ariel, and Lior Jacob Strahilevitz. 2013. Personalizing Default Rules and Disclosure with Big Data. Michigan Law Review 112:1417–1478. Posner, Eric A., and E. G. Weyl. 2017. Property Is Only Another Name for Monopoly. Journal of Legal Analysis 9 (1):51–123. 2018. Radical Markets: Uprooting Capitalism and Democracy for a Just Society. Princeton, NJ: Princeton University Press. Posner, Richard A. 1973. Economic Analysis of Law. Boston, MA: Little, Brown. 1983. The Economics of Justice. Cambridge, MA: Harvard University Press. 2011. Economic Analysis of Law. 8th ed. New York, NY: Aspen Publishers. Preston McAfee, R. 1992. Amicable Divorce: Dissolving a Partnership with Simple Mechanisms. Journal of Economic Theory 56 (2):266–293. Qiao, Shitong. 2017. Chinese Small Property: The Co-Evolution of Law and Social Norms. Cambridge, UK: Cambridge University Press.

Published online by Cambridge University Press

406

references

Ralli, Tommi, and Katja Weckström. 2010. National Report on the Real Property in Finland. In Real Property Law and Procedure in the European Union 304–391. Florence, Italy: European University Institute, www.eui.eu/ documents/departmentscentres/law/researchteaching/researchthemes/­ europeanprivatelaw/realpropertyproject/Finland.pdf. Rallo, Juan Ramón. 2019. Property Is Only Another Name for Decentralized Creation of Knowledge. European Journal of Law and Economics 47 (1):43–55. Ramseyer, J. Mark, and Minoru Nakazato. 1999. Japanese Law: An Economic Approach. Chicago, IL: University of Chicago Press. Rascha, Joseph, and Robert F. Dolana. 2021. New York Law and Practice of Real Property. 2nd ed. Vol. 1. New York, NY: WestLaw. Rauterberg, Gabriel. 2020. The Essential Roles of Agency Law. Michigan Law Review 118:609–653. Reboul-Maupin, Nadège. 2020. Droit des Biens. 8th ed. Paris, France: Dalloz. Reid, Candace. 1986. Partitions in Kind: A Preference without Favor. Cardozo Law Review 7:855–886. Reimann, Mathias. 2006. Comparative Law and Economic Analysis of Law. In The Oxford Handbook of Comparative Law 839–867, edited by Mathias Reimann and Reinhard Zimmermann. Oxford, UK: Oxford University Press. Richardson, Sally Brown. 2015. Abandonment and Adverse Possession. Houston Law Review 52:1385–1430. Ritaine, Eleanor Cashin. 2011. National Report on the Transfer of Movables in France. In National Reports on the Transfer of Movables in Europe, Volume 4: France, Belgium, Bulgaria, Poland, Portugal 1–188, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Rivers, Faith. 2007. Inequality in Equity: The Tragedy of Tenancy in Common for Heirs’ Property Owners Facing Partition in Equity. Temple Political and Civil Rights Law Review 17:1–81. Rizzolli, Matteo. 2009. Building Encroachments. Review of Law and Economics 5:661–693. Romañach Jr., Julio. 2003. Federal Civil Code of Mexico Translated into English: With an Introduction and Index. Baton Rouge, LA: Lawrence Publishing Company. 2009. Civil Code of Panama 2009: Translated into English with an Introduction and Index. Baton Rouge, LA: Lawrence Publishing Company. 2014a. Civil Code of Costa Rica and Family Code of Costa Rica Translated into English: With an Introduction and Index. Baton Rouge, LA: Lawrence Publishing Company. 2014b. Civil Code of Peru Translated into English: With an Introduction and Index. Baton Rouge, LA: Lawrence Publishing Company. 2015a. Civil and Commercial Code of Argentina Translated into English: With an Introduction and Index. Baton Rouge, LA: Lawrence Publishing Company.

Published online by Cambridge University Press

references

407

2015b. Civil Code of Brazil Translated into English: With an Introduction, Index and Glossary of Selected Brazilian Civil Law Terms. 3rd ed. Baton Rouge, LA: Lawrence Publishing Company. Ronovská, Katerina, Eva Dobrovolná, Bohumil Havel, and Vlastimil Pihera. 2020. Property and Trust Law in the Czech Republic. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Rose, Carol M. 1988. Crystals and Mud in Property Law. Stanford Law Review 40:577–610. 1997. The Shadow of the Cathedral. Yale Law Journal 106:2175–2200. 1998. The Several Futures of Property: Of Cyberspace and Folk Tales, Emission Trades and Ecosystem. Minnesota Law Review 83:129–182. 2000. Property and Expropriation: Themes and Variations in American Law. Utah Law Review 2000:1–38. 2011. Servitudes. In Research Handbook on the Economics of Property Law 296–325, edited by Kenneth Ayotte and Henry E. Smith. Cheltenham, UK: Edward Elgar. Ross, Alf. 1957. Tû-Tû. Harvard Law Review 70:812–825. Roth, Herbert. 2017. J. von Staudingers Kommentar zum Bürgerlichen Gesetzbuch: Staudinger BGB – Buch 3: Sachenrecht. Berlin, Germany: C.H. Beck. Rudolf, Claudia, Vesna Rijavec, and Tomaž Keresteš. 2009. National Report on the Transfer of Movables in Slovenia. In National Reports on the Transfer of Movables in Europe, Volume 1: Austria, Estonia, Italy, Slovenia 514–638, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Rupesinghe, Pubudini Wickramaratne, and Nuwan Rupesinghe. 2007. An Introduction to Housing and Land Laws in Sri Lanka. Geneva, Switzerland: Centre on Housing Rights and Evictions. Rustichini, Aldo, Mark A. Satterthwaite, and Steven R. Williams. 1994. Convergence to Efficiency in a Simple Market with Incomplete Information. Econometrica 62 (5):1041–1063. Sagaert, Vincent. 2011. Harmonization of Security Rights on Immoveables: An Ongoing Story. In Towards a European Civil Code 1043–1063, edited by Arthur S. Hartkamp, Martijn W. Hesselink, Ewoud H. Hondius, Chantal Mak and C. Edgar du Perron. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. 2017. Property Law. In Introduction to Belgian Law 205–226, edited by Marc Kruithof and Walter De Bondt. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Salomons, Arthur F. 2009. On the Economics of Good Faith Acquisition Protection in the DCFR. In The Politics of the Draft Common Frame of Reference 199–216, edited by Alessandro Somma. New York, NY: Wolters Kluwer. 2011. Good Faith Acquisition of Movables. In Towards a European Civil Code 1065–1082, edited by Arthur S. Hartkamp, Martijn W. Hesselink, Ewoud H.

Published online by Cambridge University Press

408

references

Hondius, Chantal Mak and C. Edgar du Perron. The Hague, the Netherlands: Wolters Kluwer. 2013. The Purpose and Coherence of the Rules on Good Faith Acquisition and Acquisitive Prescription in the Draft Common Frame of Reference: A Tale of Two Gatekeepers. European Review of Private Law 21 (3):843–862. Salomons, Arthur, and Matthias Haentjens. 2017. Property and Trust Law in the Netherlands. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Sandor, Istvan. 2017. Property and Trust Law in Hungary. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Santisteban, Sonia Martin, and Peter Sparkes, eds. 2015. Protection of Immovables in European Legal Systems. Cambridge, UK: Cambridge University Press. Schaeferdiek, Sascha. 2015. Schweden. In Handbuch Immobilienrecht in Europa: Zivil- und Steuerrechtliche Aspekte des Erwerbs, der Veräusserung und der Vererbung von Immobilien 1275–1314, edited by Susanne Frank and Thomas Wachter. Munich, Germany: C.F. Müller. Schmid, Christoph U., and Christian Hertel. 2005. Real Property Law and Procedure in the European Union. Würzburg, Germany: European University Institute Florence. Schäfer, Hans-Bernd. 2019. Rules versus Standards in Developing Countries: The Case for Clear and Precise Legal Norms on Eminent Domain Power. Law and Development Review 12 (2):425–455. Schäfer, Hans-Bernd, and Claus Ott. 2004. The Economic Analysis of Civil Law. Cheltenham, UK: Edward Elgar. 2022. The Economic Analysis of Civil Law. 2nd ed. Cheltenham, UK: Edward Elgar. Schwartz, Allan, and Robert E. Scott. 2011. Rethinking the Laws of Good Faith Purchase. Columbia Law Review 111:1332–1384. Schwartz, Shalom H. 1999. A Theory of Cultural Values and Some Implications for Work. Applied Psychology: An International Review 48 (1):23–47. Schwartz, Warren F. 1989. Objective and Subjective Standards of Negligence: Defining the Reasonable Person to Induce Optimal Care and Optimal Populations of Injurers and Victims. Georgia Law Journal 78:241–279. Schweizer, Urs. 2006. Universal Possibility and Impossibility Results. Games and Economic Behavior 57 (1):73–85. See, Alvin, Man Yip, and Yihan Goh. 2018. Property and Trust Law in Singapore. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Sentell, C. Sherburne, III. 1994. Fixing the Right of Passage from an Enclosed Estate: Deciding Where to Break out Using Louisiana Civil Code Article 692. Louisiana Law Review 54:1659–1684. Serkin, Christopher. 2013. The Law of Property. New York, NY: Foundation Press. Setten, Gerrit J. van. 2015. Finnland. In Handbuch Immobilienrecht in Europa : Zivil- und Steuerrechtliche Aspekte des Erwerbs, der Veräusserung und der Vererbung von Immobilien 265–301, edited by Susanne Frank and Thomas Wachter. Munich, Germany: C.F. Müller.

Published online by Cambridge University Press

references

409

Shaelou, Stéphanie Laulhé, Stelia Stylianou, and Karolina Anastasiou. 2010. National Report on the Transfer of Movables in Cyprus. In National Reports on the Transfer of Movables in Europe, Volume 2: England and Wales, Ireland, Scotland, Cyprus 471–608, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Shahabi, Mostafa. 2007. The Civil Code of Iran. London, UK: Wildy, Simmonds and Hill Publishing. Shavell, Steven. 2004. Foundations of Economic Analysis of Law. Cambridge, MA: Belknap Press of Harvard University Press. Sheehan, Duncan. 2011. The Principles of Personal Property Law. Oxford, UK: Hart. Simón-Moreno, Héctor. 2012. The Eurohypothec: A Perspective from Spanish Law. European Property Law Journal 1 (2):191–231. Singer, Joseph William. 2010. Property. 3rd ed. New York, NY: Aspen Publishers. Singh, Avtar, and Harpreet Kaur. 2020. Textbook on the Transfer of Property Act. 6th ed. Delhi, India: LexisNexis. Sitkoff, Robert H., and Max M. Schanzenbach. 2005. Jurisdictional Competition for Trust Funds: An Empirical Analysis of Perpetuities and Taxes. Yale Law Journal 115:356–437. Škerl, Jerca Kramberger, and Ana Vlahek. 2010. Property and Trust Law in Slovenia. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. 2020. Property and Trust Law in Slovenia. 2nd ed. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Smaliukas, Andrius. 2017. Property and Trust Law in Lithuania. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Smith, Henry E. 2000. Semicommon Property Rights and Scattering in the Open Fields. Journal of Legal Studies 29 (1):131–169. 2002. Exclusion versus Governance: Two Strategies for Delineating Property Rights. Journal of Legal Studies 31:S453–S487. 2003. The Language of Property: Form, Context, and Audience. Stanford Law Review 55:1105–1191. 2004a. Exclusion and Property Rules in the Law of Nuisance. Virginia Law Review 90 (4):965–1049. 2004b. Property and Property Rules. New York University Law Review 79:1719–1798. 2005. Governing the Tele-Semicommons. Yale Journal on Regulation 22:289–314. 2006. Modularity in Contracts: Boilerplate and Information Flow. Michigan Law Review 104:1175–1222. 2007. Intellectual Property as Property: Delineating Entitlements in Information. Yale Law Journal 116:1742–1822. 2008. Governing Water: The Semicommons of Fluid Property Rights. Arizona Law Review 50:445–478. 2009a. Institutions and Indirectness in Intellectual Property. University of Pennsylvania Law Review 157:2083–2133.

Published online by Cambridge University Press

410

references

2009b. Law and Economics: Realism or Democracy? Harvard Journal of Law & Public Policy 32:127–145. 2011a. Modularity and Morality in the Law of Torts. Journal of Tort Law 4:1–32. 2011b. Standardization in Property Law. In Research Handbook on the Economics of Property Law 148–173, edited by Kenneth Ayotte and Henry E. Smith. Cheltenham, UK: Edward Elgar. 2012a. On the Economy of Concepts in Property. University of Pennsylvania Law Review 160:2097–2128. 2012b. Property as the Law of Things. Harvard Law Review 125:1691–1726. 2013a. Why Fiduciary Law Is Equitable. In Philosophical Foundations of Fiduciary Law 261–284, edited by Andrew S. Gold and Paul B. Miller. Oxford, UK: Oxford University Press. 2015. The Persistence of System in Property Law. University of Pennsylvania Law Review 163:2055–2084. 2017. Property as Complex Interaction. Journal of Institutional Economics 13 (4):809–814. 2019. Complexity and the Cathedral: Making Law and Economics More Calabresian. European Journal of Law and Economics 48 (1):43–63. 2020. Systems Theory: Emergent Private Law. In The Oxford Handbook of the New Private Law 143–158, edited by Andrew S. Gold, John C. P. Goldberg, Daniel B. Kelly, Emily Sherwin and Henry E. Smith. Oxford, UK: Oxford University Press. 2021a. Equity as Meta-Law. Yale Law Journal 130:1050–1144. 2021b. Property beyond Flatland. Brigham-Kanner Property Rights Journal 10:9–56. Smith, Imran Oluwole. 2018. Property and Trust Law in Nigeria. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Smith, Lionel. 2013b. Chapter 5: Security. In English Private Law 307–377, edited by Andrew Burrows. Oxford, UK: Oxford University Press. Spamann, Holger. 2009a. The Antidirector Rights Index Revisited. Review of Financial Studies 23:467–486. 2009b. Contemporary Legal Transplants: Legal Families and the Diffusion of (Corporate) Law. BYU Law Review 2009 (6):1813–1877. 2009c. Large-sample, Quantitative Research Designs for Comparative Law? American Journal of Comparative Law 57 (4):797–810. 2015. Empirical Comparative Law. Annual Review of Law and Social Science 11:131–153. Sparkes, Peter. 2019. Property and Trust Law in England and Wales. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Sprankling, John G. 1994. An Environmental Critique of Adverse Possession. Cornell Law Review 79:816–884. Stöcker, Otmar. 2012. The Eurohypothec. In The Future of European Property Law 65–78, edited by Sjef van Erp, Arthur Salomons and Bram Akkermans. Munich, Germany: Sellier European Law Publishers.

Published online by Cambridge University Press

references

411

Stake, Jeffrey Evans. 2001. The Uneasy Case for Adverse Possession. Georgetown Law Journal 89:2419–2474. Stanescu, Catalin Gabriel. 2017. Property and Trust Law in Romania. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Steiner, Eva. 2010. French Law: A Comparative Approach. Oxford, UK: Oxford University Press. Stenseth, Geir. 2019. The Importance of the Social Function of Property – Norway. In Léon Duguit and the Social Obligation Norm of Property: A Translation and Global Exploration 123–142, edited by Jessica Viven-Wilksch and Paul T. Babie. Singapore: Springer Sterk, Stewart E. 1987. Neighbors in American Land Law. Columbia Law Review 87:55–104. 2008. Property Rules, Liability Rules, and Uncertainty about Property Rights. Michigan Law Review 106:1286–1336. 2012. Strict Liability and Negligence in Property Theory. University of Pennsylvania Law Review 160:2129–2156. Stoebuck, William B., and Dale A. Whitman. 2000. The Law of Property. 3rd ed. St. Paul, MN: West. Stoimenov, Dimitar. 2011. National Report on the Transfer of Movables in Bulgaria. In National Reports on the Transfer of Movables in Europe, Volume 4: France, Belgium, Bulgaria, Poland, Portugal, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Strickler, Yves. 2006. Les Biens. Paris, France: Presses Universitaires de France. Sunstein, Cass R. 2018. The Cost-Benefit Revolution. Boston, MA: MIT Press. Synodinou, Tatiana Eleni. 2020. Property and Trust Law in Cyprus. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Szilágyi, Ferenc. 2011. National Report on the Transfer of Movables in Hungary. In National Reports on the Transfer of Movables in Europe, Volume 3: Germany, Greece, Lithuania, Hungary 429–701, edited by Wolfgang Faber and Brigitta Lurger. Berlin, Germany: De Gruyter. Técnica, Ministerio de Justicia – Secretaría General. 2009. Spanish Civil Code. Madrid, Spain: Ministerio de Justicia. Taliadoros, Constantine. 1982. Greek Civil Code. Athens, Greece: Ant. N. Sakkoulas Publishers. Tang, Hang Wu. 2018. From Waqf, Ancestor Worship to the Rise of the Global Trust: A History of the Use of the Trust as a Vehicle for Wealth Transfer in Singapore. Iowa Law Review 103 (5):2263–2292. Tang, Hang Wu, and Kevin F. K. Low. 2019. Tan Sook Yee’s Principles of Singapore Land Law. 4th ed. Singapore: LexisNexis. Terré, François, and Simler Philippe. 2006. Droit Civil: Les Biens. 7th ed. Paris, France: Dalloz-Sirey. 2010. Droit Civil: Les Biens. 8th ed. Paris, France: Dalloz-Sirey.

Published online by Cambridge University Press

412

references

Terré, François, and Philippe Simler. 2006. Droit Civil: Les Biens. 7th ed. Paris, France: Dalloz. Tichy, Lubos. 2010. National Report on the Transfer of Movables in Czech. In National Report on the Transfer of Movables in Europe, Volume 6: The Netherlands, Switzerland, Czech Republic, Slovakia, Malta, Latvia 262–347, edited by Wolfgang Farber and Brigitta Lurger. Berlin, Germany: De Gruyter. Tideman, T. Nicolaus, and Gordon Tullock. 1976. A New and Superior Process for Making Social Choices. Journal of Political Economy 84:1145–1159. UN-HABITAT. 2005. Housing and Property Rights in Bosnia and Herzegovinam Croatia and Serbia and Montenegro. Nairobi, Kenya: United Nations Human Settlements Programme. Underkuffler, Laura. 2011. Open, Notorious, and Continuously Occupied: A Claim for Adverse Possession. Cornell Journal of Law and Public Policy On-line http://jlpp.org/blogzine/open-notorious-and-continuously-occupied-aclaim-for-adverse-possession/. Unknown. 1969. Pledges Law, 1967. Israel Law Review 4:443–448. 1970. Land Law, 5729–1969. Israel Law Review 5 (2):292–326. Upham, Frank K. 2013. What Are Property Rights Good For? Surprising Lessons from the Chinese Experience. In Rethinking Law and Development: The Chinese Experience 82–103, edited by Guanghua Yu. New York, NY: Routledge. 2018. The Great Property Fallacy: Theory, Reality, and Growth in Developing Countries. Cambridge, UK: Cambridge University Press. Vékás, Lajos. 1995. Law of Property and Law of Succession. In Introduction to Hungarian Law 81–93, edited by Attila Harmathy. Alphen aan den Rijn, the Netherlands: Kluwer Law. Van der Merwe, C. G., Marius Johannes de Waal, and David L. Carey Miller. 2002. Property and Trust Law in South Africa. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Van der Merwe, Cornelius, ed. 2015. European Condominium Law. Cambridge, UK: Cambridge University Press. van Vliet, Lars 2017. Transfer of Property Inter Vivos. In Comparative Property Law: Global Perspectives 150–170, edited by Michele Graziadei and Lionel Smith. Cheltenham, UK: Edward Elgar. Varadarajan, Deepa. 2014. Improvement Doctrines. George Mason Law Review 21:657–716. Vickrey, William. 1961. Counterspeculation, Auctions, and Competitive Sealed Tenders. The Journal of Finance 16 (1):8–37. von Bar, Christian, ed. 1999. Sachenrecht in Europa: Systematische Einführungen und Gesetzestexte. 1. Dänemark, England, Finnland, Schweden. Osnabrück, Germany: Universitätsverlag Rasch.

Published online by Cambridge University Press

references

413

von Bar, Christian, and Ulrich Drobnig. 2004. The Interaction of Contract Law and Tort and Property Law in Europe: A Comparative Study. Munich, Germany: Sellier European Law Publishers. Waldeck, Sarah E. 2011. Rethinking the Intersection of Inheritance and the Law of Tenancy in Common. Notre Dame Law Review 87:737–780. White, James J., and Robert S. Summers. 2007. Principles of Secured Transactions. St. Paul, MN: Thomson West. Willenzik, David S. 2015. Louisiana Future Advance Mortgages: A 20-Year Retrospective. Louisiana Law Review 75:613–695. Williams, Gordon. 2011. Property and Trust Law in New Zealand. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Wilmore, Randall L. 1986. Right of Passage for the Benefit of an Enclosed Estate. Louisiana Law Review 47:199–216. Wilson, Robert. 1985. Incentive Efficiency of Double Auctions. Econometrica 53 (5):1101–1115. Wiseman, Clark. 1974. Rezoning by Auction – A New Approach to Land Use Decisions. Utah Science 35:86–89. Wolf, Michael Allan, ed. 2021. Powell on Real Property. Vol. 7. Newark, NJ: Matthew Bender & Company. Woods, Una. 2011. Property and Trust Law in Ireland. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. 2016. Adverse Possession and Boundary Disputes: Lessons for Ireland from Abroad. International Journal of Law in the Built Environment 8:56–79. Wyman, Katrina M. 2017. The New Essentialism in Property. Journal of Legal Analysis 9 (2):183–246. 2019. Property in Radical Markets. The University of Chicago Law Review 87 (1):125–137. Xu, Guodong. 2007. Codification of Civil Codes in Comparative Perspective [比较法视野中的民法典编纂]. Beijing, China: Peking University Press. Yee, Tan Sook, and Hans Tjio. 2000. Property and Trust Law in Singapore. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Yiannopoulos, A. N. 1996. The Legal Servitude of Passage. Tulane Law Review 71:1–44. Yiannopoulos, Athanassios N. 2008. Property. In Introduction to Greek Law 153–166, edited by Konstantinos D. Kerameus and Phaedon J. Kozyris. Alphen aan den Rijn, the Netherlands: Wolters Kluwer. Zamir, Eyal, and Doron Teichman. 2018. Behavioral Law and Economics. Oxford, UK: Oxford University Press. Zerbe, Richard O. 2001. Economic Efficiency in Law and Economics. Northampton, MA: Edward Elgar. Zhang, Taisu. 2017. The Laws and Economics of Confucianism. Cambridge, UK: Cambridge University Press.

Published online by Cambridge University Press

414

references

Ziadeh, Farhat J. 1979. Property Law in the Arab World: Real Rights in Egypt, Iraq, Jordan, Lebanon, Libya, Syria, Saudi Arabia and the Gulf States. London, UK: Graham & Trotman. Ziff, Bruce. 2014. Principles of Property Law. 6th ed. Toronto, Canada: Thomson Reuters. Zimmermann, Reinhard, Kenneth Reid, and Daniel Visser. 2005. Co-ownership. In Mixed Legal Systems in Comparative Perspective: Property and Obligations in Scotland and South Africa 704–733, edited by Reinhard Zimmermann, Kenneth Reid and Daniel Visser. Oxford, UK: Oxford University Press. Živković, Miloš. 2007. Country Report Serbia. www.ius.bg.ac.rs/prof/materijali/ xivmil/vdpCountryReportSerbiaF30092007.pdf. Zweigert, Konrad, and Hein Kötz. 1998. Introduction to Comparative Law. 3rd ed. Oxford, UK: Oxford University Press. Zywicki, Todd J., and Edward Peter Stringham. 2013. Common Law and Economic Efficiency. In Encyclopedia of Law and Economics, edited by Francesco Parisi and Richard A. Posner. Northampton, MA: Edward Elgar.

Published online by Cambridge University Press

INDEX

absolutism, 17, 105, 109, 115–117 abstract principle, 101, 331 accessio, 20, 60, 169, 339, 343, 344, 346–351 accession, 7, 8, 16, 72, 160, 161, 168, 169, 178, 309, 330, 335, 336, 340, 344, 345 accessory thing, 337 acquisitive prescription, 6, 8, 16, 17, 36, 55, 58, 125–133, 135, 136, 138–143, 145–148, 150, 152, 154, 155, 157, 158, 168, 169, 262, 263, 269, 293, 294, 306, 352, 355 adverse possession, xxi, 17, 118, 125–128, 132, 133, 136, 138, 144–154, 156, 157, 159–161, 168, 169, 171, 178, 262, 272, 332, 333, 353, 354 Afghanistan, 31, 70, 72, 73, 77, 80, 103, 107, 109, 129, 135, 181, 208, 212, 233, 234, 267, 274, 294, 337, 346, 357, 363 Albania, 31, 85, 86, 103, 107, 109, 133, 181, 184, 185, 212, 235, 265, 292, 294, 297, 317, 342, 346, 349, 357, 364 Algeria, 31, 70, 77, 79, 80, 85, 102, 107, 109, 133, 167, 183, 208, 213, 214, 233, 234, 267, 274, 294, 343, 346, 357, 364 allocative benefit, 5, 48–51, 54, 55, 57, 87, 146–150, 152, 171, 177, 202, 215, 237, 270–273, 277, 281, 321, 325–328, 333 alteration, 309, 330, 333 America, 1, 14, 18, 28, 43, 62, 80, 85, 90, 118, 126, 133, 166, 186, 187, 189, 203, 206, 208, 210, 223, 231, 232, 236, 264, 294, 296, 297, 310, 312, 313, 317, 324, 332

anglophone, 1, 356 Angola, 31, 130, 132, 165, 212, 364 anticresis, 66, 79, 80, 93, 94 apparent title, 126, 128–130, 133, 134, 136, 138, 141–143, 145, 146, 157, 158, 333 Argentina, 31, 64, 69, 70, 73, 80, 87, 103, 106, 108, 111, 112, 128, 130, 131, 134, 163–165, 167, 178, 181, 206, 212, 218, 229, 234, 267, 291, 298, 311, 318, 342, 346, 349, 357, 364 Armenia, 31, 64, 72, 74, 83, 84, 103, 107, 111, 136, 168, 184, 213, 218, 236, 266, 274, 292, 298, 318, 357, 364 attornment, 36, 111, 112, 120 Australia, 1, 31, 62, 63, 65, 67, 77, 82, 84, 86, 103, 108, 135, 166–168, 185, 208, 210, 213, 215–217, 233, 236, 265, 293, 297, 306, 318, 337, 343, 344, 356, 370, 373, 375–377 Austria, 29, 31, 65, 69, 70, 72, 81, 94, 95, 104, 108, 111, 112, 136, 169, 179, 181, 208, 213, 267, 292, 298, 311, 312, 340, 345, 364, 370 Azerbaijan, 31, 69, 72, 81, 83, 84, 103, 108, 109, 130, 132, 135, 142, 167, 184, 213, 218, 234, 267, 274, 292, 298, 318, 342, 344, 357, 364 Bahrain, 31, 70, 77, 79, 80, 103, 107, 111, 112, 135, 167, 183, 208, 213, 214, 218, 267, 274, 299, 343, 346, 349, 357, 364, 376 bargaining power, 49, 57, 59, 77, 160, 172, 237–243, 245, 247

415

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

416

index

Belarus, 31, 73, 74, 77, 81, 83, 84, 103, 107, 111, 136, 168, 184, 213, 218, 236, 266, 274, 292, 298, 318, 357, 364 Belgium, 1, 31, 65, 73, 80, 84, 102, 108, 109, 134, 183, 208, 213, 233, 234, 264, 274, 298, 316, 346, 357, 364, 373 Benin, 24, 31, 73, 99, 103, 113, 135, 168, 374 Bhutan, 24, 31, 84, 105, 234, 372, 375 bilateral monopoly, 174, 242, 243, 322 Bolivia, 31, 65, 72, 80, 94, 103, 109, 130, 131, 135, 182, 212, 218, 233, 235, 256, 267, 274, 291, 294, 298, 308, 315, 342, 346, 357, 364 Bosnia and Herzegovina, 24, 31, 83, 168, 372 Botswana, 24, 31, 370 Brazil, 28, 31, 42, 43, 64, 65, 73, 77, 79, 82, 83, 94, 103, 108, 111, 112, 130, 132, 133, 164, 165, 180, 181, 212, 233, 236, 268, 287, 291, 294, 296, 298, 317, 342, 345, 346, 357, 364 Brunei, 24, 31, 105, 373 Bulgaria, 13, 31, 64, 72, 85, 103, 109, 135, 179, 180, 212, 266, 274, 291, 297, 314, 317, 337, 343, 344, 375 Burkina Faso, 31, 80, 102, 107, 109, 134, 182, 233, 235, 264, 298, 316, 346, 364 Burundi, 31, 64, 79, 102, 109, 135, 182, 213, 232, 236, 264, 317, 342, 346, 364 California, 13, 31, 41, 62, 84–86, 103, 106, 111, 114, 135, 136, 166–168, 182, 213, 233, 260, 292, 297, 315, 317, 340, 343, 345, 356, 362, 364, 370, 376 call option (call-option), 52–55, 59, 166, 175, 177, 181, 182, 184, 185, 192, 202, 204, 239, 241 Cambodia, 29, 31, 79, 103, 109, 135, 181, 212, 233, 235, 267, 274, 292, 298, 313, 315, 342, 344, 357, 364 Canada, 41, 63, 65, 67, 79, 82, 168, 181, 210, 217, 228, 233, 265, 293, 299, 318, 340, 344, 351, 372, 377

Cape Verde, 31, 130, 132, 165, 212, 364 Central African Republic, 24, 31 certificate, 108, 139, 263, 264, 293, 306, 339 Chile, 31, 65, 79, 80, 82, 87, 103, 111, 130, 131, 134, 182, 211, 233, 236, 268, 287, 291, 294, 298, 317, 339, 343, 345, 365 China, xx, 1, 9, 25, 29–31, 39, 42, 64, 71, 72, 79, 81, 82, 84, 103, 106, 108, 109, 111, 112, 125, 132, 181, 205, 212, 214, 225, 227, 228, 232, 234, 269, 289, 291, 296, 313, 318, 343, 346, 352, 357, 358, 365 collectivistic, 227 Colombia, xx, 31, 65, 80, 84, 87, 103, 111, 112, 130, 131, 134, 140, 155, 184, 211, 233, 234, 236, 263, 291, 298, 317, 336, 339, 343, 344, 365, 373 color of title, 126, 128 combination, 8, 19, 20, 48, 55, 56, 94, 136, 161, 227, 300, 302, 321, 336, 342, 344, 347 commingling, 336 commixtio, 335, 336 commonhold, 228, 354 Comoros, 31, 80, 102, 107, 109, 134, 182, 233, 235, 264, 298, 316, 346, 365 comparative law, 2–4, 13, 45, 179 international comparative law, 23 national comparative law, 23 compensation, 20, 53–56, 58–60, 116, 118, 156, 157, 161, 165, 166, 170, 172, 175, 176, 181, 190, 195–198, 202, 203, 230–234, 237, 239, 240, 242–244, 246, 248, 252, 254–256, 261, 266, 283, 287, 292, 311, 315–318, 321, 324, 327, 329–334, 337, 341, 345–350, 355 complexity, 9, 86, 203 condominium, 228, 354 confusio, 20, 309, 312, 313, 335–337, 339, 343, 344, 348, 349, 351 constitutive, 83, 101–103, 105, 106, 114, 121, 139, 145, 146, 153 constitutum possessorium, 111, 112, 120 constructive notice, 106, 115, 116, 118, 121, 139

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

index convergence, 5–11, 20, 259, 335, 352, 354, 359 cooperative (co-operative), 227, 228, 354 Costa Rica, 31, 78, 84, 87, 103, 108, 109, 112, 130–132, 158, 169, 181, 233, 236, 267, 274, 292, 298, 316, 340, 343, 346, 365, 373 cost–benefit analysis, 15, 48, 51, 177 covenant, 12, 73, 180, 205–211, 225–227, 229, 254 Croatia, 13, 31, 65, 69, 72, 81, 84, 103, 108, 111, 112, 134, 162, 164, 165, 181, 208, 213, 218, 234, 266, 274, 292, 297, 312, 341, 344, 347, 370 Cuba, 31, 65, 73, 102, 109, 134, 154, 156, 164, 182, 212, 233, 234, 256, 291, 297, 317, 336, 342, 347, 365 culture(cultural), 5, 9, 11, 12, 45, 219, 220, 259, 269, 297, 304 Cyprus, 31, 36, 64, 103, 108, 135, 182, 236, 265, 293, 306, 337, 344 Czech, 32, 64, 69, 72, 79, 81, 84, 104, 108, 109, 134, 145, 164, 172, 183, 213, 233, 236, 248, 256, 268, 292, 297, 318, 341, 346, 365 declaratory, 101, 102, 106, 118, 119, 146, 151 default rule, 17, 36, 56, 57, 69–71, 80, 83, 109, 119–121, 202, 255, 256, 310, 325, 326 delivery, 17, 36, 101, 102, 104, 110–112, 119–121, 353, 354 démembrement, 94 Democratic Republic of Congo, xxvi, 24, 32, 73, 133, 264, 373 Denmark, 25, 32, 63, 78, 81, 84, 103, 108, 110, 136, 168, 262, 291, 317, 342, 344, 358, 374 dian, 66, 79–81, 93 disparity of value test, 316 disproportionately high cost, 348 divergence, 4–6, 8–11, 359 Doing Business, 4, 98, 117, 130, 153–156, 158 Dominican Republic, 32, 65, 80, 103, 108, 109, 134, 183, 233, 235, 264, 298, 316, 346, 365, 376

417

easement, 18, 67, 69, 118, 129, 162, 163, 166, 231–233, 236, 239, 240, 244, 245, 252, 254, 371 personal easement, 66, 67, 69–71, 94, 97 real easement, 17, 66, 67, 69, 72, 92, 94, 95, 166, 353 statutory easement, 18, 19, 230–235, 238, 240, 242–248, 250–256 easement of necessity (easements of necessity), 7, 18, 19, 230–235, 251–256 East Asia, 23, 25, 28 economic value, 48–50, 54, 55, 61, 148, 150, 165, 170, 171, 193, 196, 199, 201–204, 237, 240, 244–246, 272, 287, 323, 327, 328, 330, 331, 348 economy of scale (economies of scale), 188, 189, 191, 192, 197, 202–204 Ecuador, xx, 32, 65, 80, 87, 103, 111, 130, 131, 134, 140, 182, 211, 233, 234, 236, 268, 287, 291, 298, 309, 317, 336, 339, 343, 345, 365 efficiency, 4, 5, 10, 11, 15, 17, 19, 20, 46–52, 60, 98, 114, 119, 126, 141–143, 146, 148, 152, 158, 160, 170–178, 188, 189, 192, 194, 197, 198, 216, 226, 243, 245, 246, 249, 250, 255, 260, 269–273, 275, 276, 280–288, 302, 307, 311, 321, 323, 324, 327, 329, 330, 333, 334, 350, 355, 356 Egypt, 32, 70, 77, 79, 80, 82, 103, 108, 109, 133, 167, 183, 208, 213, 214, 233, 234, 267, 274, 294, 343, 344, 346, 357, 365 El Salvador, 32, 64, 75, 80, 87, 103, 111, 130, 131, 134, 140, 182, 211, 233, 236, 291, 298, 317, 336, 339, 343, 345, 365 embezzled, 260, 273–276, 280 emphyteusis, 66, 71, 72, 74, 95 endowment effect, 147 England and Wales, 32, 214, 233, 293, 297 Equatorial Guinea, 32, 69, 80, 86, 103, 110, 130, 131, 134, 182, 212, 218, 233, 235, 256, 266, 274, 292, 298, 318, 340, 344, 346, 365 equity power, 169, 206, 226, 227

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

418

index

Eritrea, 32, 86, 103, 107, 111, 112, 136, 185, 211, 213, 233, 234, 266, 292, 297, 317, 337, 343, 347, 365 essential part, 348 Estonia, 13, 32, 64, 69, 70, 72, 78, 82, 84, 104, 106, 108, 111, 112, 130, 132, 136, 163, 164, 166, 168, 180, 185, 208, 209, 212, 218, 233, 236, 266, 292, 297, 317, 342, 347, 357, 378 Ethiopia, 32, 77, 79, 103, 108, 111, 135, 182, 213, 233, 234, 267, 291, 297, 317, 342, 345, 349, 357 EU, 6, 41, 259 ex ante, 15, 18, 19, 46, 47, 49, 58, 119, 139, 141, 152, 161, 162, 170–175, 178, 189, 190, 196, 199, 201, 225, 227, 251, 254–256, 273–275, 277, 282, 283, 286–288, 307, 322–328, 331 ex post, 18, 19, 46, 47, 50, 58, 59, 141, 161, 162, 170, 173–175, 178, 188, 254, 273, 278, 281–283, 285–288, 303, 322, 323, 325–328, 331, 333 external cost, 91 extinctive prescription, 141 extra-ordinary prescription, 131 fiduciary, 185, 214 finders, 8, 16, 19, 289–294, 296–303, 306–308, 324 Finland, 25, 32, 63, 72, 78, 79, 81, 83, 84, 103, 108, 110–112, 130–132, 158, 168, 182, 213, 266, 292, 337, 344, 345, 358, 370, 371 floating charge, 66, 84, 85, 371 foreclosure, 75, 77, 78, 80, 93 four unities, 179 fractional ownership, 50, 259, 270, 281–288, 349 France, xix, xx, 7, 28, 32, 36, 42, 65, 70, 73, 77, 79, 80, 83, 84, 86, 103, 108, 109, 114, 115, 121, 134, 140, 169, 179, 183, 213, 227, 233–235, 262–264, 293, 298, 311, 316, 336, 337, 339, 343, 344, 346, 347, 349, 358, 365 Francophone, 78, 306 functional approach, 110, 111 future interest, 63, 66, 86, 87, 90, 96, 97, 99, 353, 354

Georgia, 32, 69, 72, 78, 82, 83, 85, 103, 108, 111, 112, 130, 132, 133, 163, 166, 182, 208, 209, 212, 218, 234, 236, 263, 267, 274, 292, 298, 312, 342, 344, 357, 365 Germany, 7, 23, 25, 29, 32, 36, 39, 42, 65, 69, 70, 72, 78, 79, 81, 84, 104, 106, 108, 111, 112, 121, 127, 130, 132, 135, 147, 152, 157, 163, 164, 166, 182, 208, 213, 218, 233, 234, 236, 255, 266, 274, 292, 298, 310, 315, 342, 344, 347, 348, 365, 371 Gower distance, 24, 37–40 Greece, 32, 65, 69, 72, 83, 84, 104, 108, 111, 112, 114, 115, 134, 165, 167, 168, 182, 208, 213, 218, 233, 236, 266, 274, 292, 298, 315, 342, 344, 365 Grundschuld, 66, 78, 81 Guatemala, 32, 78, 82, 83, 86, 103, 111, 134, 182, 206, 212, 218, 233, 236, 268, 291, 298, 300, 318, 340, 344, 346, 357, 366 Guinea, 32, 102, 110, 134, 182, 235, 264, 266, 341, 345, 366 Guinea-Bissau, 32, 64, 130, 132, 165, 212, 366 Haiti, 32, 70, 80, 110, 134, 182, 235, 264, 274, 298, 316, 340, 343, 346, 347, 366 hierarchical clustering, 15, 23, 24, 37, 39 Hohfeld, 13, 14, 53, 54, 55, 70, 93, 94, 181 hold-out, 61, 220, 224, 228, 322 Honduras, 32, 65, 78, 80, 84, 87, 103, 111, 130, 131, 134, 182, 206, 212, 218, 233, 236, 291, 298, 317, 336, 339, 343, 344, 366, 375 Hong Kong, xx, 1, 32, 41, 62, 65, 78, 79, 81, 84, 86, 103, 110, 135, 156, 168, 182, 208, 210, 214, 233, 236, 265, 293, 299, 344, 371, 374, 376–378 Hungary, 32, 65, 78, 84, 104, 108, 111, 112, 134, 164, 165, 168, 182, 213, 218, 232, 234, 268, 292, 297, 318, 342, 347, 357, 366 hybrid rule, 18, 19, 237–242, 244, 246, 247, 250, 252, 256

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

index hypothec, 6, 17, 62, 63, 66, 74, 79, 80, 84, 94 in gross, 67, 69 in rem, 3, 10, 56, 67, 81, 90, 93, 130, 215, 260, 352, 357 incentive efficient, 282 India, xx, 32, 63, 80, 84, 86, 103, 110, 135, 168, 185, 211, 214, 233, 265, 291, 298, 370–373, 377 individualism, 219–221, 227, 228 Indonesia, 32, 73, 103, 108, 111, 134, 185, 235, 268, 287, 298, 315, 342, 346, 362, 366 information cost, 5–7, 17, 48, 49, 54, 56, 88–91, 94, 96, 101, 115, 116, 138, 139, 149–153, 155, 170, 223, 225, 229, 288 institution cost, 2, 5, 15, 16, 47, 48, 50, 51, 56, 88, 89, 95, 97, 113–115, 118–120, 146, 150, 153, 177, 203, 223, 237, 272–274, 280, 281, 283, 322, 326, 328, 352, 353 internal auction, 18–20, 50, 58, 60, 61, 149, 150, 180, 182–184, 186, 188, 198, 201–204, 259, 270, 281–288, 341, 346, 349, 350 Iran, 32, 74, 80, 82, 84, 86, 87, 95, 102, 111, 182, 214, 262, 290, 292, 294, 299, 306, 366 Iraq, 28, 32, 64, 72, 79, 80, 103, 110, 134, 182, 208, 211, 212, 233, 234, 266, 296, 337, 346, 357, 366 Ireland, 32, 62, 63, 65, 71, 77, 84, 86, 103, 108, 110, 135, 168, 182, 208, 210, 214, 215, 233, 264, 265, 293, 297, 312, 315, 337, 344, 373, 377, 378 Israel, xx, 25, 32, 42, 44, 64, 70, 78, 79, 83, 84, 86, 103, 108, 111, 135, 167, 169, 182, 208, 213, 218, 268, 287, 292, 297, 303, 312, 313, 342, 345, 370, 371, 373, 375, 376, 378 Italy, 32, 65, 72, 80, 82, 84, 103, 111, 114, 130–132, 135, 142, 166, 167, 184, 206, 208, 213, 218, 229, 233–235, 268, 292, 298, 315, 342, 346, 366

419

Ivory Coast (Côte d’Ivoire), 32, 80, 102, 107, 109, 134, 183, 233, 235, 264, 298, 316, 346, 366 Japan, 28, 29, 32, 36, 38, 39, 64, 72, 73, 79, 84, 98, 103, 110, 113, 115, 129, 135, 182, 212, 233, 234, 237, 248, 256, 267, 274, 292, 298, 315, 342, 345, 366, 371 joint and indivisible ownership, 358 joint tenancy, 179, 358 Jordan, 33, 73, 80, 103, 110, 129, 134, 184, 213, 233, 234, 267, 294, 299, 343, 346, 357, 366 Judge Posner, 4, 12, 53, 241, 356 judge-made law, 12, 355 jural composite, 93 just title, 128 Kazakhstan, 33, 64, 70, 73, 74, 77, 82, 85, 103, 111, 136, 168, 184, 214, 218, 234, 236, 267, 274, 292, 298, 318, 357, 366, 373 keepers, 19, 289, 291, 293, 300, 301 Kuwait, 33, 79, 80, 103, 110, 135, 167, 183, 206, 208, 213, 214, 218, 267, 274, 294, 299, 343, 346, 357, 366 Kyrgyzstan, 33, 64, 73, 74, 83, 84, 103, 111, 136, 168, 184, 214, 218, 236, 266, 274, 292, 298, 308, 318, 357, 366, 374 land charge, 66, 78 land use, 70, 106, 237, 252 landlocked, 16, 18, 50, 57, 118, 230, 231, 233, 234, 236–256 Laos, 1, 29, 33, 111, 136, 168, 184, 214, 236, 268, 292, 297, 357, 375 Latin America, 28, 78, 130, 131, 211, 357 Latvia, 33, 65, 67, 69, 79–81, 84, 104, 111, 112, 134, 145, 185, 214, 225, 267, 292, 298, 314, 318, 341, 345, 347, 357, 367 lease, 3, 70, 73, 103, 118, 155, 325 least damage, 18, 19, 231, 234, 235, 243, 248–251, 253, 255, 256 legal family, 3, 13, 23, 24, 37, 38, 357 legal origin, 12, 40, 41, 357

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

420

index

legal servitude of passage, 230–233, 237, 238, 241, 252 liability rule, 15, 18, 46, 52–54, 57–59, 81, 148, 170, 171, 175, 237–240, 244, 245, 247, 249, 256, 280, 321, 322, 326, 332 Liberia, 24, 33, 63, 103, 108, 182, 373 Libya, 33, 72, 79, 80, 82, 86, 103, 110, 129, 134, 167, 180, 208, 213, 214, 233, 234, 267, 274, 294, 298, 343, 346, 357, 367 Liechtenstein, 33, 65, 69, 72, 77–79, 81, 104, 108, 111, 112, 130, 132, 136, 154, 156, 164, 183, 206, 208, 212, 234, 267, 274, 292, 298, 317, 342, 347, 367, 378 lien, 37, 62, 63, 66, 78, 80, 85, 86, 198 life estate, 66, 73, 90, 96, 97, 100 limited property form, 3, 50, 66, 74, 87, 88, 92–95, 98–100, 353 line-of-credit, 66, 67, 78, 79 Lithuania, 33, 65, 73, 77, 79, 83, 84, 103, 107, 111, 136, 168, 184, 208, 214, 218, 236, 266, 274, 292, 298, 317, 341, 367 locus owner, 296–299, 307, 308 lost things, 289–291, 296, 299–301, 303, 307, 308 Louisiana, 14, 33, 41, 64, 69, 70, 79, 84, 86, 103, 110, 125, 128, 133, 134, 138, 154, 166–168, 182, 214, 218, 230, 233–235, 237, 260, 265, 267, 274, 292, 298, 317, 340, 343, 346, 367 Luxembourg, 33, 80, 103, 108, 109, 134, 183, 214, 233, 235, 264, 298, 316, 346, 357, 367 Macau, 1, 33, 41, 64, 72, 77, 86, 103, 108, 110, 130, 132, 135, 154, 156, 164, 165, 182, 208, 213, 218, 234, 252, 256, 268, 287, 292, 298, 318, 345–347, 357, 367, 371 (North) Macedonia, 13, 33, 64, 69, 77, 81, 85, 103, 106, 108, 111, 112, 134, 162, 164, 165, 184, 208, 213, 227, 234, 267, 268, 274, 292, 297, 312, 317, 341, 342, 347, 370

machine-learning, 23, 36, 37, 39, 45 supervised machine-learning, 30 unsupervised machine-learning, 3, 13, 15, 23 Madagascar, 33, 79, 80, 102, 107, 109, 168, 183, 233, 235, 264, 316, 346, 367, 375 majority rule, 212, 219, 220 Malawi, 24, 33, 110, 135, 168, 182, 214, 265, 376, 377 Malaysia, 33, 63, 78, 84, 86, 104, 108, 110, 135, 182, 211, 214, 232, 236, 265, 293, 297, 315, 345, 372, 375, 377, 378 Maldives, 24, 33, 297, 372 Mali, 24, 33, 70, 79, 371 Malta, 33, 65, 70, 80, 82, 84, 86, 103, 110, 112, 134, 164, 182, 214, 233, 234, 256, 267, 274, 292, 298, 311, 317, 334, 340, 343, 345, 367 manufacturing, 309 marginal cost, 88–90, 245–248, 286 marginal revenue, 245–248 market overt rule, 19, 264–269, 276, 279, 280, 287 Mauritania, 24, 33, 92 Mauritius, 33, 80, 102, 107, 109, 134, 183, 233, 235, 264, 298, 316, 346, 367 mechanism design, 19, 148, 149, 281, 284 menu, 17, 95, 96, 99, 109, 119–121 merchant dealer rule, 266–269, 279, 285, 287, 288 meta-law, 4, 11, 12, 47, 58, 126, 145, 161, 178, 226, 350, 355, 356 Mexico, 1, 33, 65, 70, 71, 77, 79, 82, 83, 86, 103, 106, 108, 110, 135, 182, 212, 218, 236, 249, 260, 267, 274, 291, 298, 306, 318, 340, 344, 346, 349, 357, 367 Middle East, 28, 293, 309, 337, 343 mistaken improvement, 309, 310 Moldova, 33, 72, 77, 83, 84, 86, 104, 108, 111, 136, 163, 164, 166, 181, 182, 206, 214, 218, 234, 266, 274, 283, 292, 294, 298, 314, 317, 334, 339, 344, 346, 349, 357, 367, 374 Monaco, 33, 102, 110, 134, 154, 156, 182, 235, 267, 292, 298, 315, 317, 340, 343, 346, 367

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

index Mongolia, 25, 33, 36, 72, 80, 83, 104, 109, 111, 112, 129, 135, 163, 166, 179, 183, 208, 209, 212, 233, 234, 236, 268, 292, 298, 306, 312, 315, 342, 345, 357, 367 Morocco, 33, 64, 79, 182, 214, 234, 235, 346, 349, 374 mortgage, 8, 17, 37, 62, 63, 66, 67, 74, 75, 77–82, 85, 86, 88, 92, 93, 96, 101, 102, 105, 107, 108, 131, 337, 353, 356 Mozambique, 33, 130, 132, 165, 212, 367 Muslim, 28 Myanmar, 24, 33, 63, 80, 104, 378 Myerson and Satterwaite Impossibility Theorem, 282 necessity, 57, 216, 230–232, 243–245, 247, 248, 250–252, 255 negligence, 279, 280, 323, 324, 334 nemo dat, 262–264, 274 Nepal, 24, 33, 80 Netherlands, xix, 28, 33, 42, 64, 73, 74, 77, 83, 84, 86, 104, 108, 109, 111, 112, 114, 136, 163, 164, 185, 206, 208, 214, 233–235, 267, 292, 298, 315, 325, 342, 345, 367 New Private Law, 2, 5 New York, 33, 41, 43, 62, 67, 77, 79, 84, 86, 103, 109, 111, 114, 135, 136, 162, 166–168, 182, 184, 213, 223, 233, 236, 260, 292, 296, 299, 315, 335, 340, 344, 345, 351, 356, 376, 379 New Zealand, xx, 33, 62, 65, 77, 82, 84, 86, 104, 109, 110, 135, 166, 167, 185, 208, 210, 214, 216, 217, 233, 236, 265, 293, 299, 306, 337, 344, 345, 370, 375, 376, 378 Nicaragua, 33, 63, 77, 78, 80, 87, 103, 109, 111, 130, 131, 134, 164, 182, 212, 218, 219, 233, 235, 256, 269, 292, 298, 318, 340, 344, 346, 368 Niger, 33, 80, 102, 107, 109, 134, 183, 233, 235, 264, 298, 316, 346, 368 Nigeria, 33, 77, 84, 105, 110, 135, 208, 210, 264, 265, 293, 345, 373, 378

421

non-causa principle, 17, 36, 101, 104, 119 Nordic, 15, 228 North Korea, 23, 28, 33, 74, 92, 102, 110, 154, 156, 182, 214, 262, 263, 291, 352, 357, 368 Norway, 25, 33, 63, 64, 78, 81, 84, 103, 108, 110, 136, 168, 182, 213, 218, 236, 266, 291, 296, 298, 342, 358, 370, 372, 374 numerus apertus, 63, 65, 89–91, 98, 353 Oman, 24, 34, 104, 135, 168, 376 Ontario, 1, 34, 41, 62, 78, 84, 86, 103, 107, 110, 135, 181, 208, 214, 260, 264, 265, 345, 373, 375, 376, 377 opportunism, 11, 12, 47, 161, 178, 226, 304, 355 opposability (opposable), 17, 77, 83, 102, 103, 106–110, 114, 121, 132, 146, 153, 156 optimal standardization, 87, 88, 90 Ottoman Empire, 28, 29, 293 ownership, 3, 6, 14, 17, 19, 60, 62–64, 66, 70, 72–74, 77, 79–81, 87, 89, 93, 94, 97, 101–110, 114, 118, 120, 121, 125–132, 134, 136, 138–158, 162–164, 168, 171, 179, 191, 205, 215, 230, 232, 234, 240, 261–263, 266, 270–272, 275, 278, 280–287, 290–302, 306, 308, 309, 311–335, 337, 339–347, 349, 350, 353, 354, 358, 359, 370, 374, 375 registration of ownership, 36, 103, 127, 129, 130, 140, 353 Pakistan, 34, 63, 79, 80, 84, 86, 104, 110, 129, 132, 183, 211, 233, 265, 291, 298, 371, 372, 376, 378 Panama, 34, 63, 77, 80, 83, 86, 103, 109, 111, 130, 131, 134, 182, 212, 218, 219, 233, 235, 291, 298, 318, 340, 344, 346, 349, 368 Paraguay, 34, 64, 71, 77, 82, 103, 111, 134, 164, 165, 212, 218, 233, 236, 267, 274, 291, 298, 311, 318, 342, 346, 349, 357, 368

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

422

index

partial partition, 18, 180, 181, 186–189, 192–196, 199–202 partition by sale, 18, 180, 181, 186, 187, 189–194, 197 partition in kind, 7, 18, 180–194, 197, 201, 202–204, 341 partnership, 179, 283, 358 personalize, 94, 120, 277, 288 Peru, xx, 34, 64, 70, 72, 77, 80, 83, 86, 103, 109, 111, 113, 134, 165, 180, 184, 208, 213, 218, 233, 236, 266, 291, 298, 313, 317, 341, 347, 349, 357, 368, 371 Philippines, 34, 69, 77, 80, 82, 86, 103, 110, 134, 182, 212, 218, 233, 235, 267, 292, 298, 318, 334, 340, 344, 346, 357, 368 pledge, 66, 74, 79, 80, 82–86, 93, 107, 372, 374 pliability rule, 56, 321, 322, 326, 330 Poland, 34, 64, 69, 71, 103, 108, 110, 135, 163, 165, 182, 206, 208, 212, 218, 234, 266, 274, 292, 297, 317, 334, 341, 345, 368, 371 Portugal, 24, 28, 34, 64, 72, 73, 82, 86, 103, 108, 110, 130, 132, 134, 143, 161, 165, 167, 182, 208, 212, 218, 234, 241, 252, 269, 287, 292, 298, 317, 318, 340, 346, 349, 351, 364, 367, 368 prevention cost, 270, 273, 275, 285, 299, 329, 330 principal thing, 337, 340, 350 priority (priorities), 358 private taking, 190, 195, 231 privilege, xx, 53, 55, 70, 94, 108 privilèges, 86 processing, 309–311, 313, 314 profit à prendre, 66, 67, 71, 95 prompt protest, 17, 36, 161, 162, 164, 166–168, 171, 173, 174, 178 property rule, 15, 18, 46, 47, 52, 53, 56, 58, 59, 170, 171, 178, 237–243, 252, 256, 321–324, 326–330 public auction, 77, 180–185, 201–204, 214, 267, 341, 350 public faith, 8, 17, 101, 105–108, 114–116, 118, 121, 139, 141

Puerto Rico, 1, 34, 64, 69, 77, 80, 87, 103, 111, 130, 131, 134, 182, 212, 233, 235, 292, 294, 298, 318, 334, 340, 344, 345, 349, 368 put option, 18, 53, 177, 178, 180–185, 190, 192, 202, 204, 331, 344–346 Qatar, 34, 73, 77, 79, 80, 87, 104, 108, 110, 129, 135, 167, 183, 208, 213, 214, 233, 234, 267, 274, 294, 299, 343, 349, 357, 368, 374 Quebec, 1, 28, 34, 41, 63, 70–73, 78, 79, 84, 87, 103, 110, 125, 127, 135, 185, 206, 208, 212, 233, 234, 236, 260, 265, 268, 287, 292, 298, 315, 342, 346, 349, 368, 377 real agreement, 17, 101, 104, 105, 119 realfolium, 14, 153 Reallast, 66, 81, 82 recording, 14, 45, 88–90, 101, 113–115, 118, 121, 126, 130, 138, 145, 147, 152, 158, 210, 254 recording of deeds, 89, 91, 114 recursive, 97, 353 recursiveness, 97 registration, 8, 14, 17, 36, 49, 60, 62, 64, 83, 88, 89, 91, 95–99, 101–106, 109, 113–116, 118, 121, 126, 127, 129–133, 138–143, 145–147, 152–158, 210, 211, 254, 288, 356, 372, 374–376 registration of possession, 127, 134, 135 registration of title, 129–131, 145, 157, 158, 376 registration of possession, 127 registry, 98, 99, 103, 105–107, 109, 115, 121, 128, 130, 136, 139, 140, 146, 155, 156, 158, 225, 236, 288 reimburse, 53, 218, 221, 261, 266–269, 279, 280, 287, 289 relativity of title, 16 rentcharge, 66, 81, 82, 376 Restatement of the Law, 14, 67, 118, 125, 127, 133, 136, 139, 144, 151, 152, 336 retention, 66, 85 revelation mechanism, 18, 196, 284 reward fraction, 294–296, 301–304, 308

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

index right of first refusal, 37, 179, 225, 357, 358 right of habitation, 30, 66, 67, 69, 71, 94–97 right of survivorship, 179, 358 right of use, 30, 36, 66, 69–71, 94, 97 Romania, 34, 65, 70, 73, 77, 79, 83, 84, 103, 109, 110, 130, 131, 135, 181, 182, 212, 218, 233, 234, 252, 256, 267, 274, 292, 298, 315, 335, 342, 346, 368 Russia, 25, 29, 34, 72–74, 77, 82–84, 104, 106, 111, 136, 168, 184, 214, 218, 234, 236, 237, 266, 274, 292, 296, 298, 310, 318, 334, 352, 357, 368 Rwanda, 24, 34, 64, 73, 165, 214, 218, 317, 342, 346, 368 safe harbor, 18, 175, 355 safety valve, 57, 145, 170, 172, 227, 249, 333 Sale of Goods Act, 110, 264, 265, 362, 377, 378 Sao Tome and Principe, 34, 212 Scandinavia, 25, 28, 29, 63, 110, 357 Scotland, 1, 25, 34, 41, 65, 69, 77, 84, 86, 87, 104, 108, 110, 130, 131, 134, 145, 154, 156, 182, 206, 211, 212, 233, 265, 292, 298, 315, 340, 345, 359, 371, 374, 376, 378 searching cost, 277, 299, 300 security interest, 62, 63, 67, 74, 83–86, 88, 95, 354 security right, 7, 66, 67, 74, 79, 82, 84–86, 107, 108, 358 self-assessment, 18, 195 Senegal, 13, 34, 110, 206, 212, 233, 235, 316, 346, 368, 371 Serbia, 13, 34, 64, 70, 72, 74, 79, 81, 104, 109, 111, 112, 134, 165, 175, 213, 218, 234, 268, 312, 317, 342, 345, 347, 373 servient land, 18, 67, 210, 211, 230–232, 238, 243, 246, 252–256 severance, 231–233, 252, 254, 358 Seychelles, 1, 34, 80, 84, 102, 107, 109, 134, 185, 214, 224, 233, 235, 264, 267, 298, 311, 316, 346, 357, 368

423

should have known, 13, 144, 150, 158, 162, 278, 279, 324, 331, 333 Singapore, xx, 34, 62, 63, 65, 67, 78, 82, 84, 86, 104, 109, 110, 132, 169, 185, 206, 211, 214, 233, 236, 264, 265, 293, 299, 315, 337, 345, 371, 373, 375, 378 Slovakia, 34, 65, 69, 74, 77, 79, 104, 108, 111, 112, 135, 182, 209, 212, 236, 291, 297, 306, 317, 337, 343, 347, 369 Slovenia, 13, 34, 65, 69, 72, 77–79, 81, 83, 84, 104, 108, 111, 112, 130, 132, 136, 167, 169, 184, 208, 209, 213, 218, 234, 268, 292, 298, 315, 335, 342, 345, 374 South Africa, 28, 34, 36, 38, 39, 42, 44, 63, 64, 69, 77, 83, 84, 104, 107, 111, 112, 127, 135, 167, 185, 208, 209, 214, 218, 233, 235, 263, 292, 298, 311, 317, 335, 337, 343, 345 South Korea, 29, 34, 64, 73, 79, 84, 104, 111, 112, 130, 132, 135, 157, 182, 212, 233, 235, 256, 267, 274, 292, 294, 298, 312, 313, 315, 342, 345, 358, 369, 371 South Pacific, 1, 24, 34, 41, 84, 110, 154, 156, 208, 232 Spain, 28, 34, 63, 64, 69, 73, 77, 79, 80, 83, 87, 103, 108, 111, 130, 131, 134, 140, 163, 182, 212, 218, 233, 235, 266, 292, 294, 298, 318, 340, 344, 346, 357, 369 specificatio, 7, 19, 20, 36, 309–313, 318, 319, 321, 322, 324–326, 328–334, 336, 337, 339 Sri Lanka, 24, 34, 85, 105, 135, 168, 185, 374, 375 stability value, 37–39 stale claims, 150, 151 stolen goods, 261–269, 274–276, 287 subjective value, 49, 58, 113, 187–194, 196–198, 200–203, 239, 240, 242–245, 247, 249, 252, 280, 310, 321, 323, 325, 327–331, 348 subordinate thing, 337, 340, 350 sunk cost, 239, 247, 302 superficies, 36, 66, 72–74, 94, 95, 100 super-majority rule, 212, 227, 228

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

424

index

sure shipwreck, 18, 175, 355 Suriname, 34, 36, 73, 77, 81, 87, 104, 111, 134, 214, 267, 298, 315, 342, 345, 347, 349, 369 Swaziland (Eswatini), 24, 34, 92 Sweden, xix, 25, 34, 63, 78, 84, 103, 108, 111, 130, 132, 185, 214, 266, 292, 318, 342, 345, 358, 370, 373 Switzerland, 7, 34, 65, 69, 70, 72, 77–79, 81, 84, 104, 108, 111, 112, 130, 132, 135, 157, 165, 183, 206, 208, 212, 234, 235, 267, 274, 292, 297, 317, 342, 345, 346, 357, 358, 369 Syria, 34, 73, 80, 104, 110, 134, 183, 208, 213, 214, 267, 274, 294, 299, 308, 343, 346, 369 tailored rule, 161, 162, 175, 178 tailored standard, 161, 162, 175, 178, 355 Taiwan, xx, 25, 28, 29, 34, 57, 61, 64, 72, 74, 77, 79, 81, 82, 104, 109, 111–113, 127, 129, 135, 147, 152, 154, 163, 165, 167, 173, 174, 178, 182, 187, 206, 208, 212, 218, 223, 224, 227, 230, 233–235, 237, 247, 248, 256, 267, 274, 292, 298, 300, 310, 312, 315, 342, 345, 347, 348, 356, 358, 363, 369, 376 Tajikistan, 34, 73, 74, 77, 83, 85, 104, 111, 135, 168, 184, 214, 218, 267, 274, 292, 297, 308, 310, 318, 357, 369 Tanzania, 24, 34, 77, 86, 105, 110, 182, 236, 265, 372, 374, 378 tenancy in common, 18, 179, 214, 216, 222, 229 Thailand, 34, 64, 70, 73, 77, 81, 104, 109, 110, 135, 163, 166, 183, 212, 233–235, 256, 266, 267, 292, 297, 312, 315, 342, 346, 369 the fragmentary land problem, 186 time-sharing, 358 Timor-Leste, 35, 64, 70, 73, 77, 80, 87, 104, 110, 130, 132, 135, 165, 208, 212, 218, 234, 235, 252, 268, 287, 292, 298, 311, 317, 340, 346, 349, 369 Togo, 35, 80, 102, 107, 109, 134, 183, 233, 235, 264, 298, 316, 346, 369 traditio, 66, 111, 120

tragedy of the anticommons, 186, 187, 205, 215, 216, 223, 229 tragedy of the commons, 18, 205, 215, 216, 223, 229 transaction cost, 10, 15, 17, 18, 48, 49, 53–56, 59, 88, 89, 91, 94, 101, 113, 115, 139, 142, 143, 148–150, 152, 155, 173, 174, 176, 215, 222, 241, 242, 245, 247, 254, 277, 322 transformation test, 19, 311–315, 318, 326–328, 330, 348 treasure trove, 290, 293, 294, 296, 297, 298, 299, 301, 306, 307, 308, 372, 378 treasuretrove, 306 trust, 6, 90, 96–98, 185, 211, 214, 224, 228, 243, 300, 353, 354, 363 Tunisia, 35, 64, 70, 79, 85, 104, 109, 110, 134, 167, 183, 212, 233, 235, 266, 299, 308, 346, 349, 370 Turkey (Türkiye), xx, 29, 35, 64, 71, 72, 81, 104, 109, 111, 112, 130, 132, 136, 163, 165, 167, 182, 208, 212, 213, 218, 234, 235, 268, 287, 297, 317, 342, 347, 357, 358, 369 Turkmenistan, 35, 69, 72, 77, 104, 109, 111, 112, 130, 132, 154, 156, 163, 166, 214, 225, 234, 236, 266, 274, 292, 298, 312, 342, 345, 357, 369 Uganda, xx, 35, 63, 77, 84, 86, 104, 109, 110, 135, 168, 211, 236, 265, 293, 315, 337, 345, 370, 372, 376, 378 Ukraine, 35, 42, 64, 69, 72, 77, 85, 104, 111, 135, 168, 184, 208, 214, 218, 267, 274, 292, 298, 317, 357, 369 unanimity, 30, 212–214, 219, 220 underinvestment, 18, 221–225, 227, 229 underuse, 18, 216–218, 222–225, 227, 229 Uniform Commercial Code, 264, 356, 379 United Arab Emirates, 28, 35, 73, 103, 233, 267, 343, 346, 369 United States, 65, 118, 187, 217, 231, 310 unjust enrichment, 12, 105, 141, 144, 145, 158, 160, 176, 221, 279, 313, 327

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press

index Uruguay, 35, 64, 70, 77, 79, 87, 104, 111, 112, 134, 182, 211, 233, 235, 267, 291, 298, 312, 317, 340, 343, 345, 369 use right, 7, 66, 67, 70–74, 95, 100, 163, 166 USSR, 25, 74, 357, 358 usufruct, 66, 67, 69–74, 80, 86, 87, 94–97, 100, 353 Uzbekistan, 35, 64, 73, 74, 77, 83, 85, 104, 111, 136, 168, 184, 214, 218, 236, 267, 274, 292, 298, 306, 357, 370 Venezuela, 35, 65, 77, 80, 87, 103, 111, 134, 182, 212, 218, 233–235, 256, 267, 274, 292, 294, 298, 312, 317, 334, 340, 343, 344, 346, 370

425

verification cost, 56, 57, 270, 277, 285, 322–325, 329, 353 Vietnam, 35, 36, 77, 80, 84, 104, 111, 136, 182, 214, 233, 235, 256, 266, 274, 290, 292, 298, 317, 342, 344, 345, 370 vindicatio, 3, 105, 160–162, 171, 173, 174, 352 waiting period, 294, 295, 301–304, 308 welfare, 15, 46–48, 50–52, 223, 241, 249 World Bank, 4, 98, 117, 130, 139, 146, 153–156, 158 Zambia, 24, 35, 104, 109, 372, 373 Zimbabwe, 24, 35, 104, 135, 168, 263, 294, 371, 375

https://doi.org/10.1017/9781009236553.021 Published online by Cambridge University Press