Secured Transactions Law Reform in Africa 9781509913077, 9781509913107, 9781509913084

Over the last few decades, many countries have reformed their secured transactions law. One of the main reasons has been

202 109 10MB

English Pages [553] Year 2019

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Secured Transactions Law in Asia: Principles, Perspectives and Reform 9781509926497, 9781509926527, 9781509926503

This collection of essays offers a unique insight and overview of the secured transactions law in many of the most impor

220 40 6MB Read more

Secured Transactions Law Reform: Principles, Policies and Practice 9781849467438, 9781509903139, 9781509903115

Secured transactions law has been subjected to a close scrutiny over the last two decades. One of the main reasons for t

180 59 4MB Read more

Secured Transactions, 2016

1,548 113 3MB Read more

International and Comparative Secured Transactions Law: Essays in honour of Roderick A Macdonald 9781849467650, 9781782257844, 9781509901142

The law of secured transactions has seen dramatic changes in the last decade. International organisations, particularly

229 23 3MB Read more

Problems & Materials on Secured Transactions [8th Edition] 9780735592490

2,376 141 1MB Read more

Availability of Credit and Secured Transactions in a Time of Crisis 9781107703704, 9781107027442

In the light of the financial crisis, it has become clear that the globalisation of financial markets has not been match

196 50 3MB Read more

Islamic Reform in Twentieth-Century Africa 9780748695447

The first comprehensive analysis of Muslim movements of reform in modern sub-Saharan Africa Based on twelve case studie

246 100 3MB Read more

Insolvent transactions in company law 9780409331707, 0409331708, 9780409331714, 0409331716

677 97 7MB Read more

Institutions and Reform in Africa: The Public Choice Perspective 0275958795

478 72 15MB Read more

Economic and Political Reform in Africa: Anthropological Perspectives 0253010934, 9780253010933

What are the local effects of major economic and political reforms in Africa? How have globalized pro-market and pro-dem

1,050 170 12MB Read more

Secured Transactions Law Reform in Africa
9781509913077, 9781509913107, 9781509913084

Author / Uploaded
Marek Dubovec
Louise Gullifer

Table of contents :
Preface
Table of Contents
List of Abbreviations
Table of Cases
Table of Legislation
Table of International Instruments
Part I: Introduction
1. Introduction
A. General Introduction
B. Reform of Secured Transactions Law
C. The Status and Background of Secured Transactions Law Reform in Africa
D. Terminology
E. Legal Issues
F. The Shape of the Book
2. A Modern Secured Transactions Law
A. Introduction
B. Scope of Secured Transactions Law
C. Creation
D. Third-Party Effectiveness
E. Registration
F. Priority
G. Rights and Duties of Parties
H. Enforcement
I. Conflict of Laws
J. Transitional Provisions
K. Conclusion
Part II: Common Law
3. Ghana
A. Ghana's Economy
B. Pre-B&L Act Legal Framework
C. The Borrowers and Lenders Act
D. Final Remarks
4. Kenya
A. Economic Overview
B. Pre-Existing Legal Framework and Registries
C. Movable Property Security Rights Act
D. Final Remarks
5. Liberia
A. Economic Overview
B. The Commercial Code
C. Chapter 5 of the Commercial Code
D. Insolvency Reform
E. Final Remarks
6. Malawi
A. Introduction
B. Warehouse Receipts Act of 2018
C. Malawi's Commodity Exchanges
D. Final Remarks
7. Nigeria
A. Economic Overview
B. Pre-Reform Secured Transactions Framework
C. Recent Reforms
D. Secured Transactions Law – The Reform
E. Final Remarks
8. Zambia
A. Introduction
B. Background to the MPSIA
C. Pre-MPSIA Framework
D. Structure of the MPSIA
E. Implementation Challenges
F. Final Remarks
9. Sierra Leone
A. Economic and Financing Overview
B. Legal Framework Affecting Secured Transactions
C. Borrowers and Lenders Act of 2014
D. The 2019 Borrowers and Lenders Act
E. Final Remarks and Future Steps
Part III: Civil Law
10. OHADA Uniform Act on Securities
A. Introduction
B. Scope: Types of Security Rights
C. Third-Party Effectiveness
D. Registration of Security Rights in the Register of Commerce and Securities
E. Priority
F. Rights and Obligations of the Parties
G. Enforcement of Security Rights
H. Conflict of Laws and Transition Provisions
I. Final Remarks
11. Burundi
A. Economic Overview
B. Pre-Reform Legal Framework
C. The 2016 Law on Security Rights and the 2016 Leasing Law
D. Final Remarks
12. Ethiopia
A. Economic Overview
B. Legal System
C. Overview of the Current Secured Transactions Regime
D. Current Security Devices
E. Proposed Reform
F. Final Remarks
13. Tunisia
A. Introduction
B. Economic Overview
C. Secured Transactions Framework - An Overview and Common Transactions
D. Creation and Third-Party Effectiveness
E. Priorities
F. Enforcement of Security Rights
G. Conclusion
Part IV: Mixed Jurisdictions
14. Rwanda
A. Country and Economic Overview
B. Secured Transactions Legal Framework
C. Law on Security Interests in Movable Property
D. Final Remarks
15. South Africa
A. Introduction
B. Background of the Legal Framework
C. Security Rights in Movable Assets
D. Final Remarks
16. Zimbabwe
A. Introduction
B. Pre-2017 Secured Transactions Framework
C. The Movable Property Security Interests Act
D. Final Remarks
Part V: Final Chapters
17. A Modern International Approach to Equipment Financing in Africa: The Cape Town Convention and its Protocols
A. Introduction
B. The Cape Town Convention Model: The Basic Legal Framework
C. Facilitating Financing in the Aviation Sector: The Aircraft Protocol in Africa
D. Facilitating Development of the Rail Sector: The Luxembourg Rail Protocol and Africa
E. The Future Mining, Agriculture and Construction Protocol
F. Final Remarks
18. The Infrastructure Underpinning a Modern Secured Transactions Law
A. Introduction
B. The Benefits of a Reformed Secured Transactions Regime
C. Components of a Modern Secured Transactions Regime
D. Credit Infrastructure Elements that Support Affordable Credit
E. Conclusion
Conclusion
A. Introduction
B. Drivers for Reform
C. The Method of Reform
D. The Substance of Reform
E. Conclusion
Index

Citation preview

SECURED TRANSACTIONS LAW REFORM IN AFRICA Over the last few decades, many countries have reformed their secured transactions law. One of the main reasons has been the clear link between reform and the a vailability of credit, and the drive to improve access to finance, particularly for micro, small and medium-sized enterprises. This book focuses particularly on developing economies in Africa, which have legal frameworks influenced by English, French, Belgian, Roman-Dutch and other laws. Reform in this area of law across African countries has taken a number of forms, which are explored and discussed in this book. Secured Transactions Law Reform in Africa is a mixture of a critical description of the pre-reform law and practice, and the reform process itself. It also includes a comparative analysis of the legal provisions and an examination of the early results of the reforms. The book sets out a road map for the future of secured transactions reform; primarily in Africa, but also in other countries that have undertaken or are contemplating similar reforms. This book is the second in a series of books about Secured Transactions Law in countries around the world, and its reform, both on a national and an international scale. The first book, Secured Transactions Law Reform: Principles, Policies and Practice, was published in 2016.

ii

Secured Transactions Law Reform in Africa Marek Dubovec and

Louise Gullifer

HART PUBLISHING Bloomsbury Publishing Plc Kemp House, Chawley Park, Cumnor Hill, Oxford, OX2 9PH, UK HART PUBLISHING, the Hart/Stag logo, BLOOMSBURY and the Diana logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2019 Copyright © Marek Dubovec and Louise Gullifer, 2019 Marek Dubovec and Louise Gullifer have asserted their right under the Copyright, Designs and Patents Act 1988 to be identified as Authors of this work. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. While every care has been taken to ensure the accuracy of this work, no responsibility for loss or damage occasioned to any person acting or refraining from action as a result of any statement in it can be accepted by the authors, editors or publishers. All UK Government legislation and other public sector information used in the work is Crown Copyright ©. All House of Lords and House of Commons information used in the work is Parliamentary Copyright ©. This information is reused under the terms of the Open Government Licence v3.0 (http://www. nationalarchives.gov.uk/doc/open-government-licence/version/3) except where otherwise stated. All Eur-lex material used in the work is © European Union, http://eur-lex.europa.eu/, 1998–2019. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication data Names: Dubovec, Marek, author. | Gullifer, Louise, author. Title: Secured transactions law reform in Africa / Marek Dubovec and Louise Gullifer. Description: Oxford ; New York: Hart, 2019. | Includes bibliographical references and index. Identifiers: LCCN 2019035722 (print) | LCCN 2019035723 (ebook) | ISBN 9781509913077 (hardback) | ISBN 9781509913091 (Epub) Subjects: LCSH: Security (Law)—Africa, Sub-Saharan. | Suretyship and guaranty— Africa, Sub-Saharan. | Law reform—Africa, Sub-Saharan. Classification: LCC KQC235.S43 D83 2019 (print) | LCC KQC235.S43 (ebook) | DDC 346.6074—dc23 LC record available at https://lccn.loc.gov/2019035722 LC ebook record available at https://lccn.loc.gov/2019035723 ISBN: HB: ePDF: ePub:

978-1-50991-307-7 978-1-50991-308-4 978-1-50991-309-1

Typeset by Compuscript Ltd, Shannon To find out more about our authors and books visit www.hartpublishing.co.uk. Here you will find extracts, author information, details of forthcoming events and the option to sign up for our newsletters.

PREFACE Reform of secured transactions law is occurring all over the globe, largely because of the economic benefits it brings, especially to developing economies. Moreover, a consensus is growing as to the principles on which a modern secured transactions law should be based, exemplified by the work of UNCITRAL in its Model Law of Secured Transactions. Both of us have actively participated in the discussions that led to the adoption of this model. This book examines the reform initiatives in Africa, including analysing the law pre- and post-reform, as well as unreformed law in some African countries, and uses these principles to frame a critical analysis of the reformed and unreformed law. It also discusses the drivers for reform, and the undisputed need for implementation initiatives to complement the reform of the law if the economic benefits are to be realised. It also contains some discussion of different legal cultures, since African jurisdictions include civil law, common law and mixed jurisdictions. All these threads are drawn together in the conclusion, which contains a thematic analysis of the issues arising in relation to secured transactions law reform, particularly in developing economies, and uses the material in the country chapters to inform and illustrate the arguments. The idea for this book came from two sources. First, since 2010, one of us, Marek Dubovec, has been involved in assisting African countries to reform their secured transactions law, setting up secured transactions (collateral) registries, and providing capacity building to various stakeholders, and has seen at first hand the benefits and challenges of reform. This direct involvement in the reforms allowed us to also explain some thinking of the local stakeholders behind certain rules and approaches taken. Some of these reforms, such as in Ghana and Sierra Leone, are entering their second phase, primarily to correct some mistakes made initially. Marek has been and is still involved in a number of other reforms in Africa, including those in Egypt, Lesotho, Mozambique, Somalia, South Sudan and Uganda, that could all learn valuable lessons from the regional experiences. New developments have continued in the countries covered in the book after it was too late to add any text, and we are pleased to report that what is referred to as the ‘draft Proclamation’ in chapter 12 on Ethiopia has now been enacted. Second, in 2013, the other of us, Louise Gullifer, had taken over as Executive Director of the Secured Transactions Law Reform Project, a project set up to examine the English law relating to secured transactions and to consider the need for and shape of future reform. In 2013, the project ran two conferences in London and Durham, examining secured transaction law reform in other jurisdictions. The papers presented at those conferences, plus others dealing with other jurisdictions, were published in a book co-edited by Louise and Orkun Akseli. The book, entitled Secured Transactions Law Reform: Principles, Policies and Practice, was published by Hart Publishing in 2016. It considered some civil law jurisdictions in Europe and common law jurisdictions in Europe, North America and Australasia. However, there was also one chapter on Malawi, which was the first A frican jurisdiction to reform its secured transactions law according to modern p rinciples,

vi Preface co-written by Marek and Cyprian Kambili. This chapter was an indication that there was secured transactions law reform going on in many parts of the world not covered by the book, and led to the idea of producing a series of books about secured transactions law reform around the world. In the future, the series will continue with edited volumes on secured transactions law reform in Asia and Central and South America. However, Africa seemed an ideal contender for the second volume, and, given Marek’s expertise in the area, a co-written volume seemed possible, with limited contribution from particular specialists. We are extremely grateful that Gavin McCosker, who has great experience of working with secured transactions registries from his work as Registrar of Personal Properties Securities in Australia and previously as a Senior Financial Sector Specialist and consultant for the World Bank Group focused on Africa, agreed to write a chapter on the infrastructure underpinning a modern secured transactions law. Equally, we are extremely fortunate that Anna Veneziano (Deputy Secretary General of UNIDROIT) and William Brydie-Watson (Senior Legal Officer at UNIDROIT) have contributed a chapter on the present and likely impact in Africa of the Cape Town Convention and its protocols. This very important private law treaty has the potential to transform the financing of high value mobile equipment in Africa, particularly aircraft and railway rolling stock as well as mining, agricultural and construction equipment, which is the subject of a new Protocol to be adopted at a Diplomatic Conference in November 2019. The rest of the book was co-written by us, with a great deal of help and advice from experts on the ground. We would like to thank Benjamin Osei-Tutu, Fred Asiamah- Koranteng (Ghana), Fonsia Donzo (Liberia), Cyprian Kambili, Chiku Namelo, Daniel Nkosi (Malawi), Yusuf Dalhatu (Sierra Leone), Jean-François Riffard (OHADA and Burundi), Aroua Farhat (Burundi and Tunisia), Don Wallace (Rwanda), and Reghard Brits (South Africa) who read and commented on individual chapters and whose assistance has been invaluable. Some of them also provided recent statistics from the use of the collateral registries, which allowed us to more deeply assess the functioning of the systems post reform. We would also like to thank Chikezie Anachu, Simon Stanley and Perla Amaya (Natlaw) who acted as research assistants and who played an important part in preparing and checking the final manuscript. Lastly, we would like to thank all at Hart Publishing for their excellent help and support. Marek Dubovec Louise Gullifer 23rd September 2019

TABLE OF CONTENTS Preface �� v List of Abbreviations�� xix Table of Cases��xxiii Table of Legislation �� xxix Table of International Instruments �� lxv PART I INTRODUCTION 1. Introduction��3 A. General Introduction��3 B. Reform of Secured Transactions Law��8 1. What is Secured Transactions Law?��8 2. Why Reform Secured Transactions Law?��8 3. Who is Secured Transaction Law Reform For?��10 4. Who Does Secured Transaction Law Reform?��10 5. The History of International Instruments and Models��11 6. Legal Transplant Theory��13 C. The Status and Background of Secured Transactions Law Reform in Africa��15 1. Secured Transactions Law Reform in Africa��15 2. African Legal Systems��16 3. African Economies��17 4. Types of Financing in Africa��18 a. Microfinance��18 b. Financing of Warehouse Receipts��19 c. Leasing��20 d. Supply Chain Finance��20 D. Terminology��21 E. Legal Issues��22 1. Common Law Concepts��22 a. Bills of Sale Acts and Companies Act��22 b. Fixed and Floating Charges��23 c. Registration Regime��26 2. Civil Law Concepts��27 3. Concepts in Mixed Jurisdictions��30 4. Common Barriers to Accessing Secured Credit��31 a. Anti-Assignment Clauses��31 b. Taking Security Over After-Acquired Assets��32 F. The Shape of the Book��33

viii Table of Contents 2. A Modern Secured Transactions Law��35 A. Introduction��35 1. Basic Principles of Any Modern Commercial Law��35 2. Modern Principles of Secured Transactions Law��36 a. Scope of the Law��36 b. Creation, Third-Party Effectiveness and Priority��37 i. Creation��37 ii. Third-Party Effectiveness��38 iii. Priority��39 c. Enforcement��39 d. Other Matters��39 B. Scope of Secured Transactions Law��40 1. Parties��40 2. Functional Approach: Types of Secured Transactions��41 a. Single Concept of Security Right��41 b. Retention of Title Devices��41 c. Other Interests that can be Included��43 3. Assets��44 4. Secured Obligations��45 C. Creation��46 1. The Security Agreement��46 2. Rights in the Asset��47 3. Scope of the Security Right��47 4. Trade Receivables��48 D. Third-Party Effectiveness��49 E. Registration��50 1. Notice Filing��51 2. Grantor-Based System��53 3. Searches��54 F. Priority��54 1. The Basic Rule��54 2. Acquisition Security Rights��55 3. Taking Free Rules��56 4. Competing Claimants with Non-Consensual Interests and Other Preferential Claimants��57 G. Rights and Duties of Parties��57 H. Enforcement��58 1. Court Enforcement��58 2. Extra-Judicial Enforcement��59 3. Surplus and Deficiency��59 I. Conflict of Laws��60 J. Transitional Provisions��60 K. Conclusion��62

Table of Contents ix PART II COMMON LAW 3. Ghana��65 A. Ghana’s Economy��66 B. Pre-B&L Act Legal Framework��69 1. Customary Pledge��69 2. The 1952 Chattels Transfer Ordinance��69 3. The 1960 Contracts Act��70 4. The 1962 Sale of Goods Act��70 5. The 1962 Incorporated Private Partnerships Act��71 6. The 1963 Companies Act��72 7. The 1974 Hire-Purchase Decree��73 8. The 1993 Finance Lease Act��73 C. The Borrowers and Lenders Act��74 1. Scope��75 2. Creation��77 3. Third-Party Effectiveness��78 4. The Collateral Registry and Regulations��82 a. Searchability and Indexing��85 b. Performance of the Collateral Registry��86 5. Priority��87 6. Enforcement��87 7. Conflict of Laws and Transition��89 D. Final Remarks��90 4. Kenya��91 A. Economic Overview��92 B. Pre-Existing Legal Framework and Registries��95 1. Chattels Transfer Act��96 2. Hire Purchase Act��97 3. Retention/Transfer of Ownership��99 4. Companies Act��100 C. Movable Property Security Rights Act��103 1. The Scope and the Functional Approach��104 2. Creation��105 3. Third-Party Effectiveness��107 4. Registration��107 5. Priority��110 6. Rights and Obligations of the Parties and Third-Party Obligors��112 7. Enforcement��113 8. Applicable Law��114 9. General Provisions��114 D. Final Remarks��117

x Table of Contents 5. Liberia��119 A. Economic Overview��119 B. The Commercial Code��121 C. Chapter 5 of the Commercial Code��122 1. Scope��122 2. Creation and Attachment��124 3. Perfection��124 4. Priorities��125 5. Registry��126 a. Development of the Registry��126 b. The Registration Provisions��127 c. Payment of Fees��128 6. Post-Implementation Issues��129 7. Enforcement��130 D. Insolvency Reform��131 E. Final Remarks��132 6. Malawi��133 A. Introduction��133 B. Warehouse Receipts Act of 2018��134 C. Malawi’s Commodity Exchanges��137 D. Final Remarks��138 7. Nigeria��139 A. Economic Overview��140 B. Pre-Reform Secured Transactions Framework��141 1. Bill of Sale��142 2. Pledge��143 3. Lien��144 4. Charge��145 5. Other Security and Quasi-Security Devices��146 C. Recent Reforms��147 1. Leasing��148 2. Central Bank of Nigeria Registration of Security Interests in Movables Regulations��149 3. Secured Transactions in Movable Assets Act of 2017��151 D. Secured Transactions Law – The Reform��151 1. Scope��151 2. Creation and Perfection��155 3. Registration��156 4. Priority��158 5. Rights and Duties of the Parties��159 6. Enforcement of Security Interests��160

Table of Contents xi

E.

7. Conflict of Laws��161 8. Miscellaneous��162 Final Remarks��162

8. Zambia��165 A. Introduction��165 B. Background to the MPSIA��167 C. Pre-MPSIA Framework��169 1. The Companies Act��169 2. Agricultural Credits Act��170 3. Trades Charges Act��171 4. Hire-Purchase Act��172 5. Other Relevant Legislation��173 D. Structure of the MPSIA��173 1. Preliminary Provisions��174 2. Collateral Registry Office and Collateral Registry��176 3. Registry Regulations��178 4. Creation��180 5. Perfection��181 6. Priority��182 7. Enforcement��185 8. General Provisions and Offences��187 E. Implementation Challenges��188 F. Final Remarks��189 9. Sierra Leone��191 A. Economic and Financing Overview��191 B. Legal Framework Affecting Secured Transactions��192 1. The Companies Act��192 2. The Registration of Business Act of 2007��194 3. Leasing Guidelines��194 4. Road Traffic Act of 2007��195 C. Borrowers and Lenders Act of 2014��195 1. Approach and models��195 2. Issues��196 3. Creation and Perfection��198 4. Priority Rules��200 5. Enforcement��201 6. Transition��203 7. Regulations��205 D. The 2019 Borrowers and Lenders Act��205 E. Final Remarks and Future Steps��208

xii Table of Contents PART III CIVIL LAW 10. OHADA Uniform Act on Securities��213 A. Introduction��213 B. Scope: Types of Security Rights��215 1. Right of Retention��216 2. Reservation of Ownership��217 3. Security Transfer of Receivable��218 4. The Fiduciary Transfer of Funds��219 5. Pledge of Tangible Assets��220 6. Pledge of Intangible Assets��220 a. Pledge of Bank Accounts��221 b. Pledge of Partnership Rights, Securities and Financial Instruments��221 c. Pledge of Intellectual Property Rights��223 d. Pledge of the Enterprise and Lien of the Seller of an Enterprise��223 7. Liens��224 C. Third-Party Effectiveness��225 D. Registration of Security Rights in the Register of Commerce and Securities��227 1. General Overview��227 2. Requirements for the Initial Registration��228 3. Amendments, Discharges and Searches��230 E. Priority��231 F. Rights and Obligations of the Parties��232 G. Enforcement of Security Rights��233 H. Conflict of Laws and Transition Provisions��236 I. Final Remarks��237 11. Burundi��239 A. Economic Overview��239 B. Pre-Reform Legal Framework��241 1. Introduction��241 2. Pre-Reform Requirements for Creation and Effectiveness against Third Parties of Non-Possessory Pledges��242 3. Pre-Reform Priority of Non-Possessory Pledges��243 4. Pre-Reform (and Post-Reform) Rules on Enforcement of possessory and Non-Possessory Pledges��243 C. The 2016 Law on Security Rights and the 2016 Leasing Law��245 1. Creation��246 2. Third-Party Effectiveness��248 a. Registration��248 b. Dispossession��250 c. Attachment of a Sign��250 d. Other Methods of Making a Pledge Effective against Third Parties��250

Table of Contents xiii

D.

e. Third-Party Effectiveness for Retention of Ownership Sales��251 f. Third-Party Effectiveness of Finance Leases��251 3. Rights and Obligations of the Parties��252 4. Priority rules��253 5. Enforcement��256 a. Enforcement of Pledges��256 b. Enforcement of Retention of Ownership Sales��257 c. Enforcement of Leases��258 6. Transitional and Conflict of Laws Rules��258 Final Remarks��260

12. Ethiopia��262 A. Economic Overview��262 B. Legal System��264 C. Overview of the Current Secured Transactions Regime��266 D. Current Security Devices��269 1. Pledge��270 2. Assignment of Rights��270 3. Advances on and Pledges of Securities��271 4. Retention of Title��271 5. Leasing��272 6. Mortgage of a Business��273 7. Pledges of Warehouse Receipts��275 E. Proposed Reform��277 1. Scope��277 2. Creation and Third-Party Effectiveness��280 3. Registry��280 4. Priority��281 5. Enforcement��282 6. Conflict of Laws��282 7. Transition��283 F. Final Remarks��283 13. Tunisia��285 A. Introduction��285 B. Economic Overview��286 C. Secured Transactions Framework – An Overview and Common Transactions��288 1. Multiplicity and Overlap of Laws��289 2. Registration Systems��290 a. Registrations��290 3. Popular Security Devices��290 D. Creation and Third-Party Effectiveness��292 1. Possessory Pledge��293 2. Pledges over Intangible Assets��293 3. Enterprise and Equipment Pledges��294 4. Finance Lease��295

xiv Table of Contents E. Priorities��295 1. A Legitimate Cause of Preference��296 2. Priority of Competing Possessory Pledges��297 3. Time of Registration��297 4. Time of Agreement��298 5. Priority of Lessors��298 F. Enforcement of Security Rights��298 1. Enforcement of Possessory Pledges��299 2. Provisional Judicial Remedies��300 3. Disposal of the Pledged Asset��301 4. Lessors’ Remedies��302 G. Conclusion��302 PART IV MIXED JURISDICTIONS 14. Rwanda��307 A. Country and Economic Overview��307 B. Secured Transactions Legal Framework��309 C. Law on Security Interests in Movable Property��312 1. Scope��312 2. Creation��313 3. Third-Party Effectiveness��315 4. Priority��317 5. Enforcement of Security Interests��321 6. Register of Security Interests in Movable Property��323 D. Final Remarks��326 15. South Africa��328 A. Introduction��328 B. Background of the Legal Framework��330 C. Security Rights in Movable Assets��332 1. Reservation of Title Devices��334 2. Pledge��335 a. Nature of the Pledge��335 b. Creation��335 c. Extinguishment and Enforcement��337 3. Cession��338 a. The Nature of Security Cession��338 b. Creation of Security Cession��339 c. Agreements Prohibiting Cession��340 d. Cession of Future Rights��342 e. Enforcement��342 4. General Notarial Bond��343 a. Scope of General Notarial Bonds��343

Table of Contents xv

D.

b. Creation and Registration��344 c. The Nature of a General Notarial Bond��344 d. Post-Default Rights��345 5. Special Notarial Bond��347 Final Remarks��350

16. Zimbabwe��353 A. Introduction��353 B. Pre-2017 Secured Transactions Framework��356 1. Pre-MPSI Security Devices��357 2. Creation and Registration Requirements��359 C. The Movable Property Security Interests Act��362 1. The Main Body of the Act��363 2. The First Schedule��367 3. The Second Schedule��369 a. Part II – Special Creation Provisions��369 b. Part III – Priorities��369 c. Part IV – Rights and Obligations of Parties and Third-Party Obligors��370 d. Part V – Enforcement of Security Interests��371 e. Part VI – Applicable Law��372 D. Final Remarks��372 PART V FINAL CHAPTERS 17. A Modern International Approach to Equipment Financing in Africa: The Cape Town Convention and its Protocols��377 A. Introduction��377 B. The Cape Town Convention Model: The Basic Legal Framework��379 1. The ‘Functional Approach’��379 2. The Role of the International Registries��380 3. Enforcement Measures��380 4. Declarations under the Convention and Protocols��381 C. Facilitating Financing in the Aviation Sector: The Aircraft Protocol in Africa��383 D. Facilitating Development of the Rail Sector: The Luxembourg Rail Protocol and Africa��384 E. The Future Mining, Agriculture and Construction Protocol��385 1. Basic Rules of the MAC Protocol��386 2. Specialised Rules of the MAC Protocol��387 a. Application of the Protocol to MAC Equipment��388 b. Immovable-Associated Equipment��389 c. Treatment of Inventory��390 3. The Relationship between the MAC Protocol and Domestic Law��390

xvi Table of Contents 4.

F.

The Importance of the MAC Protocol in Africa��392 a. The MAC Protocol as a Tool for Increasing International Private-Sector Investment in Africa��392 b. The Mining, Agriculture and Construction Sectors in Africa��392 c. Uniformity in the Legal Framework for the Financing of MAC Equipment across Africa��393 Final Remarks��394

18. The Infrastructure Underpinning a Modern Secured Transactions Law��395 A. Introduction��395 B. The Benefits of a Reformed Secured Transactions Regime��397 C. Components of a Modern Secured Transactions Regime��398 1. Legal Framework��398 a. Incentivising Registration��399 b. Incentivising Searching��400 c. The Role of Legal Infrastructure��401 2. The Registry��402 a. Variations in Approach to the Registration System in Africa��402 b. Challenges in Establishing Registries in Africa��403 3. Key Elements that Support an Effective Registry��404 a. User-Centricity��404 b. Operating Model��405 c. Administrative Processes��405 d. Access and Availability��406 e. Data Integrity��406 i. Cyberattacks and Privacy��406 ii. Incorrect Data Being Placed on the Registry Due to Human Error��407 iii. The Unauthorised Placing of Incorrect or Inappropriate Data on the Registry��407 iv. Allowing Duplicate, Out-of-Date or Redundant Data to Remain on the Registry��408 v. Incorrect Information Entered Due to Misunderstanding��408 D. Credit Infrastructure Elements that Support Affordable Credit��408 1. Prudential Regulation��409 2. Lending Practices��409 3. Secondary Markets for Collateral��410 4. Market Opportunity Assessment��410 E. Conclusion��411 Conclusion��413 A. Introduction��413 B. Drivers for Reform��413 1. General Reasons��413 a. The Desire for Economic Growth��413 b. Appearance to Global Markets��415

Table of Contents xvii

C.

D.

c. Infrastructure Improvements��417 d. Legal Structure for Specialist Financial Activity��418 i. Credit Infrastructure��419 ii. Warehouse Receipts and Commodity Trading��419 iii. Securities Held through Intermediaries��420 iv. Leasing��421 2. The State of the Unreformed Law��422 a. Complexity and Fragmentation��422 i. Parties��422 ii. Types of Devices��423 b. Lack of Functionality��423 i. Limits in what can be used as Collateral��423 ii. No Extrajudicial Enforcement��424 iii. Excessive Formalities for Creation��424 iv. Lack of Publicity��424 The Method of Reform��425 1. Use of Models��425 a. Mixture of Models��425 b. Deviations��425 c. Structure��427 d. Simplified Approach��428 2. The Timing of the Reform Process��428 a. Interrelationship between the Legislation and the Registry��429 b. Coordination within Reform Process��430 c. Hasty Drafting��431 3. Engaging Local Stakeholders��432 a. Local Legislative and Drafting Practice��433 b. Local Legal Culture and Practices��433 c. Local Conditions��434 4. Capacity Building��434 5. Interaction with Regulatory Bodies��436 6. Interrelationship with Other Areas of Law��437 a. Techniques: Express and Implied Repeal, Amendment��437 b. Interaction with Other Statutes and Common Law��439 i. Companies Act Registration Regime��439 ii. The Floating Charge and Insolvency��440 iii. Protective Provisions for Certain Parties��441 iv. Leasing Law��442 7. Statistical Information for Assessment��442 The Substance of Reform��443 1. The Scope of the Regime��443 a. The Functional Approach��444 b. Treatment of Non-Consensual Interests��446 2. Content of Security Agreement��446 a. Secured Obligation��446 b. Anti-Assignment Clauses��447

xviii Table of Contents 3. 4.

Special Rules for Consumers��447 Methods of Achieving Third Party Effectiveness��448 a. Registration as Sole Method��449 5. The Registration Regime��450 a. Manner of Registration (Notice/Document Filing)��450 b. Administrative Role of the Registry��451 c. Identifiers and Search Criteria��451 6. Enforcement��452 a. Extrajudicial Enforcement��452 b. Alternative Forms of Enforcement��453 c. Expeditious Judicial Enforcement��454 E. Conclusion��455 Index�� 457

LIST OF ABBREVIATIONS 2017 Bill

Ghana’s Draft Borrowers and Lenders Bill of 2017

ABZ

Agricultural Bank of Zimbabwe

APEC

Asia-Pacific Economic Cooperation

B&L Act

Ghana’s Borrowers and Lenders Act

BoG

Bank of Ghana

BoZ

Bank of Zambia

Cape Town Convention/ Convention/CTC

Convention on International Interests in Mobile Equipment

CBE

Commercial Bank of Ethiopia

CBL

Central Bank of Liberia

CC

Commercial Code

CCJA

Common Court of Justice and Arbitration

CR

Collateral Registry (Sierra Leone)

CSD

Central Securities Depository

DBE

Development Bank of Ethiopia

DCFR Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference EBRD

European Bank for Reconstruction and Development

ECX

Ethiopia Commodity Exchange

FinTech

Financial Technology

IACA

International Association of Commercial Administrators

IAIR

International Association of Insolvency Regulators

IBRD

International Bank for Reconstruction and Development

ICAO

International Civil Aviation Organization

IFC

International Finance Corporation

ILO

International Labour Organization

IMF

International Monetary Fund

xx List of Abbreviations Legislative Guide/LG

UNCITRAL Legislative Guide on Secured Transactions

MAC Protocol Protocol to the Convention on Matters Specific to Mining, Agricultural and Construction Equipment MFI

microfinance institution

MLE

middle-level enterprise

Model Law/ML

UNCITRAL Model Law on Secured Transactions

Model Registry Provision/MRP UNCITRAL Model Law on Secured Transactions: Model Registry Provisions MPSI

Zimbabwe’s Movable Property Security Interests Act

MPSIA

Zambia’s Movable Property (Security Interest) Act

MSME

micro, small and medium enterprise

National Registry

National Registry of Security Rights

NBE

National Bank of Ethiopia

NIALS

Nigerian Institute of Advanced Legal Studies

NPL

non-performing loans

OAS

Organization of American States

OHADA Organisation for the Harmonisation of Business Law in Africa (Organisation pour l’harmonisation en Afrique du droit des affaires) ORG

Rwanda Office of the Registrar General

OTIF Intergovernmental Organisation for International Carriage by Rail PACRA

Patents and Companies Registration Agency

PMSI

purchase money security interest

PPSA

Personal Property Security Act

RBZ

Reserve Bank of Zimbabwe

RCCM Register of Commerce and Securities (Le Registre du Commerce et du Crédit Mobilier) RDB

Rwanda Development Board

SIMP

Rwanda’s Security Interests in Movable Property Act

SME

small and medium enterprises

UAGCL

Uniform Act on General Commercial Law ()

List of Abbreviations xxi UCC

US Uniform Commercial Code

UNCITRAL

United Nations Commission on International Trade Law

UNECA

United Nations Economic Commission for Africa

UNIDROIT

International Institute for the Unification of Private Law

WBG

World Bank Group

ZAMACE

Zambian Commodity Exchange

xxii

TABLE OF CASES United Kingdom Chellaram & Sons (London) Ltd v Butlers Warehousing & Distribution Ltd [1978] 2 Lloyds’s Rep 412��144 Cosslett (Contractors) Ltd, Re [1997] EWCA Civ 2229��144 Cox Moore v Peruvian Corp Ltd [1908] 1 Ch 604��24 Dean of Ely v Bliss (1842) 5 Beav 574, 582��438 Dearle v Hall (1823) 3 Russ 1�� 70, 147, 158 Donald v Suckling [1866] LR 1 (QB) 585��144 Evans v Rival Granite Quarries Ltd [1910] 2 KB 979��146 Florence Land and Public Works Company (1878) 10 Ch D 530, CA��24 General South American Company, Re (1875) 2 Ch D 337��24 George Inglefield Ltd, Re (1933) 1 Ch 1 (CA)��218 Hamilton’s Windsor Ironworks ex p Pitman & Edwards, Re (1879) 12 Ch D 707��24 Hammond v Barclay [1802] 2 East 227��144 Helstan Securities Ltd v Hertfordshire CC [1978] 3 All ER 262��147 Holroyd v Marshall (1861–62) 10 HLC 191��23 Levy v Abercorris Slate and Slab Co [1888] 37 Ch D 260��145 Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85��147 Miller v Race (1758) 1 Burr 452, 457–58 ��437 Monolithic Building Co [1915] 1 Ch 643��103 National Provincial and Union Bank of England v Charnley [1924] 1 KB 431��52 National Westminster Bank Plc v Spectrum Plus Ltd [2005] UKHL 41��145 Panama, New Zealand and Australian Royal Mail Company, Re (1870) 5 Ch App 318, CA��24 Pilcher v Rawlins (1872) LR 7 Ch App 259, 268–69��437 Spectrum Plus Ltd, Re (2005) 2 AC 680��25, 219 Standard Manufacturing Co, Re [1891] 1 Ch 627��146 Tancred v Delagoa Bay and East Africa Railway Co (1889) 23 QBD 239��70 Tappenden v Artus [1964] 2 QB 185��144 Thoburn v Sunderland Council (2003) QB 151 (37) ��438 Thomas v Kelly (1888) 13 HL 506��23 Wheatley v Silkstone and Haugh Moor Coal Co (1885) 29 Ch D 715��24 Australia Forge Group Power Pty Ltd v General Electric International Inc [2016] NSWSC 52��401

xxiv Table of Cases Canada Bank of Montreal v I Trade Finance Inc [2011] SCC 26��153 Credit Suisse Canada v 1133 Yonge Street Holdings Ltd (1996) 28 OR (3d) 670 (ON SC); (1998) 41 OR (3d) 432��437 Innovation Credit Union v Bank of Montreal [2010] SCC 47��153 Royal Bank of Canada v Sparrow Electric Corp (1997) 1 SCR 411 ��437 Ghana Adjei v Dabanka [1930] IWACA 63��69 Kenya Civil case No 451 of 2010��100 David Mburu Kamau v National Industrial Credit Bank, Civil Suit 253 of 2007 (High Court of Kenya)��97, 98 Kenya Chemical & Allied Workers Union v Packaging Manufacturers Ltd [1976] Cause No 130 of 2012 (Industrial Court at Mombasa)��100 Lochab Brothers v Kenya Furfural Co Ltd KLR 257 (1983)��101 Mburu Gathee T/A Gathee Enterprises v National Industrial Credit Bank Ltd, Civil Appeal No 37 of 2006 (Court of Appeal in Nairobi)��98 Nigerian Industrial Development Bank Ltd v Olalomi Industries Ltd [2002] 5 NWLR (Pt 761) 532��145 Noor Abdikadir Abdalla v Grofin East Africa Fund LLC and Sood Petroleum Ltd Civil Case 316 of 2009��101 Trans National Bank Ltd v SWIFT Truckers Ltd (in receivership)��99 New Zealand Commissioner of Inland Revenue v Northshore Taverns Ltd (in liq) (2008) 23 NZTC 22,074��25 Strategic Finance Ltd v Bridgman (2013) NZCA 357��25 Nigeria Adetoyin v Bank of the North Ltd��143 Afrotec Technical Services (Nig) Ltd v MIA & Sons Ltd [2000] 15 NWLR (Pt 692) 730; [2000] LPE LR-SC132/1992�� 144, 147 Ahaneku v Iheaturu [1995] 2 NWLR (Pt 380) 770��143 AIDC v Nigeria LNG Ltd [2000] 4 NWLR (Pt 653) 494 SC; [2000] 2 iLAW /SC.174/95��156 CDC (Nigeria) Ltd v SCOA (Nigeria) Plc [2007] 6 NWLR (Pt 1030) 300��147

Table of Cases xxv Co-operative Bank of Eastern Nigeria v Maria Eke [1979] 2 FNLR 190��144 Intercontractors Nigeria Ltd v National Providence Fund Management Board [1988] 2 NWLR (Pt 76) 280�� 145, 146 International Bank for West Africa v Azubuike [1972] NC LR 37��143 Kasali A Raimi v Moshudi Funso Ogundana & 2 ors [1986] 3 NWLR (Pt 26) 97��147 Labode v Otubu [2001] 7 NWLR (Pt 712) 256��143 Mandilas & Karaberis Ltd v Anglo-Canadian Cement Company Ltd [1967] NC LR 42��146 Nze Bernard Chigbu v Tonimas Nigeria Ltd & Another [2006] 4 SCNJ 262��144 Samuel Aro v Joe Allen & Co Ltd [1979] 2 FNR 292��147 SCOA Nig PLC v Sterling Bank PLC, Appeal No CA/L/170/2013��156 United Bank of Nig Ltd v Tropic Foods Ltd [1992] 3 NWLR (Pt 228) 231��145 South Africa African Consolidated Agencies (Pty) Ltd v Siemens Nixdorf Information Systems (Pty) Ltd 1992 (2) SA 739 (C)��338 Airkel (Edms) Bpk H/A Merkel Motors v Bodenstein En’ N Ander 1980 (3) SA 917 (A)�� 335, 336 Arctocel (Pty) Ltd v Firstrand Bank Ltd (32633/2015) [2015] ZAGPJHC 244 (21 October 2015)��332 Aussenkehr Farms (Pty) Ltd v Trio Transport CC 2002 (4) SA 483 (SCA)�� 338, 339 Bank of Lisbon and South Africa Ltd v The Master 1987 (1) SA 276 (AD)��339 Barclays National Bank Ltd v Natal Fire Extinguishers Manufacturing Co (Pty) Ltd 1982 (4) SA 650 (D)�� 346, 347 Bock v Duburoro Investments (Pty) Ltd 2004 (2) SA 242 (SCA)�� 330, 331, 350 Bokomo v Standard Bank van SA Bpk (1996) (4) SA 450�� 337, 348, 350 Boland Bank Bpk v Spies 1993 (1) SA 402 (T)��345 Boland Bank Ltd v Vermeulen 1993 (2) SA 241 (E)��345 Born Free Investments 364 (Pty) Ltd v Firstrand Bank Ltd (973/12) [2013] ZASCA 166��341 Botha v Fick 1995 (2) SA 750 (A)�� 339, 340 BP Southern Africa (Pty) Ltd v Intertrans Oil SA (Pty) Ltd 2017 (4) SA 592 (GJ)��342 Britz NO v Sniegocki 1989 (4) SA 372 (D)��340 Capespan (Pty) Ltd v Any Name 451 (Pty) Ltd 2008 (4) SA 510 (C)��341 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA)��333, 344, 345, 346, 347 Cooper NO v Die Meester 1992 (3) SA 60 (A)��343, 347, 348, 349 Danzas Trek (Pty) Ltd v Du Bourg 1979 (4) SA 915��336 Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425 (SCA)�� 338, 339 Edwards v Van Zyl 1951 (2) SA 93��336 Eerste Nasionale Bank van SA Bpk v Schulenburg 1992 (2) SA 827 (T)��345 Energydrive Systems (Pty) Ltd v Tin Can Man (Pty) Ltd 2017 (3) SA 539 (GJ)��334 Farmsecure Grains (Edms) Bpk v Du Toit 2013 (1) SA 462 (FB)�� 348, 350 Fick v Bierman (1883–84) 2 SC 26��339

xxvi Table of Cases Findevco (Pty) Ltd v Faceformat Sa (Pty) Ltd 2001 (1) SA 251 (E)��346 First National Bank of South Africa Ltd v Land and Agricultural Bank and Others 2000 (6) BCLR 586 (O) 590��332 First National Bank of South Africa Ltd v Lynn No 1996 (2) SA 339 (A)��342 Firstrand Bank Ltd v Land And Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA)��333, 343, 344, 345 Freeman Cohen’s Consolidated Ltd v General Mining and Finance Corp Ltd 1907 TS 224��336 Frye’s (Pty) Ltd v Ries 1957 (3) 575 (A)��345 Goldinger’s Trustee v Whitelaw & Son 1917 AD 66��335 Goudini Chrome (Pty) Ltd v MCC Contracts (Pty) Ltd 1993 (1) SA 77 (AD)��339 Graf v Buechel 2003 (4) SA 378 (SCA)��331 Grobler v Oosthuizen 2009 (5) SA 500 (SCA)�� 337, 338, 339 Headleigh Private Hospital (Pty) Ltd T/A Rand Clinic v Soller & Manning Attorneys 2001 (4) SA 360 (W)��342 Heydenrych v Saber (1900) 17 SC 73��336 Hofmeyer v Gous (1893) 10 SC 115��335 Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA)��333, 347, 348, 349 International Shipping Co (Pty) Ltd v Affinity (Pty) Ltd 1983 (1) SA 79 (C)��346 Judes v SA Breweries Ltd 1922 WLD 1��336 Juglal NO v Shoprite Checkers (Pty) Ltd t/a OK Franchise Division 2004 (5) SA 248 (SCA)��346 Kuranda v Boustred 1933 WLD 49��342 Labuschagne v Denny 1963 (3) SA 538 (A)��340 Land and Agricultural Development Bank of South Africa v Phato Farms (Pty) Ltd 2015 (3) SA 100 (GP)��344 Lesedi Secondary Agricultural Co-operative Ltd v Vaalharts Agricultural Co-operative and Another 1993 (1) SA 695 (NC) 699��337 Lief, No v Dettmann 1964 (2) SA 252 (A)��338 Lighter v Edwards 1907 TS 442�� 335, 336 Muller NO v Trust Bank of Africa Ltd 1981 (2) SA 117 (N)��342 National Bank of SA Ltd v Cohen’s Trustee 1911 AD 235��333, 335, 338, 342 Nedbank Ltd v Norton 1987 (3) SA 619 (N)��347 Nedcor Bank Ltd v Absa Bank Ltd 1998 (2) SA 830 (W)��334 Netherlands Bank of South Africa v Stern No 1955 (1) SA 667 (W)��336 Nino Bonino v De Lange 1906 TS 120�� 331, 332 Osry v Hirsch, Loubser & Co 1922 CPD 531��331 Paiges v Van Ryn Gold Mines Estates Ltd 1920 AD 600��341 Phillips v Soqaka 1915 EDL 37��336 Picardi Hotels Ltd v Thekwini Properties (Pty) Ltd 2009 (1) SA 493 (SCA)��339 Pieters & Co v Landau Bros & the Trustees of the Insolvent Estate of I & J Goldberg 1914 SR 30��336 Reeskens v Registrar of Deeds 1964 (4) SA 369 (N)�� 333, 343 Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC)�� 339, 342

Table of Cases xxvii Rosenbach & Co (Pty) Ltd v Dalmonte 1964 2 SA 195 (N) 204G–205A��349 Roshcon (Pty) Ltd v Anchor Auto Body Builders Cc 2014 (4) SA 319 (SCA)�� 334, 335 S v Buitendag 1980 (2) SA 152��336 SA Bank of Athens Ltd v Van Zyl 2005 (5) SA 93 (SCA)��346 Sakala v Wamambo 1991 (4) SA 144 (ZH)��331 Scheltema Beleggings v Comm Truck and Trailer Sales (35420/08) ZAGPP HC 28 (2010)��349 Senwes Ltd v Muller 2002 (4) SA 134 (T)��350 Simon v Mitsui and Co Ltd 1997 (2) SA 475 (W)�� 336, 337 Smit v Carniasaad 1998 (4) SA 877 (SCA)��342 Smuts v Booyens, Markplaas (Edms) Bpk en v Booyens 2001 (4) SA 15 (SCA)��341 Standard General Insurance Co Ltd v SA Brake CC 1995 (3) SA 806 (A)�� 340, 342 Sterling NO v Landau 1921 WLD 117��336 Stratford’s Trustees v London and SA Bank (1874) 3 EDC 439��336 Trust Bank of Africa Ltd v Standard Bank of South Africa Ltd 1968 (3) SA 166 (A)��341 Van Staden NO v Firstrand Ltd 2008 (3) SA 530 (T)��342 Vasco Dry Cleaners v Twycross 1979 (1) SA 603�� 334, 335 Volhand & Molenaar (Pty) Ltd v Ruskin NNO 1959 (2) SA 751 (W)��339 Tunisia Supreme Court ruling No 11446 of 1985��297 Supreme Court ruling No 40490 of 1996��300 Supreme Court ruling No 975 of 2004��299 Supreme Court decision No 2603 of 2004��301 United States of America Feliciana Bank & Trust v Manuel & Sessions, 943 So 2d 736, 739–40 (Court of Appeals of Mississippi 2006)��313 Motors Liquidation Co, Re (2015) 777 F3d 100��402 Nichols v Branton [2014] WL4816075 (NY Sup Ct)��129 Official Committee of Unsecured Creditors of Motors Liquidation Co v JPMorgan Chase Bank, Supreme Court of Delaware. Oct 17, 2014, 103 A.3d 1010 (Del. 2014)�� 402 Security Nat Bank and Trust Co of Norman v Dentsply (1980) 617 P2d 1340��402 Zambia Sapco Fibreboard & Wood Products Ltd (in receivership), Re – An Application for Directions under s 113 of Cap 388 2001/HPC/0225��170 Zambia National Commercial Bank Ltd v Mwila (SCZ appeal no 94, 11 November 1999 and 6 April 2000)��170

xxviii Table of Cases Zimbabwe Augustine Renusu Chizikikani v Agricultural Finance Corp (SC 123/95)��359 Barkhuizen v Napier (2007) 5 SA 323 (CC)��359 Change v Standard Finance Ltd 1990 (2) ZLR 412 (SC)��331 Engen Oil Zimbabwe (Pvt) Ltd v Dowood Services (Pvt) Ltd, (2015) HC 1663/15��365 Interfin Banking Corp Ltd (in liq) v Veanarcy (Pvt) Ltd (2013) HH-388-13��365 John Nyamukusa v Agricultural Finance Corp (SC 174/94)��359 Kufandirori v Chipuriro and Others [2004] ZWHHC 12 (HH)��357 Ncube v Nyathi (2015) ZWBHC 60 (HH)��357 PG Industries (Zimbabwe) Ltd v Jones Holdings (Private) Ltd (2015) HC 10678/13, 6��357 Sakala v Wamambo��331 Sibanda v Mshapaidze HC 2961/09��365 Tembani v Republic of Zimbabwe (SADC (T) 07/2008) [2009] SADC T 3��359 Twee Jonge Gezellen (Pvt) Ltd & Anor v Land & Agricultural Development Bank of South Africa & Anor (2011) ZACC 2��365 UDC Ltd v Chawara Kapenta Fishing Co-Op Ltd (2002) ZWSC 29��361

TABLE OF LEGISLATION United Kingdom Statutes Agricultural Credits Act 1928��23, 170 s 8��170 s 9��23 Bills of Sale Act 1878�� 22, 40, 141, 142 s 4��142 s 10��23 Bills of Sale Act (1878) Amendment Act 1882��22, 23, 40, 141, 142 s 5��143 s 8��143 Sch 1��23 Companies Act 1862��27 Companies Act 1948�� 26, 145, 169 s 100��84 s 101��85 Companies Act 1985��26 Companies Act 2006��22, 26 Pt 25��23, 100 s 859A��227 s 859F��27 s 859H��27 s 859H(4)��27 s 859L��84 s 860��27 s 860(4)��27 s 860(7)��100 s 870��27 s 873��27 s 874��27 s 874(3)��27 Consumer Credit Act 1974��73, 99 s 8(3)��99 s 60��172 s 90��98, 173 s 91��98 s 92��173 s 173��173

xxx Table of Legislation Enterprise Act 2002 s 251��154 Hire Purchase Act 1938��147 Insolvency Act 1986 s 176ZA��24 s 344��23 Sch B1��26 para 14��26 para 99(3)��24 Law of Property Act 1925 s 94��46 s 136��70 Pawnbrokers Act 1877��143 Preferential Payments in Bankruptcy Amendment Act 1897 s 107��24 Sale of Goods Act 1893��70, 144 s 17��70 s 25(1), (2)��71 s 39��70 s 44��70 Sale of Goods Act 1979��70, 99 s 24��71 s 25��71 s 41��70 s 44��70 Statutory Instruments Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254) reg 1(3)��32 reg 2(1)��32 reg 4��32 Companies Act 2006 (Amendment of Part 25) Regulations 2013 (SI 2013/600)�� 100, 146 Consumer Credit (Agreements) Regulations 2010 (SI 2010/1014) Sch 1��172 Financial Services and Markets Act 2000 (Regulated Activities) Order 2001 (SI 2001/544) art 60C��99 Australia Personal Property Security Act 2009�� 61, 116, 197, 279, 404, 407 s 13��44 s 14(2)��43 s 17��46

Table of Legislation xxxi s 19(2)��156 s 46(1)�� 159, 427 s 55(2)��46 s 81(2)��32 s 109��44, 185 s 157(4)��52 s 339��116 s 339(4)��26, 116 s 339(5)��26, 116 s 340��116 s 340(1)(b)��26 s 340(5)��26 Personal Property Securities (Consequential Amendments) Act 2009��116 Burundi Code of Civil Procedure 2004 Art 3��259 Art 6��259 Art 250(8)��240 Art 258��244 Arts 307–332��244 Art 373��241 Art 598��241 Code of Commerce 2015�� 241, 242 Art 81��242 Arts 113–134��242 Art 114(1), (2)��242 Art 115��242 Art 117��242 Arts 118–125��244 Art 119��244 Art 120��244 Art 121(3)��244 Art 122��244 Arts 135–143��242 Art 137��242 Art 138��242 Art 139��243 Art 141(2), (3)��244 Art 143��242 Art 144��243 Arts 144–148��242 Art 149��242 Arts 149–154��242 Art 152��243

xxxii Table of Legislation Art 154��243 Art 157(2)��243 Art 169(2)��243 Art 352��244 Art 354��244 Art 355(2)��244 Art 356(3)��244 Art 396��244 Art 401��244 Art 407��244 Commercial Code see Code of Commerce Decree No 100/203 of July 22, 2006 related to the Regulation of Microfinance Activity��240 Decree of March 20, 1923 relating to Warrants��240 Labour Code 1991 Art 98��240 Law No 1/017 of 23 October 2003��241 Art 5��241 Art 8��241 Art 14��241 Law No 1/01 of 16 January 2015��241 Law on Leasing 2016�� 241, 245, 246 Art 1(1)��241 Art 2��246 Art 3��241 Art 7��246 Art 9��248 Art 10��258 Art 10(1)–(3)��258 Art 13��253 Art 14(1)��253 Art 14(2)�� 251, 252 Art 26(1), (3)��251 Art 28(1), (2)��251 Art 34��258 Art 36��258 Art 37��258 Law on Security Rights (Law No 10 of 12 August 2016)�� 241, 243, 245, 252, 260, 261, 438 Art 1��241 Art 3�� 246, 254 Art 4��247 Art 6��247 Art 8��247 Art 14(1)��247 Art 15��249

Table of Legislation xxxiii Art 15(1)��248 Art 15(2)��248 Art 15(3)��248 Art 17�� 245, 248 Art 18��249 Art 18(1)��248 Arts 18–22��248 Art 19��249 Art 20��249 Art 21��249 Art 22��249 Art 23��250 Art 24��250 Art 26��250 Art 26(1), (2)��250 Art 27��250 Art 28(2)��252 Arts 28–37��252 Art 29��252 Art 30��253 Art 31��253 Art 32��257 Art 32(1)–(3)��253 Art 35��253 Arts 38–41��255 Art 40��249 Art 41(1)��249 Art 41(2)��250 Art 42�� 248, 253 Art 46��255 Art 47�� 246, 255 Art 48��254 Art 49�� 249, 254 Art 50��254 Art 51��254 Art 52��257 Art 54��256 Art 55(1)–(3)��256 Art 56(1), (2)��256 Art 57(1)��257 Art 58��257 Art 59��257 Art 60��257 Art 61��257 Art 62�� 252, 257 Art 63��250

xxxiv Table of Legislation Art 63(1)��252 Art 63(2)��253 Art 63(3)��252 Art 65��247 Arts 66–80��247 Art 70(1)��257 Art 74��247 Art 75��257 Arts 75–78��247 Art 76�� 248, 250 Art 77��247 Art 79�� 247, 251 Art 80��247 Art 81�� 246, 248 Art 82��248 Art 83��251 Art 83(1), (2)��251 Art 84��258 Art 85�� 255, 258 Art 86(1)��257 Art 87(1), (2)��259 Art 88��258 Art 89��258 Canada Personal Property Security Acts (PPSAs)�� 9, 12, 13, 21, 121, 123, 197, 321, 363, 364, 416, 425, 434 Saskatchewan Personal Property Security Act 1993 �� 124, 125, 362 s 2(1)(y)��174 s 3(1)��175 s 3(1)(b)��364 s 3(2)��124 s 3(3)��175 s 4(g)��123 s 3(1)(b)��104 s 10(3)��180 s 12(1)(b)��156 s 28(3)��125 s 30(2)�� 159, 185, 427 s 30(6)��368 s 34(11)��321 s 35(1)(c)��46 s 43��177 s 43(6)��177

Table of Legislation xxxv s 45��368 s 55(2)��185 s 55(4)(b)��371 s 59(16)(c)��186 s 62�� 161, 187 Quebec Civil Code 1994��12 Ethiopia Banking Business Proclamation 2008��266 Business Mortgage Proclamation 1998�� 267, 269 Pt 1��268 Pt 2��268 art 4��274 art 5�� 268, 275 art 5(4)��275 art 9��268 art 10��268 art 11��275 art 12��268 art 13��269 art 18(2)��269 Capital Goods Leasing Business Proclamation 1998�� 267, 272 art 2(1)��272 art 2(3)(b)��272 art 2(4), (5)��272 art 2(8)��272 art 4��272 art 6�� 269, 273 art 11(2)��273 art 12��273 art 14��273 art 17��269 art 17(1)��273 art 18��269 Civil Code of the Empire of Ethiopia Proclamation 1960��265, 266, 267, 276 arts 1962–1975��270 art 1967(2)��271 art 2387�� 271, 272 art 2728��272 art 2817��276 art 2821��268 art 2825��270 art 2827��270 art 2828(2)��270

xxxvi Table of Legislation art 2829��270 art 2830(1)��270 art 2831(1)��270 art 2832��270 art 2851��269 art 2851(2)��268 arts 2851–2854��268 art 2864(1), (2)��270 art 2865��270 art 2872��270 art 3060��269 art 3347��283 Civil Procedure Code 1965��265 Commercial Code of the Empire of Ethiopia Proclamation 1960 �� 165, 266, 267 Book V�� 266, 274 art 124��274 art 125��274 art 127, 128��274 art 171��273 art 171(2)��273 art 173(2)��274 art 175��274 art 177(2)��274 art 178�� 274, 275 art 179(3)��274 art 191(2)��274 art 192(2)��275 art 192(3)��275 art 193��274 art 212��274 art 189��268 art 329��268 art 342��268 art 947��271 art 949��268 art 950��271 art 952(2), (3)��271 Commercial Registration and Licensing Proclamation 2016��281 art 4(1)��281 art 4(3)��281 Constitution of the Federal Democratic Republic of Ethiopia (1995)��266 art 47��267 Criminal Procedure Code 1961��265 Enforcement Proclamation 1998 art 8��269 art 8(2)��269

Table of Legislation xxxvii Maritime Code 1960��265 Penal Code in 1957��265 Proclamation No 14/1984��268 Proclamation to Provide for the Establishment of the Ethiopia Commodity Exchange 2007��277 art 9��268 art 24��277 Proclamation to Provide for the Payment of Stamp Duty, Proclamation 1998 art 3��267 art 11��280 Property Mortgaged or Pledged with Banks Proclamation 1998��269 Registration and Control of Construction Machinery Proclamation 1999��268 art 2(2)��268 art 3��268 Rules of the Ethiopian Commodities Exchange 2008 art 9.5.2.1��277 art 9.5.2.1(c)��277 art 9.5.2.1(d)��277 art 9.5.2.4��277 art 9.5.2.4(a)(ii)��277 art 9.5.2.4(c)��277 Secured Transactions Proclamation (draft)��277, 279, 280, 454 art 2(17)��279 art 2(29)��281 art 2(31)��279 art 2(37)��282 art 2(43)��279 art 3(2)(a)��279 art 4(1)��280 art 4(2)��279 art 4(5)��280 art 13��280 art 20��280 art 27(1)(a)��281 art 28��281 art 37��281 art 39��281 art 55�� 281, 282 art 81(2)��282 art 82(1)��283 art 89��280 art 93(1), (2)��283 Warehouse Receipts Proclamation 2003�� 267, 275, 276 art 3(1)��276 art 8��276 art 18��276

xxxviii Table of Legislation art 19��276 art 20��276 art 24��276 art 24(3)��276 art 24(6)��276 France Civil Code 1804��239 arts 524, 525��255 Commercial Code 1807�� 239, 265 Ghana Auction Sales Act 1989 s 11��88 Bodies Corporate (Official Liquidations) Act 1963 s 41��72 Borrowers and Lenders Act 2008�� 65, 66, 68, 69, 70, 71, 72, 73, 74, 75, 76, 79, 80, 81, 82, 83, 85, 87, 195, 196, 278, 426, 432, 433, 437, 438, 439, 441, 446, 447, 450, 452 s 1��74, 76 s 1(1)��75 s 2�� 74, 75, 76 s 3��75 s 4��75, 83 s 5��74, 75 ss 6–19��74 s 7��83 s 12��78 ss 13–20��74, 75 s 18��78 s 18(3)��78 s 19��78 s 21��82 ss 21–31��75 s 22��83 ss 22–25��82 s 25��83 s 25(1)��78, 84 s 25(2)��80 s 25(3)�� 78, 79, 80 s 26��84

Table of Legislation xxxix s 27��84 s 28��85 s 29��82 s 30��82 s 31��85 s 32��87 s 32(3)��201 ss 32–36��82 s 33(b)��75, 87 s 34��75 s 35��75, 87 s 37(3)��77 s 38�� 74, 76, 76, 78 Schedule��78 Borrowers and Lenders Bill 2017 �� 65, 66, 68, 69, 70, 74, 75, 76, 77, 78, 79, 80, 86, 87, 88, 89, 90, 432, 433, 440, 446, 447, 449, 453 s 1(1)(c)–(j)��89 ss 3–6��78 s 7��78 s 8��79 s 11��77 s 16��89 s 27��70 s 49��89 s 50��89 s 52��89 s 52(1)(b)��79 s 54��89 s 63��76 s 76�� 77, 79, 81, 441 s 77�� 81, 89, 440 s 79(1)��81 Companies Act 1963�� 71, 72, 76, 81, 433, 438, 441 s 86��24 s 86(2)��24 s 87�� 24, 82, 441 s 88�� 72, 82, 441 s 107��80 s 107(1)��72, 80 s 107(3)��72 s 107(6)��83 s 114��84 s 115��85 Constitution of Ghana 1992 art 11��68

xl Table of Legislation Contracts Act 1960 s 7��70 s 7(2)��70 s 7(5)��70 Finance Lease Act 1993��73 s 2(1)(d), (f)��73 s 6��73 s 6(4)��73 s 21��73 Hire Purchase Decree 1975��73 s 3(1)��73 s 7(1)��73 s 7(3)��73 s 8��73 s 8(4)��73 s 7(3)��73 Incorporated Private Partnerships Act 1962��71 s 3(1)��71 s 5��71 s 21��71 s 21(2), (3)��71 s 22��71 s 27(4)��71 Land Bill 2017 s 152��76 Land Title Registration Act 1986 s 62��77 Mortgage Act 1972��87 Ordinance to Make Provision Relating to Chattel Securities and the Transfer of Chattels (Ordinance) 1952��69 s 2��69 Rules for Effective Implementation of the Borrowers and Lenders Act 2012 ��81, 85 reg 5(3)��80, 83 reg 9��85 reg 10��85 reg 13��85 reg 16��80, 83 reg 17��88 reg 20(1), (2)��88 reg 25��83 reg 26��85 Sale of Goods Act 1962��70 s 26��70 s 31��71 s 32��71 s 36��70 s 38��70

Table of Legislation xli Kenya Agricultural Finance Corporation Act 1969��115 Banking (Amendment) Act No 25 2016 s 33B��94 Business Registration Services Act 2015��115 Central Depositories Act of 2000��105 s 42�� 105, 421 s 42(6)��105 Chattels Transfer Act 1930 (Rev 2012)�� 96, 97, 98 s 2��96 s 5��96 s 6��96 s 9��96 s 13��96 s 14��96, 97 s 15��96 Sch 1��97 Companies Act 1962 (Rev 2009)�� 100, 115 s 96��100 Companies Act (No 17) 2015�� 96, 100, 102 Pt XXXII�� 100, 102 s 3��103 s 3(1)��104 s 832��117 s 832(2)��102 s 832(3)(c)��102 s 832(4)��102 s 854(1)(h)��102 s 877(1)��100 s 878(1)�� 100, 103 s 878(4)�� 100, 102 s 878(8)��27, 103 s 885(1)(a)��100 s 889��103 s 889(1)��100 Hire Purchase Act 1982�� 97, 98, 106, 113, 115, 117, 442 s 2(1)��97 s 3(1)��98 s 4�� 97, 98, 442 s 5��98, 442 s 5(1)��98 s 5(4)��98 s 6��106 s 15(1), (2)��98 s 65(4)��106

xlii Table of Legislation Income Tax Act 2002��97 Income Tax Act Leasing Rules 2002��97 Insolvency Act (No 18) 2015�� 100, 115 s 2(1)��104 s 471��101 s 473��101 s 474��101 s 534�� 101, 102, 441 s 582��101 s 587��101 s 650��117 Sch 2��26, 101 para 4��101 Movable Property Security Rights Act (No 13) 2017��91, 98, 100, 103, 106, 107, 116, 430, 431, 432, 441 s 1(2)��103 s 2�� 22, 104, 107, 108, 111, 112 s 3(1)��98 s 4(1)��103 s 4(1)(a), (b)��104 s 4(1)(c)��105 s 4(2)�� 105, 116 s 5(2)�� 111, 113 s 6(3)��105 s 6(4)��106 s 8(3)��106 s 9��103 s 19(2)��107 s 19(2)(a)��107 s 21(1)�� 107, 109 s 27(1)(c)��108 s 27(1)(e)��109 s 30��107 s 33(5)–(7)��108 s 34�� 108, 109 s 34(1)��108 s 35��108 s 35(4)��109 s 35(6)��108 s 37��108 s 38��110 s 40��110 s 41(2)��110 s 44�� 111, 112 s 46��101 s 47��112

Table of Legislation xliii s 52��111 s 54��111 s 55��111 s 65(1)��113 s 65(4)�� 98, 113, 442 s 67(2)��114 s 67(3)��113 ss 69–78��114 s 81��114 s 88��114 s 88(2)(b)��109 s 92(1)(b)��114 s 94�� 97, 106, 115 Schedule��442 Movable Property Security Rights (General) Regulations 2017��109 reg 8(2)(h)��109 reg 11��109 reg 11(3)�� 109, 110 reg 12��109 reg 14��109 Sch 1��109 Pawnbrokers Act 1966��115 Sale of Goods Act 1930 s 19(1)��99 Stamp Duty Act 1958 (Rev 2012)�� 106, 115 s 31��106 s 38��106 Schedule��100 The Statute Law (Miscellaneous Amendments) 2007��98 Liberia Act to Repeal the Fraudulent Conveyance Act 2016��131 Commercial Code 2010�� 119, 429 Ch1 s 1.18(1)(bb)�� 122, 123 s 1.18(1)(g)��122 Ch 2��119 s 2.45(1)(b)��122 Ch 3 s 3��122 Ch 3A��121 s 3A��122 s 3A.16(2)��126 s 3A.18��126

xliv Table of Legislation Ch 4��119 Ch 5�� 119, 121, 122, 124, 131, 415 s 5.2�� 121, 123, 124 s 5.3�� 121, 123 s 5.3(a)��121 s 5.3(d)��123 s 5.4��121 s 5.5��121 s 5.6��121 s 5.8��121 s 5.9(aa)��123 s 5.9(gg)�� 123, 124 s 5.9(s)��123 s 5.10��124 s 5.12��124 s 5.13��124 s 5.14��124 s 5.24��124 ss 5.28–5.31��124 s 5.31(2)��125 s 5.31(3)��125 ss 5.34–5.36��125 s 5.37(1)(a)��125 s 5.41��125 s 5.42��125 s 5.44��125 s 5.45�� 121, 125 s 5.47��125 s 5.49��128 s 5.51(3)��126 s 5.52��127 s 5.53��127 s 5.54(3)–(7)��127 s 5.59�� 127, 129 s 5.59(1)��127 s 5.60��127 s 5.62(1)��130 s 5.62(6)��130 s 5.63��130 s 5.64��130 s 5.65��130 s 5.65(1), (2)��130 s 5.66��130 s 5.68(2)��130 s 5.70��130 s 5.75�� 122, 128

Table of Legislation xlv Ch 6��119 Ch 8��131 s 8.4��131 s 8.6��131 s 8.7��131 s 8.46��131 s 8.53(1)��131 Judiciary Law 1972, Title 17��131 Art II(1)��131 Art IV��131 Art VI��131 Regulations for Secured Transactions/Collateral Registry 2013��122 s 1(8)��128 s 6(3)��129 s 9(1)(a)��128 s 9(2)��126 s 10��128 s 10(2)��127 s 11��128 s 13(6)��128 s 14��127 s 14(1)��127 s 14(2)��128 s 14(4)��127 s 20�� 127, 128 s 20(2)��127 s 20(3)��128 s 22��128 Malawi Farmers’ Stop Order Act 1956��133 Personal Property Security Act (No 8) 2013��133, 135, 173, 179, 180 s 2��445 s 2(1)�� 135, 136 s 16��135 s 16(a), (b)��135 s 44��136 s 48��137 Warehouse Receipts Act 2018�� 134, 420 Pt V��134 s 7��136 s 7(1)��136 s 35��135 s 36��135

xlvi Table of Legislation s 36(1), (2)��135 s 36(3)�� 135, 136 s 37��135 s 37(1), (2)��135 s 38�� 135, 136 s 38(2)��136 s 39��135 s 39(1)��136 s 39(2)�� 136, 137 s 44��135 New Zealand Chattel Transfers Act 1924��115 Companies Act 1993��25, 115 Sch 7��115 para 2��25 para 3��115 Companies (Registration of Charges) Act 1993��115 Corporations Act 2001 s 51C��26 s 433��26 s 561��26 Insolvency Act 2006��115 Personal Property Security Act 1999��25, 61, 173, 279, 425 s 16��44 s 16(1)��174 s 17(3)��115 s 23(e)(x)��123 ss 35–40��46 s 40��156 s 53(1)�� 159, 427 s 66(c)��46 s 87��32 s 105��44, 185 s 108��202 s 114(2)(e)��186 s 120��202 s 133��187 Property Act 2007��115 Nigeria Banks and Other Financial Institutions Act 1991��150 s 57��150

Table of Legislation xlvii Bills of Sale Act 1923��141, 142, 143, 163 s 8��143 s 9��143 s 10��143 s 14��143 Central Bank of Nigeria Act 2007 s 2(d)��150 Companies Act 1968��145 Companies and Allied Matters Act 1990��142, 145, 152, 153, 154, 160, 162, 163, 164, 440 s 178(1)��145 s 178(1)(a)��153 s 179��146 s 180��153 s 182�� 146, 154 s 197��141 s 197(1)��146 s 197(2)�� 146, 150 s 197(11)��145 s 206��153 s 494��154 s 494(1)��146 Companies and Allied Matters Act 1990 (CAP C20, LFN 2004) Repeal and Re-enactment Bill 2018��152 Constitution 1999��142 s 2(1)��142 Sch 2 Pt I��142 Pt II��142 Item 18��142 Customs and Excise Management Act 2004��144 Equipment Leasing Act 2015��139, 140, 147, 148, 149, 163, 431 s 1��148 s 3��148 s 7��149 s 9(1)(a)��149 s 9(2)(e)��149 s 12��148 s 13��149 s 14��149 s 17��149 s 19��149 s 19(4)��149 s 20(1), (2)��149 s 21��149 s 37��149 s 38��149

xlviii Table of Legislation s 39(1)��149 s 39(3)��149 s 44��148 Sch 1 Form A��149 Hire-Purchase Act 1965��141, 142, 147, 163 s 1(a), (b)��147 s 9(1), (4)��147 Interpretation Act 1990��6 s 32(1)��144 s 45��17 Laws of the Federation of Nigeria 2004��144, 145, 147, 150 Merchant Shipping Act 2007��144 Pawnbrokers Act 1917��143 Railway Corporation Act 1955��144 Registration of Security Interests in Movable Property by Banks and Other Financial Institutions in Nigeria Regulations (Regulations No 1 of 2015)��139, 149, 150, 151, 155, 156, 157, 162 reg 2(1)��151 reg 3(3)(b)��150 reg 4(2)��150 reg 5(2)��151 reg 9(1)��151 reg 12(2), (4)��157 reg 13(1)–(3)��157 Sale of Goods Act 1893��144 Secured Transactions in Movable Assets Act 2017��139, 140, 141, 147, 148, 150, 151, 153, 154, 155, 159, 163, 164, 431, 454 Pt II��156 s 2(1)��152 s 2(1)(c)��148 s 2(3)��152, 154, 160, 440, 441 s 3��155 s 4��155 s 4(1)��155 s 4(2)��152 s 4(2)(b)��155 s 5�� 155, 156 s 5(c)��158 s 6��155 s 6(1)(b)��156 s 8(2)��156 s 10(2)��157 s 13��157

Table of Legislation xlix s 14(1)(f)��158 s 14(2), (5)��157 s 15(3)–(5)��157 s 16��157 s 17�� 157, 158 s 20��162 s 23��157 s 24(2)��158 s 27��158 s 29(2)��159 s 32(2)��159 s 33(3)��157 s 34��159 s 35(1)��159 s 36��159 s 37��160 s 38��160 s 39�� 154, 160 s 39(1)(a)��153 s 39(2)��152 s 39(5)�� 153, 160 s 40��160 s 41��161 s 42��161 s 44(3)��161 s 46��161 s 48��161 s 49��161 s 54�� 154, 162 s 57��157 s 58��162 s 58(3)��162 s 61��157 s 62��162 s 63(1)��148, 151, 155, 157, 158 Stamp Duties Act 2004�� 143, 154 s 45��143 Rwanda Companies Act 2009��311 Companies Act 2017 s 312��311 Companies Act 2018��311 Decree of 12 January 1920 on Security Interests in Business and Commercial Invoices��307

l Table of Legislation Decree of 28 July 1934 on Bill of Exchange, Promissory Note and Protest��307 Law Governing Finance Lease Operations in Rwanda 2015��309 Law Governing Finance Lease Operations in Rwanda 2015��311 Law Establishing Regulations and Conditions Governing Lease Operations 2005�� 309, 311 art 2��311 art 14��324 arts 14–17��311 art 15��326 art 16��324 art 17��324 arts 28–29��311 Law Governing Contracts 2011��308 Law on Security Interests in Movable Property 2013��307, 309, 310, 311, 312 art 1��312 art 2�� 312, 315 art 2(1)��319 art 2(3)��319 art 2(4)��316 art 2(8)��313 art 2(15)��313 art 2(20), (21)��313 art 2(22)��320 art 3��312 art 4�� 312, 313 art 4(4)��312 art 4(5)��313 art 4(6)��312 art 5�� 314, 315, 318 art 6�� 310, 314 art 7�� 314, 315 art 8�� 314, 315, 323 art 8(1), (2)��315 art 9��316 art 10��317 art 11��317 art 12�� 318, 319 art 13�� 313, 317, 319 art 14��319 art 15�� 319, 320 art 16��320 art 17�� 311, 320, 321 art 19��321 art 20��322 art 21��322 art 22�� 311, 322 art 22(2)��322

Table of Legislation li art 23��323 art 24�� 311, 323 art 25�� 323, 324, 326 art 25(1), (2)��324 art 25(4)��324 art 25(5)��325 art 25(6)��324 art 27��325 art 27(1)��325 art 27(3)��325 art 28��325 art 30��310 art 32�� 324, 325, 326 art 33��326 art 34��326 art 34(4), (5)��318 art 36��311 art 37��310 Law on Security Interests in Movable Property 2009�� 307, 311 art 35��307 Sierra Leone Borrowers and Lenders Act 2014�� 191, 192, 193, 194, 195, 197, 204, 205, 430, 432, 447 Pt II��196 Pt III��198 Pt IV��200 Pt V��200 Pt VI��196 Pt VII�� 197, 201 Pt VIII��203 s 1�� 196, 197 ss 4–8��196 s 5(3)��196 s 9��198 s 9(1)��198 s 10��198 s 10(2)��197 s 11��198 s 11(1), (3)��198 s 12��198 s 13��198 s 15��198 s 16(1)(e)��199 s 16(2)�� 199, 205

lii Table of Legislation s 17(2)��200 s 17(3)��199 s 21��198 s 22��200 s 23��200 s 24��200 s 25��200 s 26(1)��200 s 28��200 s 28(2), (3)��201 s 29��201 s 31��201 s 36(1)��196 s 38��197 s 38(6)�� 197, 201 s 38(7)��197 s 39��202 s 39(3)��201 s 40��202 s 41��202 s 41(2)��202 s 42(1)��202 s 43��202 s 43(3)��202 s 46��202 s 47��202 s 47(2)��202 s 49(1), (2)��202 s 50��202 s 51��207 s 52(2)(b)��203 Borrowers and Lenders Act 2019��191, 199, 205, 209, 426 s 1��206 s 2�� 206, 207 s 2(b)(iv)��206 s 3(1)(iv)��206 s 3(b)(vi)��7 s 3(2)��206 s 6(2)(b)��207 s 12(4)��207 s 15��207 s 18��207 s 18(3)��207 s 18(3)(b)��207 s 18(4)��207 s 19(a), (b)��207

Table of Legislation liii s 20��207 s 26(2)(b)��208 s 26(3)��207 s 35(5)��208 s 45(4)–(6)��208 s 46��208 s 48��206 s 49��208 s 51��206 s 63��208 s 64(1), (3)��208 s 77(1), (2)��206 s 78(1)��207 Borrowers and Lenders (Collateral Registry) Regulations 2016��198 Pt IV��199 Pt VII��205 reg 12��199 reg 23��205 reg 24(2), (3)��205 reg 28��193 Schedule��205 Companies Act 2009�� 192, 194, 199, 204, 209, 432 Pt VIII��192 Pt XVI��192 s 18��192 s 70(1)��193 s 145(1), (2)��192 s 150�� 192, 206 s 151�� 192, 204, 206 s 154��207 s 164(2)��193 s 172(1)��193 s 170��193 s 170(1)��193 s 170(2)�� 193, 204 s 170(2)(c)��193 s 171(1)(b)��193 s 172(1), (3)��193 s 175�� 193, 206 s 180��206 s 181�� 193, 206 s 438(1)��192 s 438(4)(b)��207 Courts Act 1965��6 s 74��17

liv Table of Legislation Goods and Services Tax Act 2009 s 99��199 Interpretation Act No 8 of 1971 s 12��203 Other Financial Services Act of 2001 s 2��194 s 15��194 s 53��194 Patent Act 2014��194 Registration of Business Act 2007�� 192, 194 s 13��194 Road Traffic Act 2007��195 s 1��195 Trademarks Act 2014��194 South Africa Companies Act 1973 s 341(2)��346 Companies Act 2008 s 134��334 Constitution 1996��330 Deeds Registries Act 1937��350 s 50��344 s 53(1)��343 s 61��333 s 61(1)��344 s 61(3)��344 s 62��333 s 62(1)��344 s 62(4)��344 s 102��333 Designs Act 1993 s 30��333 Financial Markets Act 2012 s 39(1)(a)(i), (ii)��340 s 40��340 Group Areas Act 1950��330 Insolvency Act 1936�� 334, 349 s 2�� 337, 350 s 83��337 s 83(12)��337 s 84(1)��334 s 86��343 ss 96–101�� 347, 351

Table of Legislation lv s 102�� 333, 343, 347 s 103��347 Interim Constitution 1994��330 National Credit Act 2005��331 Notarial Bonds (Natal) Act 1932�� 347, 348 Patents Act 1978 s 60��333 Security By Means of Movable Property Act 1993��343, 347, 348, 349, 350 s 1�� 333, 347, 348 s 1(1)�� 333, 335, 350 s 2��350 Trade Marks Act 1993 s 41��333 Tunisia Code of Civil and Commercial Procedures 2018��288, 289, 298, 301 art 186��302 art 216��300 arts 219–223��299 art 322��300 art 322(2)��300 art 324��300 art 360��294 art 361(1)��294 art 390��301 art 394��301 art 396(2)��301 art 398��301 arts 404–409��289 art 406��301 art 407��301 Code of Commerce 2010��288 arts 189–268��290 art 199(2)��296 art 238(1)–(3)��294 art 239(1)��294 art 240��297 art 570�� 296, 297 art 571��297 Code of Maritime Commerce 2010 arts 100–106��289 Code of Obligations and Contracts 2005��288 art 450(2)�� 292, 298 arts 1478–1531��299

lvi Table of Legislation Code of Real Rights 2011�� 288, 289, 299 art 2��292 art 9��293 art 13��293 art 14��293 art 15��293 art 17��293 arts 27–37��289 art 29(b)��297 art 192��296 art 193��296 art 194��296 art 195��296 art 199�� 296, 297 art 200��297 art 214��294 art 217��293 art 218��293 art 221��293 art 247�� 299, 300 art 250(2)��300 art 252��300 art 254��298 art 255�� 298, 299 art 256��298 art 259��297 art 269��297 Commerce Registry Law 1995��288 Commercial Companies Code 2016��289 arts 314–386��289 General Regulations of the Stock Exchange Securities of Tunis 1994 arts 186–193��301 Labour Code 1996 art 134��296 Law on Secured Transactions (draft)�� 287, 288 Law Related to Dematerialisation of Securities 2000�� 288, 289 Law Related to Fiduciary Assignments of Accounts Receivable 2000��288 Law Related to Finance Leasing 1994��288 art 1(1)–(3)��295 art 7��295 art 12��298 Law Related to Pledges of Tools, Machinery and Professional Equipment 2001�� 288, 289, 294 art 1��294 art 2��294 art 4��295

Table of Legislation lvii art 5��297 art 6��295 Personal Status Code 1956��289 UAE Federal Law 2016 on Mortgaging Moveable Properties as Security for Debts��310 Uganda Movable Property Security Interest Act 2019��310 United States of America Uniform Commercial Code Art 1 § 1-201(9)��319 Art 7��276 § 7-106��136 Art 9�� 11, 12, 13, 15, 21, 106, 121, 123, 124, 125, 143, 159, 163, 175, 177, 184, 185, 197, 201, 310, 313, 314, 316, 317, 318, 321, 325, 425, 426, 427, 428, 434, 445, 448, 449, 455 § 9-102(a)(9)��316 § 9-102(a)(44)��313 § 9-104(d)(4)��123 § 9-108(b)��106 § 9-203��314 § 9-203(b)��314 § 9-203(b)(2)��156 § 9-306(d)��32 § 9-312(c), d)��135 § 9-315(d)�� 125, 317 § 9-320(a)�� 185, 201, 427 § 9-332��159 § 9-332(a)��319 § 9-334(i)��321 § 9-406��314 § 9-514(b)��325 § 9-609��208 § 9-620(e)��99

lviii Table of Legislation Zambia Acts of Parliament Act of 1964��188 Agricultural Credits Act 1962��170 Agricultural Credits Act 1995��170 Agricultural Credits Act 2010�� 168, 170, 171 s 2�� 170, 171 s 2(1)��170 s 2A��171 s 3(a), (b)��171 s 15(2), (3)��170 s 16��171 s 16(1), (2)��170 s 17��171 s 37(3)��171 s 133��170 Bills of Sale (Registration) Act 1961�� 171, 172 Civil Courts (Attachment of Debts) Act 1940��175 Companies Act 1994�� 168, 170, 441 Pt 5��170 Pt 5.2��169 s 99(1)��170 s 101��169 Competition and Consumer Protection Act of 2010��175 Co-operative Societies Act of 1998��173 Pt XVI��173 s 129(1)��173 Hire-Purchase Act 1957��168, 169, 172, 173 Pt II��172 s 3(1)��172 s 3(1)(a), (b)��172 s 4��172 s 7(1)(b)��173 s 8��172 s 8(1)(a)��173 s 9��172 s 15(1)��173 s 17��172 s 20(1)��173 s 22��173 Movable Property (Security Interest) Act 2016�� 7, 165, 166, 167, 168, 169, 170, 171, 173 s 2��21, 22 s 2(1)��171, 174, 175, 181, 183 s 2(2)��188

Table of Legislation lix s 3��174 s 3(1)��175 s 3(1)(a)��175 s 3(1)(c)��175 s 3(2)��175 s 4��176 s 4(1)��176 s 5��171 s 6��169, 171, 174, 438 s 6(1), (2)��173 s 8(3)��176 s 11(5)��174 s 12(5)��176 s 13��179 s 15�� 176, 177 s 17�� 177, 179 s 17(1)��177 s 17(1)(a)��176 s 17(2)��177 s 17(5), (6)��177 s 18�� 177, 179 s 19��176 s 21��179 s 22��178 s 23��178 s 32(1)��174 s 33��180 s 33(4)��180 s 34(1)��180 s 35(2), (3)��180 s 36�� 180, 181 s 37��180 s 38��182 s 39��181 s 42��181 s 42(3)��181 s 44�� 181, 188 s 44(1), (2)��181 s 44(3)(a)��183 s 45��182 s 46��182 s 47��182 s 48�� 182, 184 s 49��182 s 52�� 180, 182 s 52(b)��182

lx Table of Legislation ss 53–56��182 s 57(1), (2)��183 s 59��184 s 61��182 s 62��183 s 64��184 s 66��174 s 67��184 s 67(2)��184 s 68��185 s 69��185 s 69(2)��175 s 70��185 s 71��185 s 72(1)��186 s 72(2)��186 s 72(2)(b)��186 s 72(3)��185 s 72(5)��186 s 73��186 s 74��186 s 78��186 s 79��186 s 79(2)��186 s 80�� 179, 187 s 82��187 s 83��187 s 85��187 s 86��187 s 87(1)��185 s 91��187 ss 93–99��187 s 98(1)��188 s 100��174 Sch 1��188 s 2(2)��188 s 2(5)��199 Movable Property (Security Interest) (Fees) Regulations 2016��178 Movable Property (Security Interest) (General) Regulations 2016 (Registry Regulations)��178 reg 2��179 reg 4(2)��179 reg 4(4)��179 reg 7��179 reg 8��179 reg 9��179

Table of Legislation lxi Sch 1�� 178, 179 Forms IX��179 Forms X��179 Forms XI��180 Sch 2�� 178, 179 Patents and Companies Registry Authority Act 2010��176 Trades Charges Act 1973�� 168, 169, 171 s 3(1), (2)��172 s 4(5)��172 s 5(4)��172 Trade Charges (Repeal) Act 2016��172 Trades Licensing Act 1968�� 171, 172 Zimbabwe Agricultural Finance Act 2001�� 358, 366 s 2��359 s 3(1)��359 s 5(1)��359 s 35(1), (2)��358 s 38(2)��358 s 38(4)��358 s 41(1)��359 s 42��359 s 43��359 s 48��359 s 49��359 Banking Act 2000�� 359, 366 Bills of Exchange Act 1895��366 Companies Act 1952��366 s 12(8)��360 s 106�� 360, 439, 440 s 106(2)�� 359, 360 s 106(4), (5)��360 s 107��360 s 107(1)(a)��360 Constitution 2013 Art 71(3)��361 Art 72(1)��363 Art 171(3)��362 Contractual Penalties Act 1992��361 s 4��361 s 4(2)��361 Co-operative Societies Act 2001��366 s 30��358

lxii Table of Legislation Central Securities Depository Operational Rules 2014 s 37(3)��364 Deed Registries Act 1959�� 356, 357 s 69(a)��355 Deeds Registries Act 2001��366 s 2�� 357, 359 s 3��360 s 6��358 s 55��360 Deeds Registries Amendment Act 2017��358 Finance Act 2009��359 Grain Marketing Act 1966��366 Hire-Purchase Act 1957�� 356, 358 Hire-Purchase Act 1982�� 356, 358 s 8��361 Hire-Purchase Act 1987�� 356, 358, 366 s 4��361 s 19(2)��369 s 20��361 Infrastructural Development Bank of Zimbabwe Act 2005��366 Insolvency Act 2018��366 s 89(15)�� 360, 366 Movable Property Security Interests Act 2017�� 356, 358, 362, 363, 438, 440 s 2�� 21, 363, 364 s 3��363 s 3(1), (2)��363 s 3(4)�� 360, 364, 369 s 3(5)��364 s 3(5)(a)��364 s 12(8)��360 s 4��364 s 5��364 s 6��364 s 7��364 s 8�� 365, 371, 372 s 8(1)��365 s 8(2)�� 365, 366 s 9�� 364, 367 s 9(1)��365 s 10��365 s 12��366 s 12(8)��439 s 13�� 366, 367 s 13(2)��366 Sch 1��367

Table of Legislation lxiii Pt I��367 Pt II��367 Pt III�� 367, 368 Sch 2�� 369, 372, 373 Pt I��369 Pt II��369 Pt III��369 Pt IV�� 369, 370 Pt V�� 369, 371 Pt VI�� 369, 372 Stamp Duties Act 2009��359 s 5��360 s 12��360 s 30��360 Tobacco Industry and Marketing Act 1978��366

lxiv

TABLE OF INTERNATIONAL INSTRUMENTS Basel Accords �� 409, 410 Cape Town Convention on International Interests in Mobile Equipment 2001��33, 44, 45, 61, 377, 392, 417, 418, 438, 456 Art 1(j)��379 Art 2��386 Art 2(2)��379 Art 2(3)��386 Art 3��379 Art 3(1)��387 Art 6�� 378, 386 Art 7��387 Art 8�� 380, 381 Art 8(1)��387 Art 9�� 380, 381 Art 10��380 Art 13(1)��381 Art 20��52 Art 29�� 380, 387 Art 39(1)�� 380, 382 Art 39(1)(a)��381 Art 40�� 381, 382 Art 51��386 Art 51(1)�� 386, 388 Art 54(2)�� 381, 384 Annex 1�� 388, 389 Annex 2��388 Annex 3��388 Protocol on Matters Specific to Aircraft Equipment 2001��33, 377, 378, 380, 381, 382, 383, 384, 386, 392, 394, 417 Art III��383 Art IV(1)��379 Art V��383 Art VII��383 Art VIII��383 Art IX(1)(a)��383

lxvi Table of International Instruments Art X�� 381, 383, 384 Art XI��383 Art XIII�� 383, 384 Art XIX��391 Art XXI��383 Protocol on Matters specific to Railway Rolling Stock 2007��33, 377, 378, 383, 384, 385, 386, 387, 394 Art I(2)(e)��384 Art V��385 Art VIII��381 Art IX��384 Art XIII��391 Art XIV��385 Art XVII��385 Protocol on Matters specific to Space Assets 2012��377 Art XX��381 Draft Protocol to the Convention on Matters Specific to Mining, Agricultural and Construction Equipment ��33, 378, 383, 385, 386, 387, 388, 389, 390, 391, 392, 393, 394, 456 Art I��386 Art II(1)��388 Art II(3)��388 Art VII�� 389, 390 Art VIII(1)��387 Art IX��381 Art X��384 Art XII��390 Art XVI(1)��391 Art XXXIII��389 Annex 2��389 Draft Common Frame of Reference for a European Private Law (DCFR) Art IX – 1:102(4)��43 European Bank for Reconstruction and Development (EBRD), Model Law on Secured Transactions (1994)��12 Geneva Securities Convention see UNIDROIT Convention on Substantive Rules for Intermediated Securities (2009) OHADA Uniform Act on Securities 1998 revised 2010�� 28, 29, 33, 122, 213, 238, 438 Art 10��48 OHADA Uniform Act on Securities 2011�� 213, 217, 220, 236, 238, 403, 430, 438, 445, 446, 450 Art 1��215 Art 4�� 214, 216 Art 4(4)��215 Art 5��214 Arts 5–11��216

Table of International Instruments lxvii Art 7��216 Arts 12–49��214 Art 14��214 Art 41��217 Art 50��216 Arts 50–189��214 Art 51��227 Arts 51–61�� 217, 225 Arts 51–66�� 226, 227 Art 52��236 Art 52(2)��236 Art 53��228 Art 53(2), (3)��229 Art 54��229 Art 57�� 230, 231 Art 57(1), (2)��230 Art 58(1), (2)��230 Art 60(1)��230 Art 61(1)��230 Art 64��230 Art 64(1)��230 Art 66(1), (2)��231 Art 67�� 216, 233 Arts 67–189��214 Art 70��217 Art 72��217 Art 73��217 Art 74�� 217, 225 Art 77��217 Art 79�� 218, 235 Art 80�� 218, 234 Art 80(1), (2)��218 Art 82��233 Art 84�� 219, 233 Art 86��219 Arts 86–90��222 Art 87��219 Art 87(2)��219 Art 88��219 Art 91��234 Art 92��220 Arts 92–124��220 Art 93��220 Art 96��220 Art 97��225 Arts 99–102��220

lxviii Table of International Instruments Art 102��232 Art 103��232 Art 104��234 Art 104(1), (3)��234 Art 107��231 Art 107(2)�� 216, 233 Art 108(1), (2)��232 Art 109(1), (2)��232 Art 113��232 Art 118��220 Art 119�� 220, 226 Art 120��220 Arts 120–124��220 Art 125��220 Arts 125–178��220 Art 126��220 Art 127��221 Arts 130–135��221 Art 132�� 221, 233 Art 132(2)��226 Art 133��221 Art 134��221 Art 134(1)��234 Art 135�� 221, 223 Art 136��221 Art 137��224 Arts 137–139��221 Art 138��221 Art 140��222 Art 141��222 Art 142��221 Art 145��235 Art 146��222 Art 147��222 Art 147(1)��222 Art 149��222 Art 151(1), (2)��222 Art 152��234 Art 153��235 Art 154��235 Art 154(1), (2)��235 Art 156��223 Arts 156–161��223 Art 160��226 Art 162��224 Art 162(2)��224

Table of International Instruments lxix Art 162(5)��224 Arts 162–165��220 Art 166��224 Art 172��233 Arts 174–178��227 Art 175(3)–(5)��236 Arts 179–189��224 Art 180��224 Art 181��225 Art 183��225 Art 185��225 Art 224��235 Art 225��235 Art 226��235 Art 227��237 Art 228��237 Organization of American States (OAS) Model Inter-American Law on Secured Transactions (2002) ��12, 122 UNCITRAL Convention on Assignment of Receivables in International Trade see United Nations Convention UNCITRAL Guide on the Implementation of a Security Rights Registry 2014��108, 178, 189, 326 paras 51–53��326 paras 131–135��368 Annex II��48 UNCITRAL Legislative Guide on Secured Transactions 2007 �� 12, 13, 206, 248 Recital 20��42, 44 Recital 24��42 Recital 26��44 Recital 61, 62��36 Recital 70��38 Chapter I paras 16–24��48 para 17��48 para 25��44 para 32��45 paras 36–37��48 paras 37–38��45 para 40��45 paras 110–112��36 paras 115–118��36 Chapter II��37 paras 6–70��38 paras 58–60��38 para 74��48 para 78��48

lxx Table of International Instruments paras 96–99��45, 426 para 108��32 para 110��49 Chapter III��37 para 47��49 paras 97–104��426 paras 102–107��45 para 146��50 para 148��49 Chapter IV��37 para 33��51 para 99��51 Chapter V paras 90–93��57 para 96��57 paras 99–102��57 paras 110–116��426 paras 111–114��45 para 155��55 Chapter VIII para 6��58 para 15��36 para 17��58 paras 18–20��39 para 73��59 paras 85–89��45 Chapter IX paras 52–55��55 para 79��42, 43 para 80��43 para 99��42 para 132��43 paras 154–156��45 para 181��43 para 194��43 para 195��43 Chapter X para 8��60 Chapter IX para 4��61 para 26��61 Recommendation 21��320 Recommendation 43��320 Recommendation 83��297 Recommendation 87�� 320, 426 Recommendation 88�� 320, 426

Table of International Instruments lxxi Recommendation 89�� 255, 319, 426 Recommendation 189��248 Recommendation 190�� 273, 279 Recommendation 192�� 251, 254 Recommendation 194��43, 251 Recommendation 200(a)(ii)��43 UNCITRAL Legislative Guide on Secured Transactions: Supplement on Security Rights in Intellectual Property 2011��45 UNCITRAL Model Law on Secured Transactions 2016�� 12, 13, 15, 21, 22, 32, 40, 41, 46, 79, 89, 197, 236, 262 Art 1��277 Art 1(1)�� 44, 104, 216 Art 1(2)��44, 185 Art 1(3)�� 175, 215 Art 1(3)(b)��223 Art 1(3)(c)��45, 222, 279, 421 Art 1(3)(e)�� 215, 390 Art 1(5)��41, 99, 113, 175, 448 Art 2��84, 112, 174, 267 Art 2(b)��43, 183 Art 2(bb)��48 Art 2(cc)��319 Art 2(dd)��221 Art 2(g)��50 Art 2(g)(i), (ii)��315 Art 2(h)��21 Art 2(j)��58 Art 2(kk)��41, 104, 151, 216, 445 Art 2(o)��21, 53 Art 2(s)��319 Art 2(w)��279 Art 2(z)��315 Art 3�� 36, 57, 252, 399 Art 3(d)��156 Art 3(kk)��364 Art 3(3)��161 Art 4�� 36, 57, 187, 232 Art 6��78, 105, 344, 367 Art 6(1)�� 156, 292, 314 Art 6(2)�� 247, 314, 337 Art 6(3)��46, 155, 156, 416 Art 6(3)(d)��46, 106, 109, 180, 247 Art 7��33, 45, 75, 78, 337, 368 Art 8�� 44, 46, 51, 78, 84, 156, 247, 368

lxxii Table of International Instruments Art 8(b)��247 Art 8(d)��106 Art 8(e)��109 Art 9�� 45, 46, 78, 124, 156, 218, 220, 222, 280, 368 Art 9(1)�� 247, 314 Art 9(2)��247 Art 9(3)��314 Art 10�� 47, 78, 335, 369 Art 10(1)��316 Art 10(2)��48 Art 11��48, 324 Art 12��52 Art 13�� 32, 78, 180, 218, 314, 341, 369 Art 13(2)��155 Art 13(3)��218 Art 14(1), (2)��107 Art 15��32, 78, 180, 249, 369 Art 16��368 Arts 16–21��52 Art 17��368 Art 18�� 49, 248, 298 Art 18(2)��41, 198 Art 19�� 250, 317, 368 Art 19(1)��316 Art 19(1)(b)��316 Art 19(2)(a)��107 Art 22��53 Art 23��320 Art 24��50, 54, 181, 198, 200, 216, 225, 448 Art 25��50, 198, 218, 226, 250 Art 25(c)��182 Arts 25–27��248 Art 26��205 Art 26(2)��135 Art 26(3)��450 Art 27�� 50, 181, 198 Art 29�� 54, 70, 158, 298, 317, 369 Art 29(b)��297 Art 29(c)��110 Art 31�� 110, 317 Art 32��48 Art 33��48, 255 Art 33(2)��319 Art 34�� 184, 370 Art 34(1)��48, 184 Art 34(2)�� 25, 87, 184, 232

Table of International Instruments lxxiii Art 34(3)��25, 57 Art 34(4)�� 47, 57, 159, 184, 201, 427 Art 34(5)�� 47, 57, 184 Art 34(6)��57 Art 34(9)��448 Art 35��49 Art 36�� 58, 112, 154, 185, 225, 319, 400, 404, 416 Art 37��175, 201, 370, 404 Art 37(1)��57, 175 Art 37(2)��57, 185 Art 38�� 56, 112, 125, 158, 183, 200, 217, 254, 317, 370 Art 38(2), (3)��448 Art 39��370 Art 40��175 Art 41��370 Art 42�� 320, 370 Art 44�� 111, 369 Art 44(2)��110 Art 45�� 110, 369 Art 46(2)(b)��184 Art 47��110 Art 47(1), (2), (3)��50 Art 47(6)��57, 159, 184, 319, 427 Art 48�� 57, 184, 319 Art 48(1)��427 Art 48(2)��57 Art 49��111 Art 49(1), (3)��136 Art 50��254 Art 51��50, 182 Art 51(1)��182 Art 53��57, 252, 253, 370 Art 54��57, 370 Art 55(1)(b)��232 Art 55(2)��252 Art 56�� 160, 187 Art 58��219 Art 59��219 Arts 61–65��370 Art 63��226 Art 69��370 Art 72��58, 185, 299, 371 Art 72(1)��160 Art 72(1)(b)�� 113, 160 Art 72(3)��219 Art 73(1)�� 185, 299

lxxiv Table of International Instruments Art 73(2)��58 Art 74��59 Art 75�� 59, 187, 371 Art 76��59, 233, 299, 371 Art 77�� 160, 323, 371 Art 77(2)��59, 256 Art 77(2)(b)�� 113, 186 Art 78�� 186, 256, 323 Art 78(3)��89 Art 78(4)��59 Art 78(8)�� 186, 276, 282 Art 79��89, 187 Art 79(1)��235 Art 79(2)��60 Art 79(2)(b)��59, 219 Art 79(3)��60 Art 80�� 59, 89, 187, 257, 323, 371 Art 81��89, 299 Art 81(1)��59, 322 Art 81(3)��59, 322 Art 81(5)��300 Art 82�� 186, 234 Art 83�� 234, 372 Art 84��372 Art 85�� 60, 161, 372 Art 85(1)��161 Art 85(2)��114 Art 86�� 60, 161, 372 Arts 87–90��161 Art 88��372 Art 89��372 Art 90��372 Art 91��372 Art 92��114 Art 93��114 Art 94��176 Art 96��372 Art 97��162 Art 99��60 Art 100�� 162, 236 Art 101�� 163, 169, 366 Art 104��61 Art 105�� 61, 89, 188 Art 105(1)��61 Art 105(1)(b)�� 237, 259

Table of International Instruments lxxv Art 105(2)��259 Art 106��61 Art 107��61 UNCITRAL Model Law on Secured Transactions: Model Registry Provisions 2016�� 168, 174, 178 Art 1(b)��52 Art 2��52, 249 Art 2(4)��109 Art 4��249 Art 6(4)��229 Art 7(2), (3)��52 Art 8��176, 229, 249, 275 Art 8(5)��176 Art 9��451 Art 9(1), (4)��53 Art 13(1)��176 Art 14��107, 176, 230, 325 Art 15��52 Art 15(4)��52 Art 16��84 Art 20��178 Art 20(3)��52, 84 Art 20(6)�� 107, 249 Art 23��53 Art 24�� 85, 177, 231 Art 24(1)��52 Art 24(4)��52, 177 Art 25��53 Art 26��53, 108, 369, 373 Art 27��326 Art 28(1)��368 Art 30��107 Art 32��365 Art 33��54, 326 UNCITRAL Practice Guide to the Model Law on Secured Transactions 2019�� 435, 436 paras 19, 20��399 UNIDROIT Convention on International Interests in Mobile Equipment see Cape Town Convention UNIDROIT Convention on Substantive Rules for Intermediated Securities 2009 ��45, 223, 420, 421 Arts 11–13��105 Art 12��223 Ch V��105 Art 33��235 Art 33(1)(a)(ii)��301

lxxvi Table of International Instruments UNIDROIT Legislative Guide on Intermediated Securities 2017�� 45, 105, 421 UNIDROIT Legislative Guide on Secured Transactions: Supplement on Security Rights in Intellectual Property 2011��45 UNIDROIT Model Law on Leasing 2008�� 122, 421 Uniform Act on General Commercial Law 2010 Art 20(2), (3)��228 Art 34��227 Arts 34–100��227 Art 35��227 Art 36(1)��228 Art 38(3)��227 Art 40��228 Art 51(1)��228 Art 66(1)��229 Art 135��223 United Nations Convention on the Assignment of Receivables in International Trade 2001 �� 12, 32, 49, 122, 341, 417

part i Introduction

2

1 Introduction A. General Introduction This book is the second in a series looking at secured transactions law around the world, with a particular focus on reform. The first book,1 of which one of the authors was an editor, was a general discussion of reform so far, with particular focus on Europe and the possibility of reform in European countries. It included chapters on the history of reform in the United States and its transplant into Canada, Australia, New Zealand and the first African country to undertake wholesale reform, Malawi. As mentioned below, there has been a relatively rapid take-up of the idea of reform in Africa and Latin America. This volume, therefore, concentrates on African countries, and the fourth volume in the series will concentrate on Latin America. The third volume will look at reform in Asia, particularly East Asia. These three volumes, therefore, each focus on a particular continent, and consider what reform has already taken place, the reasons for and the challenges raised by the reform process, the lessons learned and the likelihood of reform in unreformed jurisdictions. This book is both evaluative and informative. It is an academic study, involving critical analysis of the existing law (reformed or unreformed) and proposed changes, and makes a significant contribution to the debate about whether and how to reform secured transactions law, especially in developing economies. The African continent is a mixture of common law, civil law and mixed systems, all of which are represented in this book, providing resources for comparative law scholars. The book will also be informative to those involved in reforming the law, as well as those who are learning about the reformed law, whether as students or as someone who engages with the law as a matter of practice. This introductory chapter not only introduces the subject matter of the book, but provides some information about matters which will be dealt with throughout the book, such as terminology, some specific legal issues, and some background about the legal and economic situation in Africa, as well as some common types of financing. It also considers secured transactions law reform in general, so as to place the reforms taking place in Africa in a more general context. This introduction includes two tables. Table 1.1 gives the status of the reforms discussed in the book in tabular form for easy reference, plus an analysis of their Doing Business and Getting Credit rankings before and after reforms.2 The Getting Credit Indicator contains two indexes: one measures the strength of legal rights, focusing on secured transactions and 1 O Akseli and L Gullifer (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Hart ublishing, 2016). P 2 For more explanation of these rankings, see B2 below.

Ease of Doing Business rank

Getting Credit rank 2 years before reform*

Country

Year of reform and reform status

Model

2 years after reform**

Strength Depth Strength Depth Getting 2 years 2 years Getting of legal of credit Getting of legal of credit Credit Credit before after Rank 2019 credit rights information credit Rights information reporting Rank 2019 reform* reform** change rank rank index index rank index index reform? change rank

Common law Ghana

2012 – partial

Local drafting

Kenya

2017 – yes

ML

Liberia

2010 – yes

Art 9 and PPSA

Malawi

2013 – yes

Nigeria

92

67

+25

114

113

7 of 10

0 of 6

136

61

170

151

NZ

133

2016 – yes

ML and Art 9

Sierra Leone

2014 – yes

Ghana

Zambia

2016 – yes

PPSAs and LG

28

8 of 10

5 of 6

No

+85

+75

61

116

3 of 12

+19

174

135

4 of 10

164

−31

111

116

147

145

+2

146

141

147

−6

83

85

−2

73

7 of 8

8

10 of 12

8 of 8

Yes

+108

8

0 of 6

98

7 of 10

1 of 6

Yes

+37

112

7 of 10

0 of 6

151

5 of 12

0 of 8

Yes

−35

8

13

9 of 10

5 of 6

6

10 of 12

8 of 8

Yes

+7

12

163

126

7 of 10

0 of 6

152

5 of 12

0 of 8

Yes

−26

161

87

13

9 of 10

5 of 6

2

11 of 12

8 of 8

Yes

+11

3

(continued)

4 Introduction

Table 1.1 Status of reform and changes in Ease of Doing Business and Getting Credit rankings post-reform

Table 1.1 (Continued) Ease of Doing Business rank

Getting Credit rank 2 years before reform*

Country

Year of reform and reform status

Model

2 years after reform**

Strength Depth Strength Depth Getting 2 years 2 years Getting of legal of credit Getting of legal of credit Credit Credit before after Rank 2019 credit rights information credit Rights information reporting Rank 2019 reform* reform** change rank rank index index rank index index reform? change rank

Civil law Burundi

2016 – yes

Ethiopia

2019 – pending

Tunisia

ML but non−unitary approach, French law

140

164

−24

168

170

3 of 10

1 of 6

177

2 of 12

0 of 8

No

−7

178

125

159

−34

159

109

4 of 10

4 of 6

175

3 of 12

0 of 8

No

−66

175

2018 – pending ML

74

80

−6

80

126

2 of 12

5 of 8

99

3 of 12

7 of 8

Yes

+27

99

Rwanda

2013 – yes

58

46

+12

29

32

8 of 10

4 of 6

4

11 of 12

7 of 8

Yes

+28

3

South Africa

No

41

82

−41

82

28

7 of 10

6 of 6

73

5 of 12

7 of 8

No

−45

73

171

155

+16

155

104

5 of 12

3 of 8

85

5 of 12

6 of 8

Yes

+19

85

Mixed law

Zimbabwe 2017 – yes

ML and PPSA

General Introduction 5

* or 2014 if no reform. ** or most recent data.

Simplified Model

6 Introduction the rights of creditors inside and outside of insolvency, and the other measures the depth of credit information. Table 1.2 summarizes the relevant reforms measured in the latter index. Reforms of secured transactions regimes and credit information systems are taken together to establish the country’s ranking in the Getting Credit indicator. Table 1.2 Credit information reforms Country

Year of credit information reform*

Description of reform

Burundi

No reform

—

Ethiopia

No reform

—

Ghana

No reform

—

2015

Improved its credit information system by passing legislation that allows the sharing of both positive and negative credit information and establishes guidelines for the treatment of historical data

2016

Improved access to credit information by passing legislation that allows the sharing of positive information and by expanding borrower coverage

2018

Improved access to credit information by starting to distribute data from two utility companies

2019

Strengthened access to credit by introducing a new Secured Transactions Law, creating a unified secured transactions legal framework, and establishing a new unified and notice-based collateral registry

2009

Improved access to credit information by creating a nascent public credit registry in its central bank

2012

Improved its credit information system by passing a new law allowing the creation of a private credit bureau

2018

Improved access to credit information by guaranteeing borrowers the legal right to inspect their credit data from the credit bureau and by starting to provide credit scores to banks, financial institutions and borrowers (also strengthened access to credit by adopting a new law on secured transactions and establishing a modern collateral registry)

2011

Enhanced access to credit by allowing borrowers the right to inspect their own credit report and mandating that loans of all sizes be reported to the central bank’s public credit registry

2012

The private credit bureau started to collect and distribute information from utility companies and also started to distribute more than two years of historical information, improving the credit information system

Kenya

Liberia Malawi

Nigeria

Rwanda

(continued)

General Introduction 7 Table 1.2 (Continued) Country

Year of credit information reform* 2012

Improved its credit information system by enacting a new law providing for the creation of a public credit registry

2013

Improved access to credit information by establishing a public credit registry at its central bank and guaranteeing borrowers’ right to inspect their personal data

2015

Improved its credit information system by beginning to distribute both positive and negative data and by increasing the system’s coverage rate

2015 (negative effect)

Made access to credit information more difficult by introducing regulations requiring credit bureaus to remove negative credit information from their databases, such as adverse information on consumer behaviour or enforcement action accumulated on a consumer’s record before 1 April 2014

2017

Strengthened credit reporting by starting to distribute historical credit information and credit information from a telecommunications company.

2015

The credit bureau improved access to credit information by starting to exchange credit information with retailers and utilities

2016

The credit bureau began to provide credit scores.

2016

Strengthened access to credit by introducing a new Movable Property Act and by setting up a new collateral registry. The new law implemented a functional secured transactions system. The collateral registry is operational, unified geographically, searchable by a debtor’s unique identifier, modern, and notice based

2016

The credit bureau began to provide credit scores

2017

Improved access to credit information by allowing the establishment of a credit registry

2018 (negative effect)

Improved access to credit information by launching a new credit registry; however, credit scoring was discontinued, reducing access to credit information.

2019

Improved access to credit information by increasing the coverage of the credit registry and providing consumer and commercial credit scores to banks and financial institutions

Sierra Leone

South Africa

Tunisia

Zambia

Description of reform

Zimbabwe

8 Introduction

B. Reform of Secured Transactions Law 1. What is Secured Transactions Law? Very few jurisdictions have a single statute or code encompassing everything that is usually seen as secured transactions law before reform takes place; indeed, codification is one of the main characteristics of a reformed system.3 Thus, pre-reform, there is no such thing as ‘secured transactions law’. Instead, there are a number of devices that are used for the same purpose, namely, so that a creditor can have recourse to a proprietary interest in one or more assets if the obligation due to it is not performed. The interest has to be proprietary since it must be enforceable against third parties who also claim an interest in the asset or assets, including an insolvency officer. As discussed later,4 under the pre-reform law the interest could be granted to the creditor or it could be retained by the creditor in an asset it already owns, and it may not be denominated as a security interest even though it performs a security function. Secured transactions law is the law relating to various critical aspects of these proprietary interests. First, how they come about (creation), which can be either by reservation or grant, but is (usually) consensual. Secured transactions law can include provisions relating to nonconsensual interests but the focus of the law is on consensually created interests. Secondly, how they are made effective against third parties, so that they can be successfully asserted against competing claimants. Thirdly, how priority between competing claimants is determined, and fourthly, how they are enforced in the event of default of the secured obligation. Since the primary way of making a security interest effective against third parties is by making it public by registration (at least in a reformed system), the rules regarding the establishment of a registry and its operation also fall within this type of law. As with all law reform, it is necessary to consider conflict-of-laws rules and transitional provisions as well.5

2. Why Reform Secured Transactions Law? Increasing access to credit, particularly for small and medium-sized enterprises (SMEs), has been on the agenda of many international organisations that are not directly involved in secured transactions reforms. For instance, the G20 recommended a set of actions to facilitate the expansion of financial services available to SMEs in its G20 SME Finance Action Plan adopted in 2015,6 and also in the G20/OECD Effective Approaches for Implementing the G20/OECD High Level Principles on SME Financing. However, the main reasons why African countries undertake these kinds of reforms are the results of economic assessments showing a lack of access to credit, particularly for micro, small and mediumsized enterprises (MSMEs), attributable to a deficient legal framework governing secured transactions.7 The other reason, which also shapes the direction of the reform in terms of 3 Ch 2, A1. 4 Ch 2, B2. 5 See ch 2I and J. 6 See G20, ‘G20 SME Finance Action Plan Implementation Framework: Credit Infrastructure Country SelfAssessment’ (July 2016) www.g20.utoronto.ca/2016/sme-finance-action-plan.pdf. 7 The difficulties of accessing credit are noted in most chapters, see Zambia F.

Reform of Secured Transactions Law 9 particular policies and statutory solutions, is increasing the country’s ranking on the World Bank’s Doing Business index. The Doing Business Report includes an account of the business infrastructure and commercial and regulatory law of each State, but also gives an evaluation of the ‘ease of doing business’ in various categories, the most relevant of which for present purposes is the ‘Getting Credit’ indicator discussed below. It also provides a ranking of countries in these categories, so that a desire to move up these rankings was a significant driver towards legal reform.8 Although there is some merit in the Doing Business rankings, the fact that they are based on quite rigid criteria, and that they only consider the law on the books rather than how it works in practice, means that countries can rise up the list as a result of reforms, even though the actual operation of the law is still problematic.9 The Getting Credit indicator covers two aspects of the regulatory framework and infrastructure that affect access to credit: first, the availability of credit information measured in the credit information index; and, secondly, the protection of creditor and debtor rights under secured transactions and insolvency laws measured by the legal rights index.10 The first aspect, the Depth of Credit Information Index, includes the coverage, scope and quality of credit information available through credit registries and credit bureaus. The second, the Strength of Legal Rights Index, evaluates the degree to which movable property can be effectively used as collateral and the protection of creditor rights. The Getting Credit indicator scores and ranks economies based on the above indexes, which are designed to align with international best practice. However, the Doing Business Reports have been subjected to considerable criticism for their methodology, including the selection of the criteria against which laws are benchmarked, and the way in which those criteria are applied.11 One illustration of the Doing Business Reports not being reflective of the quality of secured transactions laws is the case of Zambia that in 2012 undertook to reform its framework based on the Canadian Personal Property Security Acts (PPSAs). In the 2012 Getting Credit indicator, Zambia ranked 8th, while Canada ranked 24th.12 The scoring and ranking mentioned above depends on the answers to a series of questions. The Depth of Credit Information Index had, formerly, 6 questions and now has 8 questions. The Strength of Legal Rights Index had, formerly, 10 questions and now has 12 questions. The data gathered for these indices comes from a questionnaire administered to lawyers in each country and is verified through analysis of laws and regulations as well as public sources of information on collateral and bankruptcy laws.13 Table 1.1 sets out the Getting Credit Index scoring and ranking of the countries considered in this book before and after reform, as well as their overall Doing Business rankings at those dates. Information about the countries’ Doing Business evaluations is also included in the country-specific chapters.

8 Malawi A; Rwanda A; Burundi D; Ethiopia E4; Contrast Zambia (B) where rising up the Doing Business rankings was not a driver for reform because Zambia was already ranked at 8. 9 See, eg, Rwanda A and B; Nigeria F. See also conclusion B1b. 10 See www.doingbusiness.org/en/methodology/getting-credit. 11 See, eg, G McCormack, ‘Why “Doing Business” with the World Bank May Be Bad for You’ (2018) 19 European Business Organisation Law Review 649, 650. 12 See World Bank, ‘Doing Business 2012 – Country Tables’, www.doingbusiness.org/content/dam/doingBusiness/ media/Annual-Reports/English/DB12-Chapters/Country-Tables.pdf. 13 See World Bank, ‘Doing Business Methodology Getting Credit’, www.doingbusiness.org/en/methodology/ getting-credit.

10 Introduction

3. Who is Secured Transaction Law Reform For? Every reform project studied in this book was aided by one or more economic assessments that revealed the lack of credit for businesses of all sizes, but particularly those operating in the informal economy as microbusinesses. These assessments also showed that financial institutions had identified the antiquated and complex legal framework as well as the absence of a modern electronic registration system as the main causes for their inability or unwillingness to extend credit secured with movable assets. This has had a negative impact particularly on MSMEs that do not own any land, which in many cases was the only acceptable form of collateral. The reforms enabled the use of movable assets as collateral, which facilitated loans of a few hundred pounds, allowing women to start a business or expend their microenterprises. At the same time, large corporates have also benefited from the greater certainty and predictability of the reformed frameworks, reducing their transactional costs associated with the creation of security rights, including the payment of stamp duties, or registrations where the fees were calculated based on the amount of the secured obligation. The reforms have also benefited consumers who can now acquire various types of assets on credit, either under purchase money loans or leases.

4. Who Does Secured Transaction Law Reform? Dozens of international organisations are engaged in commercial law reform. In terms of their role as lawmakers, such entities are often referred to as ‘formulating agencies’.14 Many of these entities define themselves as international standard setters either because they have acquired expertise in legislative drafting or because the composition of their membership gives them international experience15 These agencies include UNCITRAL,16 UNIDROIT17 and the World Bank,18 but also those operating on the regional level, such as the EBRD,19 the OAS,20 and OHADA.21 These institutions support the implementation of the respective instruments they adopt. In the case of the World Bank, its Principles for Effective Insolvency and Creditor/Debtor Systems, developed in collaboration with UNCITRAL amongst others, are an internationally recognised authority for norms concerning the protection of creditor rights in insolvency and debt resolution.22 The Principles address several aspects, including the laws and institutions that recognise and enforce credit agreements (including security agreements) and the legal framework for 14 L Mistelis, ‘Regulatory Aspects: Globalization, Harmonization, Legal Transplants, and Law Reform – Some Fundamental Observations’ (2000) 34 International Lawyer 1055, 1056. 15 ibid. 16 United Nations Commission on International Trade Law. See agency website https://uncitral.un.org/. 17 International Institute for the Unification of Private Law. See agency website https://www.unidroit.org/. 18 See agency website www.worldbank.org. 19 European Bank for Reconstruction and Development. See agency website https://www.ebrd.com/home. 20 Organization of American States. See agency website http://www.oas.org/en/. 21 Organisation pour l’Harmonisation en Afrique du Droit des Affaires (translates to the ‘Organization for the Harmonization of Business Law in Africa’). See agency website https://www.ohada.org/index.php/en/ and OHADA chapter A. 22 World Bank, ‘Principles for Effective Insolvency and Creditor/Debtor Rights System’ (2015), http://documents. worldbank.org/curated/en/557581467990960136/Principles-for-effective-insolvency-and-creditor-debtor-rightssystems.

Reform of Secured Transactions Law 11 risk management and informal corporate workout systems, as well as formal commercial insolvency law frameworks and the implementation of these systems through sound institutional and regulatory frameworks.23 However, for the actual drafting of local secured transactions law, UNCITRAL promotes its own instruments.24 Many other agencies are involved in a reform process, whether through the provision of financial assistance or other types of involvement, which gives them some leverage over the model chosen for the adoption. These include the United States Agency for International Development (USAID)25 as well as the Japan International Corporation Agency (JICA).26 USAID, for instance, promotes economic competitiveness in foreign countries, most recently in the Middle East and North Africa,27 through comprehensive secured transactions reform.28 JICA, in turn, has supported a number of commercial law reforms, including in the field of secured transactions, in East Asia,29 but its approach differs in that it supports reforms of civil codes that contain provisions on secured transactions. In some cases, these agencies, especially USAID, may even ‘export modified successful domestic legislation’,30 as in the case of Article 9 of the Uniform Commercial Code (UCC) in the United States.31 These agencies often promote secured transactions reforms coupled with promises of financial assistance, as was the case in Central and Eastern Europe by the EBRD.32

5. The History of International Instruments and Models The origin of the particular system of secured transactions law which has been adopted in many African countries was Article 9 of the Uniform Commercial Code, prepared by the National Conference of Commissioners on Uniform State Laws and the American Law Institute, which was first approved by the American Bar Association in 1951 and adopted by many US states during the 1950s.33 UCC Article 9 was the first system based on the functional approach,34 and stemmed from the severe fragmentation of the law relating to secured transactions in the US in the first half of the twentieth century.35 It has been through 23 ibid. 24 See B5 below. 25 See agency website www.usaid.gov. 26 See agency website www.jica.go.jp/english/. 27 C Sisko, ‘Promoting Secured Transaction Reform in the Middle East and North Africa’, Chemonics (11 May 2011) www.chemonics.com/blog/promoting-secured-transaction-reform-in-the-middle-east-and-north-africa/. 28 USAID, ‘Country Development Cooperation Strategy 2014–2018’ (April 2014) www.usaid.gov/sites/default/ files/documents/1869/Mexico%20CDCS%202014%202018.pdf. 29 Miha Isoi (Senior Advisor to JICA), ‘Conference on International Coordination of Secured Transactions Reforms II: Advancing Global Reforms and Building a Uniform System’ (16–17 October 2018). 30 ibid. 31 X Nguyen and B Nguyen, ‘Transplanting Secured Transactions Law: Trapped in the Civil Code for Emerging Economy Countries’ (2014) 40 North Carolina Journal of International Law 52, which comments that the Legislative Guide ‘represents the international best practices and epitomizes a modern secured transactions regime’. 32 D Berkowitz, K Pistor and J Richard, ‘The Transplant Effect’ (2003) 51 American Journal of Comparative Law 163. 33 For a full history of the genesis of UCC 9, and its development since the 1950s, see P Winship, ‘An Historical Overview of UCC Article 9’ in Akseli and Gullifer (n 1), 21. 34 See ch 2, B2. 35 See KN Llewellyn, ‘Problems of Codifying Security Law’ (1948) 13 Law & Contemporary Problems 687.

12 Introduction several modifications since the 1950s, and the current version, adopted by all the US states with only very minor variations, is the 2010 amendment.36 The UCC Article 9 model has now been adopted by many jurisdictions around the world. The first jurisdiction to do so was Canada,37 whose various provinces enacted PPSAs from 1976 to 2001. These PPSAs differed considerably from UCC Article 9, both in drafting style and, to a limited extent, in content, but the basic ideas were the same.38 Quebec, a Canadian civil law province, also adopted a version of UCC Article 9 in its Civil Code of 1994.39 From the 1990s, the ideas and concepts from UCC Article 9 and the Canadian PPSAs influenced secured transactions law reform in many different parts of the world. The EBRD produced a Model Law in 1994 for use in transition countries in Central and Eastern Europe,40 and, as a result, many countries in that area reformed their law. Other regional models were developed, such as the Model Inter-American Law on Secured Transactions,41 and Book IX of the European Draft Common Frame of Reference, an academic text produced in Europe as part of a harmonisation initiative.42 Some countries, such as New Zealand43 (which was heavily based on the Saskatchewan PPSA), introduced reforms based on national models, while others took the concepts and structure, but drafted from scratch, such as Australia44 and Jersey. The UCC Article 9/PPSA model influenced the United Nations Convention on the Assignment of Receivables in International Trade (2001 Receivables Convention),45 adopted in 2001 by UNCITRAL. Building on this, UNCITRAL embarked on a programme of work to produce an international model to aid reform around the world. The UNCITRAL Legislative Guide on Secured Transactions, adopted in 2007,46 drew on the experience of States which had already introduced the Article 9 model, but also had input from delegates from many other States whose systems were different. Importantly, these States were both common law and civil law jurisdictions, and the purpose of the Legislative Guide was to assist States from all legal cultures in developing modern secured transactions laws.47 Most of the recommendations in the Legislative Guide were turned into legislative drafting in the UNCITRAL Model Law on Secured Transactions,48 adopted in 2016. Some, however, were not included, and these are also discussed in this chapter as they have been adopted by some African States. 36 Legal Information Institute, ‘UCC – Article 9 – Secured Transactions (2010)’ www.law.cornell.edu/ucc/9. 37 See C Walsh, ‘Transplanting Article 9: The Canadian PPSA Experience’ in Akseli and Gullifer (n 1) 49. 38 These ideas are reflected in the ‘modern principles’ discussed in ch 2. 39 Civil Code of Québec, CQLR c C-1991, Book Six – Prior Claims and Hypothecs. 40 See F Dahan, ‘The EBRD’s Experience in Secured Transactions Reform: How Can Outsiders Help?’ in Akseli and Gullifer (n 1) 445. 41 Adopted in 2002 by the Organization of American States, and which has influenced many reforms in Latin America. See OAS, ‘Model Inter-American Law on Secured Transactions’, www.oas.org/dil/model_law_on_ secured_transactions.pdf; see also B Kozolchyk, ‘Implementation of the OAS Model Law in Latin America: Current Status’ (2011) 28 Arizona Journal of International & Comparative Law 1. 42 See H Eidenmuller et al, ‘The Common Frame of Reference for European Private Law – Policy Choices and Codification Problems’ (2008) 28 OJLS 659. Book IX has never been used as a model for reform in a particular country. 43 See M Gedye, ‘The New Zealand Perspective’ in Akseli and Gullifer (n 1) 115. 44 D Brown, ‘Australian Secured Transactions Law Reform’ in Akseli and Gullifer (n 1) 145. 45 Full text of law available at ‘United Nations Convention on the Assignment of Receivables in International Trade’ (New York, United Nations, 2004). 46 ‘Legislative Guide on Secured Transactions’ (Vienna, United Nations, 2010) (LG). 47 ibid 1, paras 1 and 3. 48 ‘UNCITRAL Model Law on Secured Transactions’ (Vienna, United Nations, 2016) (ML).

Reform of Secured Transactions Law 13 Much of the analysis in this book uses the UNCITRAL Legislative Guide and Model Law as a comparator for the reforms already enacted or proposed in African States. The Model Law is the most suitable benchmark for various reasons. First, it is very recent and embodies most of the up-to-date international consensus about secured transactions law best practices.49 Secondly, it is relatively short and drafted in a straightforward manner, while embodying a fully worked out scheme and not a ‘simplified version’.50 Thirdly, it is developed to be used in both common law and civil law jurisdictions, and, fourthly, it is truly international and therefore not tied to the economy, legal culture or idiosyncrasies of any particular region or State. The content and structure of the Model Law is discussed in chapter 2 of this book, which examines the modern principles on which it is based, and explains the reasoning behind these principles and many of the Model Law rules. In common with UCC Article 9 and the PPSAs, the Model Law is divided up into a number of chapters, each dealing with a different aspect of secured transactions law.51 Part of each chapter includes general rules, and the other part includes rules relating only to specific assets, such as receivables, bank accounts, negotiable documents, negotiable instruments and non-intermediated securities.52 The detailed provisions dealing with the registry, however, are contained in the Model Registry Provisions, which are separately numbered. Footnote references in this book, therefore, will either be to the Model Law (‘ML’) or the Model Registry Provisions (‘MRP’). The Legislative Guide will be referred to in footnotes as ‘LG’. In the text, the terms ‘Legislative Guide’, ‘Model Law’ and ‘Model Registry Provisions’ will be used.

6. Legal Transplant Theory Secured transactions reforms seek to implement an internationally recognised set of principles,53 which, as discussed in the previous section, were developed initially in the United States and eventually became the basis of various international instruments, including the Model Law. The implementation of laws based on these principles raises a number of questions from the perspective of comparative law, particularly in the context of legal transplants. This book examines secured transactions reforms in common law, civil law and mixed jurisdictions underpinned by the modern principles, thus providing a fertile ground to assess how the principles have been integrated into those legal regimes and how they operate. The term ‘legal transplant’ means the transfer of law between societies during a process of lawmaking or legal reform.54 Large-scale legal transplantation began in the colonial era when it was used as a tool for colonists to control their new settlements.55 In the post-World 49 Even this is not always the case, in that technology and ideas have, in some respects, moved on since 2016 and this is pointed out in this book where relevant. 50 See conclusion C1d. 51 This is reflected in the subheadings in ch 2 and in many other chapters considering reformed law. 52 For more discussion of the scope of the ML, see ch 2, B3. 53 See ch 2. 54 M Carfield, ‘Participatory Law and Development: Remapping the Locus of Authority’ (2011) 82 University of Colorado Law Review 739. 55 L Lin, ‘Legal Transplants Through Private Contracting: Codes of Vendor Conduct in Global Supply Chains as an Example’ (2009) 57 American Journal of Comparative Law 711.

14 Introduction War II era, however, legal transplants became an integral part of State-building among dozens of newly liberated territories, many of which borrowed extensively from Western countries.56 The legal transplantation of the colonial and post-World War II periods were typically wholesale transfers from one legal system to another, thus forming various legal families.57 The disintegration of the Soviet Union and the legal reforms pursued in the former socialist States, as well as the related movement toward global economic integration, became ‘a powerful engine pushing the wave of legal transplants to its apex’.58 The topic of legal transplantation first attracted the attention of scholars in the 1970s when two competing theories began to take shape.59 Watson’s theory60 posited that legal transplants are an essential feature of many countries’ legal systems because those in control of the legislative function identify the apparent merits that could be derived from transplanting a foreign law.61 In other words, legal transplantation is not, according to Watson, the inevitable consequence of the social structure of the transplant country, which would not have otherwise adopted the law without a model to copy.62 Otto Kahn-Freund’s theory,63 on the other hand, proposed that legal transplants cannot be separated from the social circumstances out of which those laws arose.64 Even though a country transplants a foreign law, the purpose and application of that law in the adopting country might, according to Kahn-Freund, be completely different compared to that in the country of the law’s origin.65 Building on Watson’s and Kahn-Freund’s theories, scholars of legal transplantation have identified a problem referred to as the ‘transplant effect’, which occurs when the transplanted law does not mesh well with the conditions and institutions existing in the recipient, weakening the effectiveness of the imported legal order.66 Proponents of this theory argue that although legal transplantation has accelerated the development of a formal legal order, it has also profoundly altered the pre-existing legal order, ‘and not infrequently with a detrimental outcome’.67 In order to avoid problems associated with the transplant effect, there must be demand for the law and participation among local stakeholders in the law’s uptake; this, in turn, will spur voluntary compliance and investment in the institutions required for upholding the new legal order.68 All the reform projects undertaken in Africa have engaged a variety of local stakeholders.69 Advocates for a participatory approach to law development argue that an effective legal reform strategy should ensure that measures are taken to adapt the law to the local conditions of the transplant country.70 A number of other international organisations have demonstrated sustained involvement in commercial law-making initiatives, particularly in their role as standard setters in terms of commercial law principles, including the World 56 ibid. 57 ibid. See, eg, the transplant of English company law to African countries, E1b below. See also Kenya B 4. 58 ibid. 59 Mistelis (n 14). 60 Set out in his seminal text, A Watson, Legal Transplants: An Approach to Comparative Law (Athens, University of Georgia Press, 1974). 61 Mistelis (n 14). 62 ibid. 63 O Kahn-Freund, ‘On Uses and Misuses of Comparative Law’ (1974) 37 MLR 1. 64 Mistelis (n 14). 65 ibid. 66 Berkowitz et al (n 32). 67 ibid. 68 ibid. 69 See Conclusion C3. 70 Berkowitz et al (n 32).

The Status and Background of Secured Transactions Law Reform in Africa 15 Bank Group and the International Monetary Fund.71 However, it has been argued that ‘the externally induced drive to modernize institutions of commercial law … by transnational organizations like the World Bank … that placed great stock in models offered by economists in their employ’ has ‘generated an approach to commercial law reform that treated law as a dependent variable, easily malleable under the relentless hammer of economic analysis’,72 and, consequently, the transplanted regimes created confusion and incoherence in other areas of law, which as a result had to be restructured.73 The Model Law has been the basis of a number of reforms in Africa, but in none has it been implemented as adopted by UNCITRAL. Local stakeholders not only considered the options set out in the Model Law, such as whether a security agreement must indicate the amount for which a security right may be enforced, but also alternative approaches not set out in any of the UNCITRAL instruments, such as designating registration as the sole method of third-party effectiveness74 or providing for specific expeditious remedies involving the registry.75 An important aspect of many commercial law reforms is the harmonisation of the legal principles and concepts underlying common law and civil law regimes, which often differ considerably, particularly as they relate to secured lending.76 Writing about his experience as a consultant in relation to secured transactions law reforms in Quebec and Ukraine in the early 2000s, Macdonald wrote that the three most important factors to consider in a secured transactions law reform are, first, the economic, political, social and institutional conditions that make it possible for security on property regimes to enhance access to credit; secondly, the citizenry’s confidence in the judiciary as a fair and honest arbiter of disputes; and, thirdly, the capacity building of legal actors both prior to enactment and for decades afterwards.77 Secured transactions reforms do not directly address the second factor, although in providing for a predictable and certain regime they inevitably reduce the legal risk and reliance on the judiciary. The third factor has not always been the case in Africa, particularly with unstructured and delayed deployment of capacity-building exercises.78

C. The Status and Background of Secured Transactions Law Reform in Africa 1. Secured Transactions Law Reform in Africa During the last decade, Africa and Latin America were the leading reformers of secured lending frameworks. The impetus for reform of the law in both continents was the introduction of collateral registries established on top of inadequate legal frameworks, in Ghana and Mexico, respectively. Ghana’s Collateral Registry attracted the attention of many stakeholders in African 71 ibid. 72 R Macdonald, ‘Article 9 Norm Entrepreneurship’ (2006) 43 Canada Business Law Journal 240. 73 ibid. 74 See Conclusion D4. 75 ibid D6b. 76 See M Boodman and R Macdonald, ‘How Far Is Article 9 of the Uniform Commercial Code Exportable? A Return to Sources?’ (1996) 27 Canada Business Law Journal 249. 77 Macdonald (n 72). 78 See Conclusion C2.

16 Introduction countries, who started looking at the effect of collateral registries on access to credit secured with movable assets. Gradually, they understood that collateral registries must be designed and operated pursuant to an efficient legal framework. Ghana’s redesigned, web-based Collateral Registry was the centrepiece of a pan-African peer-to-peer learning event held in Accra, Ghana in July 2012 and attended by stakeholders from 14 African countries, five of which expressed an interest in seeking assistance from the International Finance Corporation (IFC) in reforming their secured transactions law.79 This book includes chapters examining the reform process and its results in a number of African countries, as well as considering some unreformed jurisdictions. The reformed systems range from fully implemented laws (Malawi and Zambia), to those where certain elements are or have been missing (such as capacity building in Kenya). They also include jurisdictions that enacted laws but are yet to establish registries (Burundi and Zimbabwe), as well as those where a bill is awaiting enactment (Tunisia). Reform efforts are being undertaken in a number of other jurisdictions not covered in this book (Botswana, Madagascar, Morocco and São Tomé), including some in which one of the authors has been involved (Lesotho, Mozambique, South Sudan and Uganda). Many of these projects are supported by the IFC, but other entities are also involved, including the Financial Sector Deepening Trust of Kenya and Uganda, USAID, and its German counterpart Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ). One author has been involved, in some capacity, in most of the projects covered in this book. This involvement started in 2010 with a reform project in Malawi and entailed various activities, such as writing diagnostic reports identifying the deficiencies in legal frameworks and formulating recommendations for reform, assisting in the drafting of laws and regulations, as well as delivering capacity-building exercises. As a result of that involvement, the chapters include some insights into the debates on a number of issues in the reform process, and describe the policies or reasons considered by the stakeholders. In addition to secured transactions law reform, one of the authors has collaborated with other experts engaged to reform related areas of the law, especially insolvency, leasing and warehouse receipts. The greatest challenge of coordinating these reform activities has been helping the stakeholders to understand the functional approach to security rights,80 and its proper reflection, particularly in insolvency and leasing laws.

2. African Legal Systems81 There are three main types of legal family represented in Africa. The central part of this book contains chapters in three groups reflecting these legal families: common law systems, civil law systems and mixed law systems. The legal culture of many of the jurisdictions considered stems, in large part, from their colonial history.82 79 See IFC, ‘IFC and Bank of Ghana to Increase Access to Finance for Businesses in Sub-Saharan Africa through Secured Lending’ (July 2012) www.ifc.org/wps/wcm/connect/1a4ca9004bf0dce49473f71be6561834/ Summary+Ghana+P2P+Event.pdf?MOD=AJPERES. 80 See ch 2, A2a. 81 See further, E below. 82 See CN Okeke, ‘African Law in Comparative Law: Does Comparativism Have Worth?’ (2011) 16 Roger Williams University Law Review 1, 2.

The Status and Background of Secured Transactions Law Reform in Africa 17 Those countries, which were, in the past, British colonies are largely common law jurisdictions. Their law is, at least in part, based on English law, although there may be other influences, such as customary law. Typically, there will be a ‘reception clause’ in a constitutional statute providing that English law statutes of general application in force at a particular date are in force in the relevant country.83 The reception clause can also refer to other aspects of the common law, such as equity. The common law jurisdictions have, of course, developed the law by case-law and statute, but they have largely kept the common law method and, until reform, their secured transactions law broadly mirrored English law in outline, though not necessarily in detail. The common law jurisdictions considered in this book are Ghana, Kenya, Liberia, Malawi, Nigeria, Zambia and Sierra Leone. Some of the civil law countries were French colonies and their law is based on the Napoleonic Code, while others were Belgian, Portuguese or other colonies. The OHADA group of countries were all French colonies, although the Northwest and Southwest regions of Cameroon were previously British colonies and continue to apply English law. Apart from a chapter considering OHADA, the civil law countries covered in this book are Burundi, Ethiopia and Tunisia, the legal regimes of which are all inspired by French law.84 There are also mixed jurisdictions, the law of which has several influences. In this book, ‘mixed’ means that there are civil law and common law influences on the law of a jurisdiction. For example, in Rwanda, the law is a mixture of Belgian civil law and English common law; it is transitioning from a civil law to a common law system.85 The law in South Africa is a mixture of Roman–Dutch law and English common law, and this is also the case in Zimbabwe.

3. African Economies The IMF estimates that in 2019, like in 2018, sub-Saharan Africa will be home to some of the fastest growing economies in the world,86 including Ethiopia, Rwanda, Ghana and Kenya.87 The region is expected to record 3.8 per cent economic growth, slightly higher than the global forecast of 3.7 per cent.88 However, the overall growth projections are affected by the stagnation in the continent’s largest economies, Nigeria and South Africa. Excluding those two economies and Angola, aggregate growth for sub-Saharan Africa is projected to be 5.7 per cent for 2019.89 Agriculture remains the primary source of job creation, providing 60 per cent of all jobs.90 Nonetheless, Africa’s rapid population growth necessitates consequential improvements in agricultural productivity of up to 60 per cent by 2030. Yet, many African countries depend on commodities, especially from the extractive sector, for public 83 See, eg, the Courts Act 1965 (Sierra Leone) s 74; and the Interpretation Act 1958 (Nigeria) s 45. 84 See below E2. 85 Rwanda A. 86 Y Adegoke, ‘Africa Will Have Some of the World’s Fastest-Growing Economies in 2019 – and a Looming Debt Crisis’, Quartz Africa (13 January 2019) www.qz.com/africa/1522126/african-economies-to-watch-in2019-and-looming-debt/. 87 ibid. 88 ibid. 89 ibid. 90 M Diop, ‘Foresight Africa 2016: Banking on Agriculture for Africa’s Future’, Brookings (22 January 2016) www. brookings.edu/blog/africa-in-focus/2016/01/22/foresight-africa-2016-banking-on-agriculture-for-africas-future/.

18 Introduction revenue, rendering them vulnerable to fluctuations in global commodity prices.91 The informal sector accounts for a high percentage of Africa’s economic productivity, contributing 38 per cent of sub-Saharan Africa’s GDP even though the businesses operating in this sector face sizeable access to credit challenges.92 The IFC estimates that the finance gap for SMEs in Africa exceeds US$331 billion. In 2013, the IFC concluded that less than a quarter of adults in sub-Saharan Africa have access to formal financial services.93

4. Types of Financing in Africa Several types of financing products have been, generally, available to African businesses, but the options are limited by the unreformed legal regimes and the lack of expertise, although the position is changing as charted in this book. The traditional asset-based loan secured with inventory and receivables practically does not exist in Africa. Lenders extend credit in three main ways. First, credit secured with collateral that does not fluctuate, such as motor vehicles, equipment, or crops deposited in a warehouse. Secondly, credit that is dependent on there being a creditworthy large company initiating reverse factoring of its accounts payable, and thirdly, a minimal amount of wholly unsecured credit. This section explains in more detail the operation of these financing products and provides examples from African countries.

a. Microfinance Many microbusinesses, if they have external finance at all, use microfinance.94 This typically consists of very short-term loans of low amounts from microfinance institutions. Interest rates tend to be high.95 These loans are heavily based on the relationship between the borrower and the representative of the lender who makes the decision to lend, who typically is local and knows the borrower personally. Monitoring takes place through direct contact, which is frequent and regular. The underlying business model is to establish a stable lending relationship: replacing one very short term loan with the next, repeatedly, gradually increasing the loan amount based on repayment performance, providing the financial flexibility needed for the business to grow.96 This long-term relationship approach works, in a way, as a form of security; the more loans are repaid, the more certain it is that the debtor 91 K Ighobor, ‘Commodity Prices Crash Hits Africa’, Africa Renewal Online (December 2016–March 2017) www. un.org/africarenewal/magazine/december-2016-march-2017/commodity-prices-crash-hits-africa. See CTC E9, highlighting the effect of the MAC Protocol on these sectors. 92 T Collins, ‘Closing Africa’s MSME Finance Gap’, African Business Magazine (5 October 2018) www.africanbusinessmagazine.com/company-profile/4g-capital/closing-africas-msme-finance-gap/. 93 IFC, ‘Access to Finance: Sub-Saharan Africa’, IFC (2013) 2, http://documents.worldbank.org/curated/ en/599611468202144127/Access-to-finance-Sub-Saharan-Africa. 94 World Bank Group, ‘Improving Access to Finance for SMEs: Opportunities through Credit Reporting, Secured Lending and Insolvency Practices’ (May 2018) 5. 95 In Zambia, for example, the average lending rate to microbusinesses was 94.6% in 2012. Bank of Zambia introduced interest rate caps in 2013, which cut this rate in half, but repealed the caps in 2015 following a decline in microfinance lending. See A Ferrari, O Masetti and J Ren, ‘Interest Rate Caps – The Theory and the Practice’ (2018) World Bank Group Policy Research Working Paper 8398, 25. 96 See ‘Microfinance in Africa: Overview and Suggestions for Action by Stakeholders’ (UN Office of Special Advisor on Africa, February 2013) 20–21.

The Status and Background of Secured Transactions Law Reform in Africa 19 is reliable and worthy of future support. In addition, the relationship of mutual trust and friendship adds a further incentive to repay the loan, so that collateral is seldom required.97 Usually, a group structure is used, where the loans are made to several microbusinesses who, in effect, monitor, put pressure on or, even, help each other to repay.98 Since this type of finance typically operates within close-knit communities, one of the major incentives to repay the loan is the preservation of reputation. Thus, despite the fact that these loans tend to be unsecured, the percentage of non-performing loans in the microfinance industry tends to be low.99 In Zambia, the Munda Small-Holder scheme uses group pressure to ensure repayment of loans to District Farmers Associations (DFAs) secured with a 50 per cent cash collateral deposit by the DFA.100 The loans fund DFA members’ purchases of agricultural inputs. DFAs that do not repay their loans on time are prohibited from participating in the programme the following season.101 Another example comes from Tanzania. Here, a microfinance organisation accepts livestock as collateral and lends to individuals rather than groups.102 Instead of focusing on single-crop production cycles, this product is designed for smallholder farmers who have diversified incomes. It therefore allows variable repayments over the loan cycle to accommodate expected fluctuations in the borrowers’ income.103 The secured transactions reforms affect the mechanics of this type of financing in some African countries.104 Although the amounts of the microloans remain low, the security for repayment is not provided in the form of a group guarantee, but rather the microloan is given to an individual borrower who provides some of its assets as collateral for that loan. In Ghana, one of the largest users of the Collateral Registry is a microfinance institution.

b. Financing of Warehouse Receipts In a number of African countries, farmers are able to access credit, but, in the main, only once they have deposited their crop into a warehouse and obtained a warehouse receipt. However, very little financing is provided against growing crops, either because of the number of risks associated with such financing that lenders are unable to effectively mitigate (for example lack of crop insurance) or the uncertain legal framework. Secured transactions laws facilitate financing not only against growing crops, but also those stored in warehouses. However, for the warehouse receipts financing to have the potential to 97 ibid 21. 98 ibid. 99 eg a 2013 IFC survey of 26 MFIs in 12 African countries reported that only 3% of almost 1.2 million microloans were more than 90 days past due, see IFC, ‘IFC Financing to Micro, Small, and Medium Enterprises in Sub-Saharan Africa’, http://documents.worldbank.org/curated/en/728781468194050087/pdf/948700BRI0AFR00 d0Medium0Enterprises.pdf. A similar global survey in 2012 that included 113 MFIs in 53 countries found that only 3% of more than 14 million microloans were more than 90 days past due; see IFC, ‘IFC Financing to Micro, Small, and Medium Enterprises Globally (FY2013)’, www.ifc.org/wps/wcm/connect/3f845500432c1e1683feef384c 61d9f7/2013-MSME-Brochure-Global.pdf?MOD=AJPERES. 100 A Market Study on Microfinance Services in Zambia (Lusaka, Agri-ProFocus Zambia, 2014) 57. 101 ibid. 102 S Harten and S Buri, ‘Who Are the Microfinance Clients? A Case Study on Customer Segmentation and Product Development’ (IFC, 2015) 2, 4, www.ifc.org/wps/wcm/connect/97ca40004a7181c39d8edd9c54e94b00/ Field+Notes+-+MF+Clients+-+A4+Pages+-+EN+-+11-3-2015.pdf?MOD=AJPERES. 103 ibid. 104 See, eg, Ghana C3 and D.

20 Introduction grow, proper w arehouse receipts laws must also be implemented. Malawi enacted one such law in 2018,105 and both Ghana and Kenya are considering bills that should lead to the enactment of such laws in the near future. Furthermore, one of the best-known commodity exchanges for warehouse receipts operates in Ethiopia.106

c. Leasing Leases are also popular in Africa. They are used to enable businesses to acquire equipment and consumers to acquire, in particular, motor vehicles. They exist in the form of financial leases that, under reformed secured transactions regimes, are treated as security rights.107 Financial leasing is a form of specialised financing in which the lessor purchases an asset for use by a lessee for a specific period, in exchange for periodic payments. Typically, the lessee has an option to purchase the asset for a nominal price (for example 10 per cent of the asset value) at the end of the term. Unlike in an operating lease, the lessee’s obligation is not subject to termination in the absence of default. A number of African jurisdictions have recently undertaken leasing reforms.108

d. Supply Chain Finance Supply chain finance optimises the management of working capital and liquidity in supply chain processes, and facilitates the flow of capital between those participating in the supply chain. It facilitates strategic relationships among suppliers, buyers and lenders, which leads to increased trade volumes resulting from greater economies of scale.109 Reverse factoring is the financing product most commonly deployed in a supply chain.110 In reverse factoring, the debtor owing the receivable (the ‘customer’), who is usually a big and creditworthy company, arranges finance for its suppliers by arranging for a bank or other financial institution to buy receivables it owes to suppliers once they have been approved by the customer. The financier advances the purchase price (less a financing charge) to the supplier. The transaction is entered into and priced on the basis of the creditworthiness of the customer rather than the supplier. But other financing products, such as warehouse receipt loans may also support processes within the supply chain. For example, in partnership with the World Cocoa Foundation, Cargill launched a mobile payment system pilot for five farmers’ cooperatives representing 10,000 farmers in Ivory Coast.111 Farmers deliver cocoa to community warehouses where their beans are digitally weighed, assigned a bar code, and payment is made directly to the farmer’s mobile phone or e-wallet. Details of the cocoa beans are 105 Malawi C. 106 Ethiopia D7. 107 Ch 2, B2b. 108 See, eg, Burundi C; Liberia B; and Nigeria D1. 109 IFC, ‘Technology-Enabled Supply Chain Finance for Small and Medium Enterprises Is a Major Growth Opportunity for Banks’ (June 2017) www.ifc.org/wps/wcm/connect/868b5c66-827d-4f15-b56b-81b44ae107b5/ EMCompass+Note+39+Supply+Chain+Financing+FINAL2.pdf?MOD=AJPERES. 110 J Lamoureux and T Evans, ‘Supply Chain Finance: A New Means to Support the Competitiveness and Resilience of Global Value Chains’, SSRN (25 November 2012) 295, https://papers.ssrn.com/sol3/papers. cfm?abstract_id=2179944. 111 See ‘Mobile Banking in Côte d’Ivoire’, Cargill (16 October 2017) www.cargill.com/story/mobile-banking-inc%C3%B4te-d-ivoire.

Terminology 21 recorded and linked to the barcode before the beans are transferred to central warehouses. Using the bar code, Cargill can trace each individual bag of beans to the individual farmer, creating a traceable supply chain.

D. Terminology Apart from some country chapters where the terminology reflects the actual usage in that country’s legislation, the book uses the terminology of the Model Law. In some instances, this differs from the terms used in the PPSAs or UCC Article 9. The terminology of the Model Law is based on the Legislative Guide, which explained that ‘the terms used are not drawn from any particular legal system, and even if found in a particular national system, the Guide does not adopt the meaning of that term’. For instance, though the Legislative Guide and the Model Law are based on the principles of the PPSA and UCC Article 9 regimes,112 the key concept of the Model Law is the ‘security right’ rather than the ‘security interest’. Nevertheless, ‘security interest’ is the phrase used for the core concept in many reformed secured transactions laws across Africa.113 A similarly (legal system) neutral term of ‘third-party effectiveness’ has been used instead of ‘perfection’ in the Model Law and its implementation.114 The legal effect of both is to make the security right effective against third parties, as contrasted to creation upon which the security right becomes effective as between the two parties only.115 The Model Law refers to the person who creates a security right as the ‘grantor’, whether securing its own obligation or that of another person.116 It also defines the term ‘debtor’ as the person who owes payment or other performance of the secured obligation.117 The meaning of the term ‘debtor’ in the PPSA/UCC Article 9 regimes is completely different from the Model Law, as they define a debtor as a person who creates a security interest.118 The term ‘grantor’ is used in this book unless the context specifically requires use of the term ‘debtor’. The Legislative Guide introduced the term ‘notice’ as opposed to financing statement, but without changing the PPSA/UCC Article 9 approach to what must be registered in a registry, which is a set of prescribed data relating to a prospective transaction rather than a security agreement. This book uses ‘notice’ as opposed to ‘financing statement’, again, unless the country-specific legislation uses a different term. In the context of registration, the following terms are used without any difference in the meaning: ‘collateral registry’ as opposed to ‘secured transactions registry’, ‘registry’ as opposed to a ‘filing office’, ‘to register’ as opposed to ‘to file’, a ‘registrant’ as opposed to a ‘filer’. The Guide to Enactment to the Model Law explains that some terms may need to be adjusted to fit the local legal culture 112 See B5 above. 113 See Ghana D1 (2017 Bill); Sierra Leone C2; Zimbabwe C1; Zambia D1; Nigeria E1; Liberia C1; Rwanda C1; and also Malawi. 114 For legislation that uses ‘perfection’, see Liberia C3; Malawi B; Nigeria E2; Zimbabwe C1; and Zambia D4. 115 See ch 2, C. 116 ML, Art 2(o). 117 ML, Art 2(h). 118 See s 2 of Zambia’s Movable Property (Security Interest) Act 2016 and s 2 of Zimbabwe’s Movable Property Security Interest Act 2017 that refer to ‘debtor’ as the person that creates a security interest.

22 Introduction and drafting conventions.119 This has happened with a number of terms, especially the ones newly introduced by the Model Law, such as non-intermediated securities.120 Aside from the terminological difference, the meaning of the concepts and terms may also vary. In common with the Model Law,121 the core concept of ‘security right’ encompasses, in most African statutes, the rights of transferees of receivables,122 while in some that has not been the case.123 Likewise, with respect to long-term operating leases, most African laws do not reflect the approach of the Model Law, which does not apply to such leases.124 The reach of the concept of the security right then affects the meaning of the term grantor/ debtor, as it determines whether a transferee of receivables or a lessee under a long-term operating lease will be included therein. Chapters in this book, which relate to particular countries, refer to the terminology of the law as enacted in that country.

E. Legal Issues This section of the introductory chapter considers various key legal concepts to which reference is made in a number of the chapters of the book. The discussion here is reasonably generic: the issues arising in any particular African country are dealt with in the relevant chapter.

1. Common Law Concepts a. Bills of Sale Acts and Companies Act The bifurcated regime that developed under English law in relation to security interests created by companies and by individuals was adopted by many common law jurisdictions in Africa, and was often a driver towards reform.125 This section considers this regime very briefly. The Bills of Sale regime126 originated in English law in the nineteenth century, partly as a consumer protection measure and partly to provide a registration regime to avoid fraud. It built on the existing common law, which continued to govern any issue which was not dealt with in the Acts. The regime consisted of two Acts governing documents by which individuals transferred title to goods (bills of sale), whether as consumers or as unincorporated businesses. The transfer of title could be outright, as in an absolute bill, or a bill by way of security. Absolute bills are covered by the Bills of Sale Act 1878, while the Bills of Sale 119 ‘UNCITRAL Model Law on Secured Transactions: Guide to Enactment’ (Vienna, United Nations, 2017) (Guide to Enactment) 5. 120 See Kenya’s Movable Property Security Rights Act s 2 for the definition of electronic securities. 121 See ch 2, B2c. 122 See s 2 of Kenya’s Movable Property Security Rights Act 2017 and s 2 of Zambia’s Movable Property (Security Interest) Act 2016. 123 Nigeria E1. 124 For instance, see Sierra Leone C, on the exclusion of operating leases from the scope of the Borrowers and Lenders Act of 2014. Conversely, see Zambia D, on the applicability of Zambia’s Movable Property (Security Interest) Act 2016 to operating leases. 125 See Conclusion B2a. 126 See Nigeria C1; Zambia C3; Ghana C2.

Legal Issues 23 Act (1878) Amendment Act 1882 applies to security bills, incorporating some of the 1878 provisions (the 1882 Act). The 1882 Act will apply to any document that is in fact given as security, whatever its form. The 1882 Act did not attempt to cover all aspects of secured transactions: it only dealt with formalities for the creation of a valid bill, some restrictions as to which types of assets could be the subject of a security bill, registration and priorities. As a result of its consumer protection focus, the formalities for creation are very strict, and the penalty for failure to comply is severe. A security bill must be made in accordance with the precise form set out in Schedule 1 of the 1882 Act.127 Failure to comply with this renders the bill void even as between the parties.128 Only tangible goods can be the subject of a bill of sale, and these must be specifically described in the schedule to the bill. Future goods cannot be included. This means that floating charges cannot be given by an individual or an unincorporated business, and nor can fixed charges over future assets, which would otherwise be possible under the common law.129 Security bills of sale have to be registered,130 and priority is by date of registration.131 The same registration regime also applies in the United Kingdom to general assignments of book debts by unincorporated businesses.132 The Bills of Sale Acts regime does not affect the granting of security interests over assets other than goods: it is just that there is no registration regime for security interests over other assets133 and the common law applies. Moreover, the Bills of Sale Acts do not apply to devices based on retention of title, which developed in part as a means of avoiding the regime. The regime applicable to companies was also entirely based on common law, and nearly all the statutory regime related to registration of charges.134 The registration regime inherited by most common law African countries is discussed below. Other than registration, the common law governed, and was very liberal, permitting security to be taken over any asset, present or future, described generically in the charge agreement, including over the entire undertaking of the company.135

b. Fixed and Floating Charges One of the most important common law concepts, which has been a part of all African common law jurisdictions for many years, is the floating charge, which is a type of security interest that can only be granted by a company.136 The floating charge developed in English 127 One result of the restrictive drafting is that a bill of sale cannot be used by an individual to secure a guarantee of business debts. 128 S 9 1882 Act, although ‘literal conformity’ to the words of the Schedule is not required; Thomas v Kelly (1888) 13 HL 506, 516. 129 Holroyd v Marshall (1861–62) 10 HLC 191. 130 The register in the UK is very unsatisfactory, and is barely searchable; see Law Commission Report No 369: Bills of Sale (Law Commission, 2016) 2.24 et seq. 131 Bills of Sale Act 1878, s 10. 132 UK Insolvency Act, s 344. 133 In the UK, there are specialist registers for some types of intellectual property in which security interests can be registered, and several private registers for hire purchase agreements. The Agricultural Credits Act 1928 (ACA) also applies to unincorporated farmers and provides for registration by the Land Registrar of charges created pursuant to that Act (ACA, s 9). Mortgages over land are registrable in the Land Registry. 134 Now in UK Companies Act 2006, pt 25. 135 See D1b below. 136 See E1a above.

24 Introduction law in the later nineteenth century in the course of a series of cases concerning charges over the entire undertaking of companies, in which the courts found an express or implied power to dispose of the charged property free of the charge in the ordinary course of business of the company.137 There are two features of the floating charge which are critically important. The first is that the implied power to dispose enables a chargee to have a security interest in circulating assets, that is, assets which are either sold (for example the stock of a shop), made into another asset (eg raw materials used in manufacture) or consumed or destroyed in the ordinary course of business of the chargor (eg fuel such as coal or oil). On certain triggers, including an act of enforcement of the charge, the charge crystallises and becomes a fixed charge. This enables it to be enforced against the assets falling into the charge at the time of crystallisation (as well as any other future assets which fall within the charge).138 The second is that, since a charge could be taken over the entire undertaking of the company, the floating charge holder could appoint a receiver and manager139 who could take over the business, and either rescue the company by returning the business to profit or could sell the business as a going concern, thus capturing the extra value that comes from such a sale as opposed to a sale of the individual assets.140 These two features led to concerns that a floating charge holder was too powerful, since it could, effectively, take all the assets of the business on enforcement, leaving nothing for any other creditors. As a result, in 1897 the UK Parliament introduced preferential status, on enforcement by a receiver or on insolvency of the company, for employees of the company over the floating charge holder in relation to assets subject to a floating charge.141 Since then, the technique of giving priority in relation to floating charge assets to certain creditors seen as non-adjusting has been a continuous feature of English law, and therefore the law in African States based on the common law.142 The actual creditors entitled to this priority vary, but often include the tax authorities and employees. The expenses of the insolvency process are usually also paid out of floating charge assets before the claims of the floating chargee.143 It has long been the practice of lenders in common law jurisdictions to take fixed charges (or mortgages) over certain assets such as land or equipment, and for a ‘debenture’, that is, a loan to a company, to be secured by a mixture of fixed and floating charges as designated in the charge agreement.144 It is clearly in the interest of the secured lender to have a

137 Re Panama, New Zealand and Australian Royal Mail Company (1870) 5 ch App 318, 322, CA; Re General South American Company (1875) 2 ch D 337; Re Florence Land and Public Works Company (1878) 10 ch D 530, 546, CA; Re Hamilton’s Windsor Ironworks ex p Pitman & Edwards (1879) 12 Ch D 707; Wheatley v Silkstone and Haugh Moor Coal Co (1885) 29 ch D 715; Cox Moore v Peruvian Corporation Ltd [1908] 1 ch 604. 138 H Beale et al, The Law of Security and Title-Based Financing, 3rd edn (Oxford, Oxford University Press, 2018) 6.78–6.79. 139 Later, in English law, called an administrative receiver. 140 Of course, sometimes a ‘fire sale’ is inevitable, and this could also be effected by a receiver and manager. The ability to appoint a receiver and manager is not a feature of the enforcement provisions of the ML (see ch 2, H2). However, some African States with reformed law have retained this ability; see Zambia D7. 141 Preferential Payments in Bankruptcy Amendment Act 1897, s 107. 142 See the chapters in part II of this book. 143 For example, UK Insolvency Act 1986, sch B1 para 99(3) (expenses of administration); Insolvency Act 1986, s 176ZA (expenses of liquidation). 144 This practice is given legislative status in companies legislation in some African States; see, eg, Ghana C6 (discussing s 86(2) Companies Act 1963).

Legal Issues 25 fixed charge over as many assets as possible, since these assets would not be subject to the preferential claims mentioned above, and, at least under English law, considerable case-law grew up concerning the distinction between fixed and floating charges, as a result of lenders attempting to take fixed charges over all sorts of assets, including circulating ones.145 Put very simply, a fixed charge is a charge where the chargor cannot dispose of any of the charged assets without the consent of the chargee, while a floating charge is any charge that is not a fixed charge. The application of this test in the case-law is complicated. In reforming the secured transactions law of a common law jurisdiction, one major challenge is how to deal with the floating charge. The effects of a floating charge outside insolvency (or enforcement by a receiver and manager) are easily replicated in a reformed system. Under such a system, a security interest can be taken over present and future assets, and the assets need only be identified generically.146 Moreover, the secured creditor can authorise the disposal of encumbered assets (in advance or specifically) in which case the acquirer takes free of the security interest.147 In addition, buyers of goods in the ordinary course of business,148 and transferees of money and money equivalents (such as funds in a bank account or negotiable instruments) take free of security interests, provided that they do not know that the disposition is in breach of the security agreement.149 However, once a functional approach is adopted,150 the distinction between fixed and floating charges becomes legally irrelevant, and should no longer exist, since all security interests are subject to the same rules.151 This, however, leads to the problem that the distinction between fixed and floating charges needs to be replaced by a new test distinguishing between those security interests over whom the preferential creditors and the expenses have priority and those over whom they do not. The alternative is to reform insolvency law to remove all claims having priority over floating charge assets: this route could be considered if the relevant jurisdiction was also reforming insolvency law at the same time.152 In the absence of such change, the need for a distinction remains. This issue caused particular difficulties in the United Kingdom when reform to the English law of secured transactions was suggested by the Law Commission in 2005.153 Two jurisdictions have fairly recently dealt with this issue. The New Zealand PPSA 1999 introduced a provision limiting the priority of preferential creditors to security interests over receivables and inventory, but excluding purchase money security interests and transfers of receivables where new value is given.154 Although this formulation seems reasonably simple, it has also given rise to some difficulties, especially in relation to the definition of ‘receivable’.155 In Australia, references to floating charges in the 145 For discussion of the case law, which culminated in the decision of the House of Lords in re Spectrum Plus Ltd (2005) 2 AC 680, see Beale et al (n 138) 6.97–6.141. 146 See ch 2, B3. 147 See ML, Art 34(2) and (3). 148 See ch 2, F3. 149 ibid. 150 Ch 2, B2. 151 See A Duggan and D Brown, Australian Personal Property Securities Law, 2nd edn (Australia, LexisNexis Butterworths, 2016) 1.31–1.36. 152 See J Armour, ‘Should We Redistribute in Insolvency?’ in J Getzler and J Payne (eds), Company Charges: Spectrum and Beyond (Oxford, Oxford University Press, 2006) 189. 153 See Law Commission Report No 369: Company Security Interests (Law Commission, 2005) 3.165 et seq. 154 Companies Act 1993, sch 7 para 2. 155 See Commissioner of Inland Revenue v Northshore Taverns Ltd (in liq) (2008) 23 NZTC 22,074; Strategic Finance Limited v Bridgman (2013) NZCA 357.

26 Introduction relevant statutes have been replaced with references to ‘a circulating security interest’,156 that is, a security interest that has attached to a circulating asset.157 The term ‘circulating assets’ is defined by a list of possible types of assets158 but also includes the subject matter of a security interest where ‘the secured party has given the grantor express or implied authority for any transfer of the personal property to be made, in the ordinary course of the grantor’s business, free of the security interest’.159 The drafting is quite complex and there are numerous exceptions to these relatively simple rules and this approach has been the subject of some criticism.160 This particular issue raises both technical and policy considerations, and the optimal approach is not obvious.161 However, it is an issue that needs to be considered in every reform of the secured transactions law of a common law jurisdiction. Greater consideration has been given to this issue in some African jurisdictions than in others.162 There are typically other provisions referring to floating charges in the companies and insolvency legislation of a common law jurisdiction. One example is that, where the insolvency legislation has been reformed to include a corporate rescue regime similar to that of administration under UK law,163 a ‘qualifying floating charge holder’ can usually appoint an administrator out of court.164 It is relatively easy to replace this criterion with one compatible with a reformed secured transactions regime, since what is meant by a ‘qualifying floating charge holder’ is a secured creditor who has a security interest over all or substantially all of the assets of the company.165 If the jurisdiction does not include an administration regime, it is likely that a chargee with fixed and floating charges over all the undertaking of the company can appoint a receiver and manager.

c. Registration Regime It is necessary to discuss the registration regime that operated under English law until recently and which was adopted by most common law countries in Africa, since this is the regime against which the reforms in common law jurisdictions discussed in this book have taken place. This section will highlight particular aspects of the registration regime that applied under the 1948, 1985 and 2006 UK Companies Acts until it was reformed in 2013.

156 See Corporations Act, ss 51C, 433, and 561. 157 Australian PPSA, s 339(5). There are other references to floating charges in Australian statutes other than those dealing with the priority of preferential creditors and expenses, and this is the case in many common law jurisdictions. 158 Such as trade receivables, inventory, bank accounts; see Australian PPSA, s 340(5). 159 ibid s 340(1)(b). 160 See D Turner, ‘Fixed Charges over Receivables and the Personal Property Securities Act’ (2011) 19 Insolvency Law Journal 71; and the five-year review of the Australian PPSA in which it was recommended that the provisions on ‘circulating assets’ and ‘control’ be replaced with simpler wording similar to the New Zealand provisions. B Whittaker, Review of the Personal Property Securities Act: Final Report (2015) 9.2.1. No changes, however, have yet been made. 161 For a discussion of the wider implications, see S Paterson, ‘The Insolvency Consequences of the Abolition of the Fixed/Floating Charge Distinction’ (Secured Transactions Law Reform Project, January 2017) https://stlrp.files. wordpress.com/2017/01/paterson-fixed-and-floating-charges.pdf. 162 See Conclusion C6bii. 163 See UK insolvency Act, sch B1. 164 UK Insolvency Act, sch B1 para 14. 165 ibid.

Legal Issues 27 Under this regime, registration was not of a notice, but of a security interest (or interests). Only certain interests were registrable: there was a list166 including charges over land, charges over tangible assets (which were described as charges created or evidenced by an instrument which, if executed by an individual, would require registration as a bill of sale),167 charges over book debts (receivables), floating charges, charges over ships and aircraft, and charges over goodwill and intellectual property. The main exceptions were charges over securities and (probably) charges over bank accounts. Only non-possessory security interests (mortgages and charges) were registrable, and this had to be done within 21 days of creation:168 there was no registration in advance. This gave rise to a period of invisibility. If the charge was not registered for 21 days, during this period searchers would not see the registration. For this reason, advances were often not made until 21 days after creation of the charge, and after the lender had executed a clear search.169 If a charge was not registered within the requisite time period, it was void against other secured creditors and any insolvency officer.170 However, if the lack of registration was inadvertent the chargee could apply for leave of the court to register late, subject to the rights of any person who had acquired an interest in the relevant assets in the intervening period.171 The obligation to register the charge was that of the company, which stemmed from the fact that registration was originally merely in the register held by the company rather than a public register.172 Although, for many years, registration had been carried out by the secured creditor, the legislation still reflected the company’s obligation, so that a failure to register on the part of the company was a criminal offence.173 Moreover, if a charge was not registered within the required period, the obligation secured became immediately due and payable.174 This was not only a sanction on the company, it also provided the company with an incentive to create a new charge in favour of the secured creditor, which might have been an easier route to repair a failure to register in time than to ask for the leave of the court. Many of these elements remained within the companies legislation of common law African States, and some still remain post-reform.175

2. Civil Law Concepts The secured transactions legal framework of African countries belonging to the civil law tradition is fragmented and formalistic.176 The fragmentation may be partially attributed to the numerus clausus principle under which only certain rights are characterised as security rights (for example pledges), whereas others are enforced according to the specific set 166 UK Companies Act 2006, s 860 (now repealed). 167 See D1a above. 168 UK Companies Act 2006, s 870 (now repealed). 169 Beale et al (n 136). 170 UK Companies Act 2006, s 874 (now s 859H). 171 ibid s 873 (now s 859F). 172 UK Companies Act 1862. 173 UK Companies Act 2006, s 860(4) (now repealed). 174 ibid s 874(3) (now s 859H(4)). 175 The provision in the Kenyan Companies Act stating that failure on the part of the company to register is a criminal offence (s 878(8)) has not been repealed or amended as a result of the MPSR; see Kenya B4. 176 Tunisia C1.

28 Introduction of rules governing that particular transaction, irrespective of its function.177 Even the recently enacted OHADA Uniform Act takes a formalistic and individualist approach to the various security rights, providing different sets of rules for their creation, third-party effectiveness, priority and enforcement.178 The four regimes examined in this book have been largely inspired by French law.179 The frameworks include codes as well as specific legislation to provide for the use of security devices in specific contexts and transactions, such as for warehouse receipts or leases.180 Some security devices may be available only to particular types of creditors.181 The core security device is the possessory pledge, typically recognised in the civil code.182 There is still a valid pledge if the pledged collateral is possessed by the secured creditor constructively, such as when it is in possession of a negotiable document (for example, a warehouse receipt) or has appointed an agent to manage the collateral.183 A pledge of intangible assets requires a notification of the person who owes the payment or other performance, such as the debtor of the receivable or the bank that maintains the bank account subject to a pledge.184 Non-possessory security devices are limited to certain types of transactions and assets, such as pledges over motor vehicles or an enterprise.185 In addition, these frameworks recognise a number of devices where the creditor retains or acquires ownership of the asset while the grantor retains possession or some form of control. The enterprise pledge (mortgage of a business),186 which may be viewed as analogous to the common law floating charge or the Model Law ‘all assets’ security right, has a number of limitations, such as its inapplicability to inventory as well as most intangible assets,187 and the need to describe the assets of the enterprise specifically, which make it a much less effective and efficient security device.188 Creation of a security right typically requires the satisfaction of a number of formalities, such as the notarisation of signatures of the parties to a pledge agreement. Often, additional requirements apply, such as the recording of an agreement with a tax authority.189 Furthermore, claims may not be assigned when the agreement under which they arise contains an anti-assignment clause.190 Third-party effectiveness is achieved by registration only for a few categories of security rights, though the trend has been to gradually subject more security devices to registration for their third-party effectiveness.191 But the approach is not uniform. For instance, some countries require registration of financial leases for some purposes, such 177 See OHADA B, which explains why the Uniform Act does not apply to finance leases. 178 ibid. 179 Ethiopia B. 180 ibid C. 181 Under the OHADA Uniform Act, security transfers of receivables and funds may be taken only by entities conducting banking or credit operations in the regular course of business. OHADA B3. 182 Burundi B1 and Ethiopia C. 183 Ethiopia D1. 184 ibid. 185 Burundi B1 and Tunisia D2. 186 Ethiopia C. 187 Typically, an enterprise pledge covers only intellectual property rights, logo, and goodwill. See Tunisia C 2. 188 Burundi B2. 189 Tunisia D. 190 Ethiopia D2. 191 See OHADA C explaining the approach to retention of title.

Legal Issues 29 as the protection against buyers192 only, while in others, the failure to register renders the lessor’s right ineffective against all third parties.193 It is not uncommon for a law to require the registration of a particular right, but no registry has in fact been established for that very purpose.194 Civil law countries have not established general security rights registries, though the OHADA Uniform Act is a step in that direction.195 Furthermore, where available, registries require the submission of the agreement that creates the particular security right for examination by the registrar, as well as its recording. Notice registration is unknown in unreformed civil law jurisdictions in Africa.196 The registries process agreements submitted in a physical form, which creates inefficiencies. On top of that, registrations must be submitted to a country-based registry office where the collateral or the grantor may be located.197 This may require multiple registrations in different countries where the collateral may be located, as well as monitoring of the movement of the grantor and the collateral.198 The failure to register renders the security right ineffective not only against third parties, but also as against the grantor.199 The absence of a coherent legal framework creates a potential for complicated priority disputes for which the existing rules do not provide any answers, thus increasing the legal complexity.200 The rules are not set up to deal with conflicts arising when a single asset is encumbered under two or more different security devices.201 This is particularly the case for motor vehicles that may be pledged, sold under a retention of ownership, acquired under a finance lease or form a part of a larger collateral package supporting an enterprise pledge, or equipment may be encumbered under five distinct devices.202 The laws also have many gaps, such as the legal effect of movement of the assets, subject to a pledge of creditor A, to another country and the taking of a pledge in the same assets by creditor B.203 For the competing rights in the same asset established under the same law, and for which a registry has been established, generally the time of registration establishes the priority.204 However, for those where no registry exists, the time of creation may be determinative.205 Civil law jurisdictions have recognised extrajudicial enforcement in some narrow contexts (for example in some jurisdictions they are only available to banks),206 but lack a general recognition of such remedies.207 In some instances, exceptions have been made for particular transactions, such as pledges of bank accounts.208 The lack of such remedies is compensated by the availability of expeditious judicial remedies, including those that may

192 Ethiopia

D5. D4. 194 Ethiopia D4. 195 See OHADA. 196 Tunisia C2. 197 Ethiopia C and Tunisia C2. 198 Tunisia D3. 199 Burundi B2. 200 Ethiopia C. 201 OHADA I. 202 Tunisia C1. 203 ibid E. 204 OHADA E and Tunisia E3. 205 Tunisia E. 206 Ethiopia C. 207 Burundi B4 and Tunisia F. 208 Tunisia F. 193 Tunisia

30 Introduction be granted provisionally.209 The absence of the functional approach requires the application of a remedy specific to the security right used to secure the obligation. It remains to be seen whether these challenges can be successfully overcome, as the implementation of reforms in civil law countries in Africa has languished. Ethiopia, Burundi and the OHADA Member States have enacted laws, but are yet to establish centralised collateral registries.

3. Concepts in Mixed Jurisdictions210 The secured transactions legal framework of African countries with a legal system rooted in Roman–Dutch law, namely South Africa and Zimbabwe, adapted by the courts and legislature, which often drew on English law, relies on a variety of legal devices to secure obligations. The absence of a coherent and functionally oriented framework increases legal complexity and the cost of structuring many types of secured transactions, but especially those secured with fluctuating assets such as inventory. The four principal devices used to secure obligations with a right in movable assets are: (i) pledge, (ii) cession (assignment), (iii) general notarial bond and (iv) special notarial bond.211 In addition to these generally applicable devices, laws have been enacted to facilitate various retention of title arrangements, including under hire-purchase agreements and leases.212 Furthermore, the countries enacted specific legislation to support financing in certain industries or particular groups, such as for agriculture and cooperative societies.213 Tangible assets can be used as collateral for pledges, general notarial bonds and special notarial bonds, but each of these devices limits their usefulness in the modern business environment. Pledges require the pledgor to transfer possession or legal control of movable assets, tangible or intangible, to the secured creditor (pledgee).214 Constructive possession is recognised.215 A pledge over intangibles is generally made by means of cession that becomes effective when executed, and without any form of a public notice.216 The execution of a notarial bond must be attested by a notary public, and the bond must then be registered in a local deeds registry.217 In addition to notarisation, a stamp duty is payable on certain agreements.218 The situation is clouded by the uncertainty with respect to the possibility to create security rights in future assets, and the override of anti-assignment restrictions.219 The effectiveness of a notarial bond is limited, both in the types of collateral it may cover and in the protections and priority it gives the secured creditor. The special notarial bond is 209 Burundi B4. 210 Note that Rwanda is also included in the part of this book entitled ‘Mixed Jurisdictions’ but the law there is not based on Roman–Dutch law. Rather, it is based on Belgian law but is transitioning to common law. 211 South Africa C and Zimbabwe B1. 212 South Africa C1 and Zimbabwe B1. 213 Zimbabwe B1. 214 South Africa C2a. 215 ibid C2b. 216 ibid C3a. 217 ibid C4b; Zimbabwe B1. 218 Zimbabwe B2. 219 South Africa C3c–d. For further discussion of the merits of an anti-assignment override, see E4 below.

Legal Issues 31 a security interest created by statute, and is deemed to be a non-possessory pledge.220 It is limited to tangible assets that are specifically described so that they are ‘readily recognisable’ in a manner that enables third parties to clearly identify which assets of the grantor remain unencumbered.221 Thus a special notarial bond cannot be used to encumber fluctuating collateral. A general notarial bond, inspired by common law, can be granted over assets of the grantor described generally and can include future assets.222 However, it does not confer on the bondholder a real security right; instead the bondholder merely has a preference over unsecured creditors in insolvency.223 The grantor can dispose of its assets freely until its insolvency or until the secured creditor takes possession of assets and therefore becomes a pledgee.224 Moreover, the general notarial bond must cover all of the grantor’s movable property, both present and future; it cannot be limited to some portion of the grantor’s movable property.225 Retention of title devices are effective without any form of a public notice. Registration entails the submission of actual copies of agreements, and this must be done within a prescribed period of time after their execution.226 Digitisation of the registries in which bonds are registered would not solve the underlying deficiencies in their legal nature. Roman–Dutch law jurisdictions generally do not recognise extrajudicial enforcement, even with respect to pledges of intangible assets that do not raise any risk of a potential conflict with the grantor.227 However, certain exceptions have been recognised by the courts.228

4. Common Barriers to Accessing Secured Credit a. Anti-Assignment Clauses Agreements which give rise to receivables often include a term prohibiting or limiting the transfer of the benefit of the receivables, and the creation of a security right in those receivables (‘anti-assignment clause’). To the extent that these terms are intentionally included, and not just as a result of boilerplate drafting, they have several possible purposes. One is to ensure that the receivables debtor only ever deals with the receivables creditor, and not with a third-party transferee or secured creditor. Another is to eliminate the risk of paying the wrong party, if, for example, a notice of assignment is overlooked, with the result that payment may need to be made twice.229 A third is to eliminate the risk of not being able to assert a set-off against the receivables creditor, since transfer or the creation of a security right can prevent the assertion of set-off in some situations, depending on the applicable law.

220 South

Africa C5. C5. 222 ibid C4c. 223 ibid. 224 ibid. 225 ibid. 226 Zimbabwe B2. 227 South Africa B and Zimbabwe B2. 228 South Africa C3e. 229 This would depend on the applicable law. 221 ibid

32 Introduction The effect of an anti-assignment clause will vary according to the applicable law, but generally it will not only mean that the creation of a security right in the receivables (including an outright transfer) is a breach of contract by the receivables creditor, but that there will be adverse consequences for the secured creditor (or transferee). First, it may not acquire a right effective against third parties (including an insolvency officer) to the receivables. Secondly, it may be difficult or impossible for it to enforce against the receivables, especially by suing the receivables debtor. The likelihood of these adverse consequences is such that those financing against receivables (whether they use a security right or a transfer to protect their position) either refuse to finance against receivables containing an anti-assignment clause or have to use expensive workarounds, including obtaining waivers from receivables debtors in relation to the anti-assignment clauses, which increases the cost of the financing. Another concern (although perhaps overblown in the context of trade financing) is that financiers would have to examine all the contracts giving rise to the receivables to discover whether they contain anti-assignment clauses.230 Further, anti-assignment clauses are seen as interfering with the alienability of assets, thus preventing the maximisation of available collateral.231 For all these reasons, many modern secured transactions regimes, as well as international instruments such as the UNCITRAL Convention on the Assignment of Receivables in International Trade and the Model Law, include a provision overriding anti-assignment clauses. These provisions vary a little in effect: some render the clause entirely void,232 and others provide that the receivables debtor can still obtain a remedy for breach of contract, but not the remedy of termination of the contract itself.233 The scope of the override also varies, but most regimes limit it to trade receivables because different considerations apply to receivables owed to financial institutions.234

b. Taking Security Over After-Acquired Assets There are a number of good reasons why a secured transactions regime should permit and, indeed, facilitate secured creditors to include assets in which the grantor does not yet have an interest235 within the security agreement. First, it reduces costs by obviating the need to enter into a new security agreement every time the grantor acquires a new asset. Secondly, it facilitates the taking of security over circulating assets, such as inventory and receivables, which are constantly being acquired, as well as constantly being disposed of. Thirdly, and

230 One of the authors conducted empirical research in the UK, which showed: (i) that receivables financiers always look at the contracts in any event, not just to check for anti-assignment clauses but to look for all sorts of contractual provisions; and (ii) many of the contracts, especially those used by large customers, are in standard form so as long as the customer’s identity is known the terms are also known. See L Gullifer, H Beale and S Paterson, ‘A Case for Interfering with Freedom of Contract? An Empirically-Informed Study of Bans on Assignment’ (2016) 3 Journal of Business Law 203, 219. 231 G Gilmore, Security Interests in Personal Property, vol 1 (Boston, MA, Little, Brown, 1965) 212. 232 UCC §9-406(d); UK Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254) reg 2(1). 233 See, eg, Australian PPSA, s 81(2); New Zealand PPSA, s 87; and ML, Art 13. 234 LG, ch II para 108; ML, Art 13(3); UK Business Contract Terms (Assignment of Receivables) Regulations 2018 (SI 2018/1254) regs 1(3) and 4 (the UK override only applies to trade receivables owed to SMEs). 235 A grantor may not yet have have an interest in assets either because they do not yet exist, or because it has not yet acquired them.

The Shape of the Book 33 as an extension of the second point, it enables security to be taken over all the assets of the grantor.236 Fourthly, it simplifies the application of priority rules, since, instead of priority depending on when the asset was acquired by the grantor, it can be determined by reference to a single date, such as the date of the security agreement237 or, under a reformed system of secured transactions law, from the date of registration. In many unreformed jurisdictions, it is difficult or impossible to take security over future assets, at least in some circumstances. Examples include the common law Bills of Sale Acts, which prohibit the granting of security by individuals over future goods238 (although it has been possible since the nineteenth century in common law regimes for a company to be able to grant security over its future assets, and for an individual to be able to grant security over future assets other than goods), and the civil law requirement that the collateral be described specifically.239 Modern secured transactions regimes all include provisions enabling a secured creditor to take security over after-acquired assets,240 and this is replicated in reformed regimes in Africa.

F. The Shape of the Book This book is divided into five parts. The first part is introductory, and comprises this chapter and the chapter in which the principles of a modern secured transaction law, with particular reference to the Model Law, are discussed. The next three parts contain countryspecific chapters. Each discusses the current and, in reformed jurisdictions, the previous law of secured transactions. Discussion of reformed law is generally divided into the various components of such law, namely, creation, third-party effectiveness, registration, priorities, enforcement, conflict of laws, and transitional provisions, although the structure of each chapter is dictated by the legal position in the relevant jurisdiction. This paragraph describes the country-specific parts of the book. Part II focuses on common law jurisdictions, and charts the reform initiatives in seven of these. One, Malawi, was the subject of a chapter in the first book in this series, and so the chapter in this book considers developments in that jurisdiction since that first chapter was written. Part III focuses on civil law jurisdictions which have reformed their law or in which reform is actively being considered. One chapter considers the OHADA Uniform Act, which applies in 17 African civil law jurisdictions. Part IV focuses on mixed jurisdictions: the two Roman–Dutch/common law jurisdictions of South Africa and Zimbabwe, and the Belgian/ common law jurisdiction of Rwanda. Part V contains an important chapter on the current and likely future effect of the Cape Town Convention and its Protocols on equipment financing in Africa. The Convention and the Aircraft Protocol are widely ratified in Africa, and the Luxembourg Rail Protocol and future Mining, Agricultural and Construction MAC Protocol have the potential to greatly increase investment in the African rail sector and mining, agriculture and construction

236 See

E1b for discussion of some of the advantages of this. is the case under English law. See Beale et al (n 136). 238 See E1a above. 239 See E2 above. 240 See, eg, ML, Art 7; ch 2, B3. 237 This

34 Introduction industries, were they to be widely adopted. The second chapter in Part V moves away from a legal analysis to consider what infrastructure is needed to underpin a modern secured transactions law. The third chapter concludes the book by providing a thematic analysis of the information and discussion in all the other chapters, and reflects on the drivers for reform, the reform process itself, and the substance of reform, drawing on examples from all the jurisdictions considered in this book.

2 A Modern Secured Transactions Law A. Introduction This chapter addresses modern secured transactions law on three levels. First, it considers what should be the basic principles of any modern commercial law. It will then consider how these basic principles are translated into the more detailed principles which are typical of modern secured transactions legal regimes. There is a reasonable amount of international consensus on these principles, and they are exemplified in a number of international instruments,1 as well as in many national laws. This chapter makes particular reference to the Legislative Guide2 and its even more concrete offspring, the Model Law,3 including the Model Registry Provisions. The third, and longest, part of the chapter discusses the modern principles in more detail, with specific references to the Model Law. It should be pointed out that the discussion in this chapter relates primarily to transactions that take place by agreement between the parties. Most States also have rules about rights in an asset (proprietary, or which, at least, have priority in the insolvency of the owner of the asset), which can arise by operation of law to secure obligations in various circumstances. The modern principles discussed do not necessarily apply to such rights, although they are often included in a secured transactions regime for registration and priority purposes.4

1. Basic Principles of Any Modern Commercial Law There are various principles that should act as a yardstick for assessing a State’s commercial law and which should inform reform. These principles may seem obvious, but they are very difficult to achieve in practice. However, they are necessary to keep in mind when arriving at the more granular principles that apply to modern secured transactions law. They are, to a large extent, the principles which inform the modern principles of secured transactions law, and, in this chapter, are called the ‘basic principles’. First, the law should be as simple, accessible and understandable as possible. Secondly, complexity and fragmentation should be avoided wherever possible. Thirdly, and as a result of the first two principles, it should probably be codified. This principle is obvious for a system which is already codified, such as a civil law system. However, many common law systems traditionally have only used statute law where necessary, and many principles and

1 See

ch 1, B5.

2 ibid. 3 ibid. 4 See

Conclusion D1b.

36 A Modern Secured Transactions Law rules have been developed through case-law. Codification, though, is the best way of making the law as accessible as possible, and also of avoiding fragmentation. Fourthly, it should be as cheap and easy to operate as possible. This is a qualified principle, since cheapness and ease of use should not come at the expense of effectiveness, nor of protection of third parties and weaker parties, and the appropriate balancing of interests. Fortunately, developments in information technology have meant that costs can be reduced and processes made more efficient without losing other important aspects of those processes. Fifthly, the law should be flexible enough to do what it is designed to do in as many circumstances as possible. In many cases, this will involve a large degree of party autonomy, exercised against the backdrop of default rules set out in the legislation. Such a scheme maximises the chances of combining legal predictability, crucial for commercial transactions, with the necessary flexibility to meet practical business requirements.

2. Modern Principles of Secured Transactions Law These basic, and rather general, principles, are fleshed out in the context of secured transactions into modern principles which inform the various international instruments mentioned above (‘modern principles’). Before dealing with the principles relating to particular aspects of secured transactions law, it is worth mentioning that there are two general principles that permeate the whole law. The first is that, as far as is consistent with the operation of the other principles, parties are free to agree to terms that meet their precise commercial needs: this is known as the principle of party autonomy.5 The other is that parties should act at all stages of the secured transaction in accordance with good faith and commercial reasonableness.6 This principle is particularly important in relation to enforcement of security rights, when the actions of the enforcing party can be detrimental to the grantor or third parties.7

a. Scope of the Law The first group of these principles relates to the scope of the law. First, there should be one code to cover all creditors and all grantors. This means that one law covers all grantors, whatever type of person, natural or legal, they are.8 The same rules should cover corporate grantors, non-corporate businesses and consumers, although the latter may require some special rules to protect them, and there may be other limited exceptions. Further, the same rules should apply to all creditors. This contributes to the simplicity and accessibility of the law, and reduces complexity. Secondly, to the maximum extent possible, there should be one set of rules which covers all transactions where a lender or other extender of credit takes a proprietary interest in assets to which it can have recourse if not paid (‘secured transactions’). This is called the functional approach.9 This means that all types of interests that have been treated as

5 LG,

ch I paras 115–18; ML, Art 3. Art 4. 7 LG, ch VIII para 15. 8 LG, Introduction para 61. 9 LG, Introduction para 62, ch I paras 110–12. 6 ML,

Introduction 37 security rights in a State, which until reform may have been subject to different rules and requirements, would, under a modern system, be governed by the same set of rules.10 It also means that devices, such as those based on the creditor’s retention of title, are subject to the same rules as devices based on the grant of an interest by the grantor.11 These rules include the very important principle that a security right only extends as far as is necessary to secure the relevant obligation. Thus, if a junior security right is created in the asset, any surplus value in the collateral is available for the satisfaction of that security right. On enforcement, the surplus value not required to fulfil secured obligations is to be returned to the grantor.12 The functional approach can be extended further, so that the same rules also apply to other transactions which have many similarities to secured transactions, but which do not secure obligations. Thus, outright transfers of receivables usually are included in the regime, and, sometimes, long-term operating leases. In these cases, the rules about the extent of the right being limited to the amount of secured obligation do not apply because the transferees enforce as owners. Thirdly, it should be possible to take security over any type of asset or class of asset. This principle has a number of ramifications. One is that the rules as to identification of collateral need to be very broad, so that there is no need to identify specific assets.13 Another is that it should be possible to take security over future as well as present assets, and to take security over all the assets of the grantor.14 Further, the rules for all types of assets should be similar, as far as this is possible to achieve.

b. Creation, Third-Party Effectiveness and Priority The second group of modern principles relate to specific areas of secured transactions law, and reflect the need for balance between protection of third parties and ease and cheapness of operation. i. Creation15 First, it should be possible to create a security right with as little formality as possible.16 Some rules are required in order to provide certainty, both of the existence of the property interest and its subject matter. The modern principles draw a distinction between creation of a security right, which addresses its validity between the parties to the security agreement, and third-party effectiveness (which is called ‘perfection’ in some systems).17 The requirements for creation are minimal, and are directed at, first, providing some evidence of the existence of an agreement to grant security (the requirement of writing), and secondly, 10 There will be some exceptions to these rules, but they should be rationally justifiable and as limited as possible, taking into account the commercial requirements of the system and the interrelationship with other law. 11 Although devices based on retention of title are likely to create acquisition security rights, which are subject to a different priority regime. See F2 below. 12 See H below. 13 See C1 below. 14 See ch 1, E1b and E4b. 15 Note that the concept of ‘attachment’ is used in some systems instead of ‘creation’ though there are some differences; see C below. 16 See C1 below for more detailed discussion. 17 LG, chs II and III. See ch 1D on variations in terminology.

38 A Modern Secured Transactions Law very basic requirements as to the elements of the creation, such as identifying the grantor, the secured creditor, the assets to be given as security and the secured obligation; both these requirements can be described very generally. ii. Third-Party Effectiveness Secondly, the modern principles address the protection of third parties. Unlike some systems, where protection of the grantor’s creditors and other third parties is achieved, at least in part, by formal means such as the existence of a numerus clausus of property interests,18 requiring specification of encumbered assets19 or controlling over-collateralisation,20 the modern principles focus on publicity as the chief method of third-party protection, on the basis that if security rights are publicised, all third parties can adjust accordingly.21 It is a basic tenet of the modern principles that all security rights should be discoverable from a reliable source of information, and that the best way of achieving this is by registration. However, third parties who themselves take a security right in the relevant assets are only protected by registration if the rules provide that it is necessary to make a security right effective against third parties, and if the priority rules are based on a ‘first to register’ principle. If another priority rule applied (such as ‘first to create’), a third party who searched the register and found nothing could still lose priority to an earlier, hidden, security right. The link between the requirement of registration and the priority rules is critical to the operation of the modern principles, as is the rule that registration (or another form of publicity) is generally required to make a security right effective against third parties.22 Consonant with the basic principles, registration should be as cheap, easy, and efficient as possible. The modern principles achieve this by a number of means. First, rather than detailed information about a particular transaction, all that is needed to be registered is a notice, containing minimal information. Secondly, that notice can be registered at any time, including in advance of the transaction actually taking place, to enable a secured creditor to fix its priority before committing itself. Thirdly, registration and searching is primarily by grantor and not by asset:23 this is essential if encumbered assets can be identified by general description. Fourthly, the registration system should be fully electronic, with registration and searching taking place with no human intervention. This is, of course, a more recent addition to the modern principles,24 but is now seen as standard. There are many matters of relative detail in relation to making security rights effective against third parties which are open to debate and which vary among systems based on the modern principles. These are discussed later in this chapter.25

18 LG, Introduction para 70. 19 LG, ch II paras 58–60. 20 LG, ch II paras 6–70. 21 Creditors or potential creditors who are unable to adjust can, and often are, protected by insolvency law or by other legal regimes, such as compulsory insurance. 22 See D below. 23 Although this could be included in some circumstances; see E2 below. 24 For example, it is not included as a requirement in the LG or the ML. Electronic registration has, for example, been the case in Canada for over two decades; see ‘Personal Property Registry System’ (ACOL) www.acol.ca/en/ pprs/about/what-is-acol. 25 See D below.

Introduction 39 iii. Priority Thirdly, the modern principles recommend that there should be a set of clear priority rules based on rational distinctions. The rules on priority between security rights, and between security rights and other proprietary interests, are, in many systems, complex and fragmented, usually as a result of a fragmented system. The aim of the modern principles is to make the rules as simple as possible, but because these rules, above all, relate to the balancing of interests between different parties, they cannot be all that simple. Therefore, the modern principles require that where distinctions, and exceptions to the general rules, are made, they should be rational and justifiable. The basis of the priority rules, as explained above, is that priority is governed by the time of registration, but this is subject to a number of exceptions, which are discussed below.26 One is so important, however, that it is itself a principle, namely, that the ordinary operation of commerce should be facilitated by rules which provide that certain parties acquiring an interest in assets in the ordinary course of business take free of security rights.27

c. Enforcement Effective enforcement rules are critical for the economic benefits that a modern secured transactions law is likely to bring, since finance providers are more likely to lend or provide credit if they know they can enforce their security rights quickly and efficiently. Therefore, the modern principles recommend that there should be effective enforcement mechanisms, which enable secured creditors to enforce quickly, easily and cheaply, but which balance protection for the grantor and other creditors. The latter qualification is important because enforcement can potentially be detrimental to the grantor and its other creditors in two ways. First, the mere fact of enforcement can damage a grantor, and so it is important that rights to enforce only arise on default in relation to the secured obligation. Secondly, since any surplus value must go to other secured creditors or the grantor, these parties will be adversely affected if the enforcing creditor realises the asset for a sum below its reasonable value. The need for quick, easy and cheap enforcement means that the modern principles require that out-of-court enforcement is possible, as well as the possibility of expeditious judicial proceedings.28 Although one way to protect other parties is for a court to control the enforcement procedure, in out-of-court enforcement the balancing of interests required to protect other parties is dealt with by detailed rules requiring notice, enabling other parties to object or play a part in enforcement, and the overarching requirement that a secured creditor act in good faith and in accordance with commercial reasonableness.29

d. Other Matters Consonant with the basic principle that the law should be accessible and understandable, and the realisation of this in the principle of codification, the modern principles require that other matters should be included in a secured transactions law, such as provisions dealing with the

26 See

F below. a more detailed discussion, see F3 below. 28 LG, ch VIII paras 18–20. 29 LG, ch VIII; ML, ch VII. 27 For

40 A Modern Secured Transactions Law conflict of laws rules and transitional provisions. The importance of the former stems from the fact that many secured transactions involve a cross-border element, and stable conflict of laws rules provide certainty for the parties involved.30 The latter focus on the validity and priority position of security rights created under the previous regime, and are critical to the smooth introduction of a regime based on the modern principles.31 Some regimes, including the Model Law, also include more detailed provisions on the rights and obligations of the parties to a security agreement and obligors of receivables. One of the valuable aspects of the Model Law is that it sets out a logical structure for a secured transactions statute, which includes all the relevant aspects required by the modern principles, but does not include matters which are not relevant. The inclusion in the statute of other matters, such as regulatory rules or rules providing protection to weaker parties, can be confusing and is usually inappropriate.

B. Scope of Secured Transactions Law This part, and the following parts, of this chapter consider how the modern principles apply when translated into a statute creating a legal system for secured transactions. The paradigm used for discussion is the Model Law, although some aspects not included in that law are also discussed. As mentioned above, the modern principles recommend that there should be one set of rules for all types of secured transactions, between all types of parties, and in relation to all assets. Each of these aspects will now be discussed. In what follows in this chapter, and, unless the context demands a different treatment, in this book, it is assumed that the person who owes the secured obligation is the person who grants the security right.32 In theory, this need not be the case, although it very often is. However, it is simpler just to use the one word: ‘grantor’.

1. Parties Many States have separate regimes, or at least, separate statutes, for secured transactions entered into by different grantors. For example, regimes based on English common law are likely to have a statutory regime governing registration of charges created by companies, often included within a much larger Companies Act, and a separate regime based on the Bills of Sale Acts of the nineteenth century33 covering security rights created in tangible movable property by individuals, both as consumers and in the course of business, with other parts of the law being entirely based on common law. The Model Law regime applies to all types of grantors. It applies to all types of businesses, including those that are incorporated and those that are not. Application to the latter category is particularly important in Africa, where MSMEs make up by far the largest proportion of businesses, and where most of the micro and small enterprises are not incorporated. The regime also applies to 30 See I below. 31 See J below. 32 As discussed later, a security right can also arise by reservation of title by the secured creditor. The term ‘grantor’, however, covers both situations. 33 See ch 1, E1a.

Scope of Secured Transactions Law 41 consumers, bearing in mind that there can be other statutes or common law principles designed to protect individuals acting outside the course of business.34 The Model Law also applies to all types of secured creditors. No distinction is made between lenders who are banks and other regulated financial institutions, and other, nonregulated, lenders. Moreover, the functional approach35 has the result that the category of ‘secured creditors’ includes those who extend credit in ways other than a straight loan. Leasing companies, who provide equipment on finance lease or hire purchase terms, are included. So are suppliers of goods on retention-of-title terms, who can either be financiers (who buy goods from suppliers and sell them on conditional sale terms) or suppliers themselves, that is, those who manufacture or sell goods in the course of business, or even people selling goods in one-off sales outside the course of business.36 The inclusion of outright transfers of receivables also means that receivables financiers (sometimes known as ‘factors’) are also included in the regime as ‘secured creditors’.

2. Functional Approach: Types of Secured Transactions a. Single Concept of Security Right This approach is a core principle of modern secured transactions law. It has several aspects. First, the nature of the proprietary interest created in the course of a secured transaction is the same for all types of transaction. This interest is called a ‘security right’ in the Model Law.37 Conceptual differences that exist in other regimes, such as, in the common law, the distinction between a mortgage and a charge, or between the fixed and the floating charge, disappear. The difference between possessory38 and non-possessory security rights also disappears at the conceptual level, so that different consequences39 also do not apply. The taking of possession by the secured creditor is seen just as a means of making the security right effective against third parties.40

b. Retention of Title Devices The types of interests discussed in the previous paragraph are all similar in two respects. First, they are created by a grantor granting a right over its own assets. Secondly, they only extend as far as is necessary to secure the relevant obligation. Perhaps a more contentious application of the functional approach is that it applies the same regime to interests created by the reservation of title by the creditor, and which, at least in theory, are absolute and are not limited by the amount of any obligation that they functionally secure. Typical examples 34 ML, Art 1(5). 35 Discussed in B2 below. 36 An example of this last category would be a sale of used farming equipment between two farmers. 37 ML, Art 2(kk). 38 Such as a pledge. 39 For example, in common law jurisdictions, mortgagees (and chargees if in the charge document) are entitled to appoint a receiver whereas a pledgee or lienee cannot, and a mortgagee can foreclose, whereas a chargee, a pledgee and a lienee cannot. In a mixed jurisdiction, such as South Africa, extrajudicial enforcement is not possible for a non-possessory security right. 40 ML, Art 18(2).

42 A Modern Secured Transactions Law of transactions creating such interests are sales on terms that title is retained until the price is paid, hire purchase agreements and finance leases. These transactions can take place in two commercial contexts. First, in the acquisition of equipment (often called ‘asset finance’).41 Here, the creditor gives possession of the equipment to the grantor, who pays a sum equivalent to the value of the equipment plus a financing charge, in instalments over a period of time. The creditor continues to own the equipment either until the sum is paid (conditional sale) or until the grantor exercises an option to buy the equipment for a nominal price (hire purchase or finance lease).42 Secondly, in the acquisition of, typically, inventory (sale on retention of title terms). Here, the creditor agrees to sell the goods to the grantor on terms that the price is payable after a period of time, for example, 60 days. The contract provides that property will not pass until the price is paid.43 The Model Law regime treats all these transactions as creating a security right. Thus, the regime’s rules about creation and third-party effectiveness (including registration) apply. The extent of the security right is only as far as is required to secure the relevant payment obligation. One effect of this, as mentioned above, is that if a junior security right is created in the asset, any surplus value in the collateral is available for the satisfaction of that security right. In a regime where retention of title devices are not treated as security rights, the grant of a security right to be satisfied by surplus value must take the form of a security right in the intangible right of the buyer or lessee under the contract of sale or lease.44 Another effect is that, on enforcement, any surplus value realised must be paid to any junior secured creditor or other claimant or, failing this, to the grantor. It is the existence of these two features that has led some commentators to say that regimes based on the modern principles recharacterise devices based on retention of title as interests created by the grant of a security right by the grantor.45 However, although jurisprudence on this varies from jurisdiction to jurisdiction, such ‘recharacterisation’ is often for the purposes of the secured transactions law only.46 The Model Law regime’s rules on priority also apply to these transactions. However, most of them will fall within one of the main exceptions to the ‘first-to-register’ rule: the ‘superpriority’ that an acquisition security right has over a previously registered security right.47 In this regard, the inclusion of retention of title devices within the Model Law regime does not bring about a practical change to the priority position that would result in other types of regime, where the effect of the retention of title is usually that the asset in question does not fall within a previously created security right covering the after-acquired property of the grantor, so that the party retaining title has ‘super-priority’. An acquisition security right is a right in a tangible asset or in intellectual property, which secures an obligation to pay the price48 or to pay other credit extended to enable the 41 LG, Introduction para 20. 42 In some types of finance lease, the debtor never acquires ownership of the equipment. This is the case in the UK. 43 LG, Introduction para 24. 44 At least where this is a right to obtain ownership when the price is fully paid, the LG calls this an ‘expectancy right’. See LG, ch IX para 99. 45 See R Cuming, C Walsh and R Wood, Personal Property Security Law, 2nd edn (Toronto, Irwin Law, 2012) 119, 123 (in relation to Canadian law); and see A Duggan and D Brown, Australian Personal Property Securities Law, 2nd edn (Australia, LexisNexis Butterworths, 2016) 3.4–3.5 (Australian law). 46 LG, ch IX para 79. 47 This result will depend on meeting the specific requirements of the regime. These vary somewhat from regime to regime. For details, see F2 below. 48 ‘Price’ here covers any sums paid for the acquisition of the thing, if the transaction is not a sale.

Scope of Secured Transactions Law 43 grantor to pay the price.49 The beneficiary of the right can be either a party who lends the grantor money to buy the asset outright (‘lender’), or a supplier of the asset on retention of title terms using one of the devices mentioned above (‘vendor’). The Legislative Guide suggested two optional methods to deal with acquisition security rights arising from devices based on retention of title. One (‘the unitary approach’) was for these devices to be fully recharacterised for all purposes so that the grantor becomes the owner either for all purposes (in this case all relevant statutes would require consequential amendment), or just for the purposes of the secured transactions regime and related fields.50 The other (‘the non-unitary approach’) was that, although the rules should result in functional equivalence between the acquisition security right of a lender and the retention of title right of a seller, the location of ownership would be different in each case. In the former case, the grantor would be the owner, and the acquisition security right would arise by grant, whereas in the latter case, the seller would be the owner, and its right would arise by reservation. In the non-unitary approach, if the seller does not register a notice relating to its right, it loses its ownership and the ‘grantor’ (the buyer) becomes the owner.51 A State adopting this approach would also have to amend its law so that the two consequences mentioned above applied to the ‘vendor’ case, namely, the satisfaction of junior secured creditors out of the surplus value in the asset,52 and that the vendor would have to account for any surplus value on enforcement of its right.53 In addition, in order to make sure that any surplus was quantified and liquid, any person with a junior claim to the asset, or the grantor itself, would be able to require the vendor to sell the asset rather than just to retake possession.54 Moreover, as with a security right, if the proceeds of sale were not enough to discharge the outstanding price, the vendor could sue for the deficiency.55 Parties also sometimes combine outright transfers of title and retention of title devices to create the functional equivalent of a secured transaction, for example, a sale followed by a lease back to the seller. The Model Law regime treats these types of transactions as creating a security right granted by the grantor. Since the asset originally belonged to the grantor, the security right cannot be seen as an acquisition security right,56 and therefore the discussion in the previous paragraphs does not apply.57

c. Other Interests that can be Included As mentioned above, a Model Law regime can also cover other types of transactions which have similarities with secured transactions. The most typical type to include is the practice of transferring trade receivables outright in return for a sum based on the 49 ML, Art 2(b). 50 LG, ch IX para 79. 51 LG, ch IX para 181 and recommendation 194 (non-unitary approach). 52 LG, ch IX para 132. 53 LG, ch IX para 194. 54 ibid. LG, recommendation 200(a)(ii) enables the inclusion of a provision to this effect. 55 LG, ch IX para 195. 56 See Australian PPSA, s 14(2). 57 LG, ch IX para 80. Although the Model Law does not expressly deal with this point, it is implicit from the definition of ‘security right’. It is possible to deal with this point directly; see C von Bar et al (eds), Principles, Definitions and Model Rules of European Private Law: Draft Common Frame of Reference (Munich, Sellier, 2009) (prepared by the Study Group on a European Civil Code and the Research Group on EC Private Law) (DCFR) Art IX – 1:102(4).

44 A Modern Secured Transactions Law discounted face value, in order to provide cash flow financing to the person to whom the receivable is owed. This process is often called ‘factoring’ and can take a number of forms: the risk of non-payment by the receivables debtor can be borne by either the transferor or the transferee, the receivables debtor can be notified or not notified of the transfer, and so on. The Model Law rules apply to outright transfers of receivables as if they were security rights, except for the enforcement chapter.58 This is for several reasons. First, it is sometimes difficult to tell whether a transaction is effected by an outright transfer or the creation of a security right.59 Secondly, the mechanics of the two types of transactions are similar, and it makes sense for the same priority rules to apply to each. Thirdly, it is beneficial for outright transfers of receivables to be registered, as this is the most efficient way of giving them publicity.60 Another type of interest that can be included is the long-term lease. The concept of ‘security right’ in the Model Law includes any property right created to secure payment. This includes a ‘financial lease’: this term is used in the Legislative Guide to mean a lease whereby the instalments payable cover the entire value of the leased goods, plus a financing charge. Once these instalments have been paid, the lessee has the right to acquire ownership, or to exercise an option to acquire ownership, or the asset has only a nominal residual value.61 However, it is also possible for a business or a consumer to acquire the use of equipment by making periodic payments merely for the right to retain possession, with no right or intention to acquire ownership.62 This is often termed an ‘operating lease’. The line between financial and operating leases is notoriously hard to draw, and both raise the same problem of ostensible ownership, in that the lessee is in possession of goods which it does not own. For these reasons,63 some modern systems, therefore, include long-term leases (such as leases over one or two years) as if they were security rights, although the enforcement chapter does not apply to them.64

3. Assets Under the Model Law, a security right can be granted over both present and future assets, and any type of movable asset including a part or undivided right in a movable asset.65 Despite its comprehensive nature, the Model Law regime does not cover the use of all assets as security.66 Types of assets which are subject to specific registration regimes may be excluded. It is on this basis that the regime is limited to movable assets, since immovable property (land) is typically covered by a regime which prescribes specific formalities and 58 ML, Art 1(2). 59 LG, ch 1 para 25. 60 The method of notifying the receivables debtor, used in many jurisdictions, is not as efficient as an electronic public register and is burdensome when a business transfers all of its trade receivables. 61 See LG, Introduction para 20 (defining financial lease). 62 LG, Introduction para 26. 63 See B Whittaker, Review of the Personal Property Securities Act: Final Report (2015) para 4.3.5. 64 For example, the Australian PPSA, ss 13 and 109, and New Zealand PPSA, ss 16 and 105. Note that the ‘Convention on International Interests in Mobile Equipment’ (jointly adopted at a diplomatic conference in Cape Town in 2001 by UNIDROIT and the ICAO) (CTC) applies to all leases without a minimum duration. 65 ML, Art 8. 66 ML, Art 1(1).

Scope of Secured Transactions Law 45 entails an asset-based registration system.67 Nonetheless, the Legislative Guide did make some recommendations as to movables which are attached to immovable property.68 These were not included in the Model Law, but some African States have included rules based on the recommendations.69 The Model Law also gives States an option to make other exclusions on the basis that there is a specialist registration regime;70 for example, types of equipment specifically covered by national law or covered by international conventions, such as the Cape Town Convention.71 The Model Law regime also excludes assets which are subject to special legislative treatment, such as intermediated securities,72 and payment rights arising from financial contracts. The possibility that a regime covering security rights in intellectual property might be inconsistent with a national regime is acknowledged expressly, although the Model Law does expressly include intellectual property, based on the Intellectual Property Supplement to the Legislative Guide.73

4. Secured Obligations Under the Model Law regime, a security right can secure any sort of obligations, including future and conditional obligations,74 and all obligations owed by the grantor to the secured creditor at any time.75 The inclusion of future obligations is particularly important given the prevalence and utility of revolving facilities, including overdrafts. However, the unlimited scope of the Model Law in this regard raises a danger that the first secured creditor may obtain a monopoly on lending to the grantor, as it can obtain a security right over all the assets of the grantor to secure all future obligations, so that the value of a security right taken by a junior secured creditor can be eroded by further advances. The rules on the superpriority of acquisition security rights are designed to ameliorate the danger of a monopoly as they enable another financier to obtain priority over the first secured creditor in relation to the financed asset. To some extent, a rule requiring a maximum amount to be included in the security agreement and the registered notice also ameliorates this danger.76 Another way of dealing with this issue is the inclusion of rules which limit the ‘tacking’ of future obligations to an existing security right. For example, under English law, a security right cannot 67 LG, ch I para 40. 68 LG, ch II paras 96–99; ch III paras 102–07; ch V paras 111–14; ch VIII paras 85–89; ch IX paras 154–56. 69 Conclusion, C1b. African countries that have included such rules to a greater or lesser extent include Kenya C5; Zimbabwe C1, C3, C3d; Zambia D6; Rwanda C4; Liberia C1; Burundi C4; Sierra Leone C4; and Ghana D1 and E. 70 ML, Art 1(3)(c). 71 LG, ch 1 para 32. See ch 17 (discussing the Cape Town Convention on International Interests in Mobile Equipment). 72 LG, ch 1 paras 37–38. This was also because the Geneva Securities Convention covered security in intermediated securities; see ‘UNIDROIT Convention on Substantive Rules for Intermediated Securities’ (Geneva, UNIDROIT, 2009). A guide on incorporating the GSC rules into a Model Law regime has been produced and is available here ‘UNIDROIT Legislative Guide on Intermediated Securities’ (Rome, UNIDROIT, 2017). 73 ‘UNCITRAL Legislative Guide on Secured Transactions: Supplement on Security Rights in Intellectual Property’ (New York, United Nations, 2011). 74 ML, Art 7. 75 ML, Art 9. 76 See s C1 below.

46 A Modern Secured Transactions Law secure obligations arising after the secured creditor has notice of a junior security right unless the secured creditor is obliged to make further advances under the loan agreement.77 The Model Law system eliminates the considerable uncertainty resulting from such a rule.

C. Creation The Model Law regime, consistently with the modern principles, separates the concept of creation of a security right from the concept of third-party effectiveness. The difference is that, once a security right in an asset is created, it is effective just between the parties to the security agreement, whereas, once the requirements of third-party effectiveness have been met, the right is effective against third parties (including the insolvency representative of the grantor, were it to become insolvent). Some other systems use the concept of ‘attachment’ rather than ‘creation’. These systems separate out rules which govern how to make a valid security agreement from rules which determine when the security right ‘attaches’ to any particular asset.78 Attachment gives the security right some limited third-party effect, in that the priority between two security rights that are not effective against third parties is determined by the order of attachment.79

1. The Security Agreement As mentioned above, the modern principles recommend that the formalities required to create a security right are minimal. The Model Law requires the security agreement to be in writing, signed by the grantor,80 although it permits an enacting State to choose whether this requirement is constitutive or evidential.81 Whichever legal analysis is chosen, it is clear that the policy reason behind these requirements is for there to be clear evidence that the agreement has been entered into and that the grantor agreed to grant the security right. The information required in the security agreement is minimal, although most parties will include many other terms and conditions. As mentioned above, it is sufficient to describe the encumbered assets generically.82 The secured obligation must be described in a way that enables its identification, but this can also be generic – for example, it can be all obligations owed to the secured creditor at any time.83 There is no requirement to state the amount of the secured obligation: this would be pointless since the amount would change over time. However, the option exists for a State to require the agreement to state the maximum amount for which the security right can be enforced.84

77 See UK Law of Property Act 1925, s 94; L Gullifer (ed), Goode and Gullifer on Legal Problems of Credit and Security, 6th edn (London, Sweet & Maxwell, 2017) paras 5–10 and 5-17–5.24. 78 See, eg, Australian PPSA, s 17, New Zealand PPSA, ss 35–40. 79 See, eg, Australian PPSA, s 55(2), New Zealand PPSA, s 66(c), Saskatchewan PPSA, s 35(1)(c). 80 ML, Art 6(3). 81 This is likely to depend on the legal tradition of the relevant state. See ‘UNCITRAL Model Law on Secured Transactions: Guide to Enactment’ (Vienna, United Nations, 2017) para 88. 82 ML, Art 8. 83 ML, Art 9. 84 ML, Art 6(3)(d). For discussion of the link between the adoption of this approach and the World Bank’s ‘Strength of Legal Rights’ index, see Conclusion, B1b.

Creation 47 There appear to be three policy arguments in favour of this approach. One is to make it clear to others that there might be surplus value in the encumbered assets (depending on their value), out of which a junior secured creditor might be able to be satisfied: if a State chooses this option the maximum amount must also be included in the registered notice so that it is public. The second is to reduce the likelihood of one secured creditor having a monopoly on lending to the grantor.85 The third is to protect the grantor from ‘over-collateralisation’, namely, where it grants a security right over assets the value of which greatly exceeds the secured obligation.86 However, requiring a maximum amount to be stated in the security agreement and the registered notice does not necessarily achieve either of these aims, since (unless the bargaining power of the grantor is sufficient to prevent this) the secured creditor can just state a very high amount which is in excess of the likely value of the grantor’s assets in the foreseeable future. Even if a secured transactions regime does not require a maximum amount to be stated, the parties can always include one in the security agreement. However, this would only bind the parties to the agreement as a matter of contract, and would not be public, so that other potential secured creditors would not know about it.

2. Rights in the Asset Even if there is a valid security agreement, a security right in any particular asset will not be created (or, in other terminology, will not attach) until the grantor has rights in that asset. This means, in practice, that the grantor must become owner of the asset, or have some other lesser proprietary right to it, which it can encumber, such as the right of a lessee or a licensee. This analysis relates to assets which are future assets at the time of the agreement; the grantor will already have rights in the asset if it is a present asset. Alternatively, a security right can be created if the grantor has the ‘power to encumber’ but not actual rights in an asset. A grantor would have this power if the rules of the regime create an exception to the nemo dat principle87 which applies in the relevant fact situation. One example is where, in relation to receivables, the grantor has already made an outright transfer of those receivables to another party but a notice in respect to that transfer has not been registered.

3. Scope of the Security Right As mentioned above, the Model Law regime enables security to be taken over assets which may be sold or leased in the ordinary course of business,88 such as inventory, in which case the buyer or lessee takes the asset free of the security right. However, anything received by the grantor as the price, or the lease payments, will automatically fall within the security right over the asset disposed of.89 This is also the case in other circumstances when an asset is received in respect of the encumbered asset, such as insurance proceeds. In fact, the definition 85 See above B4. 86 For the possible dangers of ‘over-collateralisation’ see LG, ch II paras 68–69; and L Gullifer and I Tirado, ‘A Global Tug of War: A Topography of Micro-Business Financing’ (2017) 81 Law and Contemporary Problems 109, 133. 87 For a definition of this principle, see F1 below. 88 ML, Art 34(4)(5). 89 ML, Art 10.

48 A Modern Secured Transactions Law of proceeds in the Model Law is very wide.90 In some situations, the security right may persist in the asset (eg when it is sold not in the ordinary course of business)91 but the security right also extends to the proceeds. There are various reasons for the extension of the security right to the proceeds of the encumbered asset. First, it is thought necessary to protect the interests of the secured creditor if the encumbered asset is disposed of.92 However, arguably, it is only when the disposition is unauthorised that the secured creditor needs this protection.93 When the disposition is authorised, or in the ordinary course of business, arguably it is up to the parties to agree whether the secured creditor should have a security right in the proceeds (in which case this should be made clear in the security agreement) or whether the disposition is for the benefit of the grantor, so that the proceeds can be used to run its business. These arguments, however, have not influenced the regimes based on the modern principles, probably because it is thought too difficult to distinguish between authorised and unauthorised dispositions. Where proceeds take the form of money or funds in a bank account, the Model Law provides that the grantor can, in any event, dispose of them free of the security right in the ordinary course of business,94 so the commercial running of the business is not impaired. A second reason is that it is generally accepted in many jurisdictions that a security right should extend to ‘fruits’ of the asset, and it is often difficult to differentiate between fruits and proceeds.95 Thirdly, the rule is also said to give effect to the commercial expectations of the parties.96 In addition, the Model Law provides for situations where the asset has been mixed with other assets to make a mass, or has been made into a product, with or without other assets.97 In all these cases, the Model Law regime makes it unnecessary to include possible proceeds as encumbered assets in the security agreement, since not only does the security right extend to the proceeds, mass or product but the priority of the extended security right is the same as that of the original security right,98 despite the fact that the proceeds, mass or product will not be identified in the security agreement or in the registered notice.

4. Trade Receivables One particular feature of regimes based on the modern principles is that the principle of freedom of alienability is given precedence over the principle of freedom of contract in relation to the transfer of, or the creation of security rights in, trade receivables. 90 ML, Art 2(bb). 91 ML, Art 34(1). See also the definition of proceeds in ML, Art 2(bb) – ‘proceeds’ means whatever is received in respect of an encumbered asset, including what is received as a result of sale or other transfer, lease, licence or collection of an encumbered asset, civil and natural fruits, insurance proceeds, or claims arising from defects. On some limitations of the extension of the security right into proceeds, see LG, ch I paras 36–37. 92 LG, ch I para 17. 93 See LG, ch II para 74. In this case, the rule effectively replaces the rules on tracing and, in fact, rules about tracing into a bank account are included in ML, Art 10(2). 94 See F3 below. 95 LG, ch I paras 16–24. 96 LG, ch II para 78. 97 ML, Art 11. 98 ML, Arts 32–33.

Third-Party Effectiveness 49 Transfer or creation of security rights is effective even if there is an agreement between the grantor and the debtor owing the receivable prohibiting or limiting the grantor’s right to do this.99 The policy behind this provision is the promotion of receivables financing.100

D. Third-Party Effectiveness While creation means that the security right is effective between the parties, this is unlikely to be useful in situations when the secured creditor will want to rely on the security right. For this, the security right must be effective against all third parties, including the insolvency representative of the grantor (so that the secured creditor has priority in relation to the encumbered assets over all the unsecured creditors of an insolvent grantor).101 As mentioned above, making a security right effective against third parties largely consists of methods which give publicity to the security right: primarily registration, but also other methods, including the secured creditor being in possession of the encumbered asset.102 Possession by the secured creditor has traditionally been seen as a type of publicity of a security right, since, although the secured creditor’s possession could be for a reason other than the existence of a security right in the relevant asset, the lack of possession of the grantor indicates to third parties that the grantor may not have unencumbered rights in the asset.103 While this lack of possession may alert a third party to a potential problem, however, it does not provide the level of information provided by registration. In specific situations, methods other than registration and possession may be used to make a security right effective against third parties. These typically exist as an alternative to registration. One method is called control, and has several iterations, described below, which largely depend on the type of asset involved. As will be seen from the discussion in the next paragraph, control does not give full publicity of the security right but only limited publicity, the strength of which depends on the circumstances. One reason, other than publicity, given for the existence of ‘control’ as an alternative to registration (at least in relation to a security right over a bank account) is that it tracks existing banking practices, and facilitates lending by deposit-taking banks, who will normally insist on the borrower having at least one bank account held with it over which it will take security.104 A similar market based reason could apply to uncertificated non-intermediated securities. Another possible reason is that if a secured creditor has control, it is sometimes more difficult (though not impossible) for another person to take a security right over the asset. Yet another possible reason is that taking control of an intangible asset is the ‘equivalent’ to possession of a tangible asset (including a negotiable instrument): this argument, however, fails to take account of the fact that control gives even less publicity than the taking of possession. It is, 99 For further discussion, see ch 1, E4a. 100 LG, ch II para 110. The ML mirrors the UN Assignment Convention in this regard, see ‘United Nations Convention on the Assignment of Receivables in International Trade’ (New York, United Nations, 2004). 101 See ML, Art 35 confirming that a security right that is effective against third parties retains its effectiveness and priority on the insolvency of the grantor. 102 ML, Art 18. 103 LG, ch III para 47. 104 LG, ch III para 148.

50 A Modern Secured Transactions Law however, supported by the approach of the Model Law to priority in relation to negotiable instruments or negotiable documents.105 In relation to a bank account,106 the strongest form of control is where the secured creditor (who is, ex hypothesi, not the bank itself) becomes the account holder. This gives only limited publicity: there is no publicly searchable record, but if a third party asks the grantor to produce evidence that it, the grantor, is the account holder, the grantor will not be able to do this. This form of control gives priority over any security right made effective against third parties in any other way.107 The second type of control is where the bank itself is the secured creditor. Here, although there is no publicity as such, third parties are not prejudiced, since even if there were no security right, the bank would have a right of setoff, which would not be publicised.108 This type of control also gives priority over security rights made effective against third parties in any other way, except a security right held by a creditor who has become the account holder.109 Another type of control is conclusion of a ‘control agreement’ between the bank, the grantor and the secured creditor whereby the bank agrees to follow the secured creditor’s instructions as to the bank account without the need for consent from the grantor.110 This gives limited publicity only if the bank is authorised by the grantor to tell third parties who enquire about the grantor’s account that the control agreement exists. This type of control gives priority over a security right made effective against third parties by registration.111 A similar regime applies to uncertificated non-intermediated securities.112 Another possible ‘method’ of making a security right effective against third parties is for the law to provide that no steps at all are required to make a security right effective against third parties. This is sometimes called ‘automatic perfection’ (in regimes that use the term ‘perfection’). An acquisition security right in consumer goods (up to an amount specified by the enacting State) is made effective against third parties in this way under the Model Law.113 This is because even the cheap registration procedure under the Model Law is costly in relation to low value consumer transactions.

E. Registration As mentioned above, an efficient registry is at the heart of the Model Law regime. Registration and searching should be as cheap and easy as possible. It should provide a means of publicity which is reliable, and the rules of which reflect the principle that registration enables searching third parties to adjust in the light of the registered information, as well as conduct further investigation. A fully electronic system, with no human intervention between the entry of information and the registration itself, and which is searchable w ithout

105 See

below F1 explaining this in detail. ML, Art 25. 107 ML, Art 47(1). 108 LG, ch III para 146. 109 ML, Art 47(2). 110 ML, Art 2(g). 111 ML, Art 47(3). 112 ML, Arts 27, 51. 113 ML, Art 24. 106 See

Registration 51 human intervention, fulfils some of these aims, while the others are fulfilled so long as the priority rules entirely align with the rules and operation of the registry. This section will discuss some key aspects of a modern secured transactions registry, many of which are outlined above.114

1. Notice Filing Neither transactions nor security rights are actually registered. What is registered is a notice indicating that a security right may have been granted by the named grantor to the named creditor over the identified assets. To discover whether a security right has actually been granted, it is necessary for a person searching to make further investigations. This system, often known as ‘notice filing’ has a number of consequences. First, a notice can be registered at any time, including before any secured transaction takes place. This has the advantage that a secured creditor can fix its priority before committing to the transaction, and that one notice can cover any number of security agreements between the parties as long as they create security rights in the identified assets. This is particularly advantageous in relation to sales on retention of title terms, which could take the form of many different sale agreements.115 Moreover, because a security right can be granted over a future asset, the notice filing system means that no problems arise from a registration relating to a security right which has not yet been created because the grantor does not yet have rights in the relevant asset.116 Secondly, the notice needs to include only a small amount of information, namely, the identifier and address of the grantor and the secured creditor117 and a description of the encumbered assets. This description can be generic, in the same way as in the security agreement.118 It is also possible for a registration regime to require, or make optional, the inclusion of a serial number for some narrowly defined types of encumbered asset, typically motor vehicles. In this case, the regime will also enable a search to be made using a serial number as a search criterion. This is particularly useful in asset finance, which tends to relate to particular pieces of equipment.119 It allows searchers to discover notices in relation to a particular asset even when it has been transferred from the grantor to someone else and an amendment notice has not been registered.120 If the enacting State requires a maximum amount to be included in the security agreement,121 this will also need to be included. Thirdly, a system which permits registration of a notice in advance of a transaction taking place must have rules to ensure the grantor has agreed to the registration, and which enable it to compel removal of a registration which does not relate to an actual 114 See A2b above. 115 LG, ch IV para 99. 116 See C2 above. Registration of a future security right could be seen as problematic in some legal cultures. 117 ML, Art 8. There are rules about how these parties are to be identified, which could be by name or by some other means, such as an ID number. 118 See C1 above. 119 See B2b above. 120 See LG, ch IV para 33, and E2 below. 121 See C1 above.

52 A Modern Secured Transactions Law or intended transaction. Otherwise, the grantor could be prejudiced as searchers could obtain the impression that the grantor’s assets were encumbered when they were not. There are various ways of protecting the grantor, which are used in different systems. The Model Law uses two methods. First, a notice is ineffective unless it is authorised by the grantor in writing.122 Although this provides an incentive to a secured creditor not to register without the grantor’s consent, it does not actually prevent it, and searchers will not know whether a notice is effective or not. Therefore, the second method requires the registry to send a ‘verification statement’ to the registrant123 after registration of a notice, and the registrant must send a copy to the grantor.124 On receiving this statement (or at any time), the grantor can require that the notice be cancelled by the registrant if it was not authorised or if, even though it was authorised, the intended transaction did not proceed.125 However, to be successful, this method requires the registrant to comply with the obligation to send a copy of the statement to the grantor. Failure of the registrant to do so will, typically, only result in a potential liability for damages, and perhaps a nominal fixed penalty,126 which is unlikely to deter a person making an unauthorised registration. A third, much more effective, method is for electronic consent of the grantor to be required.127 Fourthly, there is no question of the content of the notice being checked by the registry. This is, first, because there is no actual transaction against which to check128 and, secondly, because the information is entered directly into the online system by the registrant and is effective without human intervention.129 The risk of inaccuracy in the registered information is therefore on the registrant, as opposed to (sometimes) being on the searcher in other systems.130 An error in the grantor identifier renders the registration of the notice ineffective as against that grantor,131 as does an error in any other information which would seriously mislead a reasonable searcher.132 Fifthly, any registry system has to include means by which the registered information can be kept up to date. Despite the minimal amount of information in a registered notice, this is still the case with the Model Law system, which provides for amendment and cancellation notices133 which can be registered by the identified secured creditor. Thus, for example,134 122 MRP, Art 2. This authorisation can be given at any time, either by the grantor entering into a security agreement with the secured creditor or by a separate authorisation. 123 That is, the person identified in the notice as the secured creditor, who, in the situation being discussed, will not actually be a secured creditor. 124 MRP, Art 15. 125 MRP, Art 20(3). The same applies if the security right has been extinguished by payment of the secured obligation; see also ML, Art 12. 126 See MRP, Art 15(4). The exact consequences will vary according to the system; see Australian PPSA s 157(4) which merely gives the possibility of an action for damages. 127 See CTC, Art 20. 128 Unlike a document filing system or a system where registration takes place after the creation of the security right; see Conclusion D5a. 129 See A2b above. See MRP, Art 7(2) and (3). 130 This used to be the position under English law, see National Provincial and Union Bank of England v Charnley [1924] 1 KB 431. 131 MRP, Art 24(1). 132 MRP, Art 24(4). 133 See ML, Arts 16–21. 134 Amendment notices can be registered in relation to other information included in the notice as well; see MRP, Art 1(b).

Registration 53 an amendment notice can be registered if the secured creditor changes its identifier (including when the security right is transferred to another party), if the grantor changes its identifier, or if any of the encumbered assets are transferred in circumstances where the transferee takes subject to the security right and is therefore a new ‘grantor’.135 Since the last two situations could mislead searchers, the Model Law contains provisions to balance the interests of the secured creditor and potential searchers. Under some options the secured creditor is given a ‘grace period’ within which to register an amendment notice, failing which it will lose priority to another secured creditor who has registered its security right before any amendment notice is registered.136

2. Grantor-Based System As mentioned above, the registry of a system based on the modern principles must be a debtor-based (or, more accurately, grantor-based) system, that is, the notice must identify the grantor and the primary method of search must be based on the grantor’s identifier.137 This is because the whole point of the system is to enable searchers to find out what security rights a particular grantor may have granted, and also because assets can be described generically in the notice, so that an asset-based search would not work. For this system to work, the grantor’s identifier has to be robust and as unique as possible. Some systems, including the Model Law, require the name of the grantor (legal or natural)138 as the identifier. The drawbacks of this approach are, first, that a name may change,139 secondly, that there may be more than one person with the same name and, thirdly, that it is quite possible to get the name wrong when searching.140 Another approach is to use a nationally recognised ID number, as has been done in several African countries. As long as the national ID system is robust, this system avoids the drawbacks of the ‘name’ system and reduces any need for a ‘close match’ searching system.141 As mentioned above, some systems also permit a registrant to include the serial numbers of any encumbered assets in the notice, and to search using the serial number as a search criterion. This is a useful addition to the system of registering and searching by grantor identifier, since it enables a person intending to take an interest in a specific tangible asset to discover whether there are (or are likely to be) any security right affecting that asset. This is particularly useful for prospective buyers, who are more likely to search using the serial number, whereas a prospective secured creditor is more likely to search using the grantor’s identifier.142 135 See definition of grantor in ML, Art 2(o). 136 MRP, Arts 25 and 26. Art 26 gives states two options as to the details of this provision and includes a similar priority rule relating to a buyer, as opposed to a secured creditor. 137 See ML, Art 22. 138 The state must specify an official document that must be the source of the name used; see MRP, Art 9(1)(4). 139 See E1 above. 140 See E3 below. 141 Such a system is an option under MRP, Art 23 but has not been used in Africa. 142 For this reason, in some jurisdictions, an error in the registered serial number will only render the registration ineffective against buyers and lessees and not secured creditors. See Kenya C4; Zimbabwe C2. Oddly, Liberia, despite having a sophisticated system of serial number registration in other respects, does not limit ineffectiveness for error in this way; see Liberia C5b.

54 A Modern Secured Transactions Law

3. Searches According to the modern principles, searching the register should be as cheap and easy as possible. This is fulfilled by a fully electronic system of registration and searching, and by searching being either free or at a very minimal cost.143 The principle of publicity means that there should be no restrictions on who can search, and that there should be as few barriers to access as possible. Searching will only be effective if the searcher can discover all registered notices which relate to the grantor in which the searcher is interested. The Model Law provides that an error in the grantor identifier in a notice only renders the notice ineffective if the information would not be retrieved by a search using the correct identifier.144

F. Priority Many of the priority rules have already been discussed above, so this section will highlight some particularly significant points.

1. The Basic Rule As mentioned above, the basic priority rule is ‘first to register’. Since registration can take place before a security right is created, this priority rule can operate in some circumstances as an exception to the nemo dat quod non habet principle (that a person cannot give what they have not got): that principle means that the first security right to be created has priority. This can be illustrated simply. On day 1, A registers a notice identifying itself as the creditor and B as the grantor, and the asset as X. On day 2, B grants a security right over asset X to C. On day 3, C registers a notice identifying itself as creditor and B as grantor and the asset as X. On day 4, B grants a security right over asset X to A. Under the first-to-register rule, A has priority over C, whereas under the nemo dat principle, C has priority over A. As mentioned earlier, it is also possible to make a security right effective against third parties by other means. The priority rules relating to control have already been discussed.145 Priority between a security right where the secured creditor has possession and a registered security right is determined by whichever event occurred first: the taking of possession or the registration (on the basis that each event gives publicity to the security right to which it relates).146 This basic rule is subject to an exception that is similar to the priority rules relating to control. A secured creditor with a security right in a negotiable instrument or document who has possession of the relevant asset has priority over a secured creditor with a registered security right. The reason for this exception is to preserve the negotiability of

143 MRP,

Art 33 which gives an option for charging fees and one for not charging fees. Art 24. 145 See D above. 146 ML, Art 29. 144 ML,

Priority 55 the instrument or document, which would be limited if a negotiation was subject to a priorregistered security right.147

2. Acquisition Security Rights The ‘super-priority’ of acquisition security rights148 has been discussed already.149 There are various reasons for this exception to the ‘first to register or take possession’ rule. The first is that the exception encourages lenders other than the first secured creditor to extend credit to the grantor. Otherwise, the ability to take security over all future assets gives rise to the danger of one secured creditor having a monopoly on all lending to the grantor.150 Because all subsequent security rights would potentially be junior to that first secured creditor, no other potential lenders would want to lend.151 Another reason is that the advance secured by the acquisition security right directly enables the grantor to acquire a new asset:152 in the absence of ‘super-priority’ the first secured creditor would obtain a first-ranking security right in this new asset, thus swelling its encumbered assets without making any further advance. This state of affairs, it is said, gives the first secured creditor an unjustified windfall, whereas the advantages in the acquisition secured creditor obtaining priority are that the monopoly is broken and, particularly in the case of asset finance, the specialised nature of the acquisition secured creditor may well mean that it will advance the credit at a lower cost than the original lender.153 Another relevant factor is that, in regimes other than those based on the modern principles, acquisition financing is predominantly effected by devices based on retention of title, which automatically have ‘super-priority’ status because title does not pass, and so the ‘super-priority’ rule merely retains the status quo.154 However, the acquisition security right super-priority regime applies to acquisition financing whether it is provided by a credit seller, a lessor, or a lender who lends the money to the grantor, who then buys the asset, paying the purchase price to the seller immediately.155 The priority rules relating to acquisition security rights can be seen as quite complex, but the complexities are there to deal with a need to balance the interests of the acquisition secured creditor and the previous secured creditor, who needs to have some way of discovering that the new asset is subject to an acquisition security right so that it does not rely on that asset as part of its collateral, and so that it has the option of refusing to make 147 LG, ch V para 155. 148 Elsewhere often called purchase money security interests. 149 See B2b above. 150 See B4 above. 151 See T Jackson and A Kronman, ‘Secured Financing and Priorities among Creditors’ (1979) 88 Yale Law Journal 1143. 152 Even in the case of a lease structured so that the grantor never obtained ownership of the asset, it still obtains the use of the asset which can be used to advance its business, see Duggan and Brown (n 45) 8.5. 153 For further discussion of the relevant policy considerations, see C Walsh, ‘The Floating Charge Is Dead: Long Live the Floating Charge’ in A Mugasha (ed), Perspectives in Commercial Law (London, Prospect Publishing, 2000) 129, 134. 154 See B2b above. 155 See LG, ch IX para 52–55.

56 A Modern Secured Transactions Law future advances. The Model Law offers States two options in this regard.156 Option B states the basic rule, which is that the acquisition security right has priority over a security right registered earlier if either the acquisition secured creditor still has possession of the asset, or if the notice relating to the acquisition security right is registered within a short grace period of the grantor taking possession of the asset.157 However, option A applies this rule only to equipment. In relation to inventory, two things must happen before the grantor takes possession of the asset. Not only must the relevant notice be registered, but also the acquisition secured creditor must give notice to the previous secured creditor that it has, or intends to obtain, an acquisition security right in the asset. The reason for the difference between equipment and inventory is that the acquisition of equipment could be a one-off transaction, and the acquisition itself might be aborted (in favour of another supplier) if it cannot take place straight away without waiting for registration. No such constraints apply in relation to inventory, where the supplier and the grantor are likely to have an ongoing relationship, and only one notice needs to be registered, and served on the previous secured creditor, in respect of all the parties’ dealings.158

3. Taking Free Rules Commercial activity would be unreasonably delayed and costly if every participant in such activity needed to check the register in order to adjust to the presence of security rights. One way that this is facilitated is for the secured creditor to agree with the grantor that the grantor can dispose of encumbered assets under certain circumstances (eg to buyers in the ordinary course of business). This agreement could be generic, in the original security agreement,159 or specific in relation to a particular disposition. A party taking under a disposition authorised by the secured creditor takes free of the security right.160 In addition, the Model Law provides default rules so that some types of persons acquiring an asset from a grantor will take free from all security rights in that asset. Such commercial actors will not then need to worry about searching the register. One such group are buyers, lessees or licensees of encumbered assets in the ordinary course of business, although such a commercial actor will not take free if it has knowledge that the disposition is in breach of the secured creditor’s rights.161 Other such groups are buyers (and other transferees) of negotiable instruments,162 a transferee of funds from a bank account163 and a transferee of money:164 in all these cases, again, the buyer or transferee will not take free if it has knowledge that the disposition is in breach of the secured creditor’s rights. 156 ML, Art 38. 157 The rules described here relate to tangible property, but the same rules apply, with appropriate modifications, to intellectual property. 158 Duggan and Brown (n 45) 8.35–8.36. 159 This is similar to the English law floating charge. The permission in a floating charge in theory includes the creation of security rights as well, but this is nearly always prohibited by a negative pledge clause. See ch 1, E1b. 160 ML, Art 34(3). 161 ML, Art 34(4)(5)(6). 162 ML, Art 48(2). 163 ML, Art 47(6). 164 ML, Art 48.

Rights and Duties of Parties 57

4. Competing Claimants with Non-Consensual Interests and Other Preferential Claimants The Model Law permits States to provide for priority of specified claims competing with security rights on grounds of policy (eg the claims of employees for unpaid wages).165 Although not addressed in the Model Law, it is possible for a system to permit notices of at least some types of these claims (eg tax liens) to be registered, and even for priority to depend on date of registration (although this would be unlikely to accord priority to such claims over secured creditors because such claims are likely to arise, and be registered, when the grantor is close to going into insolvency proceedings).166 Judgment creditors will normally have the right to acquire rights in the grantor’s assets by a process of execution. The rules of a secured transactions regime must balance the interest of the judgment creditor in knowing what assets will be available to satisfy the judgment if it goes through the execution process, and the interest of the secured creditor in its security right being fully enforceable. This balance is achieved in the Model Law by providing that the judgment creditor has priority if execution is completed167 before the security right is made effective against third parties.168 This provision, which is consistent with the ‘first to make effective’ rule, is justified on the basis that the judgment creditor will not be able to find out about the security right before it has executed. If the security right is made effective against third parties before execution is complete, the secured creditor has priority. However, once the secured creditor knows about the executed judgment, it is in a position to stop making further advances. Thus, the Model Law provides that the judgment creditor can serve a notice of execution on the secured creditor, and the secured creditor will not obtain priority for credit extended after receiving the notice, except to the extent that it had a pre-existing irrevocable obligation to extend a fixed amount of extra credit.169

G. Rights and Duties of Parties Although the rights and obligations of the parties to the security agreement will largely depend on the terms of that agreement, many systems based on the modern principles provide for some default rules, which, largely, can be adapted by the parties for their particular circumstances. In the Model Law, the parties cannot derogate from the default rules, which impose obligations based on commercial reasonableness.170 Indeed, commercial reasonableness is a standard which all adaptations of the default rules must meet.171

165 ML, Art 36; LG, ch V paras 90–93. 166 LG, ch V para 93. 167 What amounts to completion of execution will depend on other laws in the enacting State. 168 ML, Art 37(1); LG, ch V para 96. 169 ML, Art 37(2); LG, ch V paras 99–102. 170 The obligation on a party in possession of an encumbered asset to take reasonable care to preserve it (ML, Art 53) and the obligation of a secured creditor in possession to return the asset once the secured obligation is extinguished (ML, Art 54). 171 ML, Arts 3–4.

58 A Modern Secured Transactions Law Many of the default rules in the Model Law relate to the rights and obligations of parties in relation to security rights over receivables, and, therefore, also include rules about the rights and obligations of obligors.172

H. Enforcement An effective enforcement regime is critical if the secured transactions law is to encourage the extension of credit and a reduction in its cost, since the whole point of taking security is to increase the chance of repayment of the secured obligation in circumstances when the grantor does not (or cannot) repay voluntarily. Enforcement must be available at low cost, and speedily, and the available processes must be predictable and certain. However, it is also necessary to provide some protection to the grantor, and also to third parties who may have an interest in the assets against which enforcement will take place.173 One very important feature of a secured transactions regime based on the modern principles is that it enables enforcement of security rights to take place either by applying to a court or extra-judicially. Different principles apply to each of these, as the method of achieving the necessary balance between the policy issues mentioned above differs. The Model Law sets out in some detail the available methods of extra-judicial enforcement, and rules designed to achieve this balance, so that secured creditors can know in advance what they will have to do. However, the parties can provide for other means of enforcement in the security agreement if they are not inconsistent with the Model Law provisions.174 The parties can agree to vary any of the rights in the enforcement chapter, but only after default;175 this is so that the parties can have the freedom to negotiate a method of realising the value of the encumbered assets in a way that benefits all parties.176 An important protection given by the Model Law to the grantor is that enforcement can only take place after default.177

1. Court Enforcement Court proceedings provide considerable protection to the grantor and other parties, but can be costly and slow. The Model Law therefore requires there to be, in the enacting State, expeditious court enforcement proceedings which can be used by an enforcing creditor,178 supplementing, if necessary, the existing procedures. It also requires the court to distribute the proceeds of enforcement in accordance with the Model Law rules on priority. 172 ML, ch VI ss IB and II. 173 LG, ch VIII para 6. 174 An example could be an agreement for the appointment of a receiver, although this could depend on the relevant insolvency law. 175 ML, Art 72. 176 LG, ch VIII para 17. 177 ‘Default’ includes not only failure to pay the secured obligation but also any other event that the parties have defined as a default in the security agreement (ML, Art 2(j)). 178 ML, Art 73(2).

Enforcement 59

2. Extra-Judicial Enforcement Extra-judicial enforcement is usually quicker and cheaper than court enforcement, but requires more protection for the grantor and other interested parties. Therefore, each method of enforcement provided for by the Model Law includes protective measures. The Model Law also requires States to provide for expeditious court proceedings if the secured creditor does not comply with these protective measures.179 There are various types of protective measures depending on the mode of enforcement. One, for example, is the requirement of notice of enforcement to be given to the grantor and interested parties.180 This enables any party to object to enforcement, and also enables claimants with an interest in the relevant asset to notify the enforcing secured creditor of their claims. Another example is the requirement of consent to extra-judicial possession from interested parties.181 A third is that a higher-ranked creditor has the right to take over enforcement.182 Further, any person with a right in the encumbered asset (including the grantor) can terminate an enforcement process by paying the secured obligation.183 The methods of enforcement provided for in the Model Law are, first, the taking of possession of an encumbered asset; secondly, the disposal of an encumbered asset; thirdly, the acquisition of an encumbered asset in full or partial satisfaction of the secured obligation; and fourthly, the right to collect payment from the obligor in relation to a receivable, a negotiable instrument, funds in a bank account or a non-intermediated security. One method missing from this list is the right to take over the business and to sell it as a going concern, which is common in some jurisdictions. This method is not precluded by the Model Law if the security right covers all the grantor’s assets, and the parties agree to it.184 It may be an advantageous method of enforcement, since the ‘going concern’ value of the business is captured.

3. Surplus and Deficiency The Model Law also sets out clearly the consequences (on enforcement) of a security right being limited to the extent necessary to secure the relevant obligation. An enforcing secured creditor must apply the proceeds of enforcement to satisfaction of its secured obligation (plus costs of enforcement) and then must pay any surplus to anyone with a subordinate claim185 who has notified the enforcing creditor of its claim to the asset (in accordance with the amount of their claim) and pay any balance to the grantor. There is also the alternative 179 ML, Art 74. 180 This is required for the taking of possession (ML, Art 77(2)) and the disposal of the encumbered asset (ML, Art 78(4)). A proposal to acquire the asset in satisfaction of the obligation must also be sent to all interested parties (ML, Art 80), who all have the opportunity to object and prevent the acquisition. 181 ML, Art 77(2). The grantor must have consented to extrajudicial possession in writing, and any party in possession must also consent. 182 ML, Art 76. 183 ML, Art 75. 184 LG, ch VIII para 73. 185 ML, Art 79(2)(b). Note that a person acquiring an asset under an extrajudicial disposition will take subject to the rights of secured creditors senior to the enforcing creditor (ML, Art 81(3)), whereas a similar person under a court-imposed disposal takes free of all claims if the enacting state so provides (ML, Art 81(1)).

60 A Modern Secured Transactions Law of a court or other authority taking over the disposition of proceeds, which can include distribution to senior secured creditors.186 Not only is a security right limited so that there is an obligation on a secured creditor to account for any surplus value in the encumbered asset, but also if the enforcement proceeds are not enough to satisfy the secured obligation, the debtor is liable for any balance.187

I. Conflict of Laws Many secured transactions involve parties from different jurisdictions or assets located in different jurisdictions, and it is critical for parties to know from the outset which State’s rules will apply to each legal element of the transaction. In particular, a secured creditor will need to know where to register, and a prospective secured creditor or other interested party will need to know in what registry to search.188 The Model Law, therefore, includes conflict of laws rules,189 in order to provide certainty and predictability to parties.190 Broadly, the conflict of laws provisions cover which law applies to the creation, thirdparty effectiveness, priority and enforcement of a security right in various types of assets. For tangible assets, it is the law of the location of that asset.191 For intangible assets, it is the law of the location of the grantor.192 For intellectual property, it is the law of the State in which the intellectual property is protected.193 For uncertificated non-intermediated securities, it is the law under which the issuer is constituted.

J. Transitional Provisions The transition from an existing secured transactions legal system to one based on the modern principles raises a number of important issues. Some are practical, and are addressed elsewhere in this book.194 This section addresses the legal issues that arise from replacing an existing regime with a new regime. The transitional provisions in the Model Law envisage that, in relation to transactions creating security rights within the scope of the Model Law, previous law will be repealed and replaced with the new law. Accordingly, there needs to be a set of provisions dealing with the effectiveness of a security right created and made effective against third parties under the previous law, as well as the

186 The enacting state has to stipulate which authority can do this. ML, Art 79(2). 187 ML, Art 79(3). 188 Conflict of laws rules for these purposes would still be required even if every state enacted a Model Law system. LG, ch X para 8. 189 These will only apply, in the case of a dispute in court, if the forum is a state that has enacted these provisions of the ML. 190 LG, ch X para 6. 191 ML, Art 85. 192 ML, Art 86. 193 ML, Art 99. 194 See Conclusion C.

Transitional Provisions 61 priority of such rights vis-a-vis security rights created and made effective against third parties under the new law. The approach adopted in the Model Law is one that has been followed in many national and other systems.195 First, the date of entry into force of the new law is set as a specific date in the statute itself or to be determined according to the practice of the enacting State.196 It is envisaged, however, that there will be a reasonable period of time between the enactment of the statute and its coming into force, in order to enable market participants to prepare for the new regime and for the new registration system to be fully implemented.197 Secondly, a security right created under the previous regime remains effective between the parties under the new regime.198 Thirdly, during a transitional period a security right created and made effective against third parties under the old system remains effective between the parties themselves, and against all third parties. During this period, the secured creditor can make the security right effective against third parties according to the new regime, and it will then remain so. If the secured creditor does not do that, the security right stops being effective against third parties at the end of the transitional period.199 If the secured creditor later makes it effective against third parties, this effectiveness will not be retroactive, but will date from the date on which it occurred.200 There are two situations that can arise in relation to priority. The first situation is where both rights arose (and were made effective against third parties) under the previous law. Here, the previous law determines priority providing that the priority status of neither right has changed since the new law came into force. There would be such a change in relation to one of the rights if it were not made effective against third parties under the new law within the transitional period (so that its effectiveness against third parties ceased), or if it were not made effective against third parties under the previous law and only became effective against third parties under the new law.201 In both cases the new law applies. In the former case, the right that is effective against third parties has priority over the right that is not. In the latter case, the rule discussed in the next paragraph applies. The second situation is where one right arose (and was made effective against third parties) under the previous law and one arose (and was made effective against third parties) under the new law. Here, the new law governs priority, but the priority point for the former right dates from when it was made effective against third parties under the old law, whereas the priority point for the latter right would be the priority point under the new law.202

195 See, eg, the Australian PPSA, the New Zealand PPSA, and the CTC. 196 ML, Art 107. 197 LG, ch XI para 4. The Australian PPSA, for example, did not come into force until over two years after the Act had been passed. See Duggan and Brown (n 45) 15.1. 198 ML, Art 104. 199 This is also the case if the effectiveness against third parties under the previous law terminated under that previous law before the end of the transition period. For example, if the period of registration had expired (ML, Art 105(1)(a)). 200 ML, Art 105. 201 ML, Art 106. 202 LG, ch XI para 26. This conclusion arises from a combination of Arts 105 and 106 of the ML.

62 A Modern Secured Transactions Law

K. Conclusion This chapter discusses a secured transactions legal system based on the modern principles, as defined in section A2, with particular reference to the Model Law. The Model Law provides a fully worked-out system, drafted in a way suitable for legislation, and designed to be suitable for implementation in both civil and common law jurisdictions. It is therefore suitable for detailed consideration by African countries, and many of these that have reformed their law have used it as a model. There are, however, other models, including national laws, which can be used or taken into account in conjunction with the Model Law. Some differences between the Model Law and these models are matters of drafting, whereas others are substantive. Some of the substantive differences have been discussed in this chapter.203 As discussed elsewhere in this book, it is very important indeed that legal reform in this area of law takes account of local practices and local economic and commercial conditions, as well as dovetailing properly with all the other law in the reforming State. The Model Law is just that: a model.204 However, this chapter also discussed the modern principles of secured transactions law on which the Model Law is based. These principles have been worked out, not just as a theoretical experiment, but by experience in many jurisdictions, to provide a legal environment that facilitates credit at a low cost by providing clarity and certainty to lenders and borrowers, as well as providing cheap and easy ways to create security rights, make them effective against third parties, and enforce them. These principles are designed to work as a whole, and derogation from them may jeopardise the beneficial impact of reform. Having said this, all principles are a starting point for discussion, albeit that a very high bar is set for derogation from them, and local conditions can make it imperative to qualify some principles in order for a reform to work at all. The point of this chapter is therefore threefold. It is to provide a summary of a regime based on modern principles so that the descriptions of reformed regimes in subsequent chapters are more easily understood. It is to provide a yardstick against which to compare the actual reforms in Africa, and to assess whether the derogations from the Model Law, and from the modern principles, are justified by local conditions. Finally, it is to provide a basis on which to consider future reform, either in unreformed jurisdictions or in those jurisdictions where legislation is being considered to amend existing reformed law. To that end, it includes discussion of the policy reasons behind many of the rules in the Model Law, which can help inform choices as to the details of future reformed law.

203 For example, the inclusion of long-term leases in the system and the inclusion of serial numbers in the registered notice. 204 For further discussion of models, see Conclusion C1.

part ii Common Law

64

3 Ghana In 2010, Ghana became the first African economy to establish a collateral registry, which put into operation a secured transactions law enacted in 2008. The registry was subsequently modernised with the support of the International Finance Corporation (IFC), and relaunched as a web-based system in 2012.1 This inspired other African economies to consider reforming their own secured transactions frameworks and establishing modern electronic registration systems for notices of security rights. In conjunction with the launch of the registry, in July 2012, the IFC organised a pan-African event in Accra, the capital of Ghana, to assess the status of secured transactions reforms in the region and to showcase Ghana’s Collateral Registry.2 However, many of the approaches taken in Ghana, particularly in relation to the legal framework, were conceptually flawed, inadequately implemented and uncoordinated with other legislation. With the exception of Sierra Leone, other African economies have managed to avoid some of the pitfalls of the Ghanaian approach (the main one being the insertion of a new Act into the existing legal framework) while also drawing positive lessons from Ghana’s reforms, especially in relation to the design and establishment of its Collateral Registry. One such lesson is the importance of involving central banks in the establishment, hosting and operation of an electronic notice-based registry. Registries of this nature had been traditionally operated by registrar general’s offices under ministries of justice. Ghana’s Central Bank-operated model has been replicated in Liberia, Nigeria, and Sierra Leone, and is being considered in a number of other African economies (eg Ethiopia and Zimbabwe). This chapter examines the framework that became an inspiration for many economies in Africa to reform their secured transactions systems. The legal framework underpinning Ghana’s Collateral Registry fails on almost all counts to meet the best international standards, including those now enshrined in the Model Law. The Borrowers and Lenders Act of 2008 (B&L Act) was developed and drafted without the assistance of a donor agency or foreign expert(s).3 From its structure, approaches and drafting, it is clear that the recommendations of the then available Legislative Guide were not taken into account nor was any other foreign model, such as one of the PPSAs. Its enactment simply added another element to Ghana’s secured lending

1 IFC, ‘Secured Transactions and Collateral Registries’, www.ifc.org/wps/wcm/connect/793e79804ac10fff9ea69e 4220e715ad/Secured+Transactions+and+Collateral+Registries+Brochure-English.pdf?MOD=AJPERES. 2 IFC, ‘IFC and Bank of Ghana to Increase Access to Finance for Businesses in Sub-Saharan Africa through Secured Lending’ (July 2012) www.ifc.org/wps/wcm/connect/1a4ca9004bf0dce49473f71be6561834/Summary+ Ghana+P2P+Event.pdf?MOD=AJPERES. 3 The Borrowers & Lenders Act 2008.

66 Ghana legal framework, which was already comprised of a patchwork of statutes that provided for a variety of outdated, inefficient and practically cumbersome security devices that are inconsistent with one another.4 Furthermore, many of these Acts contemplated the registration of charges and other security devices, but the relevant registries were never established. Although the B&L Act and its Collateral Registry has increased access to credit primarily for individuals and micro businesses, it has failed SMEs. A comprehensive amendment to the B&L Act – in substance tantamount to a wholesale replacement – was drafted by the Bank of Ghana (BoG) and presented to the Deputy Governor of the BoG in November 2011. This amendment, as revised significantly in 2017, still awaits enactment. As of early 2019, Ghana can no longer be considered an inspiration for African reformers. Regardless, the Collateral Registry continues to perform an important educational function for future registrars in the region who visit its offices in Accra to learn about the steps, procedures, as well as challenges, in setting up and operating an electronic notice-based registry.

A. Ghana’s Economy Ghana is a lower middle-income economy in West Africa.5 Well endowed with natural resources, agriculture, industry and services are the mainstays of the economy, employing the entire labour force.6 Previously, agriculture was the dominant economic activity; but over the last decade the services sector has become increasingly important.7 The services sector is the largest economically, accounting for almost 60 per cent of the GDP, while industry and agriculture contribute 24.5 and 18.3 per cent, respectively.8 The main i ndustry is (gold) mining.9 Ghana’s top exports include gold, crude oil, cocoa and manganese ore.10 Ghana is Africa’s second largest gold producer11 – and gold constitutes 44 per cent

4 See C below. 5 World Bank, ‘World Bank Country and Lending Groups: Lower-Middle-Income Economies ($1,006 To $3,955)’ (2019) https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups. 6 Central Intelligence Agency (CIA), ‘Economy: Ghana’ (The World Factbook, 3 April 2019) www.cia.gov/ library/publications/the-world-factbook/geos/gh.html. In 2013, it was estimated that agriculture employed 44.7 per cent of the workforce, while services employed 40.9 per cent of the workforce. 7 JB Powell, ‘Factors Affecting Growth in Ghana’s Services Economy’ (US International Trade Commission, O ctober 2015) 1, www.usitc.gov/publications/332/executive_briefings/powell_ghana_ebot10-15-2015.pdf; Chatham House, ‘Financial Sector Development in Ghana: Enabling Efficiency and Broad-based Growth’ (14 July 2015) 2, www. chathamhouse.org/sites/files/chathamhouse/events/2015-07-14-financial-sector-development-Ghana-meetingsummary.pdf (summary of speech delivered at Chatham House by Kweku Bedu-Addo, Chief Executive Officer of Standard Chartered Bank Ghana Limited). 8 CIA (n 6). 9 ibid; Oxford Business Group, ‘Ghana Aims to Leverage Mining Revenues’ (14 March 2016) www.oxfordbusinessgroup.com/news/ghana-aims-leverage-mining-revenues. 10 ‘Gold, Cocoa and … Vegetables? Surprise Ingredient of Ghana’s Growing Economy’ CNN (2 February 2015) www.cnn.com/2015/01/30/africa/ghana-economy-vegetables/index.html. Ghana’s main export partners are Switzerland, India, the United Arab Emirates, China, and Vietnam. CIA (n 6). 11 R Jadhav, ‘Ghana’s Gold, Diamond Output to Drop as Govt Curbs Small-scale Mining’ Reuters (12 August 2017) www.reuters.com/article/ghana-gold-output/ghanas-gold-diamond-output-to-drop-as-govt-curbs-small-scalemining-idUSL4N1KY096. South Africa is the largest gold producer in Africa. ‘Ghana Maintains Rank as World’s 10th Biggest Gold Producer’ Ghana Web (30 March 2016) www.ghanaweb.com/GhanaHomePage/business/ Ghana-maintains-rank-as-world-s-10th-biggest-gold-producer-426993; ‘Mali’s Gold Production up by 0.4 Tonnes

Ghana’s Economy 67 of exports.12 Cocoa is the main cash crop: Ghana provides about 25 per cent of the global supply, supporting one million Ghanaian households.13 Ghana’s economy grew rapidly from the mid-1990s14 after the return to democratic rule,15 which, in large part, led to sound economic management and the creation of a competitive business environment.16 Between 1995 and 2014, economic growth dipped below 4 per cent only once.17 From 2010 to 2013, the economy averaged over 8 per cent economic growth per annum, making Ghana one of Africa’s most vibrant economies.18 In 2015 and 2016, economic growth declined further to 3.9 and 3.3 per cent respectively, due to low oil prices, the implementation of tight fiscal and monetary policies, technical issues related to oil production, high inflation, a ballooning public deficit and a weakening currency.19 As of 2018, top economic issues include a lack of affordable electricity, no solid domestic revenue base and a US$31 billion debt burden.20 The economy is susceptible to negative developments because of its reliance on commodity exports.21 From 2012 to 2016, Ghana’s GDP fell from 9.29 per cent to 3.72 per cent; in 2017, it increased to 8.44 per cent but was projected to fall to 6.29 per cent in 2018 and then to 5.48 per cent in 2020, 5.11 per cent in 2021, and finally to an all-time low of 3.19 per cent in 2022.22 The implementation of government industrial policies, such as the Industrial Sector Support Programme (ISSP, 2011–2015), has been crippled by the long-standing public sector resource crunch, the high cost of credit and limited access to start-up financing. The cost of access to finance is a particularly significant limitation, as the cost of borrowing23 for businesses and individuals stood at 28 per cent in 2016.24 The informal sector25 is in 2016’ Reuters (12 January 2017) www.reuters.com/article/mali-gold/malis-gold-production-up-by-0-4-tonnes-in2016-idUSL5N1F23MA. 12 ‘Gold, Cocoa and … Vegetables?’ (n 10). 13 S Knott, ‘Swapping Cocoa for Karats’ US News (21 July 2017) www.usnews.com/news/best-countries/ articles/2017-07-21/illegal-gold-mining-threatens-ghanas-lucrative-cocoa-industry. 14 The Commonwealth, ‘Ghana: Economy’, www.thecommonwealth.org/our-member-countries/ghana/economy. 15 The Constitution of 1992 was Ghana’s fourth attempt at creating a republican democratic government since its independence. As such, it is often referred to as the Fourth Republic. See Country Studies, ‘The Fourth Republic’, www.countrystudies.us/ghana/100.htm. 16 CIA (n 6). 17 See World Bank, ‘GDP Growth (annual %)’, https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG? locations=GH. The economy grew by just 3.7 per cent in 2000. This drop was attributed to a decline in the price of gold and cocoa, Ghana’s main export commodities. African Economic Outlook (OECD/African Development Bank, 2002) 155. 18 E Boh, ‘Ghana’s Economy in “Bad Shape” – President Akufo Addo’ Africa News (22 February 2017) www. africanews.com/2017/02/22/ghana-economy-in-a-bad-way-nana-akufo-addo//. 19 R Lal, E Okudzeto and K Sedegah, African Economic Outlook: Ghana (OECD/UNDB/African Development Bank, 2017) 264; Boh (n 18); ‘Ghana Country Profile’ (BBC, 1 May 2018) www.bbc.com/news/ world-africa-13433790. 20 CIA (n 6). 21 Lal et al (n 19); E Dontoh and MM Dzawu, ‘Ghana Finds a US$1.6 billion Hole in Budget’ Bloomberg (1 February 2017) www.bloomberg.com/news/articles/2017-02-01/ghana-vice-president-reveals-1-6-billion-undisclosed-spending. 22 Statista, ‘Ghana: Growth Rate of the Real Gross Domestic Product (GDP) from 2012 to 2022 (Compared to the Previous Year)’ (2019) www.statista.com/statistics/447479/gross-domestic-product-gdp-growth-rate-in-ghana/. 23 ‘Cost of borrowing’ refers to interest and other charges that have to be paid when borrowing. 24 E Rumney, ‘Ghana Should Use $2.2bn in Borrowing to Diversify Economy, Says ODI’ Public Finance International (19 April 2017) www.publicfinanceinternational.org/news/2017/04/ghana-should-use-22bn- borrowing-diversify-economy-says-odi. To curb inflation and strengthen the cedi (the Ghanaian currency), the Central Bank raised interest rates to 26 per cent, further stymieing the economy. Deloitte, ‘Africa from the Inside: Spotlight on Retail in Ghana’ (3 March 2016) https://www2.deloitte.com/content/dam/Deloitte/za/Documents/ strategy/ZA_AfricafromtheInside-%20GhanaRetail_290216.pdf. 25 That is, unregistered businesses.

68 Ghana responsible for 60 per cent of all employment in Ghana26 and comprises 62 per cent of all businesses,27 but only contributed a maximum of 30 per cent (in 2016) to the country’s GDP between 2013 (29 per cent) and 2017 (28.6 per cent). Accordingly, the high cost of borrowing negatively affects the growth of MSMEs.28 The high cost of loans affects the high default rates.29 In 2015, non-performing loans (NPLs) rose to 14.7 per cent;30 they decreased to 10.5 per cent in 2017 but rose again to 11.4 per cent in October 2018.31 Ghana’s banks32 are reluctant to lend to MSMEs, citing as an excuse the high rates of default, lack of verifiable credit history and the absence of adequate corporate governance structure.33 As of October 2018, the private sector accounted for 95.5 per cent of NPLs, whereas indigenous private enterprises were responsible for 73.2 per cent of total NPLs.34 The mining and quarrying sector received the least bank credit and, it follows, contributed the least to NPLs, at 1.4 per cent in October 2018.35 For want of better options to expand their businesses, Ghanaian MSMEs are often at the mercy of private (individual) lenders who charge exorbitant interest rates.36 This significant lack of access to credit was the impetus for the enactment of the B&L Act.37 However, the high cost of credit for Ghanaian businesses (discussed above) was not improved by the implementation of the Collateral Registry, as evidenced by the statistics from the BoG for 2011, which indicated that the average annual interest rate for loans secured with charges registered with the Collateral Registry was 40.17%. The registration form for Ghana’s Collateral Registry required an indication of the interest rate, since the BoG collected this information for statistical purposes. Current information is not available because the Collateral Registry’s Annual Reports no longer include information on interest rates. Moreover, the

26 P Nyarko (ed), Integrated Business Establishment Survey: National Employment Report (Ghana Statistical Service, September 2015) x. 27 ‘Economy Remains Largely Informal – GSS Report Confirms’ Ghana Web (14 February 2017) www.ghanaweb. com/GhanaHomePage/business/Economy-remains-largely-informal-GSS-report-confirms-509880#. 28 One 2013 survey found that SMEs account for around 90 per cent of all businesses in Ghana. Oxford Business Group, ‘Efforts to Improve Credit Access for SMEs’, www.oxfordbusinessgroup.com/analysis/efforts-improvecredit-access-smes. 29 See generally, AN Issahaku, ‘Reducing the Cost of Finance in Ghana’ (Bank of Ghana, 26 April 2016) 2–3, https://bog.gov.gh/privatecontent/Speeches/ADDRESS%20BY%20DR.%20ABDUL-NASHIRU%20ISSAHAKU,%20 GOVERNOR,%20AT%20THE%20SME%20FINANCING%20FAIR%202016,%20ON%20TUESDAY,%20 APRIL%2026,%202016.pdf. 30 Issahaku (n 29). BA McTernan, ‘Can Ghana Stop its Borrowing Binge?’ The Africa Report (29 August 2016) www.theafricareport.com/946/can-ghana-stop-its-borrowing-binge/. 31 ‘Banks Write-Off GH¢1.1 Billion as Bad Debt’ Ghana Web (11 January 2019) www.ghanaweb.com/ GhanaHomePage/business/Banks-write-off-GH-1-1-billion-as-bad-debt-714356#. 32 There were 37 licensed banks in Ghana as of August 2017. See Bank of Ghana, ‘Licensed Banks, Representative Offices & Their Registered Offices in Ghana’ (30 August 2017) www.bog.gov.gh/privatecontent/Banking_Supervision/LICENSED%20BANKS%20%20ADDRESSES_AUGUST%2030%202017.pdf. The number fell to 30 as of October 2018. A Ishaq, ‘BoG Releases Banking Sector Report’ Ghana Live (1 January 2019) www.ghanalive. tv/2019/01/01/bog-releases-banking-sector-report/. 33 See ‘Efforts to Improve Credit Access for SMEs’ (n 28). 34 ‘Banks Write-Off GH¢1.1 Billion’ (n 31). 35 ibid. 36 See BA McTernan, ‘Ghana’s Big Business and Informal Traders Alike Hit by Crumbling Economy’ The Guardian (29 August 2014) www.theguardian.com/global-development/2014/aug/29/ghana-business-traderseconomy-development. 37 IFC, Report on Modernization of the Secured Financing Law of Ghana (11 June 2008) 6 (on file with authors).

Pre-B&L Act Legal Framework 69 B&L Act did not have a positive effect on banks’ credit practices, as evidenced by 14.7 per cent of all bank loans being NPLs in 2015, almost 8 per cent higher than in 2008.38

B. Pre-B&L Act Legal Framework Ghana’s pre-B&L Act framework was (and remains) fragmented, cumbersome to apply, and supported by inefficient or non-existent registries. Unlike the Model Law, which is premised on a wholesale replacement of the pre-existing legislation, the B&L Act took a completely different approach, leaving the entire framework intact and even duplicating certain requirements; thus increasing the cost for parties to a secured transaction. After its adoption, the application of the various laws and their security devices continued to be determined based on whether the borrower was an individual, registered or unregistered company, or limited liability partnership. These borrower-type secured transactions laws were supplemented by transaction-type laws that govern sales of goods, hire-purchase agreements and finance leases. The following section lists the statutes making up this legal framework and comments on areas likely to give particular difficulty if they coexist with a modern secured transactions regime.

1. Customary Pledge The Ghanaian customary pledge entails ‘the delivery of a thing or chattel to a creditor as security for money advanced or due on condition of his restoring it to the owner after payment of the debt and subject to a conditional power of sale if the loan or debt be not paid at a certain specified time’.39 The customary pledge is still used in some Ghanaian rural communities.40

2. The 1952 Chattels Transfer Ordinance The Ordinance to Make Provision Relating to Chattel Securities and the Transfer of Chattels (Ordinance) was enacted in 1952 to apply to ‘instruments that secure the payment of money or the performance of some obligation and any documents that transfer or purport to transfer the property in or right to the possession of chattels’.41 This Ordinance was never brought into force. It is not atypical for African jurisdictions to have enacted 38 A Kovanen, ‘Monetary Transmission Policy in Ghana: Does the Interest Rate Channel Work?’ (November 2011) IMF Working Paper WP 11/275, 7–8. 39 See JM Sarbah, Fanti Customary Law (London, William Clowes and Sons, 1879) 83; and Adjei v Dabanka [1930] IWACA63, where the West African Court of Appeal held that it ‘is an essential characteristic of a pledge for the creditor to be placed in possession’; and NA Ollennu, Principles of Customary Land Law in Ghana (London, Sweet & Maxwell, 1962) 95. 40 M Dubovec and B Osei-Tutu, ‘Reforming Secured Transactions Laws in Africa: The First African Collateral Registry in Ghana’ (2013) 45(1) UCC Law Journal 77. See also Constitution of Ghana 1992, Art 11 of which provides that ‘[c]ustomary law is part of the laws of Ghana’. 41 The Ordinance to Make Provisions Relating to Chattel Securities and the Transfer of Chattels 1952, s 2.

70 Ghana laws providing for some type of a security device and even a registration system, but then fail to operationalise them. The proposed amendments to the B&L Act will repeal the Ordinance.

3. The 1960 Contracts Act Section 7 of the 1960 Contracts Act governs the assignment of legal rights. It appears that an assignment can be absolute or by way of security. This is because subsection 2, which sets out the criteria for an assignment that can be enforced by the assignee in its own name,42 states that it applies to an assignment that is ‘absolute and not by way of charge only’. This suggests that there can be an assignment not falling within subsection 2, which is by way of charge.43 Section 7(5) deals with priority of two competing assignments, and provides that the assignee second in time will have priority if the assignee first in time had not given notice of its assignment to the debtor at the time the second assignment is taken. This provision is strange, since it does not even replicate English law,44 and certainly does not comply with the requirements for a modern secured transactions law, where priority is determined by order of registration.45 None of this Act has been repealed by the B&L Act.

4. The 1962 Sale of Goods Act This Act is clearly based on the UK Sale of Goods Act 1893.46 Under section 26 of the Sale of Goods Act, ownership to goods passes when the parties intend it to pass.47 The parties may thus agree that ownership will pass only once the buyer has paid the purchase price in full, that is, a sale on retention of title terms. The seller has some protection against nonpayment, even without agreement. While the seller is still in the possession of the goods, he has a lien to secure payment of the price,48 and a right to stop them in transit if the buyer becomes insolvent.49 The Act also includes various exceptions to the nemo dat rule (that a

42 This is the equivalent of s 136 of the Law of Property Act 1925 in English law. 43 In relation to the similarly worded English law section (s 136 of the Law of Property Act 1925), a mortgage of a debt can fall within the section as it involves an ‘absolute’ assignment (Tancred v Delagoa Bay and East Africa Railway Co (1889) 23 QBD 239), but a charge does not. For discussion of the English law position, see L Gullifer (ed), Goode and Gullifer on Legal Problems of Credit and Security, 6th edn (London, Sweet & Maxwell, 2017) para 3-09. 44 Priority under English law is determined by the rule in Dearle v Hall (1823) 3 Russ 1, namely, by the order in which notice is given to the debtor, with the exception that, irrespective of the order of notice, the first assignee will have priority if the second assignee knew of the first assignment at the time the second assignee took its assignment. If no notice has been given to the debtor, the priority rule is first in time. 45 ML, Art 29. See also the draft Borrowers and Lenders Bill 2017, s 27. 46 Now replaced by the Sale of Goods Act 1979. 47 This precisely replicates s 17 of the 1893 Act. 48 s 36, which replicates s 39 of the 1893 Act (now s 41, 1979 Act). 49 s 38, which replicates s 44 of the 1893 Act (now s 44, 1979 Act).

Pre-B&L Act Legal Framework 71 person cannot pass a title it does not have). These include a provision enabling a seller in possession after property has passed to the buyer to pass the title of his buyer to a good-faith purchaser or pledgee without notice,50 and a similar provision enabling a person who has agreed to buy goods who obtains possession before property has passed to pass the title of his seller to such a purchaser or pledgee.51

5. The 1962 Incorporated Private Partnerships Act The Incorporated Private Partnerships Act provides for the incorporation and registration of partnerships, which are defined as ‘the association of two or more individuals carrying on business jointly for the purpose of making profits’.52 Section 5 requires registration of a partnership agreement, as well as a statement signed by all partners that indicates, inter alia, the particulars of any current charges against the partnership or a covenant that the partnership has not created any charges in its assets. As well as its ability to create mortgages and fixed charges, under section 21, a partnership may create a floating charge over the whole or part of its assets.53 With respect to registration of charges, the Act mirrors the provisions of the 1963 Companies Act that apply to registration of charges by companies. The partnership may continue dealing with its assets in the ordinary course of business until one of several specified events occurs,54 which causes the floating charge to crystallise, that is, becomes a fixed charge.55 The particulars relating to a charge (Form 9) must be registered within 28 days of its creation.56 Form 9 must be accompanied by the original or certified copy of the charge instrument, which is returned to the registrant after the registration process has been completed.57 The Registrar-General’s Department, under the Ministry of Justice and Attorney General, administers and operates the registry of partnerships. Charges over a partnership’s assets may be registered electronically.58 An electronic registration is completed upon approval of the Registrar-General’s Department and payment of the registration fee, which may be made online or at a bank designated by the Registrar-General’s Department. The provisions on the registration of charges under this Act were not repealed by the B&L Act.

50 s 31, which replicates s 25(1) of the 1893 Act (now s 24, 1979 Act). 51 s 32, which replicates s 25(2) of the 1893 Act (now s 25, 1979 Act). 52 See Ghana’s the Incorporated Private Partnerships Act 1962, s 3(1). 53 C6 below and ch 1, E1b for further discussion of the floating charge. 54 The appointment of a receiver or manager by a chargee or the Court, the chargee going into possession or the firm going into liquidation. 55 Incorporated Private Partnerships Act 1962 (n 52) s 21(2)–(3). 56 ibid s 22. 57 ibid s 27(4). 58 ‘RGD Portal User Guide – Online Filing’ (Registrar General’s Department) www.egovonline.gegov.gov.gh/ RGDPortalWeb/ShowPropertyServlet?nodePath=%2FUCMRepository%2FContribution+Folders%2FRGD% 2FServices%2Ffileresolution%2FRGD+Portal+User+Guide+-+Filing+002.pdf&_pageLabel=portal_ RGDHome_ServiceLogin_page.

72 Ghana

6. The 1963 Companies Act The 1963 Companies Act is currently being reviewed with the objective of reforming it in the near future. Unfortunately, this reform process has not been coordinated with the amendments to the B&L Act. In Part L, the Companies Act provides the rules for the registration of charges.59 Section 107(1) states that every charge is subject to registration and the failure to do so within 28 days from its creation renders the charge void, and the debt immediately payable.60 Charges created by companies may be registered electronically through the RegistrarGeneral Department’s online registration portal.61 To register a charge, a company must enter certain details into the relevant section of the portal, such as the type of instrument (charge, mortgage or debenture), the date of the instrument, the amount of the secured obligation, and information about the chargee and the chargor.62 The particulars of the charge (Form 9) must be uploaded to the portal. As with the registration of a charge created by a partnership, registration of a charge over a company’s assets is not complete until it is approved by the Registrar-General and the registration fee is paid.63 The registration fee may be paid online or at a bank designated by the Registrar-General’s Department.64 Prospective secured creditors may undertake searches of the Companies’ Registry.65 However, as is the case with partnerships, there is no clear procedure for searching the Companies Registry. A registration may be discharged by filing Form 10 – Declaration Verifying Memorandum Relating to a Registered Charge.66 As discussed below, none of the provisions in this Act on registration of charges have been repealed by the B&L Act. Section 86 of the Companies Act provides that debentures can be secured by fixed and floating charges, and section 87 defines a floating charge. When a floating charge becomes enforceable, the court can appoint a receiver and manager of the company,67 and section 89 provides for priority of preferential creditors in relation to assets subject to a floating charge if a receiver is appointed or if possession is taken of charged assets.68 This is also the case if the company is liquidated.69

59 Companies Act 1963. 60 ibid ch 1, s 107(3) provides limited exceptions from the registration requirement that relate to possessory security interests in goods (pledges, possessory liens) or any charge by way of pledge, deposit, letter of hypothecation or trust receipt, of various instruments (bills of lading, dock warrants or other documents of title to goods, or of bills of exchange, promissory notes or other negotiable securities or money). 61 See ‘Change Particulars’ (Registrar General’s Department) https://egovonline.gegov.gov.gh/RGDPortalWeb/ portal/RGDHome/eghana.portal?_nfpb=true&_st=&_pageLabel=portal_RGDHome_ServiceLogin_page&linksP ageTitle=Change+Particulars&service=ChangeInParticulars#wlp_portal_RGDHome_ServiceLogin_page. 62 ‘RGD Portal User Guide – Online Filing’ (n 58) 9–18. 63 ibid, 16–18. 64 ibid. 65 Oxford Business Group, The Report: Ghana 2016 (2016) 252. 66 ‘Form Pack – Change in Particulars of Company (All Categories)’ (Registrar General’s Department) https:// egovonline.gegov.gov.gh/RGDPortalWeb/portal/RGDHome/eghana.portal?_nfpb=true&_st=&_pageLabel=rgd_ portal_fomspack_content&formId=5203#wlp_rgd_portal_fomspack_content. 67 Companies Act 1963, s 88. See ch 1. 68 See ch 1. 69 s 41 of the Bodies Corporate (Official Liquidations) Act 1963.

Pre-B&L Act Legal Framework 73

7. The 1974 Hire-Purchase Decree Hire-purchase agreements may be used by individuals and businesses to acquire consumer goods and equipment.70 Section 3(1) of the Decree sets out the formal requirements for a hire-purchase agreement and conditional sales.71 One of the requirements is a description or list of the goods to which the agreement relates sufficient to identify them, but there is no further requirement for a specific or other detailed description. The seller’s or owner’s ownership rights to goods in the possession of buyers or hirers are effective against third parties without any form of public notice. The remedies available to the seller or owner are limited, as compared to the Model Law. For instance, repossession of protected goods72 may be effected only pursuant to a court order.73 Additionally, under section 7(3) the seller or owner may not dispose of the goods until the expiration of a period of 28 days after repossession, in order to give the buyer or hirer the opportunity to exercise its right in section 7(1) to complete the purchase by payment of the price.

8. The 1993 Finance Lease Act Section 21 of the Finance Lease Act defines a finance lease as ‘an agreement under which the lessor shall retain full title to the asset and the lessee may exercise an option to purchase the asset at the expiration of the leased period’.74 The Act requires the lessee to acknowledge in the finance lease agreement that it will hold the asset as a bailee and will not claim any property right thereto.75 The lessee has no right to return the asset prior to the lease expiration and thus cannot terminate the agreement.76 Although section 6 authorises the BoG to designate a registering authority for finance leases, no entity has been so appointed. This is therefore another piece of Ghanaian legislation that contemplates the establishment of a registration system, but for which no such system has been put into operation. An unregistered finance lease remains valid between the parties, but not against third parties acting in good faith.77 Neither this Act, nor the Hire Purchase Act, was repealed or amended by the B&L Act.

70 Ghana has a number of hire-purchase finance companies, such as Goldman Capital, F& D Finance and Leasing Limited, Consumer Finance Company (CFC) Ghana, Quicken Microfinance, and Syndicated Capital. Many household goods suppliers also offer hire-purchase packages for the acquisition of consumer and household goods. These include, among others, City Hire Purchase, Credit Mall Limited, Home and Family Support Ltd, Gentrac Investments Ltd and Rowi Limited. 71 A conditional sale is a sale where the purchase price is payable by instalments and title does not pass until the price is paid, and a hire-purchase agreement is an agreement for a bailment where the hirer has the option to buy the goods. See Hire-Purchase Decree 1974, s 24. 72 ‘Protected Goods’ is defined in s 8(4) and are goods in respect of which half the price has been paid or tendered. This protection mirrors that provided in other common law countries – see, eg, Nigeria B5, Zambia C5 – and reflects similar provisions in the UK Consumer Credit Act 1974. 73 Hire-Purchase Decree 1974, s 8 provides ‘[t]he owner or seller shall not enforce any right to recover possession of protected goods from the hirer or buyer otherwise than by action’. 74 Finance Lease Act 1993. 75 ibid s 2(1)(d). 76 ibid s 2(1)(f). 77 ibid s 6(4).

74 Ghana

C. The Borrowers and Lenders Act This section examines the B&L Act of 2008 from the perspective of the Model Law. As noted earlier, the B&L Act was comprehensively reviewed, resulting in a proposal for a major overhaul (hereinafter ‘the 2017 Bill’).78 Some relevant provisions from the Amendments will be examined below to illustrate the change in approach and how the consequences of this differ dramatically from the current B&L Act. The B&L Act is comprised of 38 sections and a schedule that includes a ‘pre-agreement truth in lending disclosure statement’. Given the broad range of issues that it covers, it is not a typical secured transactions law. First, it includes extensive coverage of regulatory matters such as the supervisory and enforcement role of the BoG79 and borrowers’ rights in relation to discrimination, confidentiality and disclosure of information.80 The Model Law deals only with rights and duties that are commercial in nature and that relate to security rights. In contrast, the matters governed by sections 6–19 of the B&L Act are regulatory in nature and do not necessarily deal with any aspects of securing obligations; rather, they touch on broader issues relating to access to credit. Second, modern secured transactions laws are structured around the core concepts of creation, third-party effectiveness, registration, priority and enforcement of security rights. The B&L Act addresses (in a highly unsatisfactory manner) only registration and enforcement. Leaving aside the 14 ‘regulatory sections’, the B&L Act is merely a skeleton of a secured transactions statute and is rife with significant gaps. The entire approach of the B&L Act is further muddled by the inclusion of provisions that purport to apply to unsecured undertakings, such as guarantees. Section 1 outlines the scope of the B&L Act as applying to ‘credit agreements’ and section 2 sets out a list of credit agreements, including credit guarantees. Section 5 defines credit guarantee as ‘an agreement whereby a third party undertakes or promises to satisfy on demand an obligation of a borrower’, essentially implying that the B&L Act applies to an unsecured promise by a third party to repay a debt that does not create any interest in the property of the borrower or the guarantor. Moreover, ‘borrower’ is defined in section 38 as ‘a person who has entered into a credit agreement with a lender’ and the term ‘credit agreement’ includes a credit guarantee. Thus, the protective ‘borrower rights’ in sections 13–20 apply to a guarantor as well as to a borrower. However, close scrutiny of the provisions of the B&L Act reveals that it does not include the regulation of many aspects of guarantees that one would expect to find in a law on guarantees. These and many other issues, examined in the following subsections, have plagued the uptake of the B&L Act and continue to present challenges to this day. In 2011, while interviewing financial industry stakeholders in Accra, one of the authors learned that a number of financial institutions do not register their charges covered by the B&L Act. One interviewed bank blamed technological problems, others interpreted the B&L Act as not requiring the registration of finance leases, and a group of non-bank institutions stated that they do not trust the Collateral Registry.81 These challenges were documented and formed 78 See Borrowers and Lenders Bill 2017 (‘2017 Bill’), available at https://www.bog.gov.gh/privatecontent/Public_ Notices/DRAFT%20BORROWERS%20AND%20LENDERS%20BILL.pdf. 79 See ss 6–12 of the B&L Act. 80 See ss 13–19 of the B&L Act. 81 Dubovec and Osei-Tutu (n 40).

The Borrowers and Lenders Act 75 the basis of the argument to reform the B&L Act. The BoG was an early advocate of the need for reform and continues to play a critical role in interacting with stakeholders and formulating the Amendments. The Amendments were extensively discussed during the summer and autumn of 2011 with the Collateral Registry Reform Committee of the BoG and a group of local stakeholders. These discussions continued over the years and culminated in a set of intensive meetings, including a large stakeholder workshop held in Accra in 2017. The 2017 Bill includes a number of fundamental changes inspired by the Model Law, and strive to achieve conceptual clarity and precise drafting, as well as to fill the gaps in the B&L Act. From a structural standpoint, the 2017 Bill is divided into several chapters, including rules addressing the individual aspects of a secured transaction; namely, the scope, creation, third-party effectiveness, registration, priority, enforcement, and conflict of laws. The following subsections critically consider the B&L Act and make some comparative comments about the 2017 Bill.

1. Scope The challenges presented by the B&L Act begin with the provisions on its scope. The B&L Act applies to ‘a credit agreement between parties who deal at arm’s length’ and ‘a credit guarantee where the guarantee is only in respect of a credit facility or credit transaction covered by this Act’.82 A credit agreement is defined in section 2 as an agreement in the nature of a credit facility, a credit transaction, a credit guarantee or any combination of these. Sections 3–5 then specify the meaning of credit facility, credit transaction and credit guarantee. A credit facility is basically defined as a loan. The meaning of credit guarantee was already examined above.83 Section 4 defines a credit transaction to include: (i) a pawn transaction, (ii) an instalment agreement, (iii) a mortgage or secured loan, and (iv) a finance and operating lease. Although a credit transaction is defined to include a secured loan, the substance of the definition of credit facility clearly also includes a secured loan. Setting out the scope through a combination of multiple definitions, which are ambiguous and overlapping, makes it very complicated for the user to determine whether the B&L Act applies to a particular transaction. Furthermore, the B&L Act does not provide rules specific to credit facilities, credit transactions, credit guarantees, and their subtypes, except for the generic borrower protection provisions in sections 13–20, which can apply to unsecured credit as much as to secured credit. Therefore, defining these credit products is unnecessary. Instead of focusing on specific types of credit products, the B&L Act should have set out the scope by focusing generally on obligations whose performance is secured with some collateral, as in Article 7 of the Model Law. For sections 21–31, 32(4) and 33(b)–35 of the B&L Act to apply, credit agreements must be secured by a charge. A charge is defined in section 38 as a ‘charge, mortgage, security, interest, lien, pledge, assignment by way of security, covenant, restriction, reservation, lease, trust, order, decree, judgment, title defect (including retention of title claim), or any other 82 Under s 1(1)(a), credit agreements covering an amount of less than 100 Ghana cedis or an amount determined by the BoG, and any other credit agreement that may be exempted by the BoG are excluded from the B&L Act’s scope of application. 83 Introduction to D below.

76 Ghana encumbrance of any nature other than liens arising by operation of law’. This definition raises a number of questions. First, it refers to claims without specifying whether they affect either movable assets, immovable assets or both. For instance, ‘covenants’ and ‘restrictions’ do not usually secure obligations and their inclusion may well relate to immovable assets only. The intention behind requiring the registration of covenants and restrictions in the Collateral Registry is unclear, especially since the Companies Act does not include similar provisions in connection with the registration of charges in land in the Companies Registry. The 2017 Land Bill contemplates the registration of restrictive agreements affecting enjoyment of the land with the Land Registrar.84 Second, it covers other property rights that do not necessarily secure obligations, such as a trust, which may also be inspired by the land law position. Third, it applies to nonconsensual rights in property, such as orders, decrees and judgments, but excludes liens arising by operation of law. Finally, it does not cover outright transfers of receivables, in contrast to the Model Law. In all these aspects, the B&L Act departs from the Model Law, creating a conceptually unsound framework that is difficult for local lenders to understand and deploy in practice. During the interviews that one of the coauthors conducted in Accra in 2011, a number of lenders expressed confusion as to whether the B&L Act applies to operating leases; and if so, whether to all of them irrespective of their duration. There was also uncertainty as to whether operating leases would be subject to registration in the Collateral Registry. Many lenders were also unsure of the application of the B&L Act to a common Ghanaian arrangement in which a loan given to an individual is guaranteed by a family member whose guarantee is secured by the assignment of wages.85 Another limitation in the B&L Act’s scope relates to the question of who can act as a lender. Section 38 defines a lender as ‘a person who, as part of business, advances loans and other credit facilities including micro credit facilities’. Although section 38 could in theory cover a credit sale in the course of a supplier’s business, in fact, the Collateral Registry can only be accessed by financial institutions regulated by the BoG. Further, the regulatory sections of the B&L Act could not possibly apply to sellers on credit. That this understanding of the limitation was correct was confirmed during the deliberations of the 2017 Bill that extends the scope to any type of borrower and lender, but limits the application of the regulatory provisions only to lenders supervised by the BoG.86 This lender-specific orientation of the B&L Act further contributes to the fragmentation and complexity of the Ghanaian secured transactions framework. One of the unusual features of the B&L Act is its application to charges over immovable property; section 38 defines property to include movable and immovable property. Accordingly, charges over immovable property are registrable in the Collateral Registry. Presumably, the drafters took inspiration from the Companies Act, which also contemplates the creation and registration of charges over immovable property. However, the legal effect of registering such a charge in the Collateral Registry on the concurrently applicable requirements of the land law was not defined, particularly as no changes were made to land law as a result of the reforms. As the underlying policy of the B&L Act is to leave the provisions

84 The

Land Bill 2017, s 152. and Osei-Tutu (n 40). 86 See, eg, s 63 of the 2017 Bill. 85 Dubovec

The Borrowers and Lenders Act 77 of other laws unaffected, it would seem that a charge over immovable property requires registration in the land registry. Section 62 of the Land Title Registration Act clearly states that ‘instruments recorded in the Land Register shall have priority according to the order in which the instruments were presented to the Registry irrespective of the dates of the instruments’. Accordingly, registration of a charge relating to land in the Collateral Registry is a validity requirement, but its priority is established by the time of presentation of the instrument to the Land Register. In 2016, of the 43,504 registrations completed in the Collateral Registry, 7,670 identified some immovable collateral.87 In 2018, 4.1 per cent of all assets identified in registrations were immovable. The two numbers differ because the 2018 statistics aggregate the number of collateral types from a drop-down list. Many registrations select multiple types of collateral. The 2017 Bill introduces a number of changes to the scope provisions, including the redefinition of the notion of a charge (and its replacement with the term ‘security interest’) in broad accordance with the Model Law’s concept of a security right, as well as the elimination of guarantees. Although not entirely clear, it appears that, applying the scope provisions in sections 1 and 2, and the definition of ‘security interest’ in section 76, the 2017 Bill will apply to a finance lease, which begs the question of whether some or all of the Finance Lease Act88 will be impliedly repealed.89 The 2017 Bill also fashioned an innovative set of rules governing security interests in immovable property. Under section 11, registration of a security interest over an immovable in the Collateral Registry will make it effective against other lenders and insolvency representatives. Section 38 also establishes a priority point for the registration in the Collateral Registry as against any competing interest in the immovable made effective against third parties under other laws. So, a first-in-time registration in the Collateral Registry shall have priority over a later-in-time registration in the Land Registry. However, buyers for value will take free of the security interest under section 37(3), if the lender has not registered its interest in the immovable asset under the applicable land law. If enacted, the amended B&L Act will be the first of its kind in Africa to apply to security interests in both movable as well as immovable property, contemplating the registration of a notice with respect to security interests in immovable property in the Collateral Registry. Such a registration would have the legal effect of establishing priority. Sierra Leone is pursuing an overhaul of its own Borrowers and Lenders Act to apply to security interests in immovables as well.

2. Creation The B&L Act does not include any provisions whatsoever on the creation of a charge, so that, presumably, the relatively liberal rules of the common law still apply – namely, that a charge can be created by any form of agreement provided that the subject matter is identified, although this can be done in a very general way.90 There are no statutory rules that provide certainty as to the requirements to constitute a charge, such as those providing

87 Annual

Report of the Collateral Registry on file with the authors. above B8. 89 There appears to be no express repeal. 90 See Gullifer (n 43) 2-03 et seq. 88 See

78 Ghana a standard for the description of encumbered assets and secured obligations, nor is there provision for the continuation of a security right into the proceeds, or an override of antiassignment clauses, which exist in the Model Law and the PPSAs. Section 18 requires the lender to provide to the prospective borrower a pre-agreement disclosure statement in the form set out in the Schedule to the B&L Act, before concluding a credit agreement. However, this duty, and that in section 19 relating to marketing information, appear to apply both to secured and unsecured credit, and are really a matter of borrower protection: they are not matters that relate to the creation of a charge, and the lender’s failure to do so has no effect on the validity of the charge, although there is an administrative sanction of a fine provided for in section 12.91 The 2017 Bill, which is based on the Model Law,92 contains a separate part on the creation of a security interest, including provisions on the minimal requirements for a credit agreement, the standards for describing collateral and secured obligations in credit agreements, as well as on the continuation of a security interest into proceeds.93 The 2017 Bill also reflects a different policy to the B&L Act with respect to the effect of anti-assignment clauses. Section 7 follows the Model Law94 closely, and provides that a security interest in an account receivable95 or a bank account is effective notwithstanding any anti-assignment clause.

3. Third-Party Effectiveness The B&L Act lacks any rules expressly dealing with third-party effectiveness as understood under the Model Law. Section 25(1) requires ‘a borrower or a person interested in a charge to register a certified copy of a charge or collateral created by the borrower … within twentyeight days of the creation of the collateral or charge’,96 and section 25(3) provides that an unregistered charge is ‘of no effect as security’. This provision recognises registration as the sole method of third-party effectiveness under the B&L Act, even where possession is taken by the secured creditor, since the definition of ‘charge’ includes pledges and liens, which are possessory security interests.97 This is unusual in comparison to both reformed and unreformed systems generally, although not unique in Africa.98 The 2017 Bill contains a separate part on third-party effectiveness, but the sections included therein continue to recognise registration as the sole method of third-party effectiveness. The other alternatives, including control, possession and automatic effectiveness, were rejected by the working group. The discussions with stakeholders confirmed this position, as lenders generally preferred to rely on the public records found in the Collateral Registry, and did not see any utility in possession or control. Registration in the Collateral Registry will also be the sole method of third-party effectiveness for security interests in immovable property. 91 s 18(3) of the B&L Act authorises the BoG to impose an administrative fine. 92 ML, Arts 6–10. 93 2017 Bill, ss 3–6. The Amendments continue to refer to credit agreement rather than the Model Law’s security agreement, although the definition of ‘credit agreement’ in s 3 is limited to a transaction that creates a security interest. 94 ML, Art 13. 95 Unfortunately, this term is not defined, unlike in the Model Law. 96 See C4 below or our analysis of the language in this subsection, especially ‘registration of collateral’. 97 B&L Act, s 38. 98 Kenya C3, Sierra Leone C3, Zimbabwe C1.

The Borrowers and Lenders Act 79 Unlike the Model Law, neither the B&L Act nor the Amendments permit registration in advance of creation of the security interest. This means that a lender cannot protect its priority position by registration before the security interest is created, although it can, of course, refrain from advancing any funds until the registration is completed and a search has confirmed that there are no competing security interests. This provision has been inserted at the suggestion of the BoG to ensure that regulated financial institutions properly protect themselves against the risk of extending credit before their security interest is properly perfected. It also means that, since reservation of title sales are included in the definition of ‘security interest’ in section 76 of the 2017 Bill, each one of such agreements will have to be registered separately, which could be onerous. Unlike the Model Law, the B&L Act does not distinguish clearly between creation and third-party effectiveness. Under section 25(3), the failure to register a charge within 28 days of creation causes the charge to be of ‘no effect as security … and the money secured shall immediately become payable’. As mentioned above,99 the second part of this provision mirrors English law, and originates in the fact that the primary obligation to register was on the debtor company, although it has been retained in order to incentivise the debtor to agree to the grant of a new charge if the first one is not registered.100 In drafting the 2017 Bill, questions were raised as to the effect of the provision that an unregistered charge was ‘of no effect’ and why it was needed. The BoG argued that the 2017 Bill should reinforce the lender’s obligation to register notices of charges by preserving the requirement of registration within a specified timeframe, with the same consequences for a failure to do so. The BoG’s objective was to ensure that regulated financial institutions extend credit in a prudential manner, that is, for such institutions to make their security interests effective against third parties properly before extending loans. After lengthy discussions, section 16 included in the 2017 Bill imposes an obligation on lenders to register a security interest before they disburse credit or within 28 days after the creation of a security interest. Although the failure to do so has no effect on the validity of the security interest between the parties, it implicates the BoG supervisory powers that may levy an administrative penalty, but only on those lenders subject to its supervision. The failure to register does, however, render the security interest ineffective against third parties, which appears to include an insolvency administrator.101 Another concern was with the enlarged scope of application of the B&L Act after the enactment of the Amendments, which would allow anyone to act as a lender. The Collateral Registry Department accepted that it could not effectively ensure that registrations will be completed on time by unregulated lenders. Eventually, the working group settled on an approach whereby a security interest that has not been made effective against third parties could not be enforced extrajudicially.102 This approach was mainly informed by the prudential regulatory concerns of the BoG. The potential (and highly likely) risk of losing priority to a later-in-time lender for failure to register a notice was rejected as an insufficient motivation for completing registrations. Although there are no sound commercial or conceptual reasons to condition availability of extrajudicial remedies on

99 See

ch 1, E1c. ibid. 101 2017 Bill, s 8. 102 ibid s 52(1)(b). 100 See

80 Ghana the completion of registration in the Collateral Registry, the Ghanaian experience illustrates some profound lessons. One such lesson is that the involvement of central banks in reform activities and hosting collateral registries introduces a different perspective that working groups and funding agencies must consider.103 In Ghana, the desire to ensure that regulated financial institutions extend secured credit in compliance with the BoG’s prudential regulations coloured the discussion on the third-party effectiveness and enforcement rules of the 2017 Bill. As mentioned above, the B&L Act neither replaced any pre-existing legislation nor affected already existing registries, including those for companies and partnerships. In the context of third-party effectiveness of charges, the relationship between the Collateral Registry and the Companies Registry, operated by the Registrar-General’s Office under the Ministry of Justice and Attorney General, is very problematic. Section 25(2) of the B&L Act provides that ‘where a charge is created by a company, the requirement to register charges with the Collateral Registry under this Section is in addition to the requirement under Section 107 of the Companies Act, 1963 (Act 179) to register charges with the Registrar of Companies’. These provisions clearly require double registration of charges for borrowers that operate as registered companies under the Companies Act. Rather than filling some gap, this misguided approach of the B&L Act increases the burden on parties to a secured transaction. From the process standpoint, duplication of records increases transactional costs and elevates the risk of error, as well as complicating searching. From the legal standpoint, section 25(2) generated multiple questions that were discussed by the working group and were used to persuade the relevant stakeholders of the need for reform. For instance, what is the legal effect of a charge that has been registered in only one of these registries? It would appear from the wording of section 25(3) of the B&L Act and section 107(1) of the Companies Act that such a charge would be invalid since each provide that this is the consequence of non-registration in the relevant registry. What if one of the registrations includes information not reflected in the other registration or both registrations are inconsistent with one another in some other fashion? In addition to these questions on perfection of charges, one could envisage numerous scenarios with respect to the priority of charges. Suppose that Bank A registers in the Collateral Registry on Day 1, but in the Companies Registry on Day 5, and Bank B registers in the Companies Registry on Day 1 and in the Collateral Registry on Day 5. How would priority be determined? Neither the B&L Act, nor the Companies Act, provides a clear answer. One answer could be that priority is determined by the general law (eg order of creation) since neither Act purports to govern priority issues. However, the Rules for Effective Implementation of the Borrowers and Lenders Act, adopted in 2012 (Registry Regulations),104 provided that registration in the Collateral Registry was a priority point.105 Once this is the case, the double registration requirements raise difficult issues of priority where Bank B’s charge was created before that of Bank A: under the Registry Regulations Bank A would win, whereas under the general law Bank B would win. However, regulations issued by a central bank are subsidiary to any law, not limited to the statute that they purport to implement, so the

103 See

conclusion C5. below. 105 Rules for Effective Implementation of the Borrowers and Lenders Act 2012 (Registry Regulations), s 16. 104 See

The Borrowers and Lenders Act 81 priority provision of the Registry Regulations arguably has not altered any priority rule established by the general law. Regardless of these legal questions, the operational design of the Companies Registry effectively renders the Collateral Registry system inoperative for the registration of charges created by companies in the Collateral Registry. Although the Companies Registry operations are not entirely paper based, as of February 2019 the online portal for registering charges did not indicate that electronic search services were available.106 One of the coauthors visited the Companies Registry in 2013 and was told that on average the Companies Registry receives 20 submissions per month for the registration of a charge, but is forced to reject over half of them because it cannot locate the file of the company that purportedly granted the charge. In essence, due to the poor indexing and storage of records relating to companies, the Companies Registry could not determine whether a company formally existed or not. By leaving untouched the requirements of the Companies Act, the B&L Act rendered the registration of companies’ charges in the Collateral Registry meaningless. The 2016 Annual Report of the Collateral Registry indicates that of the 43,504 registrations, 34,056 were against individuals and 5,681 against SMEs, many of which were expected to be companies.107 This illustrates that the Collateral Registry primarily facilitates credit for consumers and borrowers that do not operate as formal enterprises, especially individual entrepreneurs operating as microbusinesses. This aspect of the relationship between the two registries is addressed in the 2017 Bill, though not satisfactorily. First, section 79(1)108 does not repeal any provisions of the Companies Act, which would have been a straightforward solution and consistent with the Model Law and similar reforms elsewhere.109 Secondly, the relationship is addressed through section 77, which provides: Except as otherwise provided in this Act, the provisions of this Act on effectiveness of security interests against third parties, registration, priority and enforcement shall, on the commencement of this Act, supersede any provisions in existing enactments applicable to security interests in movable and immovable property.

The effect of this provision on the Companies Act is not entirely clear. If repeal of the provisions on charges in the Companies Act had been intended, then they would be identified in section 79. It appears, then, that the Companies Act continues to apply. The uncertainty is increased by leaving it up to the judges to determine which provisions of the Companies Act, as well as other affected legislation, are to be superseded: an approach that is inferior to indicating the provisions to be repealed.110 It also means that the floating charge would retain a role in a system that purports to implement the functional approach.111 This means that the difficult distinction between 106 See Registrar General’s Department, www.rgd.gov.gh/. 107 Report is on file with the authors. 108 This section is mis-numbered as 77 in the 2017 Bill. 109 For example, Malawi, see M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in L Gullifer and O Akseli (eds) Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016) 183, and Zambia, see Zambia C. 110 See also conclusion C6a. 111 See also s 76 of the 2017 Bill which specifically includes floating charges in the definition of ‘security interest’. For more discussion of the problems that retaining the concept of the floating charge can cause, see ch 1, and Kenya B4 and C9, Nigeria E1.

82 Ghana a fixed and floating charge would continue to have to be drawn112 in order to determine whether a receiver and manager could be appointed,113 and also to determine over which assets preferential creditors had priority if a receiver is appointed or the company is wound up.114 As in other African countries,115 this reform approach presents a number of legal risks that are expected to increase the cost of credit. The Companies Act has been under review for many years, and introducing consequential amendments of a material nature would significantly delay the reform of the B&L Act. An earlier draft provided that a charge may be registered in either the Collateral or Companies Registry; but a registration in the Collateral Registry is superior, irrespective of the time of the competing registrations.116 Accordingly, the relationship between the Collateral and Companies Registry would have been addressed as a question of priority, as a secured creditor who seeks to establish priority would register its security right over a company’s assets in the Collateral Registry. This solution, though not ideal, would appear to create fewer risks and ambiguity than the current draft, especially as to which provisions of the Companies Act would be superseded on the account of being inconsistent with the enactment of the 2017 Bill.

4. The Collateral Registry and Regulations The B&L Act contains a Part dedicated to the Collateral Registry and its various functions. Section 21 specifically provides the legal grounds for the establishment of the Collateral Registry. This Part also contains provisions that do not relate to registration. For example, section 29 provides for the appointment of a receiver or manager, although this is followed by section 30 which deals with the procedure for registration of, and the giving of notice of, such appointment. This structure, in common with other parts of the B&L Act, is confusing. The content of section 29 could have been included in the part dealing with enforcement (sections 32–36) and section 30 could have cross-referred to it. Section 22 sets out the objective of the Collateral Registry, which is ‘to register charges and collaterals [sic] created by borrowers to secure credit facilities provided by lenders’. This statement is troublesome for a number of reasons. First, unlike land and motor vehicle registries that record ownership rights to assets, the Collateral Registry does not register collateral.117 The objective of the Collateral Registry under the B&L Act was implemented by designing the registration form so that registrants are required to select the applicable type(s) of collateral that describe(s) the encumbered assets. The selections from this drop-down menu are tabulated and enable the Annual Reports of the Collateral Registry to indicate the total volume of registrations relating to each type of collateral. In 2016, the 43,504 registrations identified 142,640 categories of movables used as collateral; most of

112 Based, presumably, on the definition in Companies Act 1963, s 87 as well as the common law. See ch 1, E1b, Kenya B4 and C9. 113 s 88 of the Companies Act 1963. 114 See C6 above. 115 Kenya B4 and C9, Nigeria E1, conclusion C6bii. 116 s 33 of an earlier draft. 117 This point also applies to ss 23–25 of the B&L Act.

The Borrowers and Lenders Act 83 which were equipment, inventory/receivables and household goods.118 In 2018, 224,583 categories of collateral were identified in 82,302 registrations. Household assets represented over 27 per cent of the categories. Secondly, borrowers do not ‘create collaterals’: this appears to be a drafting error. More importantly, section 22 refers only to registration of charges securing credit facilities. As explained above, credit facilities are one of the three types of credit agreements to which the B&L Act applies. Literal interpretation of section 22 would lead to the illogical conclusion that charges securing credit transactions (defined in section 4 to include secured loans) are not registrable in the Collateral Registry. The intention of the drafters must have been for section 22 to require registration of charges created by credit agreements. The Collateral Registry was launched in February 2010 on a proprietary network of the BoG not available over the Internet. Only those banks and financial institutions that were connected to the Central Securities Depository of Ghana could access the Collateral Registry. In June 2012, the system was redesigned and the Collateral Registry became accessible over the Internet. In 2018, 270 lenders registered their charges with the Collateral Registry. To facilitate the transition to the new web-based platform, the BoG published ‘Rules for Effective Implementation of the Borrowers and Lenders Act, 2008 (Act 773)’ (‘the Registry Regulations’) pursuant to its powers to issue notices under section 7 of the B&L Act.119 However, the main purpose of the Registry Regulations was not to implement the provisions of the B&L Act, but to address its gaps and ambiguities. As earlier noted, the B&L Act has many deficiencies, including the absence of any priority rules. In fact, the priority rule of first-to-register was initially established in a document published by the Collateral Registry entitled ‘Collateral Registry, All You Need to Know’, which was essentially an informational guide for users.120 The Registry Regulations then reflected this critical rule in Regulation 16. This process is far from ideal, as priority rules should be comprehensively set out in the substantive secured transactions law to provide predictability and certainty, and Regulation 16, as explained above,121 probably does not override the effect of a general priority rule of first created. Even though the B&L Act does not prescribe the manner in which registrations and searches may be performed, from its inception the Collateral Registry operated as an electronic registration system. The electronic access design was reinforced in Regulation 5(3) of the Registry Regulations, which requires users to establish accounts in order to submit registrations. However, with respect to the notice-based nature of the Collateral Registry’s operations, section 25 of the B&L Act requires the registration of ‘a certified copy of a charge or collateral created by the borrower … within twenty-eight days after the date of creation of the collateral or the charge’.122 Nevertheless, since it was established, the Collateral Registry has registered only a standard form that includes summary information about the charge entered, rather than a certified copy of the charge. Thus, despite the clear legal requirement

118 The Reports treats inventory and receivables as a single class of collateral, disregarding the fact that they are very different types of collateral. 119 Rules for Effective Implementation of the Borrowers and Lenders Act (n 105). 120 See Bank of Ghana, ‘Collateral Registry – All You Need to Know’, https://www.bog.gov.gh/privatecontent/ Collateral_Registry/All%20You%20Need%20to%20Know_Updated.pdf. 121 See above C3. 122 The B&L Act does not specify what qualifies as a certified copy. Arguably, the Companies Act 1963, s 107(6), which defines a certified copy as ‘a copy which has endorsed thereon a certificate to the effect that it is a true and complete copy of the original under the seal of the company or under the hand of some person, interested therein otherwise than on behalf of the company’ could be applicable to certified copies under the B&L Act.

84 Ghana to operate as a document-registration system, the Collateral Registry has largely operated as a notice-based system, despite the fact that advance registration is not permitted. It does require submission of supporting documentation for certain registrations, including those submitted outside of the 28-day window and cancellations. Section 26 sets out the information that a registrant must provide to complete a registration. The required information includes (i) the name of the borrower, (ii) the name of the lender, (iii) the nature of the charge, (iv) the date of creation of the charge, (v) the amount secured by the charge, (vi) short particulars of the property charged, and (vii) in the case of a floating charge, the nature of any restriction on the power of the borrower or chargor to grant further charges. In several aspects, these requirements depart from the Model Registry Provisions under the Model Law. An indication of the date of creation enables the Collateral Registry to reject late registrations, that is, those submitted more than 28 days after their creation, as prescribed in section 25(1). This is in contrast to the Model Law under which a notice may be registered before a security right is created.123 The B&L Act requires an indication of the amount secured by the charge as opposed to the statement of a maximum amount for which the security right may be enforced, which is an optional feature of the Model Law.124 The meaning of ‘short particulars’ is uncertain, failing to provide a safe harbour standard, unlike the Model Registry Provisions’ requirement of a description that reasonably identifies the asset. The last particular refers to a negative pledge clause, which is commonly found in relation to floating charges. It refers (correctly in normal usage) to ‘the power … of the chargor’, but the definition of chargor in section 38 is an obvious drafting error because a chargor is defined as ‘a lender in whose favour a charge is created’,125 whereas the term ‘chargor’ actually refers to the person who creates the charge. The last particular also refers to the ‘power of the borrower or chargor to grant further charges that rank in priority to or with the charge created’. It makes no sense to include the borrower here: if it is a different person from the chargor, the borrower cannot create a charge ranking in priority to that created by the chargor. Section 27 provides for the registration of a memorandum of satisfaction of debt or release of the property charged in whole or in part. This section is clearly based on section 114 of the Companies Act 1963, which itself was based heavily on section 100 of the UK Companies Act 1948.126 The purpose of these sections was to enable either a lender or borrower to ‘clean’ the register if the transaction was completed, and keep it up to date if the transaction was not. There was, however, no duty on either the lender or the borrower to make such a registration, and this is also the case under the B&L Act. This provision can be criticised on the basis that, while it is understandable to require the borrower to provide proof that the debt has been satisfied or the charge released, it is unduly onerous to require a lender to do this. Under the Model Registry Provisions only the secured creditor can register a cancellation notice127 (which has broadly the same effect), but can be compelled by the grantor to do so in certain circumstances, including extinguishment of the security right.128 123 See ML, Art 2. 124 See ML, Art 8. 125 Furthermore, the definition of chargor is incorrect, as the term ‘chargor’ is typically associated with the person that creates a charge. 126 Which is replicated, though differently drafted, in the current s 859L of the UK Companies Act 2006. 127 MRP, Art 16. 128 MRP, Art 20(3).

The Borrowers and Lenders Act 85 All charges must be registered within 28 days of their creation, but the Registrar may grant an extension unless it would adversely affect the rights of third parties.129 The Model Law does not mandate that a registration be effected within a particular period, and the failure to complete a registration has no effect on the validity of the security right as between the two parties. The appointment of a receiver or manager must be notified to the Collateral Registry in a prescribed form, after which the Registrar must then publish a notice of appointment in the Gazette.130

a. Searchability and Indexing The B&L Act does not identify the criteria governing indexing and searches (for registrations) in the Collateral Registry. These issues were first addressed in the ‘Collateral Registry, All You Need to Know’ document.131 This document indicated that the searcher must provide the borrower’s details, as well as some description of the collateral against which a search could be conducted. Accordingly, it was impossible to conduct a search only against an identifier of the borrower, and there is no legal support for this kind of search in the B&L Act. This approach to searching was reflected only later in the Registry Regulations.132 Modern secured transactions laws do, however, provide for such a search. The drafters of the B&L Act might have thought that a searcher is more likely to be interested in a specific asset (eg a tractor) or class of assets (eg inventory) of the borrower; however, that is not the case in many transactions. Furthermore, a prudent prospective lender should determine whether some other assets of the loan applicant are encumbered in order to adequately assess the overall creditworthiness of the loan applicant. Moreover, the availability of other assets will enable the prospective secured creditor to take measures if certain assets (eg receivables) are already encumbered, but other assets which the lender is considering to take as collateral (eg inventory) can be disposed of to generate receivables. In addition, the B&L Act does not provide any rule on the legal effect of mistakes in registration information,133 exacerbating the confusion and legal uncertainty, and posing a significant legal risk since the lender cannot be sure ex ante whether a judge might find its registration invalid, even for trivial errors. Regulation 13 of the Registry Regulations introduced the concept of seriously misleading errors that would render the registration invalid, along the lines of Article 24 of the Model Registry Provisions. Regulation 26 of the Registry Regulations prescribes that a search with legal effect can only take place against one of three criteria: the name of the borrower, the collateral ID number or the registration number. A search against the identification numbers of borrowers may be made, but it lacks legal effect.134 The name of the borrower is to be determined in accordance with Regulations 9 and 10, which prescribe what documents are authoritative 129 B&L Act, s 28. This replicates s 115 of the Companies Act 1963, based on s 101 UK Companies Act 1948 (now s 859F Companies Act 2006). 130 B&L Act, s 31. 131 ‘Collateral Registry – All You Need to Know’ (n 120) 5. 132 Rules for Effective Implementation of the Borrowers and Lenders Act 2012 (Registry Regulations), s 26. 133 See ch 2, E1. 134 See Collateral Registry, ‘Collateral Registry System: Search’, www.collateralregistry.gov.gh/Search/Search/ Create.

86 Ghana to do so. An authoritatively determined name includes the name of a citizen stated in their national ID and the name of a legal person as stated in the document constituting the legal person. Absolute precision in these kinds of rules is essential not only for the proper functioning of the Collateral Registry, but of the entire secured transactions system. However, being able to make a legally effective search only against a name is problematic. Given the popularity of certain names in Ghana, searching in this way could generate a long list of results that exactly match the search criterion entered. In one case, one of the coauthors was told that individuals belonging to a particular village use the name of the village as their last name, and many have similar or very common first names. Further, such individuals lack any form of official identification. The 2017 Bill will move Ghana towards a system where borrowers will be identifiable in registrations by their unique numbers, which has become the norm across Africa.

b. Performance of the Collateral Registry From the moment the Collateral Registry was launched, non-bank financial institutions have been some of its largest users and have driven the increase in registration volumes in subsequent years. Statistics indicated that 17,582 charges were registered in the Collateral Registry in 2011, which was an increase of 21.57 per cent from the preceding year. For 2018, 82,302 registrations were completed as compared to 57,509 in 2017. Of the total number of registrations in 2018, 71 per cent were performed by non-bank financial institutions. In 2018, the number of lenders that registered charges increased to 370, as compared to 163 in 2015. This increase in the number of users, searches and registrations is largely attributable to the BoG’s public awareness, education and training campaigns. The registration statistics reveal a number of unsound lending practices. Since the Collateral Registry collects fees for searches, it is able to track the number of searches conducted by users. It is telling that, over the years, the number of registrations has significantly exceeded the number of searches.135 In 2016, 16,064 searches of the Collateral Registry records were performed against 43,504 registrations. In 2018, 36,889 searches were performed as compared to 82,302 registrations. One would assume that a prudent lender in loan negotiations with a customer offering some assets as collateral would search the Collateral Registry to ensure that the customer has not previously charged the same assets to another lender. The number of searches should, at a minimum, equal the number of new registrations and in a healthy credit market the number of searches should exceed the number of registrations because not all searches will return satisfactory results leading ultimately to loans. Besides, prospective buyers of assets should also conduct searches. Finally, it is also a good practice to conduct a post-registration search to ensure that no other registration has been recorded in the interim. The overhaul of the B&L Act will bring much needed legal clarity, which is expected to further increase the utilisation of the Collateral Registry. In the meantime, the BoG should focus its training and awareness efforts on explaining the function of searches. These efforts will also help with the general objective of ensuring that regulated financial institutions extend credit in a prudentially sound manner.

135 Annual

Reports are on file with the authors.

The Borrowers and Lenders Act 87

5. Priority As noted above, the B&L Act does not include any rules on the priority of charges, a significant oversight that is remedied in the 2017 Bill, which would codify the first-in-time, first-in-right rule. In other words, a lender that registers a notice of its security interests first will have priority over any other subsequently registered security interests. Certain exceptions to this general temporal priority rule have also been included in the 2017 Bill, particularly to reflect the special nature of purchase-money security interests (acquisition security rights) where a later-in-time lender provides credit to the borrower to enable it to acquire a new asset. The 2017 Bill also provides ‘taking free’ rules that protect buyers in the ordinary course of business, and transferees of money, negotiable instruments and securities, as well as providing for the priority of service providers’ retention liens. Priority rules addressing specific situations, such as when movable assets subject to multiple security interests are commingled or accessed to one another have been included as well. All of these priority rules reflect the approaches and, in many cases, the actual drafting of Chapter V of the Model Law, although it is odd that there is no equivalent of Article 34(2), which provides that a buyer takes free if the sale is authorised.136 This, together with the ‘taking free’ rules referred to above, preserves the effect of the floating charge in the Model Law scheme.137

6. Enforcement The B&L Act includes four sections dealing with the enforcement of charges.138 These sections empower lenders to enforce their rights extrajudicially. However, the requirements that lenders must satisfy practically neutralise the usefulness of extrajudicial enforcement. Section 32 generally requires the lender to send a notice of default to the borrower. The breadth of this section is limited in two significant ways. First, default is defined only as the failure to make a payment on the due date. It is unclear how the lender should proceed if the borrower breaches an obligation other than the duty to make timely payments (eg when the borrower damages or fails to insure the collateral). The B&L Act does not state whether these rules are mandatory or whether they can be modified by parties’ agreement specifying further instances of default, such as damage to the collateral. Ghanaian lenders interviewed by one of the coauthors stated that their default clauses in credit agreements include acts and omissions other than the failure to make timely payments as events of default. Second, the borrower has 30 days after receipt of the notice of default to cure it. Thus, the B&L Act effectively grants the borrower a period of 30 days to pay the amount outstanding. This exposes the lender both to the uncertainty as to whether that might happen, and, more importantly, to the risk of loss or deterioration in the value of the collateral. Additionally, on the expiration of the grace period, the lender must send yet another notice to the borrower indicating its intention to proceed extrajudicially.139 During a number of workshops that 136 Of course, there can always be express authorisation of a particular transaction under general law. 137 See ch 1, E1b. 138 B&L Act, s 35 governs the enforcement of mortgages providing that the Mortgage Act of 1972 does not apply to the rights of a lender under this Act. 139 B&L Act, s 33(b).

88 Ghana one of the coauthors conducted in Accra, local lenders repeatedly voiced their frustration with these requirements, stating that they are close to impossible to satisfy in the present lending environment. Routinely, borrowers avoid receiving the default notice, knowing that it is a precondition for the lender to proceed extrajudicially. Consequently, the 30-day grace period to cure the default may never commence, and the lender has no choice but to file a petition with the court. Judicial involvement is a very inconvenient remedy in Ghana considering that, according to lenders interviewed by one of the co-authors, the average court case takes two to three years to resolve. Moreover, disposal of collateral through public auctions are very expensive, as the fees charged by auctioneers (up to 7 per cent of the proceeds) and valuers increase the cost of disposal. Under section 11 of the Auction Sales Act, the owner of a property may also request an auctioneer to sell it. Unconventionally, the Collateral Registry imposes further hurdles to extrajudicial enforcement. This is quite uncommon since the business of registries of security rights is not to interfere with the enforcement process, whether judicial or extrajudicial. The Ghanaian Collateral Registry developed the odd practice of requiring lenders to register a notice indicating the date when default occurred and providing some evidence that a notice of default has been delivered to the borrower. This practice was given legal backing with the promulgation of the Registry Regulations.140 Further, Regulation 20(1) of the Registry Regulations requires a lender who intends to enforce a registered charge extrajudicially to register a notice of that intention 30 days after the borrower receives the notice of default, and under Regulation 20(2), ‘[t]he Registrar ‘shall certify the realisation process by issuing a certificate to that effect’ known as a ‘memorandum of non-objection’. Upon registration of the notice of the intention to enforce extrajudicially, the Registrar would verify that the lender notified the borrower of a default and that a 30-day period has elapsed. If both conditions are satisfied, the Registrar issues a memorandum of non-objection.141 There is no justification to vest such powers in the Registrar, especially when one of the underlying policies of the B&L Act is to facilitate extrajudicial enforcement. The fact that there was resistance to giving up this power until the late stages of developing the 2017 Bill reflects the highly borrowerfriendly mindset of the Ghanaian regulators, as evidenced by their desire to protect the borrower (understood to be either a consumer or a microbusiness, and which comprise the bulk of borrowers against whom charges in the Collateral Registry are registered). One of the Registrars of the Collateral Registry was transferred from the Consumer Protection Department of the BoG, and that background might have inspired the push to preserve this function. During the interviews and meetings carried out by one of the coauthors, many lenders mentioned that borrowers simply ignore their requests to surrender the collateral based on the memorandum of non-objection. The memorandum has no legal and practical value, as it is not a court order empowering the lender to repossess the collateral or authorising the lender to procure assistance of the police. As such, it is an unnecessary and illogical administrative burden that stifles extrajudicial enforcement. After some discussion, the 2017 Bill no longer includes this function. In addition to these deficiencies, the B&L Act is silent on certain key issues. These include how the collateral may be disposed of (eg by public or private auction), whether remedies 140 Registry Regulations, s 17, which requires that a lender register the date of default, the date of receipt of the default notice and a notice of any cure of the default within the specified period. 141 Registry Regulations, reg 20(2).

The Borrowers and Lenders Act 89 other than disposal (eg acceptance of the collateral in satisfaction of the obligation) are available, how the proceeds must be distributed and the effect of disposal on competing claims. All of these aspects are addressed comprehensively in the 2017 Bill142 along the lines of Chapter VII of the Model Law.143

7. Conflict of Laws and Transition Unlike Chapter VIII of the Model Law, the B&L Act does not include any provisions that determine the law applicable to various aspects of a charge when the transaction involves a ‘foreign element’, such as when the borrower or the collateral are located outside Ghana. Given the application of the B&L Act primarily to charges created by individuals in favour of non-bank financial institutions licensed by the BoG,144 the absence of such rules has not caused any practical challenges since the transactions mostly do not involve any foreign elements. The absence of such rules may be another factor behind the stymied growth of the lending market in Ghana, which remains predominantly low-value transactions. The 2017 Bill will significantly redesign the secured transactions framework, including by incorporating most of the provisions of Chapter VIII of the Model Law. Section 1(1)(c)–(j) of the 2017 Bill provides that the B&L Act determines issues of creation, third-party effectiveness, priority and enforcement if the tangible collateral is located in Ghana or, for intangible collateral, the borrower is located in Ghana. However, unlike the Model Law provisions that determine the applicable law (whether that of the enacting State or a foreign country), the rules of the 2017 Bill determine only when the B&L Act applies. The 2017 Bill also includes special rules applicable to security interests in bank accounts and immovable property, but again establishing only when the B&L Act will apply to security interests in such assets. The B&L Act does not provide any transitional rules, unlike Chapter IX of the Model Law. This is a significant gap in any secured transactions legislation. The absence of such rules in the B&L Act did not cause legal or practical challenges in Ghana because of the limited scope of application of the B&L Act, its ‘non-effect’ on other legislation, and the short-term nature (typically three months) of the kinds of transactions covered by the B&L Act (eg a microloan secured with household assets). Accordingly, providing a period of 6–12 months for prior security rights to be transitioned into the new regime, especially by their registration in the Collateral Registry, would have no practical relevance since most of those transactions would expire during the transitional period anyway.145 Nevertheless, the Amendments draw inspiration from the Model Law and include provisions dealing with transition as well as their effect on other legislation. The 2017 Bill repeals only the Chattels Transfer Ordinance and the B&L Act, and addresses its relationship with the other laws by way of a supremacy rule in section 77. Therefore, unlike the Model Law that contemplates repeal and consequential amendments to affected legislation, the 2017 Bill will leave most of the affected legislation intact creating unpredictability.146 Although this is an unconventional 142 2017 Bill, ss 49, 50, 52, and 54 respectively. 143 ML, Arts 78(3) (which leaves the method of sale up to the enforcing creditor, subject to the overriding requirement of commercial reasonableness), 79, 80, and 81, respectively. 144 See C1 above. 145 cf ML, Art 105. 146 2017 Bill, s 16.

90 Ghana approach that raises a number of questions, the political environment was not amenable to the kind of approach enshrined in the Model Law.

D. Final Remarks The objective of the B&L Act and the Collateral Registry was to provide a modern framework facilitating access to credit secured with movable property. Ghana was lauded as the model for Africa and, in fact, has become an inspiration for many African economies to reform their frameworks and establish central bank-run collateral registries. However, beyond inspiring reforms in Africa, there is very little that is positive to be drawn from the B&L Act. Instead of constituting a veritable model, it is a prime example of a legal framework that economies should avoid replicating. It is conceptually unsound, has an uncertain scope of application, and coexists with a mixture of outdated laws without any attempt to coordinate their relationships. In addition, it contains too many gaps, leading to an inhospitable environment for movables lending in Ghana. Moreover, its structure and content, especially the parts governing regulatory aspects and issues of general lending, are not found in secured transactions legislation elsewhere in Africa other than Sierra Leone, whose framework unfortunately models Ghana’s B&L Act in many aspects. Both of these Acts are presently the subject of proposed major revisions aligning them more closely with the Model Law. However, problems persist in Ghana, which have prevented a reform for over eight years. Nonetheless, the BoG has managed to operate the Collateral Registry in a manner that facilitates some level of secured lending for many Ghanaian financial institutions. Particularly noteworthy is the extent to which non-bank financial institutions have used the Collateral Registry to extend credit to individuals and micro-entrepreneurs. Yet it is important to acknowledge that the growth of the Ghanaian secured credit system and the use of the Collateral Registry has been stifled by the deficient legal framework. Lending to types of prospective borrowers other than individuals and micro-entrepreneurs is exposed to a number of risks stemming from the B&L Act, especially the absence of priority rules and its dependence on the Companies Registry for the charges created by companies. Because of the many legal impediments, extrajudicial remedies cannot be effectively deployed. As a whole, enforcement has been made inefficient by the intervention of the Collateral Registry, which must certify the use of extrajudicial remedies. When Ghana adopts the 2017 Bill, the existing first-class registration system will then be ready to support lending activities. Ghana is once again attempting to innovate and perhaps provide a template for the second phase of secured transactions reforms by making registration in the Collateral Registry the perfection mechanism, as well as a priority point, for security interests in immovable assets. Concurrent registrations in the land registries will no longer be required other than to achieve an effect against buyers of immovable assets. Hopefully, enactment of the 2017 Bill in the near future will restore Ghana’s reputation as a leader in secured transactions law reform in Africa. Any such improvement in Ghana’s status would be attributable to UNCITRAL whose Legislative Guide was the basis for the 2017 Bill, with further refinement from the provisions of the Model Law.

4 Kenya Kenya is one of the most important economies in Africa. It also significantly influences legal developments in the region, particularly in neighbouring countries such as Uganda. Moreover, Kenya is a hub for innovation, having developed and implemented the most widely adopted mobile payment system in Africa, and is now a magnet for all kinds of blockchain applications. Like other economies in Africa, several studies have confirmed that Kenya’s secured transactions system is outdated and inadequate to fulfil the credit needs of Kenyan businesses.1 The deficiencies of Kenya’s legal framework for secured lending were generally similar to the challenges experienced by other African economies that inherited English law. Kenya’s Vision 20302 provided the necessary policy framework to move forward with legal and institutional reform to change the status quo, including in the area of credit infrastructure. This framework was originally adopted in 2007, at which point it was estimated that only 15 per cent of the population had access to finance.3 One of the aims of Vision 2030 was to decrease ‘the share of the population without access to finance from 85% to below 70%’.4 Recent reports suggest that since 2007, the number of Kenyans with access to finance has risen considerably. A 2016 FinAccess survey, for example, revealed that only 17 per cent of Kenyan adults do not have access to financial services.5 The picture in relation to SME finance, however, is more mixed, as discussed below.6 In 2014, Kenya embarked on a reform process that resulted in the enactment of the Movable Property Security Rights Act (MPSR) in March 2017. The MPSR is based on, and largely implements, the Model Law, albeit with some deviations, both minor and substantial, which are examined below. Although this was a great legislative achievement for Kenya, its operationalisation undermined the trust of financial institutions in the new secured transactions framework. The Collateral Registry was launched about two months after the enactment. The Registry Regulations, adopted to coincide with the establishment of the Collateral Registry, suffered from a number of deficiencies, as a result of which they were revised within months of their coming into force. However, two years later, the revision is still not finalised. No structured and intensive capacity-building programs were conducted, especially for lenders, and the launch of the Collateral Registry was not publicised through a media campaign. Both capacity building and a media campaign should be 1 See National Law Center for Inter-American Free Trade, ‘Securing Credit for Growth: The Case for a New Asset Register in Kenya’ (1 March 2014) www.fsdkenya.org/publication/securing-credit-for-growth-the-casefor-a-new-asset-register-in-kenya/. 2 VDS Project ‘Kenya Vision 2030’ (2007) https://vision2030.go.ke/. 3 ibid, s 4.6, 15. 4 ibid. 5 2016 FinAccess household survey (18 February 2016) https://fsdkenya.org/publication/finaccess2016/. 6 See A below.

92 Kenya integral components of any reform project. Kenya is a case study on how years of diligent and arduous work developing the substantive law can be undermined by the motivation to implement a registry in a hasty manner. During the reform process, alignment of the secured transactions legislation with the recently enacted Companies and Insolvency Acts was considered. Both these statutes were based on English law. It was a challenge to make the concepts of a secured transactions law based on the functional approach consistent with the conceptual structure of English law.7 Nonetheless, the new secured transactions regime in Kenya shows great promise, and the mistaken approaches taken could and should be corrected. One factor that has played an important role in the operationalisation of the reform is political risk. The uncertainty surrounding the 2017 elections led to the suspension of aid from the International Monetary Fund (IMF).8 As the political picture becomes clearer, the prospect for a proper, though delayed, implementation of the secured transactions law reform brightens. However, public policies, such as the statutory cap on interest rates, may also continue to affect secured lending in a negative way. Nonetheless, reform in this area of law is continuing. For example, the Warehouse Receipts Bill was lodged with the National Assembly in February 2018.9

A. Economic Overview Kenya is an East African, lower-middle-income economy, which has experienced an average of 5 per cent GDP growth per year for the past 10 years.10 In 2017, GDP growth slowed down to 4.8 per cent, the lowest in five years. This was attributable to a slowdown in the growth of credit to the private sector, and to the political uncertainty surrounding the 2017 elections.11 GDP growth is expected to rebound and is projected to be 6.1 per cent by 2020.12 The economy relies largely on agriculture, which, in 2017, made up 35.3 per cent of Kenya’s GDP and employed about 75 per cent of the labour force.13 Over 75 per cent of Kenya’s agricultural output comes from small-scale, rain-fed farms. The most important agricultural products are tea, coffee, corn, wheat, sugarcane, fruit, vegetables, dairy, beef,

7 See L Gullifer and M Racyznska, ‘The English Law of Personal Property Security: Under-Reformed?’ in O Akseli and L Gullifer (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Hart Publishing, 2016). 8 See D Miriri, ‘IMF Stops Kenya’s $1.5 Billion Standby Credit after Deficit Clash’ Reuters (20 February 2018) www.reuters.com/article/us-kenya-imf/imf-stops-kenyas-1-5-billion-standby-credit-after-deficit-clash-idUSKCN1G41IQ. 9 See Kenya Gazette Supplement, ‘The Warehouse Receipt System Bill, 2018’, No 6 (7 February 2018) http:// kenyalaw.org/kl/fileadmin/pdfdownloads/bills/2018/WarehouseREceiptSystemBill_2018.pdf. 10 Central Intelligence Agency, ‘Africa: Kenya’ (The World Factbook, 1 March 2019) https://www.cia.gov/library/ publications/the-world-factbook/geos/ke.html. 11 C Awiti, A Dennis, C Mutie, et al, World Bank Group, ‘Kenya Economic Update: Policy Options to Advance the Big 4 – Unleashing Kenya’s Private Sector to Drive Inclusive Growth and Accelerate Poverty Reduction’(1 April 2018) http://documents.worldbank.org/curated/en/327691523276540220/pdf/125056-WP-P162368-PUBLICKenyaEconomicUpdateFINAL.pdf. 12 ibid. 13 Central Intelligence Agency, ‘Africa: Kenya’ (n 8).

Economic Overview 93 fish, pork, poultry and eggs.14 Kenya’s major industrial products mainly include low-value products (such as plastic, furniture, batteries, textiles, clothing, soap, cigarettes and flour).15 Vegetable products dominate Kenya’s exports, making up 54 per cent of all exports in 2016 (US$2.56 billion).16 Export data from 2016 shows that Kenya’s chief export partners are mostly from Europe (35 per cent), followed by Africa (27 per cent) and Asia (25 per cent).17 In 2016, Kenya’s major imports were machines, textiles, mineral products, chemicals, metals and transportation.18 Kenya is also a hub for innovation in Africa, as well as globally. Its most famous innovation is probably the M-Pesa system,19 one of the world’s first mobile money applications, which is used by over 20 million Kenyans. Thanks to M-Pesa, 70 per cent of all transactions in Kenya are already digital.20 Similar services have attempted to emulate M-Pesa’s success. For example, BitPesa, which was established in Nairobi in 2013, provides a cryptocurrency exchange and wallet.21 The widespread adoption of mobile payment systems has opened the door for other mobile banking services. For example, M-Shwari, a mobile savings and credit provider, has made more than 20 million loans since its launch to almost 3 million borrowers.22 M-Shwari was the world’s first widespread lending product to calculate credit limits based on the client’s history of payments for mobile phone airtime.23 Technology also permeates commercial activities and farming. For instance, Hello Tractor, a start-up operating in Nigeria and Kenya, sells 15-hp (three wheeled) tractors for $4,000. The tractors are embedded with GPS/telematics, which connect the smart tractors to cloud-based software.24 Hello Tractor has developed an SMS booking and payment system whereby farmers can request, schedule and prepay for tractor services from nearby smart tractor owners.25 Farmers using the smart tractors have reported a 200 per cent improvement in their farm yields.26 The familiarity and acceptance of alternative payment solutions in Kenya has facilitated exploration of blockchain solutions and the relatively small number of legacy systems has minimised the transition costs to distributed ledger technologies.27 14 ibid. 15 ibid. 16 Observatory of Economic Complexity, ‘Kenya’, https://atlas.media.mit.edu/en/profile/country/ken/. Tea is by far the largest single export good, comprising 23 per cent of all exports in 2016 (US$1.09 billion). ibid. 17 ibid. 18 ibid. 19 T Harford, ‘Money via Mobile: The M-Pesa Revolution’ BBC News (13 February 2017) www.bbc.com/news/ business-38667475. In 2011, 72 per cent of Kenyans living on less than $1.25 per day used M-Pesa. D Runde, ‘M-Pesa and the Rise of the Global Mobile Money Market’ Forbes (12 August 2015) www.forbes.com/sites/ danielrunde/2015/08/12/m-pesa-and-the-rise-of-the-global-mobile-money-market/#46f7f9a75aec 20 See International Finance Association (IFC), ‘Blockchain in Financial Services in Emerging Markets Part II: Selected Regional Developments’ EMCompass Note 44 (August 2017) http://documents.worldbank.org/curated/ en/981091507180912887/pdf/120262-BRI-PUBLIC-EMCompass-Notes-44-FINAL-8-21.pdf. 21 See BitPesa, ‘Fast Easy, Cost-Effective Financial Transactions for Africa’, www.public.bitpesa.co/. Bitpesa now operates in seven African markets. ibid. 22 T Cook and C McKay, ‘How M-Shwari Works: The Story so Far’ (10 April 2015) 1–3, http://s3-eu-central-1. amazonaws.com/fsd-circle/wp-content/uploads/2015/08/30095111/Forum-How-M-Shwari-Works-Apr-2015. pdf. M-Shwari was developed by Safaricom in partnership with the Commercial Bank of Africa (CBA). ibid. 23 ibid. 24 See Echoing Green, ‘Jehiel Oliver’ (2015) www.echoinggreen.org/fellows/jehiel-oliver. 25 https://www.hellotractor.com/home. 26 ibid. 27 IFC (n 20).

94 Kenya As of January 2019, 42 commercial banks, 1 mortgage finance company and 8 authorised non-operating bank holding companies operate in Kenya.28 Other non-bank financial institutions include pension funds, insurance companies, hire-purchase companies and microfinance banks.29 In late 2018, the Central Bank of Kenya (CBK) maintained a 9.0 per cent key interest rate.30 Banking fraud, and failures on the part of the supervisory authorities, resulted in a minor banking crisis in 2016. As a consequence, CBK took over three undercapitalised banks, including Chase Bank of Kenya, which held US$1.43 billion in assets shortly before being placed under receivership.31 In 2016, the government enacted legislation to regulate interest rates on loans and deposits. This was primarily to address the high cost of credit and also to enhance financial inclusion.32 The law sets the maximum lending rate at no more than 4% above the CBK rate and sets the minimum interest rate on deposits held in interest-earning accounts at 70% of the maximum lending rate.33 In August 2018, the Kenyan Parliament voted to retain these rate caps, despite strong criticism from the IMF.34 The rate caps have resulted in a sharp reduction in available credit, reaching the lowest level in ten years, as well as a shortening of loan maturity periods and a sizeable increase in non-performing loans.35 Commercial banks started directing their lending operations to large corporate borrowers and government instrumentalities, while eschewing smaller and riskier borrowers.36 In a September 2017 CBK survey, 54 per cent of respondents confirmed that the interest rate caps had negatively affected lending to SMEs.37 Many of the typical credit products supported by a modern secured transactions framework were unavailable in Kenya for one reason or another, which may still be the case today. For instance, financial leases are unattractive because lease payments are subject to a value-added tax. Similarly, the growth of factoring was hindered because of uncertainty surrounding the rights of the factor: these rights came from common law but had not been the subject of any court decisions.38 A great deal of uncertainty also existed with respect to the application of the relevant stamp duty, which was set at 0.2 per cent for assignments by

28 Central Bank of Kenya, ‘Directory of Licensed Commercial Banks, Mortgage Finance Institutions and Authorized Non-Operating Holding Companies’, www.centralbank.go.ke/wp-content/uploads/2017/05/Directory-of-Licenced-Commercial-Banks-Mortgage-Finance-Institutions-and-NOHCs.pdf. 29 Central Bank of Kenya, ‘Statistical Bulletin’ (December 2017) www.centralbank.go.ke/uploads/statistical_ bulletin/95183200_Statistical%20Bulletin%20Revised.pdf. 30 Trading Economics, ‘Kenya Interest Rate’, www.tradingeconomics.com/kenya/interest-rate. 31 Standard Digital, ‘SBM Concludes Chase Bank Acquisition’ (18 April 2018) www.standardmedia.co.ke/ business/article/2001277336/sbm-concludes-chase-bank-acquisition. 32 See The Banking (Amendment) Act No 25 of 2016, s 33B. See also M Safavian and B Zia, ‘The Impact of Interest Rate Caps on the Financial Sector: Evidence from Commercial Banks in Kenya’, World Bank Policy Research Working Paper 8393 (2 April 2018) 2. 33 Central Bank of Kenya, ‘The Impact of Interest Rate Capping on the Kenyan Economy’ (March 2018) 4, www.centralbank.go.ke/wp-content/uploads/2018/03/Interest-Rate-Caps_-March-2018final.pdf. 34 See J Ndiso and H Malalo, ‘Kenya’s Parliament Approves Retaining Interest Rate Cap, Against IMF Wishes’ (30 August 2018) https://uk.reuters.com/article/uk-kenya-economy/kenyas-parliament-approves-retaining-interestrate-cap-against-imf-wishes-idUKKCN1LF1L4. 35 See Safavian and Zia (n 32). 36 ibid. 37 See M Safavian and B Zia, ‘To Cap or Not to Cap? What Does Kenya’s Experience Tell Us About the Impact of Interest Rate Caps on the Financial Sector?’ (25 June 2018) http://blogs.worldbank.org/allaboutfinance/ cap-or-not-cap-what-does-kenya-s-experience-tell-us-about-impact-interest-rate-caps-financial-sector. 38 Costs of Collateral in Kenya: Opportunities for Reform (FSD Kenya, 2009) 26.

Pre-Existing Legal Framework and Registries 95 way of security, but at 2 per cent for conveyances of sale.39 These examples illustrate that there are significant obstacles beyond the secured transactions law that should be addressed as part of a broader reform project. In common with many countries, there is a financing gap in relation to SMEs in Kenya. Such financing gaps exist where traditional banks meet the lending needs of large enterprises, while microenterprises are mostly served by microfinance institutions, but neither of these sources of finance are willing or able to meet the needs of SMEs.40 About 27 per cent of SMEs in Kenya experience funding-related constraints.41 A survey by the Kenya Bankers Association showed that, of SMEs facing funding constraints, 57 per cent blame the cost of credit, 34 per cent blame access to credit (ie the unavailability of credit) and 10 per cent blame both.42 Similarly, 56 per cent of SMEs reported that interest rates, fees and commissions were the single most important challenge in securing credit, followed by collateral requirements (9 per cent) and repayment terms (8 per cent).43 In a CBK 2015 survey, it was noted that the cost of credit for SMEs remains high due to several factors, including inefficiencies in the collateral registration process, cost of judicial enforcement, high overhead costs, and limited information and exchange of information on borrowers between banks.44 Secured transactions reform has attempted to deal with some of these deficiencies.

B. Pre-Existing Legal Framework and Registries The pre-reform legal framework that governed secured lending in Kenya was typical of the other jurisdictions that inherited English law. Secured transactions were governed by many sources of law: statute, common law and equity. There were a number of devices achieving the purpose of security, that is, enabling a creditor, in the event of non-payment, to have recourse to assets out of which the secured obligation could be paid, comprising ‘true’ security interests (charges, mortgages, liens and pledges), and those based on retention and outright transfer of title. In some cases, the particular security device was void against other secured creditors and on insolvency unless registered within a specified time period (eg mortgages and charges). In other cases, no registries had been established (eg for financial leases and for sales on retention of title terms). This meant that the rights of creditors were effective against third parties without any form of public notice. More over, the registration regimes for security devices were different depending on whether the borrower was a company, on the one hand, or an individual, including an unincorporated business, on the other. This fragmentation of the legal framework forced lenders to investigate the business form of the loan applicant (ie whether it was a registered company or was an informal and unincorporated business). Lenders often used several security 39 ibid. 40 Kenya Bankers Association, ‘Financing Small and Medium Enterprise: The Reconciliation of Borrower– Lender Expectations’ (July 2016) www.kba.co.ke/downloads/SME%20Financing%20Survey.pdf. 41 ibid. 42 ibid. 43 ibid. 44 Central Bank of Kenya, ‘FinAcess Business – Supply: Bank Financing of SMEs in Kenya’ (September 2015) www.centralbank.go.ke/images/docs/Bank%20Supervision%20Reports/BankFinancingSMEsKenya.pdf.

96 Kenya devices in order to satisfy the credit needs of the borrower and had to comply with the formalities applicable to each.45 Additionally, some types of assets could be used as collateral in a number of different ways. Take, for example, a business seeking to acquire a motor vehicle. The business could acquire the vehicle pursuant to a hire-purchase agreement. Alternatively, it could borrow the purchase price, and secure this loan by either granting a chattel mortgage over the vehicle under the Chattels Transfer Act (CTA), or it could grant a fixed charge to the lender which would have to be registered under the Companies Act.46 Another option would be for the lender of the purchase price to register itself as the co-owner of the motor vehicle in the motor vehicle registry, and take delivery of the log book. Thus, there could be registration in three, or four, different registration systems. This fragmentation complicated initial due diligence, since a lender would have had to request the log book from the loan applicant and to search both the Chattels and Hire-Purchase Registries, and, maybe, the Company Charges register.

1. Chattels Transfer Act The CTA was passed in 1930 and amended in 2012 to accommodate certain market changes.47 The CTA was based on New Zealand legislation and thus provided detailed provisions on mortgaging sheep fleece and wool clips.48 It allowed grantors, who could be both individuals and legal entities (though not companies), to execute an instrument to secure the payment of a loan creating an interest in ‘chattels’.49 Section 2 of the CTA set out a list of transactions excluded from its scope, including hire-purchase agreements, mortgages, charges and debentures issued by companies or other corporate bodies, and transfers of (or agreements to transfer) instruments by way of security, as well as transfers of chattels in the ordinary course of business.50 Instruments required attestation by at least one witness.51 The CTA provided for registration of instruments, rather than registration of the security interests themselves. Sections 2 and 5 of the CTA defined the Registry as a transactionregistration system whereby the submission of the instrument, as well as a Schedule (list) of collateral,52 was required within 21 days of its execution,53 although late registration was possible with the leave of the court.54 Unregistered instruments were void against specified third parties, including the official receiver or trustee in bankruptcy, the assignee or trustee acting under any assignment for the benefit of the creditors, and any person entitled to seize an asset pursuant to a court order.55 Unregistered instruments were not ‘valid or effectual’ against bona fide purchasers or mortgagees for valuable consideration who acquired their 45 See above n 1. 46 The Companies Act, ch 486, pt IV ‘Registration of Charges with Registrar’ (rev edn, 2009). 47 Chattels Transfer Act, 1930 (rev 2012). 48 M Rogers, ‘A Note on Hire-Purchase and Chattels Mortgages in Kenya’ (1975) 19(1–2) Journal of African Law 154. 49 CTA, s 2, definitions of grantee and grantor. 50 CTA, s 2, definitions of chattels and instruments. 51 CTA, s 15. 52 CTA, s 13. 53 CTA, ss 6 and 14. 54 CTA, s 9. 55 CTA, s 13.

Pre-Existing Legal Framework and Registries 97 rights without notice.56 However, the Kenyan courts confirmed that the failure to register an agreement did not affect its enforceability as between the parties.57 The single Registrar’s office for the entire country was located in Nairobi, which, given the need to submit the instrument in person, significantly complicated the registration of instruments executed in distant locations. The CTA also provided, in its Schedules, the content and information to be lodged with the Registry. The Registry Book (database), which was kept manually, included a field for ‘Nature and Date of the Instrument’58 in which registrants commonly inserted the actual amount of the secured obligation.59 More recently, registrations were also entered into an Excel spreadsheet. The Registry Book did not include a field in which a description of the collateral could be entered. Therefore, in order to identify the collateral, a searcher had to physically examine the copy of the instrument kept at the Registry. However, the files of the Registry were disorganised, effectively precluding the searcher from locating the relevant instrument.60 A visit to the Registry office revealed that very few control mechanisms were in place to avoid the loss of the Registry Book pages. Other aspects of the registration process that exposed creditors to risks included61 the improper transcription of information from the document submitted by the lender on to the Registry Book and the lack of a timestamp which would determine priority among lenders who might have submitted documents for registration on the same day.62 The reform resulted in the repeal of the CTA.63 The Registry thus ceased to perform all functions.

2. Hire Purchase Act The 1982 Hire Purchase Act (HPA) regulates hire-purchase agreements, which are ‘agreements for the bailment of goods under which the bailee may buy the goods or the property in the goods will or may pass to the bailee’.64 The goods, therefore, remain in the ownership of the supplier throughout the currency of the agreement until the full price is paid and, in some cases, until the bailee opts to purchase the goods. Some tax aspects relating to hire-purchase arrangements and leases, particularly with respect to equipment, were also governed by the Income Tax Act of 2002.65 The HPA established a registry of hire-purchase agreements.66 56 CTA, s 14. 57 David Mburu Kamau v National Industrial Credit Bank Civil Suit 253 of 2007 (High Court of Kenya). 58 CTA, First Schedule, Register Book. 59 See above n 1. 60 Notes from a visit to the Registry on file with the authors. 61 See above n 1. 62 The CTA does not have specific provisions about priority, but the basic rule is, presumably, priority by date of creation (the standard nemo dat rule) plus the usual common law exceptions, most particularly the priority of a bona-fide purchaser of the legal interest without notice of an earlier equitable interest. Section 16 provides that an instrument takes effect from the time when it is registered and s 4 provides that registration is constructive notice to everyone. Thus, if an instrument creating an equitable interest (eg a declaration of trust) is followed by one creating a legal interest, the holder of the latter interest would have notice of the earlier interest and so the priority rule would still be first in time. 63 MPSR, s 94. 64 HPA, s 2(1). 65 Income Tax Act Leasing Rules (2002). 66 HPA, s 4.

98 Kenya The HPA was targeted at hire-purchase agreements entered into by vulnerable individuals, either for small-scale business purposes or as consumers. Thus, it did not apply to corporate hirers or to agreements where the hire purchase price exceeded 4 million shillings (approximately US$49,000).67 It provided some protection for these vulnerable hirers, including the implication of conditions and warranties and the avoidance of contractual provisions relating to enforcement, which are inconsistent with the protective enforcement provisions of the HPA, discussed below. If, however, the hire-purchase price of the object exceeded the statutory threshold, the hirer could not invoke the protection of the HPA and on default the owner could enforce any remedies provided for in the agreement.68 This monetary threshold could be varied by the Minister, and was increased from 300,000 to 4 million shillings in 2007 by Act No 7 of 2007. A hire-purchase agreement had to be stamped and delivered for registration within 30 days of its execution.69 The stamp duty was payable in addition to the registration fee. Unlike under the CTA, the failure to register an agreement also affected its enforceability against the hirer; in such cases, the owner lost the right to enforce the agreement or any related contract of guarantee against the hirer and, consequently, could not repossess the asset.70 Some dicta in Kenyan court decisions ignored this statutory language, stating that an unregistered hire-purchase agreement remains enforceable against the hirer.71 As part of the protection that the Act extended to vulnerable hirers, extrajudicial enforcement by repossession of the asset was allowed, but only until the hirer repaid twothirds of the hire-purchase price,72 after which, if an owner repossessed without a court order, the hirer could recover all the sums paid and obtain a release from any liability under the agreement.73 The MPSR did not completely repeal the HPA. The intention was to preserve the licensing provisions for hire-purchase businesses contained in the HPA, but for hire-purchase agreements to be governed by the MPSR, including their registration in the MPSR Registry. Eventually, upon a broader review, only the first paragraphs of sections 4 and 5, which provided for the registration of hire-purchase agreements, were repealed. The intention of the drafters was to preserve the application of the enforcement protective provisions in section 5(4) for vulnerable persons who enter into such agreements,74 thus extending these protections post-reform. However, these protections should have been incorporated directly into the text of the MPSR rather than leaving much of the HPA in force. Furthermore, it is unclear why the remaining paragraphs of sections 4 and 5 of the HPA have not been repealed as they all directly relate to the registration of hire-purchase agreements. It is not uncommon for modern secured transactions laws to limit post-default enforcement rights of secured creditors with respect to consumer or otherwise vulnerable classes

67 HPA, s 3(1). 68 Mburu Gathee T/A Gathee Enterprises v National Industrial Credit Bank Ltd, Civil Appeal No 37 of 2006 (Court of Appeal in Nairobi). 69 HPA, s 5(1). The agreement also had to be in English. 70 HPA, s 5(4). 71 David Mburu Kamau v National Industrial Credit Bank (n 57). 72 HPA, s 15(1). 73 HPA, s 15(2). These provisions mirror ss 90 and 91 of the UK Consumer Credit Act 1974 where the protection is triggered by payment of one-third of the purchase price. 74 MPSR, s 65(4).

Pre-Existing Legal Framework and Registries 99 of debtors.75 Although the Model Law does not provide such protections, it defers to consumer protection legislation that may restrict the enforcement rights of secured creditors.76 Notwithstanding, there is a difference between the two approaches as the Model Law and other modern secured transactions legislation aim to protect consumer grantors, but the HPA protects unincorporated business hirers as well as consumer hirers, albeit only in relation to fairly small extensions of credit. In this, it mirrors the UK Consumer Credit Act 1974, which applies both to consumer credit agreements and business credit agreements entered into with an individual up to a value of £25,000.77

3. Retention/Transfer of Ownership Under section 19(1) of the Sale of Goods Act, a buyer acquires title to the goods when the parties intend for the title to pass from the seller to the buyer.78 Furthermore, under section 21, the seller may reserve the right of disposal until the buyer satisfies certain conditions. A seller may reserve a right of disposal with respect to the goods until certain conditions are fulfilled, and only then shall the property pass to the buyer.79 Thus, the seller can use the reservation of the right of disposal as a security for the payment of the purchase price. The Sale of Goods Act is a copy of the English Sale of Goods Act, which contains equivalent provisions.80 A uniform pre-MPSR practice for the financing of motor vehicles was for the lender to name itself as the co-owner of the motor vehicle in the records of the National Safety and Transport Authority.81 On registration, the Authority would issue a logbook evidencing the co-ownership right and deliver it to the creditor. Kenyan creditors were comfortable with this mechanism, particularly because processing times (approximately one day) were significantly shorter those of other registries.82 Furthermore, the retention of the logbook effectively precluded the borrower from selling the vehicle without any prior authorisation from the creditor. The naming of the creditor as a co-owner also limited the possibility of the debtor fraudulently claiming the loss of the logbook and procuring a new logbook without any notice to the creditor. Receivables could be financed under various factoring arrangements where the factor would purchase the receivables outright. One such arrangement, in the form of an invoice factoring facility, provided to a transport and haulage company operating in East Africa, was considered by the High Court of Kenya in Trans National Bank Limited v SWIFT Truckers Limited (in receivership). In that case, the borrower argued that the agreement was 75 See, eg, UCC §9-620(e). 76 ML, Art 1(5). 77 Consumer Credit Act 1974 s 8(3) and Art 60C of the Financial Services and Markets Act 2000 (Regulated Activities) Order 2001 (2001/544). 78 Laws of Kenya, ‘Sale of Goods Act’ ch 31 (2009). 79 H Beale, M Bridge, L Gullifer and E Lomnicka, The Law of Security and Title-Based Financing, 3rd edn (Oxford, Oxford University Press, 2018) 4.22. 80 These provisions are the basis for a body of English law on sales of raw materials and stock for resale. Under this body of law, retention-of-title clauses worked in relation to the goods themselves but not in relation to products or proceeds because the interest created there was a registrable (but universally unregistered) charge. ibid ss 7.12–7.25. 81 See further http://ntsa.go.ke/index.php?option=com_content&view=article&id=227&Itemid=719. 82 See above n 1.

100 Kenya unenforceable for want of stamping and registration.83 The court rejected all arguments questioning the enforceability of the factoring agreement, and concluded that the Schedule to the Stamp Duty Act did not refer to a factoring agreement requiring stamping and that the transaction did not create a charge that would require registration. The effect of the functional approach of the MPSR is that ownership rights of sellers of goods and transferees of receivables are characterised as security rights.

4. Companies Act In 2015, Kenya adopted a new Companies Act No 17, which replaced the previous legislation enacted in 1962.84 The reform package included a new Insolvency Act.85 Both statutes were based on English legislation. The Companies Act has 1,026 sections that, together with the Schedules, contain over 1,600 pages. It is, by far, the most extensive legislation ever enacted in Kenya.86 Part XXXII specifically dealt with company charges, including mortgages.87 The MPSR was intended to expressly repeal much of Part XXXII, particularly the sections covering charges in movable property, and initial drafts of the repeal provisions of the MPSR identified the specific sections of the Companies Act that would be repealed. However, the MPSR did not, eventually, include those provisions. This approach differs from that followed elsewhere, including in the Canadian Provinces, Australia and Malawi, where the affected provisions of the companies legislation was expressly repealed. The Kenyan approach thus left some doubts as to the actual effect of the MPSR’s consequential amendments. The Companies Act was adopted to apply only to charges88 specifically enumerated in section 878(4), which sets out a detailed, though reasonably comprehensive, list which mirrors that pertaining in English law until 2013.89 A charge must be registered within 30 days (previously 42 days)90 of its creation91 by lodgement of the particulars of the charge, prescribed in the Regulations, and the document that created or evidences the charge.92 The Registrar registers the particulars in the register, having checked these against the document, and issues a conclusive certificate verifying that the charge has been effectively registered.93 The failure to register a charge renders it void as against a liquidator, an administrator and a creditor of the company.94 Thus, for example, in one case a Kenyan court held that an unregistered charge was void against a creditor who had completed execution.95 83 Civil case No 451 of 2010. 84 Companies Act, 1962 (Rev 2009). 85 Insolvency Act, No 18 of 2015. 86 See R Harney, ‘The New Companies Act 2015 Has Come into Operation in Kenya’, www.bowmanslaw.com/ wp-content/uploads/2016/08/The-New-Companies-Act-2015-Has-Come-into-Operation-In-Kenya.pdf. 87 Companies Act s 877(1). Part XXXII of the Companies Act is based on part 25 of the UK Companies Act 2006, but before that section was amended by the Companies Act 2006 (Amendment of Part 25) Regulations 2013/600. 88 It should be noted that the word ‘charge’ in part XXXII of the Act includes a mortgage, s 877(1). 89 UK Companies Act 2006, s 860(7), now repealed. 90 Companies Act 1962, s 96. 91 ibid s 885(1)(a). 92 ibid s 878(1). 93 ibid s 884. This was the registration system in English law until the 2013 reforms. 94 ibid s 889(1). 95 Kenya Chemical & Allied Workers Union v Packaging Manufacturers Ltd [1976] Cause No 130 of 2012 (Industrial Court at Mombasa).

Pre-Existing Legal Framework and Registries 101 Companies may create floating or fixed charges.96 Assets subject to an uncrystallised floating charge may be disposed of by the chargor company in the ordinary course of business, thus enabling a chargee to take a floating charge (or, more usually, a combination of fixed and floating charges) over all the assets of a company.97 The priority position of a floating charge holder is quite precarious, particularly before crystallisation of the charge. For example, it has priority over an execution creditor only if its charge crystallised before the creditor has completed execution over the assets.98 An example of how the floating charge works is found in Noor Abdikadir Abdalla,99 where a company sold a trailer, which was subject to an uncrystallised floating charge, to a third party. As the sale was in the ordinary course of business, the court held that the buyer took free of the floating charge. Further, the priority of a floating charge which secures a debenture is subordinated to preferential creditors if the holder of the debentures take possession of the charged assets.100 On the liquidation of the chargor, the priority of the floating charge is similarly subordinated to preferential creditors,101 and also to the expenses of the liquidation102 and, in relation to a proportion of floating charge assets, to a fund ring-fenced for unsecured creditors.103 If the chargor company goes into administration, the administrator has the right to dispose of floating charge assets without leave of the court.104 The holder of a ‘qualifying floating charge’105 can appoint an administrator out of court, which is a very significant advantage. Whether a charge is fixed or floating is an issue which has been the subject of a great deal of English case-law, and is complicated and uncertain.106 Since the MPSR adopts a functional approach, it applies equally to all types of security rights in movables, however denominated by the parties, including fixed and floating charges.107 The distinction between the two, then, disappears for the purposes of the MPSR. Thus, for example, the rule in relation to priority between an execution creditor and a floating charge holder, mentioned above, no longer applies and the priority position is governed by section 46 of the MPSR, which applies to all security rights. However, the sections of the Companies Act and Insolvency Act which provide for the priority of preferential creditors have not been changed, so that it is still necessary to determine whether a security right creates a floating charge in order to discover over which assets the preferential creditors have priority. This is particularly odd, since the tax authorities are included in the list of preferential creditors in the Second Schedule of the Insolvency Act,108 yet tax claims can now be registered in the MPSR Registry and their priority position is governed by section 46. Moreover, the criterion for determining which creditor can appoint an 96 For further discussion, see ch 1, D1b. 97 Noor Abdikadir Abdalla v Grofin East Africa Fund LLC and Sood Petroleum Ltd Civil Case 316 of 2009. 98 Lochab Brothers v Kenya Furfural Co Ltd KLR 257 (1983). 99 See above n 97. 100 Insolvency Act s 582. 101 ibid s 471 and Second Schedule. 102 ibid s 473. 103 ibid s 474. 104 ibid s 587. The charge extends to any proceeds of the disposition, but if the assets are used to pay pre-existing debts there may be nothing left for the floating chargee. 105 Defined in Insolvency Act s 534: a qualified floating charge holder is a person who has security rights over all or substantially all the assets of the company. 106 See ch 1, D1b. 107 See MPSR, s 4(1)(b). The functions of the floating charge are replicated in a reformed system, see ch 1, D1b. 108 Insolvency Act, Second Schedule para 4.

102 Kenya administrator out of court remains whether that creditor is the holder of a qualifying floating charge, despite the fact that the MPSR amended the relevant section of the Insolvency Act109 to make the MPSR determinative of the priority of floating charges if there is more than one. The background to the lack of amendment of the Companies and Insolvency Acts in this regard is discussed below.110 The recent enactment of the Companies Act led some to question, in the early stage of designing the approach for the secured transactions reform, whether (mainly for political reasons) the MPSR should exclude company charges and apply only to security interests created by non-corporate grantors. This limited approach would have ensured that the bill passed more smoothly through parliament. However, the Technical Working Group (TWG), tasked with the development of the bill, ultimately acknowledged that this would be a serious departure from international best standards, and would undermine the economic benefit of the reform. After careful discussions, the TWG decided to introduce changes to the Companies Act, first, to replace and (where appropriate) supplement the concept of a charge with a security right; secondly, to discontinue, for security interests covered by the MPSR, the requirement of registration of charges with the Registrar of Companies; and, thirdly, to align some terminology with that of the MPSR. The intention was for the Companies Act no longer to provide for the registration of charges that have been absorbed by the MPSR concept of a security right, and for the registration functions previously performed by the Registry of Companies with respect to such charges to be taken over by the Collateral Registry. However, section 832(3)(c) of the Companies Act, which listed as one of the component parts of the Registry of Companies ‘certificates of registration of charges’ was not repealed; instead, the reference to ‘charges’ was changed to a reference to ‘security rights’. The Registrar’s duty thus remains to maintain the Registry and the information recorded on it.111 This could either mean that double registration (in the Companies Registry and in the Collateral Registry) is required if the security right is created by a company and would constitute a charge, or that an interested party must simply provide a certificate of registration under the MPSR to the Registrar of the Companies Registry for information purposes only. The latter interpretation is more closely aligned with the intentions of the TWG. The repeal of section 854(1)(h) of the Companies Act requiring the Registrar not to disclose ‘the contents of a document creating or evidencing a charge and lodged with the Registrar for registration under Part XXXII’ may support that interpretation in that it demonstrates that the drafters of the MPSR assumed that there would no longer be any such documents. It is troubling that section 878(4) of the Companies Act (which requires registration of company charges) has not been amended to delete the references to those registrable charges which now constitute security rights under the MPSR. This omission creates confusion because one provision of the Companies Act now requires the Companies Registry to maintain a register of certificates of registration of security rights, presumably certificates issued by the Collateral Registry under the MPSR, and another provision requires the registration of charges that would constitute security rights under the MPSR. Section 878(4) should have been amended to no longer require the registration of charges that constitute MPSR security rights.

109 ibid

s 534. C9 below. 111 Companies Act, s 823(2) and (4). 110 See

Movable Property Security Rights Act 103 The sanction for non-registration, in section 889, has been amended,112 so that a security right, which is registered in the Collateral Registry, will not be void against the liquidator, administrator or (secured)113 creditors. This is further evidence that a dual registration scheme was not intended. However, section 878(1), which states that the obligation to register is that of the grantor, has not been amended, and if the grantor does not comply with that obligation it commits a criminal offence under section 878(8). These provisions also should have been amended in relation to MPSR security rights. The MPSR, in section 94, also introduced a number of additional changes, such as in the definition of ‘retention of title agreement’, adding words to change its effect into an acquisition security right. In the Companies Act, the retention of title agreement was defined as an agreement ‘under which, if the seller is not paid and the company is wound up, the seller will have priority over all other creditors of the company with respect to the goods or any property representing the goods’.114 Under section 4(1) of the MPSR, the retention of title agreement would create a security right of an acquisition nature.

C. Movable Property Security Rights Act The National Assembly passed the Movable Property Security Rights Bill on 21 March 2017 as Act No 13 of 2017, and the MPSR was given assent on 10 May 2017. Unlike in Sierra Leone or Zambia,115 its entry into force was expressly deferred until the Cabinet Secretary submitted a notice to that effect in the Gazette.116 This occurred on 15 May 2017, naming the commencement date as 16 May 2017, so that the Act came into force on 16 May 2017. The dates and the order of events leading to the operationalisation of the Collateral Registry raise many questions.117 Most importantly, on 16 May 2017 when the MPSR was brought into force, the Collateral Registry was not operational, and this did not occur until 30 May 2017, and even then not all of its functions were made available to the users. As a result, the MPSR was, in effect, without a registry for a period of two weeks, which created considerable legal uncertainty and effectively shortened the transitional period to ‘re-register’ prior security rights.118 The Registry Regulations, in turn, entered into force on 24 May 2017. The Collateral Registry should not have been launched so soon after the enactment of the substantive law. Typically, the reform process should allow time for the proper design and testing of the registry system as well as delivery of a robust capacitybuilding programme. The MPSR was based on the Model Law, and is arguably the most faithful implementation of that model to date, at least in Africa. In addition to the challenges in coordinating its effect with the Companies and Insolvency Acts, it does contain some approaches and provisions that depart from those of the Model Law, which are explained below. Although 112 By the addition of the words ‘unless the requirements of the applicable laws for the charge or security right to become effective against third parties have been satisfied’. 113 This word does not appear in the statute but is taken to be included; see Re Monolithic Building Co, [1915] 1 Ch 643. 114 Companies Act, s 3. 115 See Sierra Leone C6 and Zambia B and E. 116 MPSR, s 1(2). 117 See further C4 below. 118 See further C9 below.

104 Kenya the substantive law withstands the test of the international best standards, the registry regulations and design were put together too hastily for the registry provisions of the MPSR to be implemented properly.

1. The Scope and the Functional Approach The MPSR follows the functional approach, which subjects all transactions that secure an obligation with a movable asset to its scope.119 It does so explicitly through the scope provision in section 4(1)(a), which replicates Article 1(1) of the Model Law, and the definitions, including that of ‘security right’ which replicates Article 2(kk) of the Model Law and includes outright transfers of receivables.120 During the deliberations on the bill, one of the coauthors of this book advised the TWG. The functional approach121 was one of the most difficult aspects of the reform for the TWG to understand, and agreement on its inclusion was difficult to reach, as it would have a dramatic impact on the classification of property rights, changing the then existing (common law based) legal landscape. Subjecting security rights to registration was not an issue, since company charges, hire-purchase agreements and instruments creating chattel mortgages were already registrable.122 Moreover, at the same time, Kenya was considering passing a leasing statute that contemplated registration of the ownership rights of lessors, but without altering the nature of the lessor’s ownership right. Upon registration, the lessee would not have any rights in the leased object or any power to create a competing interest. Taking into account this concurrent reform effort, and after some discussions, the TWG agreed that the MPSR should recharacterise financial leases as security rights, incorporating the definition of financial lease from the UNCITRAL Legislative Guide.123 The MPSR does not rely solely on the generality of the provisions that provide for the functional approach along the lines of Articles 1(1) and 2(kk) of the Model Law, but, rather, includes some very specific provisions to avoid any confusion as to its scope. Section 4(1)(b), which was based on the PPSAs,124 provides that the MPSR applies to security rights in movable assets ‘without limiting the generality of paragraph (a), a chattel mortgage, credit purchase transaction, credit sale agreement, floating and fixed charge,125 pledge, trust indenture, trust receipt, financial lease and any other transaction that secures payment or performance of an obligation’. The definitions of ‘credit purchase transaction’ and ‘credit sale agreement’,126 included in section 2 of the MPSR,127 were copied from section 2(1) of the Insolvency Act, although the MPSR refers to ‘credit sale agreement’ rather than ‘conditional sale agreement’. This slight misalignment of the terminology should not be interpreted as a substantive difference, as the intention of the drafters was to fully align the terminology and

119 MPSR,

s 4(1)(a). s 2. 121 See ch 2, B2. 122 See above, B1, B2, B4. 123 MPSR, s 2. See Legislative Guide 9. For a discussion of recharacterisation, see ch 2, B2b. 124 See Saskatchewan PPSA, s 3(1)(b). 125 See discussion above at B4. 126 An identical definition also appears in s 3(1) of the Companies Act. 127 Trust indenture and trust receipt are also defined in s 2 of the MPSR. 120 MPSR,

Movable Property Security Rights Act 105 concepts of the two statutes. Moreover, section 4(1)(c) provides that the MPSR, with the exception of the Enforcement Part, also applies to outright transfers of receivables. Section 4(2) lists the exclusions from the scope of the MPSR along the lines of the Model Law, including security rights in intermediated securities (book-entry securities) covered by the Central Depositories Act of 2000 (CDA). Representatives of the Kenyan central securities depository128 participated in the discussions of the TWG. Accordingly, Kenya did not blindly implement the rule of the Model Law on this aspect. Section 42 of the CDA provides for the creation of charges over book-entry securities, and states that the chargor may create only a single charge, and any subsequent charge shall be void. Floating charges, however, are not governed by the CDA and may be created under the common law.129 The creation of charges under the CDA is very cumbersome and requires the parties to submit the charging instrument to the depository agent for registration in the registry of charges maintained by the central depository. The charged securities will be debited from the securities account of the chargor. The framework prescribed by the CDA for creating charges in book-entry securities differs from the international best standards as provided in the UNIDROIT Convention on Substantive Rules for Intermediated Securities (Geneva Securities Convention).130 Only central depository agents may open securities accounts with the Central Depository & Settlement Corporation (CDSC) that they maintain for customers or as proprietary accounts. When a customer wishes to pledge securities, it must complete a standard pledge form and submit it for approval to the agent that opened the relevant securities account and the lender. The completed form must be delivered to the CDSC, which ‘freezes’ the securities.131 This process is much more complicated than one which complies with international best standards and the UNIDROIT Convention. As part of the overall secured transactions reform process, reformers in Africa and elsewhere should take the opportunity to review the rules applicable to securing obligations with rights in intermediated securities, particularly the procedural requirements. In most cases, such rules are inadequate and desirous of modernisation along the lines of the UNIDROIT Legislative Guide on Investment Securities.132

2. Creation As under the Model Law,133 a security right is created under the MPSR upon execution of a security agreement that includes the mandatory information listed in section 6(3). The MPSR did not incorporate the optional feature of requiring a statement of the

128 See Central Depository & Settlement Corporation (CDSC), ‘Company Profile’, www.cdsckenya.com/about-us/ company-profile. 129 Central Depositories Act, s 42(6). 130 See UNIDROIT, ‘Convention on Substantive Rules for Intermediated Securities’ Arts 11–13 and Chapter V of ‘the Geneva Securities Convention’ (25 September 2013) www.unidroit.org/instruments/capital-markets/ geneva-convention. 131 See Central Depository & Settlement Corporation (CDSC), ‘Depository Services’, www.cdsckenya.com/ products-services/depository-services. 132 See UNIDROIT Legislative Guide on Intermediated Securities, ‘Implementing the Principles and Rules of the Geneva Securities Convention’, www.unidroit.org/instruments/capital-markets/legislative-guide. 133 ML, Art 6.

106 Kenya aximum amount for which a security right may be enforced.134 However, as will be m explained below, this feature caused some confusion and challenges in the drafting of the Registry Regulations.135 Unlike its model, the MPSR requires all security agreements to be in writing, whether or not the secured creditor is in possession of the collateral. This approach was chosen to minimize the risk of disputes, which tends to be higher when agreements have not been recorded in writing. The drafters did not wish to assign a significant role to possession within the context of the MPSR, which also does not recognise it as a method of third-party effectiveness.136 Section 6(4) was inserted to ensure that the requirements of other legislation, similar to the Stamp Duty Act, would not apply to security agreements. The sections of the Stamp Duty Act affecting secured transactions under the MPSR were repealed on the entry into force of the MPSR.137 Prior to the MPSR, agreements creating security rights had to be stamped. Although the fee itself was relatively modest, the stamping was done manually and generally required the presentation of the agreement at an office empowered to affix stamps. The MPSR thus minimises transactional costs associated with the creation of security rights. Such costs remain a challenge in many other African economies.138 However, the effect of section 6(4) remains unclear with respect to the HPA. As mentioned earlier,139 the intention of the reformers was to repeal only those provisions of the HPA that provided for the registration of hire-purchase agreements, preserving the protective provisions in enforcement. Section 6 of the HPA, which has not been repealed, provides for a number of requirements for hire-purchase agreements that go beyond those prescribed by section 6 of the MPSR. By virtue of this section, it appears that a hire-purchase agreement creates a security right even when it does not comply with section 6 of the HPA. However, such a security right might not be enforceable since section 65(4) allows the enforcement of a security right created by a hire-purchase agreement to proceed only under the HPA. This analysis illustrates the complexity of navigating through a framework in which parts of the pre-reformed legislation that should have been repealed remain in force. Following the Model Law,140 the MPSR permits a super-generic (all assets) description in a security agreement. Taking inspiration from the UCC Article 9,141 it provides further guidance by stating that description can reasonably identify the collateral if it identifies it by specific listing, category, type or quantity.142 The MPSR faithfully implements the remaining articles from the creation chapter143 of the Model Law, including the provision overriding contractual limitations on the creation of security rights in receivables.

134 ML, Art 6(3)(d). 135 See below C4. 136 See below C3. 137 See s 94 of the MPSR that repealed ss 31 and 38 of the Stamp Duty Act. 138 M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in Akseli and Gullifer (n 7) 200. 139 B2 above. 140 ML, Art 8(d). 141 UCC §9-108(b). 142 MPSR, s 8(3). 143 ML, ch II.

Movable Property Security Rights Act 107

3. Third-Party Effectiveness The initial draft of the MPSR reflected all the methods of third-party effectiveness prescribed in the Model Law. However, after some discussions, and after learning about the approaches in other African economies undertaking similar reforms, the TWG decided that registration should be the sole method of third-party effectiveness, as the alternatives were considered impractical (possession), not sufficiently transparent (automatic), or unknown in the local environment (control). Even for security rights in negotiable documents and instruments that are commonly delivered to secured creditors, registration is the sole method of third-party effectiveness. Accordingly, the third-party effectiveness chapter of the MPSR is significantly shorter than that of the Model Law.144 For security rights in ‘noncash proceeds’, the MPSR designates a period of 10 days during which the security right will remain effective against third parties before registration is required for such effectiveness.145

4. Registration Part IV of the MPSR provides a comprehensive basis for the establishment and operation of a modern, electronic, notice-registration system. In addition to the registration of notices with respect to security rights, the MPSR contemplates the registration of notices with respect to rights of non-consensual creditors.146 ‘Non-consensual creditor’ is defined in section 2 to include not only holders of judgments, but also any authority that is owed taxes. Accordingly, the MPSR subjects ‘tax liens’ to the registration and priority scheme applicable to security rights in order to enhance the transparency and credit-friendliness of the framework (Kenya’s Treasury Department was one of the co-sponsors of the reform project). The registration provisions of Part IV closely follow the registration provisions of the Model Law. Nevertheless, the following paragraphs highlight and explain some deviations as well as the options chosen by the MPSR. Some articles of the registry provisions of the Model Law (Registry Provisions) allow enacting States to implement one of several options. For instance, Article 30 invites States to choose either an option whereby the registry must remove a registered notice on registration of a cancellation notice or the expiry of its effectiveness, whichever is earlier, or an option requiring removal only on the expiry of effectiveness. The MPSR chose the latter approach in section 21(1). Section 30 of the MPSR implements Option C of Article 14 of the Registry Provisions, permitting registrants to select the period of effectiveness, which cannot exceed 10 years.147 An amendment to continue the effectiveness of the registration may be registered only within the last six months prior to the expiry of the registration.148 Article 20(6) of the Registry Provisions contemplates a summary judicial or administrative procedure to facilitate the e nforcement of the rules for

144 ML,

ch III. Art 19(2)(a) invites enacting states to designate an appropriate short period of time. 146 MPSR, s 19(2). 147 ML, Art 14(1) Option C. 148 ML, Art 14(2) Option C. 145 ML,

108 Kenya compulsory amendments and cancellations. Section 33(5)–(7) of the MPSR provides for such an administrative process whereby the grantor may petition the Registrar when it believes that a registered notice should have been amended or cancelled. The Registrar must then notify the secured creditor of the request. Any decision by the Registrar may be appealed to the court. The other significant choice facing enacting States is with respect to the search logic – either strict or flexible. Sections 34 and 35 of the MPSR opted for strict search logic, under which any error in the identifier of the grantor (as defined in the Registry Regulations) will not retrieve the registered notice, rendering it ineffective. The MPSR does not permit searches by the registration numbers of notices because modern registry design wherein all registered notices submitted by a registrant are stored in a user account rendered that search functionality unnecessary. With respect to the three approaches concerning the effect of a transfer of an encumbered asset in Article 26 of the Registry Provisions, section 37 of the MPSR implements Option B, which designates a period of 10 days within which the secured creditor must register an amendment notice adding the transferee after acquiring knowledge of the transfer in order to preserve its priority position against security rights created by the transferee and transfers of the collateral to a buyer. As for the approaches not contemplated by the Model Law, the most fundamental one, though foreshadowed in the UNCITRAL Registry Guide, is the provision for indexing and searching by the serial number of specific items of collateral (accompanied by the relevant priority rules). In this regard, Kenya chose to follow the approach common to other African reformers that implemented the PPSA models.149 Section 2 defines ‘serial number’ and ‘serial-numbered collateral’ as limited to motor vehicles and trailers, thus resisting the temptation to broaden the categories of assets beyond those for which serial number indexing/searching has practical utility. As part of the reform process, the relevant office for the registration of rights to motor vehicles150 was visited, its processes studied, and the contemplated approach for the MPSR was discussed with the relevant officers. It is important for motor vehicle registration agencies to be apprised of the approaches in secured transactions laws that would affect their operations. Under section 27(1)(c) of the MPSR, serial-numbered collateral does not need to be described in a registered notice by a serial number if it is held as inventory, but, otherwise, if the registrant selects a motor vehicle or trailer from the drop-down list of the types of collateral, it must enter a serial number. When the registrant selects ‘all assets’ or ‘equipment’ from the drop-down list, and the collateral also includes a motor vehicle, a serial number need not be entered. An error in the serial number renders the security right ineffective, but only against buyers and lessees of the relevant asset.151 An error should be understood to include any omission of a serial number, such as in the case of a registration that relates to all assets of the grantor. A serial number is one of the two search criteria under which a search may be conducted.152

149 Nigeria E3; Sierra Leone C3; Liberia C5b; Zambia D2; Rwanda C6. 150 This includes title rights and also other claims against the vehicle. 151 MPSR, s 35(6). This is because a buyer or lessee would be expected to search using the serial number as identifier, whereas a (prospective) secured creditor or insolvency trustee would be expected to search using the grantor’s identifier. 152 MPSR, s 34(1).

Movable Property Security Rights Act 109 The other approach deviating from the Model Registry Provisions, but consistent with many reform projects in Africa,153 is the requirement to enter statistical information154 relating to matters such as the industry in which the grantor operates, the size of the business (if the grantor is an entity) and whether the business is majority owned by women. The statistical information is removed from the public registry record, and thus cannot mislead a searcher. For the sake of clarity, section 35(4) of the MPSR attaches no legal significance to errors in statistical information, but any lack of information will lead to the registration being rejected by the system. Finally, section 24 of the MPSR does not permit registration of a notice prior to the grantor providing authorisation to do so, unlike Article 2(4) of the Registry Provisions.155 Even though the Collateral Registry does not verify whether an authorisation has been given, the TWG felt that providing an express statutory authority for registrations that are not authorised by the grantor would be politically challenging, and that critics of the reform could use this issue as a ground for impeding the enactment of the MPSR Bill. Part IV of the MPSR is supplemented by the Movable Property Security Rights (General) Regulations 2017 (‘the Registry Regulations’).156 The Registry Regulations were not properly drafted; rather, they were hastily put together only a couple of months after the adoption of the MPSR, with a view to operationalising the Collateral Registry as quickly as possible. This hasty implementation process may have been motivated by the desire for the reform to be taken into account in the 2018 World Bank Doing Business Report, which had a cut-off date of 30 May 2017. The Registry Regulations included a number of deviations from the MPSR, both in substance and language. For instance, regulation 8(2)(h) requires the registrant to enter, in an initial notice, a statement of the maximum amount secured by the security interest. Including statements of the maximum amount of the secured obligation in security agreements or notices, along the lines of the optional features of the Model Law,157 was discussed and rejected during the deliberations on the MPSR Bill.158 Furthermore, section 27(1) of the MPSR clearly states the information that must be included in an initial notice, making regulation 8(2)(h) plainly inconsistent with the policy as well as the drafting of the MPSR. Other provisions of the Registry Regulations are unclear. An example of this is found in regulation 11(3), which provides ‘[t]he Registrar may provide specific categories of the collateral’. The effect of this provision is not entirely clear: one possible interpretation is discussed in the next paragraph. The Registry Regulations also departed from the MPSR with respect to wording, such as regulation 12, which refers to ‘applying for registration in a specific form’ rather than ‘submitting a notice’.159 An application for registration implies that the recipient of that application, who is the Registrar, evaluates it and either accepts or rejects it, which is no longer the case under the MPSR. The Registry Regulations include forms (notices) in the First Schedule, but the reason for this is unclear as the system is fully electronic.160 Furthermore, the forms 153 See Sierra Leone C3; Nigeria F; Malawi A, D; Rwanda C6; Liberia C5a. 154 MPSR, s 27(1)(e). 155 This subsection is not included in the MPSR. 156 Kenya Gazette Supplement No 80, Legislative Supplement No 39. 157 ML, Arts 6(3)(d) and 8(e). 158 See C2 above. 159 See also Regulation 14 that provides ‘A registrant … shall apply in Form 9 … ’. 160 It should be noted that s 88(2)(b) provides that the regulations may provide for the format of notices to be registered in the Registry.

110 Kenya contain boxes for information which is not required by either the MPSR or the Registry Regulations, such as a Kenya Revenue Authority (KRA) PIN number and a date of birth. In part 4 of the initial notice form (Form 3), it provides a field for the registrant to enter ‘details of the collateral’, including a list of checkboxes for various types of assets. The purpose of regulation 11(3) is likely to have been to give the Registrar power to specify these categories of collateral. Unfortunately, the list contains some overlaps, such as entries for both ‘inventory’ and ‘stock in trade’, as well as some items which are not categories of collateral, such as future proceeds. The drop-down menu does not include a category for equipment, which has been added in the proposed amendments. Finally, the forms allow registrants to check-off a box for immovable property, even though the MPSR does not apply to immovable property at all. The forms thus add to the confusion caused by the many departures of the Registry Regulations from the MPSR. The second schedule to the Registry Regulations prescribes a fee of 500 shillings for searches. All registrations are free of charge. The troublesome implementation of the Collateral Registry, coupled with the absence of a training and a public awareness campaign, has already produced some negative, though not unexpected, results. For example, in the first few months, the number of registrations exceeded the number of searches 100-fold.161 This is a staggering ratio: under normal circumstances the number of searches would exceed the number of registrations at least two-fold because prudent lenders first search, then register, and then search again before disbursing funds to ensure there is no other notice registered in the interim. A considerable majority of all registered notices covered household items and furniture, which illustrated the reluctance of lenders to engage in commercial transactions. Thousands of registrations also designated immovable property as collateral, providing clear evidence of the need to align the registry system fully with the MPSR as well as for capacity building.

5. Priority Given that registration is the sole method of third-party effectiveness, the priority provisions of the MPSR are much simpler than those of the Model Law. A number of provisions were simplified, such as the general approach of ‘first to register or perfect has priority’,162 which section 38 of the MPSR states as ‘first to register has priority’.163 Additionally, some Model Law provisions, such as the asset-specific rules, did not need to be included.164 Aside from the effect of this approach on third-party effectiveness, the priority rules of the MPSR implement the provisions of the Model Law.165 Although delivery of possession of highly transferable assets such as negotiable instruments, non-intermediated securities or money does not make the security right effective against third parties, a person taking 161 Statistics for October–December 2017 are available with the authors. 162 ML, Art 29(c). 163 Other ML Articles, eg Art 31 on the effect of changing the method of third-party effectiveness on the priority, became unnecessary as well. 164 See, eg, Art 47 of the ML providing the priority structure for security rights in bank accounts because this depends on control being an available method of third-party effectiveness. 165 See, eg, s 40 on the irrelevance of knowledge in determining priorities (ML, Art 45) and s 41(2) on the priority of a security right with respect to future collateral (ML, Art 44(2)).

Movable Property Security Rights Act 111 possession may acquire the status of the transferee and take free of the security right.166 The MPSR does not contain any priority rule protecting transferees of negotiable documents, unlike Article 49 of the Model Law. The status of transferees of warehouse receipts is being addressed through a warehouse receipts reform project, but that of transferees of other types of negotiable documents is still governed by common law. This omission may leave some questions open with respect to the other type of negotiable document used in practice, which is the bill of lading. Banks take possession of bills of lading, issued or endorsed to their order, for the financing of import/export transactions under letters of credit. Since the priority of a holder of a bill of lading as against a competing security right made effective against third parties by registration in the Collateral Registry is not governed by the MPSR, banks that finance imports of goods should in addition register a notice in the Collateral Registry. Otherwise, the priority of their rights as holders of bills of lading against security rights would be determined by the common law. Nonetheless, implementation of Article 49 of the Model Law would have provided for a more predictable solution. In some other respects, the MPSR deviates from the UNCITRAL models. For example, the Model Law did not incorporate the specific recommendations of the Legislative Guide regarding attachments to immovable property. However, under section 44 of the MPSR a security right may be created in an attachment to immovable property, whether pre- or post-affixation, and a security right made effective against third parties under the MPSR has priority over any right or interest made effective against third parties under immovable property law. Section 2 defines an attachment to immovable property as ‘a tangible asset that, despite the fact that it is physically affixed to immovable property, it [sic] is treated as movable property’. In contrast, the Legislative Guide defines an attachment to immovable property as ‘a tangible asset that is so physically attached to immovable property that, despite the fact that it has not lost its separate identity, it is treated as immovable property’. It is unclear whether the section 2 definition provides that attachments of this nature will be treated as movable property under other law, especially the land law of Kenya. Furthermore, movable property is not defined in the MPSR. This definition raises additional questions as to which types of attachments will continue to be treated as movable assets, and whether this is irrespective of the nature of the damage their removal may cause. Irrespective of what the land law provides with respect to attachments, this provision should not be read to allow the creation of security rights in building materials or assets, the removal of which would cause irreparable damage. Although no limits are expressly imposed on the creation and enforcement of security rights in such attachments to immovable property, secured creditors have an overarching duty to proceed diligently and in good faith imposed by section 5(2), and this sets the parameters for enforcement actions with respect to removing attachments. The duty to exercise rights and perform obligations ‘diligently and in good faith’ corresponds to the standards of commercial reasonableness imposed under Article 4 of the Model Law. The relevant priority rule of section 44, which presupposes a conflict between a security right under the MPSR and immovable property law, is inconsistent with the definition in section 2. A security right in an attachment of this kind may be created only under the MPSR and not under the Kenyan immovable property law since the attachment is defined as an asset which 166 See ss 52, 54 and 55. Similarly, transferees of electronic securities and funds may acquire free of the security right. These provisions mirror those of the ML.

112 Kenya is treated as movable property (presumably under Kenyan general law), and so there cannot be a conflict. The drafters recognised the dire legal and registration situation with respect to land rights and sought to establish a priority scheme that would give primacy to those who make their security rights in attachments effective against third parties by the registration of a notice (electronically and at a low cost) under the MPSR. However, that objective could have been achieved without the section 44 priority rule. The other substantial deviation is in section 47, which provides for the priority of an acquisition security right. The Model Law, in Article 38, provides two options that enacting States may choose from: the first one distinguishes between inventory and the assets held in a different manner, whereas the second one provides a single rule irrespective of the manner in which the collateral to be acquired by the grantor is held.167 The MPSR chose the latter approach. The Model Law invites enacting States to designate a grace period of time after the grantor takes possession of the tangible collateral or enters into an agreement for the sale of intellectual property, within which a notice of the acquisition security right must be registered to retain its ‘super-priority’. The MPSR takes a different approach and requires a notice to be registered before the grantor acquires possession or enters into an agreement in order for the acquisition secured creditor to have ‘super-priority’. This approach could create some practical challenges because it does not allow acquisition-secured creditors a grace period in which to register a notice, and thus could delay the execution of certain transactions. It was felt by the drafters of the MPSR that a grace period would create the risk of invisibility. However, this decision was made only after it had been agreed that the Collateral Registry would be an electronic and easily accessible system, which may ameliorate the difficulties for acquisition secured creditors. The other departure from the Model Law is that the MPSR did not expressly provide for the definition of an acquisition security right to include licences of intellectual property.168 An acquisition security right is defined in section 2 of the MPSR as ‘a security right in a tangible asset or intellectual property’, and ‘intellectual property’ is defined in the same section to include any right related to copyright, industrial rights and trademark. Arguably, a licence is a right related to those types of intellectual property rights, and therefore the relevant priority on acquisition security rights would apply to licences. The MPSR does not include an equivalent of Article 36 of the Model Law, which instructs enacting States to list preferential claims and their maximum amounts. This is because tax claims are treated as non-consensual claims subject to the third-party effectiveness and priority provisions of the MPSR, and ‘wage claims’ would ordinarily arise only in insolvency situations, although some uncertainty remains as to the effect of wage claims on the liquidation of a grantor outside of insolvency.

6. Rights and Obligations of the Parties and Third-Party Obligors Part VI of the MPSR includes nine of the 19 comparable sections of the Model Law. With respect to these provisions, the view was that such detailed coverage was unnecessary. Some members of the TWG also felt that the full inclusion of all articles would have made this Part 167 Ch 2, F2. 168 The definition of ‘acquisition security right’ in Art 2 of the ML expressly includes licences of intellectual property, but this is omitted in the definition in s 2 of the MPSR.

Movable Property Security Rights Act 113 and the entire Bill too long for the parliament. The nine sections in this Part, which include most of the important provisions concerning the rights and obligations of the third-party obligors where a security right is taken over receivables, do not deviate from the Model Law.

7. Enforcement Part VII of the MPSR deals comprehensively with the enforcement rights of secured creditors, following closely the provisions of the Model Law. It facilitates extrajudicial enforcement, although it should be noted that some self-help remedies were available to Kenyan creditors pre-MPSR under the common law and under specific statutes such as the HPA.169 In some respects, it deviates from the text of the Model Law, but those deviations are not substantial. For instance, an informal assessment as to whether the HPA should be repealed concluded that one of its main purposes was to protect vulnerable debtors against the abusive conduct of creditors, especially after default.170 It was agreed that the enforcement provisions of that Act served this purpose, and that, therefore, Part VII should not apply the same set of rules to entity grantors as to individual grantors acting as consumers and entity grantors entering into certain types of low-value business transactions. Article 1(5) of the Model Law contemplates that States may preserve or enact certain laws protecting debtors who used some of their assets as collateral in ‘consumer transactions’. Section 65(4) of the MPSR, which preserves the limitations on enforcement under the HPA,171 mirrors that policy. However, as explained above, that policy should have been implemented by a repeal of the HPA, and incorporating the protective enforcement provisions directly into Part VII of the MPSR. In commercial transactions, as opposed to those involving consumers or other vulnerable debtors, the MPSR is more protective of the rights of secured creditors than the Model Law. Section 65(1) entitles secured creditors to avail themselves of the remedies provided in this Part of the MPSR, but also those that may be agreed to in a security agreement or in some other ‘written law’,172 but, unlike the Model Law,173 it does not circumscribe their application by requiring their consistency with the MPSR. Nevertheless, there is some circumscription as secured creditors have an overarching duty to exercise their rights, including post-default remedies, ‘diligently and in good faith’, in accordance with section 5(2) of the MPSR. Section 67(3) provides an illustrative list of remedies that secured creditors may use, including appointment of a receiver,174 taking possession, leasing and selling an asset. Arguably, this list was not necessary as section 65(1) enables the secured creditor to use any remedies mentioned in section 67(3). The MPSR fleshes out certain aspects of enforcement treated only generally in the Model Law. For instance, Article 77(2)(b) of the Model Law requires that a notice of default and intention to repossess the encumbered asset be given to the grantor, without specifying the 169 See B2 above. 170 ibid. 171 ibid. 172 That is, a statute. 173 ML, Art 72(1)(b). 174 Appointment of a receiver is not provided for in the ML. The MPSR does not specifically provide for it either, but presumably permits secured creditors to include a provision empowering them to appoint a receiver.

114 Kenya content of the notice of default. Unlike the Model Law, section 67(2) of the MPSR provides for the minimum content of the notice of default. The substance of the provisions in sections 69–78 faithfully implements the relevant articles of the Model Law. These cover terminating enforcement, taking over the enforcement process, obtaining possession of the collateral, the disposal of collateral and the distribution of proceeds. In addition, they cover the acceptance of the collateral in satisfaction of the secured obligation, the rights of transferees of the collateral upon disposal, the collection of payment under receivables, negotiable instruments, non-intermediated securities and bank accounts, as well as the collection of payment under receivables transferred outright.

8. Applicable Law The conflict of laws aspect of the secured transactions reform raised some challenges for stakeholders, some of whom even suggested the deletion of this entire Part of the MPSR, which was faithfully copied from the Model Law. Some of their arguments revolved either around the nature of transactions to which the MPSR was expected to apply (largely lowvalue deals without any foreign element) or that the principles for the application of the MPSR could be found in the common law. Eventually, in the interest of certainty and predictability, a significant majority of the provisions from the conflict of laws chapter of the Model Law were retained. The MPSR incorporates the Model Law approach with respect to providing freedom to the parties to modify their contractual rights and obligations, while providing for mandatory rules applicable to the creation, third-party effectiveness, priority and enforcement of security rights. The elimination of the alternative methods of third-party effectiveness rendered the inclusion of some of the conflict of laws provisions of the Model Law unnecessary.175 Some provisions of the Model Law were also incorporated using a different structure. For instance, section 81 of the MPSR states not only the rule applicable to the creation, third-party effectiveness and priority of a security right in intangible assets, but also the rules on these matters with respect to special types of intangible assets, including deposit accounts, non-intermediated securities and intellectual property, which are dealt with separately in the Model Law. Some provisions of the Model Law, such as Article 92 on renvoi and Article 93 on public order, were not incorporated.

9. General Provisions Part IX, entitled General Provisions, governs three matters: the power to issue regulations; the transition of prior security rights; and the repeal/amendment of other laws. Some provisions of the Registry Regulations, issued pursuant to the power granted to the Cabinet Secretary under section 88 of the MPSR, have already been examined above.176 The transitional provisions replicate the rules of the Model Law. Section 92(1)(b) designated

175 ML, 176 See

Art 85(2). discussion in C3 above.

Movable Property Security Rights Act 115 a transitional period of nine months within which any prior security rights must be made effective against third parties to retain their pre-MPSR third-party effectiveness. As mentioned above, only two months elapsed between the enactment of the MPSR and the launch of the Collateral Registry. Two months is not adequate to design and test a registry system properly, nor is it sufficient to provide capacity-building exercises to the affected stakeholders. It is highly doubtful that the registry system operated properly straight after the launch, as it is inconceivable that all of its functions had been properly tested. The hasty launch also significantly reduced the time necessary for large lenders, in particular, to effect a transition of their prior security rights into the MPSR regime, as preparatory work commonly commences long before the transitional period starts running. Section 94 provides for amendment of various statutes, the details of which are set out in a schedule. The schedule provides that, first, the CTA is to be repealed in its entirety and, second, that various provisions of other statutes are to be amended, on the entry into force of the MPSR. The statutes to be amended are the Agricultural Finance Corporation Act, the Stamp Duty Act, the HPA, the Pawnbrokers Act, the Business Registration Services Act, the Companies Act, and the Insolvency Act. The traditional English law security devices referred to in the Companies and Insolvency Acts caused coordination challenges for the drafting of the MPSR because of the functional approach of the Model Law. Since these two statutes were enacted recently and were not fully operational, there was very little political appetite to make any changes that would even align the concepts and terminology with the MPSR, in particular, the provisions relating to the floating charge.177 However, the TWG recognised the consequences of disconnection between the secured transactions legislation on the one hand, and the Companies and Insolvency Acts on the other, as any uncertainty could hinder a successful implementation and application of the MPSR. At that time, there was no guidance to be obtained from similar reforms in Africa. For example, Malawi enacted a complete package of laws at the same time, whereas Zambia’s insolvency reform was in its infancy. The TWG thus looked elsewhere and identified two potential models: New Zealand and Australia.178 The New Zealand PPSA repealed most statutory provisions relating to secured transactions, including the Chattel Transfers Act 1924 (relating partly to retention of title devices) and the Companies (Registration of Charges) Act 1993, which covered registration of non-possessory security interests created by companies. However, some references in the Companies Act 1993, the Property Act 2007 and the Insolvency Act 2006 to mortgages and charges were left unchanged,179 although it was made clear that the PPSA applied to such interests.180 As discussed above,181 the most critical issue is the use of the term ‘floating charge’ to delineate the category of charged assets over which preferential creditors have preference. In New Zealand, the Seventh Schedule to the Companies Act set out which claims are preferential, and the criterion in that Schedule as to the assets over which preference exists was amended to align it with concepts in the PPSA.182

177 See

discussion in B4 above. ch 1, D1b. 179 See M Gedye, ‘The New Zealand Perspective’ in Akseli and Gullifer (n 7) 137. 180 New Zealand PPSA 1999, s 17(3). 181 See B4. 182 Companies Act 1993, Schedule 7 para 2. See further, ch 1, D1b. 178 See

116 Kenya In contrast, the Australian PPSA originally included provisions to redefine various statutory terms,183 including ‘charge’, ‘floating charge’ and ‘fixed charge’.184 The PPSA used the defined concept of ‘circulating assets’185 to provide that a floating charge was a ‘security interest that has attached to a circulating asset’ and a fixed charge was a ‘security interest that has attached to personal property that is not a circulating asset’.186 Subsequent legislation has amended all relevant statutes to include the new definitions of ‘fixed charge’ and ‘floating charge’,187 so that section 339 is now redundant.188 However, the approach of the Australian PPSA was not an ideal model for Kenya for various reasons. First, Australian insolvency law is contained in the Companies and Bankruptcy Acts, which has not been the case in Kenya since 2015 when a stand-alone Insolvency Act was enacted. Secondly, the Australian PPSA recognises multiple methods of third-party effectiveness, including possession and control, which the MPSR does not. Thirdly, the Australian PPSA applies to certain assets to which the MPSR does not (eg securities held in accounts maintained by the Central Securities Depository). Fourthly, the Australian PPSA has been subjected to a detailed review, which has recommended amendments to the PPSA’s insolvency-related provisions, including the definition of ‘circulating assets’ and the term ‘circulating security interest’, which replaced the previous words ‘floating charge’ in various pieces of legislation.189 Finally, the approach of redefining the relevant terms in the PPSA itself has now been superseded, as mentioned above. The unavailability of a model that could be readily applied in the Kenyan context required a careful consideration of the suggested modifications. The alignment of the notion of security right with the pledge in the Insolvency Act did not cause any consternation among the TWG members. Similarly, the TWG members were satisfied with altering the effect of retention of title devices, which were previously fully effective in insolvency without registration with respect to the goods to which the devices related,190 conditioning their effectiveness on the satisfaction of the third-party effectiveness provisions of the MPSR and essentially treating them as acquisition security rights. However, difficulties arose with respect to aligning the concept of a security right under the MPSR with that of a charge under the Insolvency Act. A wholesale replacement of the references to ‘charge’ with a ‘security right’ was not justified because the MPSR expressly excludes several assets from its scope over which charges may still be created under other laws, such as land, ships and aircraft.191 To address this, where appropriate, the Insolvency Act was amended to add the words ‘or a security right’ to a reference to ‘charge’. Given the absence of a relevant model and the challenges with the implementation of the recent reforms in Kenya, it remains to be seen whether these ‘coordination approaches’ will cause any practical difficulties. 183 See A Duggan and D Brown, Australian Personal Property Securities Law, 2nd edn (LexisNexis Butterworths, 2016) 417. 184 Australian PPSA, s 339. 185 Australian PPSA, s 340. 186 Australian PPSA, s 339(4) and (5). 187 Personal Property Securities (Consequential Amendments) Act 2009. 188 See B Whittaker, Review of the Personal Property Securities Act 2009, Final Report (2015) s 9.3.2.1. 189 ibid s 9.2.1.1.1. The reviewer called some of these provisions as ‘very detailed and technical, and at times difficult to apply’. 190 If a retention-of-title clause purported to extend to products or proceeds, this would have created a registrable charge. See Beale (n 79) ss 7.12–7.25. 191 MPSR, s 4(2).

Final Remarks 117 Problems are most likely to occur because of the retention of the fixed/floating charge distinction, which still governs the insolvency-related consequences discussed above.192 The reasoning behind the functional approach is that fine distinctions between different types of security interests become irrelevant.193 Although replacing the ‘floating charge’ trigger for insolvency consequences is not at all easy, it is unfortunate that finding a replacement more consonant with the principles of the MPSR was not explored further, as the retention of the concept of ‘floating charge’ is likely to give rise to complicated case-law in the long term. The other potential problems raised above194 in relation to the registration provisions in the Companies Act, while untidy, are unlikely to cause major problems in practice: there is the remote possibility that dual registration could be required, but this can be avoided if the interpretation of section 832 Companies Act suggested in section B4 above is adopted.

D. Final Remarks Kenya joined the growing list of African economies that undertook assessments of their credit markets and attributed the failures to channel credit, particularly to SMEs, to antiquated laws and inefficient registry systems. In contemplating, discussing the policies and drafting the MPSR, the Kenyan stakeholders proceeded very methodically. This approach to adopting a reformed secured transactions law may be a model for the rest of Africa. The MPSR faithfully implements the Model Law and should be studied by future African reformers. Although some deviations exist, they do not contradict any of the fundamental policies or principles of the UNCITRAL model instruments. However, the coordination with related legislation leaves a number of questions open, some of which raise significant concerns. A faithful implementation of the functional approach would have required a wholesale repeal of the HPA, which has not happened because the stakeholders wished to retain the provisions that protect vulnerable grantors in enforcement. Those provisions should have been directly incorporated into the MPSR. The definition of attachment to immovable property is ambiguous as to its effect on Kenyan land law, although it is likely that its reach is limited by the provisions of the MPSR that require parties to proceed in a commercially reasonable manner when removing an attachment from the land. The framework that governs security rights in intermediated securities should also have been considered in the light of the need for a closer alignment with international standards. The most significant challenges that remain concern the MPSR’s coordination with the Companies and Insolvency Acts, as discussed above.195 Furthermore, the actual implementation, including the entry into force of the MPSR before the Collateral Registry was launched, the hasty launch of the Collateral Registry, the publication of Registry Regulations containing provisions inconsistent with the MPSR, the lack of robust capacity-building programmes and a public awareness campaign should be a warning to prospective reformers in Africa, as well as globally. These implementation

192 See B4 above. Note that some amendments do take account of the new regime, such as the deletion of Insolvency Act, s 650, which relates to the effect of a moratorium on an uncrystallised floating charge. 193 Gedye (n 179) 118; Duggan and Brown (n 183) 4.45. 194 See B4 above. 195 See B4 and C9 above.

118 Kenya challenges were reflected in the early results of registration activity, especially the highly disproportionate ratio of registrations to searches and a significant majority of registrations being made against commercially insignificant assets, such as household goods, or even against assets not subject to the MPSR at all. Despite these challenges, Kenya’s future prospects are bright. Kenya has a vibrant economy fuelled by entrepreneurial spirit, as reflected in the creation and implementation of mobile payment platforms in the past, and now in various blockchain applications. The Collateral Registry system has been upgraded and now fully supports all the functions envisaged by the MPSR. A concrete proposal to amend the Registry Regulations has been formulated to correct the mistakes identified in this chapter. In addition, legislative reform activity continues to bolster certain aspects of the MPSR, especially in the area of warehouse receipts. Robust capacity-building exercises as well as public awareness campaigns are planned for 2019. Finally, the political situation has stabilised, providing a fertile environment for the growth of credit.

5 Liberia In 2010, Liberia adopted a comprehensive Commercial Code (CC) to overhaul its commercial legal infrastructure.1 To a large extent, the structure as well as the specific provisions of the CC, divided into seven chapters, are based on the US UCC. For example, sales are dealt with in Chapter 2, negotiable instruments in Chapter 4 and secured transactions in Chapter 5. However, the CC also covers areas of law not covered by the US model; thus, land mortgages are dealt with in Chapter 6 and arbitration in Chapter 7. This chapter examines the provisions relating to transactions secured with personal (movable) property and their implementation through the establishment of a Collateral Registry that was officially launched by the Central Bank of Liberia (CBL) in June 2014. As with Sierra Leone, the effect of this reform has been muted because of the Ebola crisis, which broke out shortly after the launch of the Collateral Registry. In the first six months of operations, the Collateral Registry received a total of 33 registrations. Although these covered various types of collateral, they predominantly related to inventory and motor vehicles, each type being included in about a third of the registrations. After the Ebola crisis came to an end, the number of registrations grew, which indicates that the framework is being deployed in practice.

A. Economic Overview In 2005, two years after the end of a long-running civil war,2 Liberia became a democracy and averaged 7 per cent real GDP growth from 2005 to 2013.3 This economic growth was cut short from 2014 to 2015 by the onslaught of the Ebola crisis and by significant falls in the global commodity prices of Liberia’s major exports.4 The mainstays of the Liberian economy are agriculture, services and industry.5 Agriculture employed 46 per cent of the 1 The Liberian Commercial Code of 2010 (CC 2010). 2 Between 1989 and 2003, Liberia was engaged in a costly civil war. The war led to the loss of hundreds of thousands of lives and the destruction of most of Liberia’s infrastructure. For an in-depth historical and anthropological discussion of the Liberian conflict and its underlying causes, see JI Levitt, The Evolution of Deadly Conflict in Liberia: From ‘Paternaltarianism’ to State Collapse (Durham, NC, Carolina Academic Press, 2005). 3 IMF, Liberia: Country Report No 16/239 (2016) 3. 4 Economic growth slowed drastically to 0.7 per cent in 2014; it dropped further in 2015 to 0.3 per cent, and fell once more in 2016 to –1.6 per cent, until it began to recover in 2017, when it was at 2.5 per cent. CIA, ‘Economy: Liberia’ (The World Factbook, 3 April 2019) https://www.cia.gov/library/publications/the-world-factbook/geos/ li.html. In 2019, the GDP annual growth rate is projected to be 2.5 per cent, with an outlook of 3.87 per cent for 2020. 5 ibid. Manufacturing constituted just 7.6 per cent of 2015 GDP. Ministry of Finance and Development Planning, Annual Economic Review 2015 (Government of Liberia, 2016) 19.

120 Liberia labour force in 2015,6 while mining activities provided the largest source of foreign direct investment.7 Services accounted for 46.6 per cent of the monetary value of all goods and services produced in the country in 2015.8 Following the Ebola crisis, Liberia’s real GDP increased by 2.5 per cent in 2017,9 by approximately 3.2 per cent in 2018,10 but decreased by 2.5 per cent in 2019.11 Liberia’s economic outlook is positive, and GDP is projected to increase by 4.7 per cent by the end of 2019 and by 4.8 per cent in 2020.12 However, a UN peacekeeping mission to Liberia was completed in 2018, which may have an effect on the country’s economic outlook.13 As in Sierra Leone, Liberian lenders rely primarily on two types of assets as collateral: land and motor vehicles. Most Liberian businesses do not generate any receivables, and any inventory they have is of low value. During a 2013 meeting with the local banking industry, one of the authors of this book enquired as to how present-day Liberian banks carry out inventory financing. After a pause, one participant asked: ‘What is inventory?’ Needless to say, inventory financing was non-existent in Liberia. After the establishment of the Collateral Registry in June 2014, banks, in particular, began changing their attitudes and practices towards inventory financing. Before then, banks’ credit policy manuals did not allow them to take goods as the only collateral for loans. This was because, first, such assets depreciate quickly; secondly, it is difficult for the bank to exercise adequate control; and thirdly, such assets require strict monitoring, which is both expensive and difficult to arrange. In addition, goods that are specialised equipment are difficult to dispose of in the secondary market. Goods, and other personal property, were therefore only used as secondary assets to further secure credit facilities; land was relied upon as the primary collateral. Some receivables financing did, however, exist, particularly in the rubber industry, where local growers of rubber for sale to Firestone Natural Rubber Company often assigned their receivables to a bank. Receivables financing was also possible for some arrangements where the funding was provided by the United Nations. Shortly after the establishment of the Collateral Registry, local banks asked the Registrar whether they could register financing statements relating to stock pledge agreements, assignments of contracts, assignment of mineral rights, assignments of land contracts and leasehold rights, and mortgages of capital assets, thus indicating a growing interest in asset-based financing.14

6 Annual Economic Review 2015 (n 5). 7 ibid viii. Iron ore, diamonds and gold are the main minerals, although many other minerals, such as bauxite, copper, oil, and zinc, are found across the country. African Development Bank Group, Liberia: Infrastructure and Inclusive Growth (Tunis, Temporary Relocation Agency, 2013) 31. 8 Annual Economic Review 2015 (n 5) 23. 9 World Bank, ‘Overview: Political and Security Update’ (12 October 2018) www.worldbank.org/en/country/ liberia/overview. 10 African Development Bank Group, ‘Liberia Economic Outlook’ (2019) www.afdb.org/en/countries/westafrica/liberia/liberia-economic-outlook/. The recovery was driven in large part by the mining and manufacturing sectors. ibid. 11 ‘Liberia: Economic Forecasts: 2019–2021 Outlook’ Trading Economics (2019) www.tradingeconomics.com/ liberia/forecast. 12 ibid. The positive projections are supported by growth in the agriculture, fisheries and services sectors. 13 ibid. The IMF projected in 2016 that Liberia’s economic growth would stabilise at around 6 per cent over the medium term. IMF, ‘Press Release No 16/196: IMF Staff Completes 2016 Article IV Mission to Liberia’ (4 May 2016) www.imf.org/external/np/sec/pr/2016/pr16196.htm. Inflation, depreciation and other issues dampened expected growth somewhat. ‘Liberia Economic Outlook’ (n 10). 14 Correspondence on file with the authors.

The Commercial Code 121 Microcredit remains the dominant form of finance, but in Liberia lenders do not rely as much on collateral as on daily incremental repayments of the microloan.15 A typical microbusiness such as a car repair shop could generate daily revenue of US$3–5 from which US$1 would be remitted to the microfinance institution through a mobile phone transfer. Such microloans are also regularly issued by crediting the mobile phone of the borrower.

B. The Commercial Code Chapter 5 (Secured Transactions) of the CC is a mixture of provisions drawn primarily from the US UCC Article 9 and Canadian PPSAs. The combination of these two models, in some cases, did not work well, creating some risks for lenders. Further, there are some tensions between Chapter 5 and Chapter 3A (Financial Leases),16 and Chapter 5 contains some drafting errors that could have easily been avoided.17 However, on balance, Chapter 5 provides a modern legal and registration framework for security interests in personal property that is far superior to the frameworks adopted by Liberia’s Ghanaian and Sierra Leonean neighbours. The Legislative Drafting Notes to the CC indicate that the key stakeholders, including the Bankers’ and Bar Associations and the Chamber of Commerce, had identified the lack of a commercial law framework and efficient dispute-resolution mechanisms as the main constraints to doing business in Liberia.18 The World Bank’s Enterprise Survey, based on interviews with 150 Liberian businesses in 2008–2009, concluded that only 14 per cent of them had access to a loan or line of credit;19 in 2017, 18.9 per cent of businesses in Liberia had access to finance.20 Unlike in other African economies, stakeholder consultations identified a need for a comprehensive legal reform rather than one limited to transactions secured with personal property. The stakeholders also noted that local financial institutions had sufficient liquidity but were effectively precluded from using it for productive purposes by the unpredictable legal framework. For example, bank deposits were often the only acceptable form of collateral for business loans.21 Apart from the need to provide a better legal framework, the following principles were identified in the Legislative Drafting Notes to the CC as guiding the reform process: (i) globalisation and commerce in the twenty-first century; (ii) modernity and currency in the twenty-first century; (iii) textual simplicity and clarity; (iv) transparency in commercial transactions; (v) legislative coherence; (vi) legislative responsiveness; (vii) transactional efficiency and equity; (viii) context; and (ix) consistency with regional harmonisation initiatives.22 In fulfilment of the first and the last principles, the Notes further explain that the

15 See ch 1, C4a. 16 See C1 below. 17 For instance, CC 2010 (n 1) s 5.2 provides that it is ‘[s]ubject to Section 5.2 of this Chapter’, thus referring to itself, as opposed to s 5.3 which seems to be what was intended. Much of the other internal cross-referencing is wrong; eg it is not at all clear what the references to 5.27 and 5.45 in s 5.3(a) relate to; the references to 5.3 and 5.4 in ss 5.6 and 5.8 clearly refer to ss 5.4 and 5.5, and so on. 18 The Legislative Drafting Notes to the Commercial Code are on file with the authors. 19 Enterprise Surveys: Liberia Country Profile 2009 (Washington, DC, World Bank, August 2010). 20 Enterprise Surveys: Liberia 2017 Country Profile (Washington, DC, World Bank, 2017) 11. 21 Legislative Drafting Notes (n 18). 22 ibid 4–5.

122 Liberia drafting took into account the models developed by UNCITRAL, UNIDROIT, and OHADA, as well as, in some cases, national law models. The chapters on leasing23 were influenced not only by the US UCC, but also by the UNIDROIT Model Law on Leasing.24 The main motivation for drafting a modern leasing framework was to unlock Liberia’s potential for the extractive industry, and the work was further spurred on by an expectation of demand, in particular, for mining equipment.25 Until the adoption of the CC in 2010, the New Financial Institutions Act of 1999 provided only a licensing and regulatory framework governing leasing. For Chapter 5, the Legislative Drafting Notes state the following: The Secured Transactions chapter is based substantially on the Canadian Personal Property Security Act with due modification to suit the Liberian commercial context. This was supplemented with best practice provisions from several other statutes, including the UCC (Article 9, revised), the Inter-American Model Law on Secured Transactions, the UNCITRAL Legislative Guide on Secured Transactions and the New Zealand Personal Property Security Act.26

In relation to secured transactions, although the OHADA Uniform Securities Act might have been considered during the development of Chapter 5, the final approved text of this Chapter does not reflect the fundamental principles and approaches underlying the OHADA model. Interestingly, the United Nations Convention on the Assignment of Receivables in International Trade, ratified by Liberia in 2005, was not listed as a model. It is noteworthy that the ratification was not motivated by a desire to introduce a modern framework for receivables financing, but rather by the government’s decision to ratify a large number of international conventions to make Liberia’s economy attractive to foreign investors.

C. Chapter 5 of the Commercial Code Chapter 5, together with the Regulations for Secured Transactions/Collateral Registry (2013) (‘the Regulations’) issued by the CBL under the authority given to it in section 5.75 of the CC, governs security interests in personal property. Certain other provisions of the CC, not included in Chapter 5 or the Regulations, also affect secured transactions. For instance, section 1.18(1)(g) defines a buyer in the ordinary course of business, and section 2.45(1)(b) recharacterises any retention or reservation of title by a seller as a security interest. The structure of the Chapters and the allocation of various rules that affect security interests within them follow that of the UCC.

1. Scope In order to determine the scope of Chapter 5, one needs to examine several provisions. First, section 1.18(1)(bb) defines the term ‘security interest’ in a manner that conforms Chapter 5 with the functional approach of its international models. A security interest is defined as

23 CC

2010 (n 1) s 3 on leases and s 3A on financial leases. Model Law on Leasing’ (UNIDROIT, 13 November 2008). 25 Legislative Drafting Notes (n 18) 13. 26 ibid 21. 24 ‘UNIDROIT

Chapter 5 of the Commercial Code 123 an interest in personal property or fixtures which secures payment or performance of an obligation. ‘Security interest’ includes any interest of a consignor and a buyer of accounts, chattel paper, a payment intangible, or a promissory note in a transaction that is subject to the Secured Transactions Chapter. ‘Security interest’ does not include the special property interest of a buyer of goods on identification of those goods to a contract. The right of a seller or lessor of goods to retain or acquire possession of the goods is not a ‘security interest’, but a seller or lessor may also acquire a ‘security interest’ by complying with the Secured Transactions Chapter.

This definition is somewhat confusing in relation to outright assignments of accounts receivable. Read literally, the first sentence appears to limit the definition of ‘security interest’ to an interest which secures payment or performance. Although this clearly includes functional equivalents such as sales on reservation of title terms, chattel paper, consignments of goods and security assignments of accounts receivable, it does not include outright assignments of accounts receivable, which do not perform the function of security. The second sentence, on first reading, appears to apply only to those interests described which fall within the definition in the first sentence and, therefore, would not include outright assignments of accounts receivable. However, the second sentence includes any interest of a buyer of accounts ‘that is subject to the Secured Transactions Chapter’. Section 5.2 provides that the Chapter applies to every transaction ‘that in substance creates a security interest’, and section 5.9(gg) (in contrast to section 1.18(1)(bb)) expressly includes ‘the interest of a transferee of an account or chattel paper’ within the definition of ‘security interest’. Although one could quibble with the imprecise drafting, the meaning is relatively clear: outright assignments of account receivables are included in the scope of the Chapter. This conclusion is, of course, subject to section 5.61, which limits the operation of the subchapter on enforcement to security interests that secure payment or performance of obligations, thus excluding outright assignments of accounts receivable. This pattern follows that of all the international models. Section 5.3 excludes certain types of outright assignments from the scope of the Chapter, following the model of the US and Canadian statutes, but section 5.3(d) lacks clarity. It provides that ‘except as otherwise provided under this Chapter, this Chapter does not apply to a sale of accounts or chattel paper’, which appears to add nothing to the rest of section 5.3. The most likely explanation is that this was intended to exclude sales of accounts as part of the sale of a business out of which they arose, as provided in UCC Article 9 and the PPSAs,27 and some words were omitted by mistake. Despite the slightly imperfect drafting, Chapter 5 should be read as following its international models in that it applies to sales of accounts receivable with the exception of its enforcement provisions. The influence of the Canadian PPSAs presents itself in the scope of the CC, which defines a security interest to include the interest of a lessor of goods under a lease for a term of more than one year, whether or not the interest secures payment or performance of an obligation.28 Neither UCC Article 9 nor the Model Law applies to long-term operating leases. The CC includes a specific definition of a lease for a term of more than one year that classifies it as a type of purchase money security interest (PMSI).29 Accordingly, lessors should be mindful of the impact of the definition of PMSI on long-term operating leases.

27 New

Zealand PPSA s 23(e)(x); Saskatchewan PPSA s 4(g); UCC, §9-104(d)(4). 2010 (n 1) s 5(9)(gg). 29 ibid ss 5(9)(s) and (aa). 28 CC

124 Liberia It would have been clearer for section 5.2 to set out distinctly in its scope the application of Chapter 5 to outright assignments of accounts receivable and to long-term operating leases rather than having to do so through a series of definitions, particularly given the infelicitous drafting of section 5.9(gg).30 However, it should not be too difficult for the courts to interpret the scope of Chapter 5 in the way suggested here. It is also noteworthy that Chapter 5 expressly applies to security interests in fixtures, whereas the Model Law has not incorporated the specific recommendations of the Legislative Guide relating to attachments to immovables.

2. Creation and Attachment Section 5.10 provides that a security interest is created on execution of a security agreement, creation in this case meaning that the interest is effective as between the two parties to the agreement. For it to become enforceable against third parties, it must ‘attach’, a concept borrowed from UCC Article 9 and PPSAs but not reflected in the Model Law.31 As in its North American models, section 5.12 imposes three requirements for the security interest to attach, namely: (i) value was provided by the secured creditor; (ii) the debtor has rights in the collateral or a power to transfer it; and (iii) a security agreement was signed by the debtor and sufficiently describes the collateral for it to be identified. A security agreement may cover after-acquired property, with two exceptions related to crops and consumer goods, and the collateral described therein may secure future advances.32 Following its UCC Article 9 and PPSA models, Chapter 5 does not specifically prescribe a standard for the description of secured obligations.33

3. Perfection Perfection is covered in subchapter 3. The structure of this subchapter follows the Saskatchewan PPSA in first presenting the consequences of an unperfected security interest on competing claimants before actually prescribing the methods of perfection.34 Sections 5.28–5.31 set out the methods of perfection, including: (i) by possession, (ii) by registration and (iii) temporarily. Unlike its international models, Chapter 5 does not recognise control as a method of perfecting a security interest. Registration is a method available to perfect a security interest in any collateral, and possession is limited to tangibles and documentary intangibles, such as instruments and negotiable documents. There are also various situations in which perfection continues (permanently or temporarily) where the original perfection method no longer applies to the collateral itself or to proceeds of that collateral. For example, under section 5.30, a security interest in an

30 cp the Saskatchewan PPSA, ch P-6.2 of the Statutes of Saskatchewan (1993, as amended) s 3(2). 31 See ch 2C. 32 See CC 2010 (n 1) ss 5.13–5.14. 33 The ML requires that the secured obligations are described in a way that reasonably allows their identification, but this can be a generic description, Art 9. 34 CC 2010 (n 1) s 5.24.

Chapter 5 of the Commercial Code 125 instrument, negotiable document or goods covered by a negotiable document perfected by possession remains perfected for 10 days after the collateral has been returned to the debtor for one of the purposes stated in that section, including loading, unloading and shipping of the goods. Where the security interest in the original collateral is perfected by registration, the security interest in the proceeds remains continuously perfected.35 Where the security interest in the original collateral is perfected by possession, section 5.31(3) provides that ‘the security interest in the proceeds becomes unperfected ten (10) days after the debtor acquires an interest in the proceeds unless the security interest in the proceeds is perfected under this Chapter’. This section, modelled on UCC Article 936 and Saskatchewan PPSA37 sections, seems to have misunderstood the purpose of those sections, which only relate to non-cash proceeds. The Liberian law does not differentiate between cash proceeds (which should not require a new act of perfection) and non-cash proceeds (which should, but only after a grace period).

4. Priorities Subchapter 3 on perfection also includes three sections, 5.34–5.36, that cover priorities of claimants whose interests are in competition with those of secured creditors, such as buyers and lessees in the ordinary course of business, purchasers of chattel paper, and transferees of instruments. Subchapter 4 deals comprehensively with other priority conflicts, including those involving secured creditors, PMSI creditors and lien holders. The general rule for security interests perfected by registration in section 5.37(1)(a) provides that priority is determined by the order of registration, irrespective of the time of perfection. This general rule is subject to some specific rules of subchapter 4 that may confer priority on security interests perfected by later-in-time registrations, such as section 5.42 that covers PMSIs. For a PMSI in inventory to gain priority over an earlier perfected security interest, the PMSI holder must register a financing statement before the debtor obtains possession of the inventory and notify the secured creditor with a security interest in inventory of the same debtor perfected by registration. This approach to the priority of PMSI in inventory mirrors Option A of Article 38 of the Model Law, which was based on UCC Article 9 and the Canadian PPSAs. For a PMSI in an asset other than inventory to gain priority over competing security interests in the same asset, the PMSI holder must register a financing statement within 10 days of delivering the asset to the debtor and does not have to notify any other secured creditor with a competing security interest in the collateral. Section 5.41 provides for a similar ‘super-priority’ for secured creditors who advance funds to the debtor for the purposes of growing crops. Sections 5.44, 5.45 and 5.47 address three situations: first, where the collateral is affixed to real property such that it constitutes a fixture, secondly, where it becomes an accession and, thirdly, where it is commingled with some other asset so as to lose its identity. Sections 5.44 and 5.45 also regulate the right of the secured creditor to remove the fixture or accession from the asset to which it has been affixed or accessed, and they provide the priority

35 Thus,

as long as the registration remains effective, under ibid s 5.31(2), the security interest is perfected. §9-315(d). 37 See Saskatchewan PPSA 1993 (n 30) s 28(3). 36 UCC

126 Liberia rule for a conflict between the secured creditor and the person who acquired the real or personal property to which the asset has been affixed or accessed for value.

5. Registry a. Development of the Registry Section 5.51(3) of the CC empowered the CBL to establish a temporary registry until the Collateral Registry became operational. In early 2013, during the initial phase of the implementation project, the possibility of running a temporary registry in the form of an Excel spreadsheet was discussed, but it was eventually rejected because it would have created numerous operational problems and the Collateral Registry was expected to be launched within months. In the absence of the Collateral Registry and a temporary registry, lenders developed the cumbersome practice of registering their security interests with the probate court, though registrations had no legal effect under the CC. In addition to Chapter 5 (which contemplates the establishment of the Collateral Registry), section 3A.18 provides for the establishment of a registry for finance leases. As the establishment of both registries is governed under separate chapters of the CC, in the implementation process a question arose whether the ‘Leasing Registry’ should be separate from the Collateral Registry. Chapter 5 also characterises some finance leases as security interests and deems finance leases for a term of more than one year to be security interests. The Finance Lease Registry is yet to be established. A registration with respect to a finance lease that constitutes a security interest or is deemed to be a security interest does not constitute the registration of a finance lease for the purposes of section 3A.18 because it would not meet the requirements prescribed therein, which include submission of the lease agreement to the Registrar and an indication of a lease term in the registration. The failure to register a finance lease in the Finance Lease Registry does not have an impact on the rights of the lessor against third parties. Under section 3A.16(2), ‘The title of the lessor or his assignees to the asset leased shall prevail at all times over claims by third parties including creditors of the lessee.’ The establishment of the Collateral Registry involved interesting discussions about the requirement for registrants to enter various information, not required by Chapter 5 or the Regulations, for the sole purpose of allowing the CBL to collect statistical information.38 In Liberia, one item of statistical information considered for inclusion was the loan-to-collateral ratio. Some felt that this data was relevant because loans in developing economies are often over-collateralised by significant margins. The counter-argument, which eventually prevailed, was that this data would require secured creditors to value the assets taken as collateral, which would be difficult, particularly in relation to used equipment. Eventually, section 9(2) of the Regulations identified the gender and birth date of the individual grantor, whether the business was majority owned by a woman, and the maximum amount for which the security interest may be enforced as required statistical information.

38 See

conclusion C7.

Chapter 5 of the Commercial Code 127

b. The Registration Provisions Under the CC, a financing statement may be registered prior to the execution of a security agreement unless the collateral is consumer goods.39 Within 30 days of registration, the secured creditor must provide the debtor with a copy of a statement verifying the registration.40 The financing statement may be registered for the duration indicated therein or perpetually, except for a financing statement that covers consumer goods, in which case its duration may not exceed five years.41 The Registry system has been designed to require the registrant to select the appropriate box if the collateral includes consumer goods, thereby preventing the registrant from choosing a period that exceeds five years. For consumer goods collateral, the CC also imposes an automatic duty on the secured creditor to discharge a registration,42 as opposed to other collateral for which the secured creditor’s duty to discharge a registration is triggered only on receipt of a demand from the debtor.43 The Model Law’s Registry provisions do not distinguish between consumer and non-consumer collateral, but the Liberian provision can be justified as providing extra protection for consumer debtors. CC section 5.52 provides for the issuance of certificates by the Registrar in response to requests to search the Collateral Registry. In common with the North American models and the Model Law, in the Liberian system the name of the grantor is the main indexing and search criterion; and is required to be included in the financing statement, otherwise the registration will be rejected.44 The Collateral Registry may also be searched according to a serial number and an identification number of the debtor.45 Under section 14(1) of the Regulations, a financing statement is effective only if it provides the correct name of the debtor. The name of an individual debtor is to be determined pursuant to section 10(2) of the Regulations that provides a hierarchic order of identification documents, including a valid passport, a voter registration card and a driver’s licence. However, in the Liberian system, the registrant must also include an identification number for the debtor in the financing statement. For individuals, the identification number is the number stated in one of the identification documents listed in section 10(2) of the Regulations. The identification numbers will allow searchers to distinguish individual debtors if a search against a name of the debtor retrieves multiple results. Section 20 of the Regulations also prescribes that searches may be made against an identification number or against a combination of a name and an identification number. Paragraph 2 of section 20 then states that a search against these criteria produces a legal effect under section 14 of the Regulations. However, section 14 of the Regulations treats names and identification numbers differently, requiring that names be stated correctly under paragraph 1, whereas, for identification numbers, paragraph 4 applies the ‘seriously misleading’ test. In other words, a financing statement that correctly identifies the debtor’s name may be rendered ineffective only if the identification number of the debtor indicated in that financing statement seriously misleads a reasonable searcher. For legal persons, their name is

39 CC 2010 (n 1) s 5.53. 40 ibid s 5.54(3). 41 ibid s 5.54(4)–(7). 42 A discharge is effected by the registration of a financing change statement. See ibid s 5.59(1). 43 ibid ss 5.59–5.60. 44 See ‘Regulations for Secured Transactions/Collateral Registry’ (XII Liberia Official Gazette 33, 26 April 2013) s 14(1). 45 ibid s 20.

128 Liberia the name under which they are registered in the Liberian Business Registry. The identification number is the legal person’s enterprise code, assigned upon registration in the Liberian Business Registry. The possible effect is that a foreign company that has not been registered in the Liberian Business Registry cannot be identified in a financing statement.46 The Collateral Registry may be searched by a name or an identification number. Section 20(3) of the Regulations authorises the Registrar to provide additional search functions. For instance, holders of user accounts with the Collateral Registry are also able to search the records according to the birth date of the debtor, the date of registration, etc. However, only searches conducted by a name or an identification number are deemed by the Regulations to have legal effect.47 In addition, the Regulations provide for serial-number indexing with respect to motor vehicles, unless they are held as inventory.48 However, unlike the PPSAs, an error in the serial number renders the registration ineffective as against all competing claimants,49 rather than a narrower category of buyers and lessees only. Again, this is an important substantive aspect that should have been comprehensively addressed in the CC, which should also have included a set of associated priority rules. During the implementation process, local stakeholders raised several questions and made suggestions for the Collateral Registry to include certain features that they felt would minimise some of their risks. For instance, some queried whether the Collateral Registry could ensure that another financing statement is not registered against the same debtor or at least build a function to notify the first secured creditor if that occurs. The former would require a change in the substantive law that does not restrict the ability of the debtor to create competing security interests.50 The latter is technologically possible if, for instance, all users have access to the system through dedicated accounts and the system communicates with them electronically. This feature may enhance the ability of the secured creditor to monitor the debtor’s collateral. Other stakeholders queried whether a Registry function could be designed that would effectively prevent movement of the collateral out of the country. It is impossible to design such a function for a registry; this risk should be addressed by proper monitoring of the collateral.

c. Payment of Fees CC section 5.75 confers authority on the Registrar to set the fees for various Registry functions. This authority has been exercised by the CBL subject to section 22 of the Regulations, which prescribes that there shall not be any fees for searches. The exercise of these powers by the CBL is a model for authorities in other States to follow in terms of the amount of registration fees. Since the Collateral Registry requires registrants to enter the maximum amount secured, the CBL set three registration fees – US$1, US$3 and US$5 – corresponding to 46 ibid ss 9(1)(a), 10 and 11. 47 ibid s 20. 48 See ibid s 1(8) that defines ‘serial number’, s 13(6), which prescribes the serial-number description of motor vehicles, and s 14(2), which determines the legal effect of an error in a serial number. 49 ibid s 14(2). 50 CC 2010 (n 1) s 5.49 provides ‘[t]he rights of a debtor in collateral may be transferred voluntarily or involuntarily despite a provision in the security agreement prohibiting transfer or declaring a transfer to be a default, but no transfer prejudices the rights of the secured party under the security agreement or otherwise’.

Chapter 5 of the Commercial Code 129 the amount indicated in the relevant field of the financing statement. If the registrant is a financial institution, it can have a deposit account with the CBL out of which the fees will be debited monthly. If the registrant does not have a deposit account with the CBL, the Registry system allows the registrant’s user account to be credited with funds, which may be deposited at designated financial institutions. During the debates leading to the agreement on the fee structure, the idea of allowing users to pay annual subscription fees that would allow them to perform an unlimited number of transactions was proposed. This payment model is used by credit bureaus, but eventually the CBL chose a fee-per-transaction structure.

6. Post-Implementation Issues Before, during and after the implementation of a reformed law, discussion about the challenges and the best practice in the implementation of such reforms is likely to take place between local stakeholders and those in neighbouring countries who have been involved in the reform of the secured transactions regimes of their respective countries. The operation of such dialogue is illustrated by the following anecdote. At the 2015 Annual Conference of the International Association of Commercial Administrators (IACA),51 attendees from Liberia learned about the phenomenon of bogus filings in the United States.52 This discussion raised the concern that the Liberian Collateral Registry and the legal framework are not sufficiently equipped to deal with bogus filings. The Liberian attendees learned at the IACA Conference that in some US states the filing officer has the power to determine whether a financing statement was submitted with malicious intent and reject it or, after it has been filed, remove it from the record.53 A discussion ensued as to whether similar powers should be granted to the Liberian Registry staff. Eventually, it was agreed that the powers under the Liberian law and also the Registry framework were adequate to deal with potential bogus filings.54 First, the Collateral Registry allows access for the purposes of submitting financing statements only through user accounts.55 Accordingly, unlike in the United States where the filer need not be identified in the financing statement and may remain unidentifiable, the Liberian Collateral Registry is able to tie a particular financing statement to a particular user account. This should disincentivize bogus filings, since the filer can be easily identified. Second, section 5.59 of the CC provides a mechanism whereby the person named as debtor in a bogus filing can demand that the secured creditor register a financing change statement discharging the filing and, in case of the failure of the secured creditor to act, empowering the aggrieved person to obtain a court order that both awards damages to the aggrieved 51 See International Association of Commercial Administrators (IACA), ‘About Us’, www.iaca.org/. 52 P Hodnefield, E Marshall and T Timmons, ‘Bogus Filing Tactics: A Primer’ IACA 38th Annual Conference, Savannah, Georgia (2015) www.iaca.org/wp-content/uploads/Bogus-Filing-Tactics-IACA-2015.pdf. A bogus filing is a registration that is completed with malicious intent and without any authorisation of the debtor, such as are made, for example, by prisoners against judges and prosecutors. Nichols v Branton [2014] WL4816075 (NY Sup Ct). 53 In 2014, 14 US states had statutes that empowered the filing officer to nullify or expunge from an official record a false or fraudulent lien or instrument. See State Strategies to Subvert Fraudulent Uniform Commercial Code (UCC) Filings (Washington, DC, National Association of Secretaries of State, April 2014) 9. 54 See ch 2, E1. 55 See Regulations 2013 (n 43) s 6(3).

130 Liberia person and orders the secured creditor to discharge the registration.56 Moreover, under section 5.70 the court has the power to issue any order to ensure compliance with Chapter 5, which could include an order served on the Collateral Registry to remove a registration. Third, there has been no evidence of malicious registrations in the Collateral Registry or other Liberian registries. A provision authorising the Registry to reject a financing statement or expunge one that has been registered not only undermines the administrative function of the Registry (which does not include scrutinising information contained in financing statements)57 but also delays the completion of registrations. It also raises concerns for users that some of their financing statements may be rejected or removed on the suspicion that they have not been properly authorised when they, in fact, relate to legitimate commercial transactions. This is a particular concern in developing economies where lenders view the administrative role of registries as a key component of reformed infrastructure, which alleviates their frustrations with the traditional registries that are fraught with inefficiencies and, at times, with unpredictable, and even fraudulent, conduct of registry staff.

7. Enforcement Subchapter 6 provides for the rights and remedies after default of the debtor. According to section 5.62 (1), the secured creditor may use the remedies set out in this subchapter, which include extrajudicial enforcement in other law, or in the security agreement. If the collateral includes both personal and real property, the secured creditor may also proceed under the real property law with respect to the personal property collateral.58 Finally, section 5.63 provides that nothing in this law precludes the secured creditor from appointing a receiver or receiver and manager.59 Collateral may be exempt from seizure under other laws that protect low-value assets of consumer debtors, which would prevent enforcement of a security interest that has not been perfected by possession or taken to secure the purchase price of the collateral.60 The permitted enforcement methods depend on the nature of the collateral. For example, the secured creditor may collect receivables, instruments or chattel paper, and may take possession of tangible assets.61 Section 5.65(1) expressly authorises the secured creditor to seek an ex parte court order mandating the sheriff to either seize collateral or immobilise it at the debtor’s premises.62 Some protection for the debtor is built into many enforcement provisions. For example, section 5.66 authorises disposal of the collateral in its present condition or after some commercially reasonable repair, in a public or private sale or other disposition and at any time and place as long as it is commercially reasonable, and not less than 15 days’ notice is given to the debtor and other interested parties. Section 5.68(2) provides for the remedy of accepting the collateral in satisfaction of the secured obligation. 56 Note that these powers can be used in other situations, such as where the description of the collateral in the financing statement is too broad. 57 See Conclusion D5b. 58 CC 2010 (n 1) s 5.62(6). 59 See ch 1, E1b. 60 ibid s 5.65(2). 61 ibid ss 5.64–5.65. 62 ibid s 5.65(1) provides that, upon default, collateral capable of immediate possession may be seized and equipment, vehicles or plants can be rendered immobile.

Insolvency Reform 131 To enable the swift resolution of disputes relating to secured transactions, but primarily to facilitate the enforcement of security interests, Liberia adopted An Act to Provide for the Establishment of a Commercial Court.63 The structure and operating procedures of the Court were inspired by a similar court in Ghana.64 The Commercial Court, established in September 2010, has jurisdiction to adjudicate any commercial disputes that involve a claim of at least US$15,000.65 The parties may appeal a decision of the Commercial Court, but enforcement of the Commercial Court decision will not be stayed pending the appeals process.66 Article VI of this Act gives the Commercial Court the power to authorise the secured creditor, upon an ex parte application, to possess or repossess property that is subject to a security interest or mortgage.67

D. Insolvency Reform In December 2016, the Liberian Parliament adopted a law to reform Chapter 8 of the CC, the Insolvency and Restructuring Act.68 The Commercial Court has exclusive jurisdiction over insolvency cases under Chapter 8.69 It defines a secured claim in section 8.4 as ‘a claim held by a creditor the repayment of which is secured by a lien, pledge, hypothecation or other interest in moveable or immoveable property’. Since the entry into force of Chapter 5, any pledge or hypothecation of movable property creates a ‘security interest’ within Chapter 5, in other words, the specific categories of interest have been abolished. Although the definition in section 8.4, by its reference to ‘other interest in moveable property’, would include a Chapter 5 security interest, the fact that Chapter 8 was enacted six years after the enactment of Chapter 5 should be reflected in its approach and terminology, and specific types of security interest should not have been mentioned. Section 8.46 provides for the treatment of secured claims and imposes an automatic stay on any enforcement actions of secured creditors, with the exception of finance leases. Therefore, even though finance leases constitute security interests under Chapter 5, they are treated differently under Chapter 8. Finance leases include leases that in substance create security interests as well as long-term operating leases.70 It is unclear why Chapter 8 provides for special treatment for all finance leases rather than just long-term operating leases. In liquidation, secured claims are paid first ahead of administrative claims as well as unpaid wages owed to employees.71 63 An Act to Amend the Judiciary Law 1972, Title 17, Liberian Code of Laws Revised, To Provide for the Establishment of a Commercial Court. 64 See IFC, ‘Liberia’s Landmark Commercial Code to Strengthen the Economy’, www.ifc.org/wps/wcm/connect/ news_ext_content/ifc_external_corporate_site/news+and+events/news/liberia_commercial_code. 65 EM Morgan, ‘The Commercial Court of Liberia: Enabling the Investment and Commerce Sector’ (Liberia Chamber of Commerce Business Forum, 24 April 2015); Act to Amend the Judiciary Law 1972 (n 63) Art II(1). 66 Morgan (n 65); Act to Amend the Judiciary Law 1972 (n 63) Art IV. 67 Morgan (n 65); Act to Amend the Judiciary Law 1972 (n 63) Art VI. 68 An Act to Repeal the Fraudulent Conveyance Act, Chapter 8 of the Liberian Commercial Code, Title 7 of the Liberian Code of Laws Revised, and to Enact in Lieu Thereof a New Chapter 8 of the Liberian Commercial Code Entitled the Insolvency and Restructuring Act. 69 ibid s 8.6–8.7. 70 See C5a above and ch 2, B2b. 71 ibid s 8.53(1). Contrast the approach in many common law jurisdictions where these claims have preference over floating charges, or the equivalent under reformed law, see ch 1, E1b and conclusion C6bii.

132 Liberia

E. Final Remarks Liberia has enacted a modern legal framework for security interests in personal property and established an excellent Collateral Registry that is accessible electronically and at a low cost. Its low ranking of 112 in the World Bank’s Doing Business Report, the Getting Credit indicator, is attributable to the lack of credit bureaus and credit reference systems that are also taken into account under this indicator.72 Although the drafting of some provisions included in Chapter 5 could have been clearer, reasonable interpretation should lead to their proper application in practice. Liberia’s reform process, which takes into account various models, is a lesson for those contemplating a similar approach to carefully ensure that the provisions drawn from different sources work together in a coherent manner. The CC also provides some useful provisions for consideration by those States that wish to legislate beyond the Model Law – for example, by including provisions on security interests in fixtures and applying the registration and priority regime to long-term operating leases. However, when looking at the Liberian model, States may wish to address some important questions in the substantive legislation (rather than in the regulations) that are fundamental to the operation of the new secured lending regime. It is hoped that the secured transactions framework becomes one of the driving forces for the growth of the Liberian economy.

72 See

Doing Business 2019: Liberia (Washington, DC, World Bank, 2018).

6 Malawi A. Introduction Malawi enacted the Personal Property Security Act (PPSA) in 2013,1 and it entered into force in November 2015 when the Collateral Registry became operational.2 Prior to the reform, Malawi was one of the worst performers in the Getting Credit Indicator of the World Bank’s Doing Business Report, ranking 151 out of 189 economies.3 Malawi performed much more poorly than the sub-Saharan Africa average, which was 122. The implementation of the PPSA saw a remarkable improvement in Malawi’s performance in the indicator.4 In the 2018 Doing Business report, Malawi ranked sixth out of 190 economies.5 Growth in GDP accelerated from 2.5 per cent in 2016 to 4 per cent in 2017.6 As of February 2018, for the first time in years, inflation was contained at a single-digit level of 7.8 per cent.7 Malawi’s secured transactions framework, including the PPSA and some post-adoption developments, has been examined in detail in the first volume of this series.8 This chapter provides an update to that volume. Malawi faced some transitional issues. Post-launch, the registry established pursuant to the Farmers’ Stop Order Act repealed by the PPSA was supposed to stop accepting registrations of stop orders that constituted security interests as defined in the PPSA, but despite this, a number of farmers’ stop orders were registered. A question arose whether those registrations have any legal effect and whether they could be transitioned into the PPSA regime under the transitional rules. The PPSA does not contemplate such a situation and the transitional rules cover only those security interests perfected or created prior to its entry into force. This occurrence illustrates the complexity of the transitional process. It is important 1 Personal Property Security Act (No 8, 19 April 2013). 2 See Personal Property Security Registry System of Malawi, ‘Home’ MPPSR.gov (2015) ppsr.registrargeneral. gov.mw/; see also Government of the Republic of Malawi, ‘Personal Property Security Registration’ (Department of the Registrar General, 2015) www.registrargeneral.gov.mw/services/personal-property-security-registration.html. 3 For the five features on which Malawi recorded positive scores, see World Bank, Doing Business 2015: Going Beyond Efficiency: Economy Profile 2015: Malawi (12th edn, 1 June 2014) 48–49; for rankings, see ibid 11–15. 4 ibid 57. 5 World Bank, Doing Business 2019: Training for Reform (16th edn, May 2018) 34. 6 The World Bank in Malawi, ‘Overview’ (World Bank, 10 October 2018) www.worldbank.org/en/country/ malawi/overview. 7 ibid. 8 M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in L Gullifer and O Akseli (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016) 183; see also M Dubovec and C Kambili (eds) A Guide to the Personal Property Security Act: The Case of Malawi (Pretoria, Pretoria University Law Press, 2015); and see M Dubovec and C Kambili, ‘Using the UNCITRAL Legislative Guide as a Tool for a Secured Transactions Reform in Sub-Saharan Africa: The Case of Malawi’ (2013) 30 Arizona Journal of International & Comparative Law 163.

134 Malawi that registries tasked with registering rights in movable assets, such as the motor vehicle registry with respect to liens on registered motor vehicles or the companies registry with respect to charges created by companies, stop accepting such registrations. The number of registrations in 2017 totalled 13,074, and users conducted 10,635 searches.9 These numbers decreased in 2018 both for registrations (6,696) and searches (5,591). Out of the total number of 19,770 registrations in 2017 and 2018, 6,918 were against individual debtors and 8,141 against microbusinesses. These numbers illustrate that the framework has supported the extension of credit to individuals and microbusinesses who were expected to be the primary beneficiaries of the reform. The decline in registration numbers experienced in 2018 may be attributed to the deceleration of economic growth and stagnation in agriculture.10 However, the statistics also reveal some areas that should be targeted by those tasked with the implementation of the reform. First, the number of searches is lower than the number of registrations in both 2017 and 2018. Although the number of registrations is high compared to the number of searches in most African collateral registries, in Malawi the ratio is much closer to the optimum, as compared to Ghana11 and Kenya12 for instance. The second issue is the type of collateral taken to secure obligations to which the registrations relate. For example, out of the 8,141 registrations against microbusinesses, 7,327 identified the collateral as ‘money’. Accordingly, secured parties were not financing any assets used in or generated by a business activity, such as equipment, inventory or receivables. In the same vein, out of the 19,770 total registrations, only two identified accounts receivable as collateral. Arguably, receivables may have been taken as collateral under facilities for which the registrations identified the collateral as ‘all assets’, but such registrations totalled only 87 in those two years. Inventory was identified as collateral in only 42 registrations, which illustrates the lack of classical asset-based lending that relies on inventory and receivables as collateral. In chapter 8 of the first book in this series,13 it was noted that perfection of a security interest in an electronic warehouse receipt would require the registration of a financing statement, but that process has since been affected by the enactment of a warehouse receipts law, which was already being developed at that time. This chapter examines the effect of the Warehouse Receipts Act of 2018 (WRA) on the perfection of security interests in electronic warehouse receipts, and its interaction with the PPSA.

B. Warehouse Receipts Act of 2018 Malawi enacted the WRA in 2018, but it is yet to enter into force pending the adoption of subsidiary regulations on warehouses and related matters. Part V, entitled ‘Perfection and Priority of Security Interests in Warehouse Receipts and Goods Covered by Warehouse 9 The registry statistics were provided by the Malawi Collateral Registry and are on file with the authors. 10 The World Bank in Malawi (n 4). 11 See Ghana D4b. 12 Kenya C5. 13 M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in Gullifer and Akseli (n 8) 183.

Warehouse Receipts Act of 2018 135 Receipts’, contains five sections dealing with various aspects of perfection and priority of security interests. These sections were coordinated with the PPSA by the relevant teams who worked on both laws. Section 35 recognises the application of the PPSA, but subject to the special rules provided in Part V. Sections 36–38 provide the rules on the perfection and priority of security interests in paper-based negotiable warehouse receipts, paper-based non-negotiable warehouse receipts and electronic negotiable warehouse receipts. Section 39 concerns the rights of purchasers of warehouse receipts as well as buyers of goods covered by warehouse receipts. Under section 16 of the PPSA, a security interest in goods in the possession of a bailee can be perfected in a number of ways. One is by registration, which would apply whether a warehouse receipt or other document of title had been issued or not.14 A second is attornment by the bailee (where the bailee holds the goods on behalf of the secured creditor).15 A third, which will only work where the goods are represented by a negotiable document of title,16 is by delivery of possession of the document to the secured creditor. A fourth is where the bailee issues a document of title in the name of the secured creditor; this document of title can be non-negotiable or (presumably) negotiable. These provisions are based on the UCC Article 9-312(c)–(d), which provides for the methods of perfection with respect to security interests in goods covered by negotiable and non-negotiable documents. The Model Law provides that where a tangible asset is covered by a negotiable document, possession of the document by the secured creditor perfects the security interest.17 Apart from this, the Model Law does not provide any special rules on the perfection of security rights in documents of title, but does provide that a security right can be perfected by possession of the goods, and possession includes actual possession by an independent person that acknowledges holding goods for the secured creditor. Therefore, the second and fourth methods in the Malawi PPSA are indirectly included in the Model Law. Sections 36(1)–(2) and 37(1)–(2) of the WRA replicate the corresponding PPSA perfection rules, clarifying that these rules apply only to documents issued ‘in paper’. Paragraph 3 of section 36 of the WRA provides a priority rule that does not have a counterpart in the PPSA. This rule addresses a conflict between a security interest perfected by possession of a negotiable document of title and an earlier security interest perfected by some other method: the security interest perfected by possession has priority while the goods are in the possession of the issuer of the document of title. In contrast, the only rule in the PPSA under which a person taking possession of a document of title can gain priority over an earlier perfected security interest is the ‘taking-free’ rule in section 44 of the PPSA, which applies to ‘purchasers’ of a document of title. The protection provided by this rule is broader than in its New Zealand model, in which section 99 is limited to holders of documents of 14 PPSA 2013 (n 1) s 16(a). 15 ibid s 16(b). 16 Under the PPSA, a document of title is defined as: ‘a writing issued by or addressed to a bailee (a) that covers goods in the possession of the bailee … and (b) in which it is stated that the goods identified in it will be delivered to a named person, or to a transferee of that person or to a bearer or to the order of a named person’. This definition clearly includes warehouse receipts. In the WRA, a warehouse receipt is defined as ‘a document of title issued under this Act and section 2(1) of the PPSA by a warehouse operator’. 17 ML, Art 26(2).

136 Malawi title that are negotiable.18 Under the PPSA, for a person to become a ‘purchaser’, it must give value, take possession of the document, and acquire the document without knowledge that the transaction is in breach of the security agreement to which the security interest relates. This rule is the equivalent to the taking-free rule in Article 49(3) of the Model Law, which applies to negotiable documents,19 and section 39(1) of the WRA. In addition, under Article 49(1) of the Model Law, a security right in an asset made effective against third parties by possession of the negotiable document has priority over a competing security right made effective against third parties by some other method.20 As seen from the preceding discussion, the PPSA did not incorporate an equivalent of Article 49(1), but this has been added in the WRA in section 36(3). Accordingly, under the WRA, a secured creditor in possession of the negotiable warehouse receipt may acquire priority over an earlier-in-time security right even if it knew that the taking of its security right would be in breach of the right of a competing secured creditor, which would disqualify it from purchaser status. Thus, the WRA provides broader protection to those who lend against negotiable warehouse receipts than the PPSA. The PPSA provides perfection rules applicable to both paper and electronic warehouse receipts, as well as special priority rules applicable to documents of title issued in paper.21 Under the PPSA, a security interest in an electronic warehouse receipt, or any other document of title issued electronically, may be perfected by registration or issuance of the document in the name of the secured creditor. Section 38 of the WRA adds the perfection mechanism of ‘control’22 for electronically issued warehouse receipts. It is noteworthy that, other than warehouse receipts, registration or the issuance of a document of title in the name of the secured creditor remain the only available perfection mechanisms for security interests in documents of title that may be issued electronically, especially bills of lading, as provided for in the PPSA. Also, the special priority rule of section 36(3) of the WRA (discussed above) does not apply to bills of lading. Section 7 of the WRA sets out how control over an electronic warehouse receipt may be acquired; namely, where a person is recorded in the relevant system as either the transferee of the warehouse receipt or the person to whom the warehouse receipt was issued.23 The latter form of perfection by control was previously possible under the PPSA. The rules to establish control replicate the provisions in UCC Article 7-106 for control of electronic documents of title. Like section 36(3) of the WRA, section 38(2) provides that a security interest perfected by control has priority over any competing security interest perfected by another method. Many warehouses are not mere storage facilities, but actively trade the goods stored in their warehouses, whether owned by them or belonging to third parties. Under a special rule provided in section 39(2), buyers of fungible goods from a warehouse operator take free of any security interests, if: (i) the warehouse operator is in the business of buying and selling such goods; (ii) the goods are delivered to the buyer; and (iii) the buyer takes without knowledge that the sale constitutes a breach of the security agreement under which the

18 PPSA

2013 (n 1) s 44; see also ibid s 2(1) that defines ‘purchase’. Art 49(3) has no requirement that the transferee must provide value. 20 See ch 2, F1. 21 See the definition of ‘document of title’ and ‘writing’ in the PPSA 2013 (n 1) s 2(1). 22 See ch 2, D. 23 WRA 2018 (n 12) s 7(1). 19 Although

Malawi’s Commodity Exchanges 137 security interest was created. Such buyers are in a position functionally similar to buyers in the ordinary course of business protected under section 48 of the PPSA. Arguably, buyers of goods from warehouse operators, under the circumstances contemplated in section 39(2) of the WRA, would take free of security interests under section 48 of the PPSA in any case. However, WRA section 39(2) provides further clarity for any transactions that warehouse operators may engage in the ordinary course of business.

C. Malawi’s Commodity Exchanges The recent enactment of the WRA will affect the two commodity exchanges operating in Malawi.24 AHL Commodities Exchange (AHCX), an agricultural commodities exchange established in 2013, is a fully electronic platform for the trading of agricultural commodities under standardised contracts, with a guarantee mechanism that assures the quality, quantity and payment for rights to commodities traded on its platform.25 Under its warehouse receipt financing initiatives, financial institutions can extend loans to depositors of commodities in AHCX-managed warehouses where the AHCX also serves as the collateral manager.26 In this regard, the AHCX collaborates with New Finance Bank to offer warehouse receipt financing to depositors and farmers.27 In 2016, AHCX adjusted its minimum tonnage requirements downwards (to half a tonne) in order to accommodate small-scale farmers.28 The Malawi Agricultural Commodities Exchange (ACE) operates 53 warehouses with 225,884 tonnes of storage space.29 It also operates a nationwide network of 29 rural warehouses owned by partner farmer associations or private-sector partners.30 ACE gathers data from farmers to integrate into its SMS messaging database so that it can send crop prices and trading opportunities to farmers throughout Malawi.31 ACE offers an input-financing arrangement (the ‘Chithumba Model’) where producers can acquire farming inputs, such as fertiliser, seed and chemicals, on credit with the expectation that they will repay the loans in kind at the end of the harvest season.32 Both of these exchanges are governed by internal rules and regulations that provide for the pledge of warehouse receipts traded therein. These rules were already affected by the enactment of the PPSA, and the pledging mechanisms will be further shaped by the WRA, under which the exchanges may develop mechanisms to facilitate the acquisition and transfer of control over electronic warehouse receipts.33 24 Gullifer and Akseli (n 8) 198. 25 AHL Commodities Exchange, ‘About Us: Company Profile’ (2016) www.ahcxmalawi.com/about-ahcx/; see also AHL Commodities Exchange, ‘Risk Management’ (2016) www.ahcxmalawi.com/risk-management/. 26 AHL Commodities Exchange, ‘Warehouse Receipt Financing’ (2016) www.ahcxmalawi.com/ warehouse-receipt-financing/. 27 AHL Commodities Exchange, ‘AHCX in Warehouse Receipt Financing Agreement with New Finance Bank’ (1 August 2018) www.ahcxmalawi.com/ahcx-in-warehouse-receipt-financing-agreement-with-new-finance-bank/. 28 AHL Commodities Exchange, ‘AHCX in Move to Attract More Small Scale Farmers’ (13 September 2018) www.ahcxmalawi.com/ahcx-in-move-to-attract-more-small-scale-farmers/. 29 ibid. 30 ACE, ‘Agricultural Commodity Exchange for Africa’ Aceafrica.org (2019) www.aceafrica.org/#. 31 ibid. 32 ibid. 33 See B above.

138 Malawi Further to this, the Financial Services (Regulation of Commodity Exchanges) Directive, adopted in April 2018, provides for the regulation and supervision of commodity exchanges. In section 2, it defines ‘commodity exchange’ to include a warehouse receipts system. It took effect in April 2019 pursuant to section 20. This is to give time for commodity exchanges to adjust to the new regulatory environment. Although the Directive deals with supervisory matters and does not directly affect any transactions involving commodities, or related contracts that could be used as collateral, it characterises the provision of a loan against an exchange-traded warehouse receipt as a financial service. Lenders taking exchange-traded EWRs as collateral, as opposed to other types of WRs, are subject to this Directive.

D. Final Remarks Although Malawi’s legal framework has incorporated the modern principles of best international standards, there remain a few obstacles to unlocking the full potential of the PPSA framework, including the low capacity of local lenders. The relative unfamiliarity or reluctance to take certain assets as collateral is reflected in the statistics of the Collateral Registry which, for 2017 and 2018, recorded thousands of registrations covering ‘money’ as collateral but fewer than 50 covering inventory and receivables, not including the registrations against ‘all assets’. To address this deficiency, the IFC launched a training program on the fundamentals of asset-based lending for both lenders and borrowers. Some of the training sessions focused on teaching women fundamental entrepreneurship skills because Collateral Registry statistics revealed that only about 0.03 per cent of loans secured with rights in movable assets perfected by registration have gone to women small-scale entrepreneurs.34 Malawi remains a success story in terms of being able to adopt and implement a modern secured transactions regime, which is now being complemented and refined by new legislation facilitating warehouse receipts transactions with electronic varieties of warehouse receipts.

34 O Chiphwanya, ‘Collateral Registry Sidelines Women’s Firms’ The Malawi Nation Online (29 November 2017) www.mwnation.com/collateral-registry-sidelines-womens-firms/.

7 Nigeria In 2014, after years of sustained economic growth,1 Nigeria’s economy became the largest in Africa, having maintained an average of 5 per cent growth in its GDP from 2011 to 2013.2 Because Nigeria’s economy is heavily dependent on oil,3 the recent volatility in oil prices has inspired efforts at diversification, including supporting development of local manufacturing and MSMEs.4 To address the problem of a lack of access to credit for a significant majority of MSMEs, a number of reform activities have been undertaken, particularly with the assistance of the World Bank Group.5 These efforts culminated in 2015 when the Central Bank of Nigeria issued the Registration of Security Interests in Movable Property by Banks and Other Financial Institutions in Nigeria Regulations (CBNR).6 Two years later, Nigeria’s legislature, the National Assembly, passed the Secured Transactions in Movable Assets Act (STMAA).7 Pursuant to the CBNR, an online notice-based collateral registry was established in May 2016.8 However, these reforms have not resulted in a coherent legal framework for secured transactions, and the current regime departs in a number of significant aspects from the Model Law. Nigeria also embarked on reforms to the legal framework for leasing, which led to the enactment of the Equipment Leasing Act (ELA).9 The Equipment Leasing Association of Nigeria (ELAN) has been the main policy advocate for the enactment of the ELA since

1 See C Aylward et al, ‘Nigeria Economic Growth and Poverty’ EPAR Technical Report #327 (22 March 2016) www.evans.uw.edu/policy-impact/epar/research/nigeria-economic-growth-poverty. Between 1992 and 2009, Nigeria’s GDP per capita grew at almost 70 per cent. 2 World Bank, Nigeria Economic Report no 2 (July 2014) 3–4. Upon a rebasing of Nigeria’s economy carried out by Nigeria’s National Bureau of Statistics, Nigeria’s GDP for 2013 was revised to US$509 billion, making it the 26th largest economy in the world and the largest in Africa. 3 PWC Economics & Policy, ‘What Next for Nigeria’s Economy? Navigating the Rocky Road Ahead’ (May 2015) 2, www.pwc.com/ng/en/assets/pdf/economy-watch-may-2015.pdf. Oil accounted for over 90 per cent of exports and 70 per cent of government revenue in 2013. 4 M Buhari, ‘The Three Changes Nigeria Needs’ Wall Street Journal (New York, 13 June 2016). 5 M Lucas, ‘IFC, CBN to Launch Collateral Registry’ The Nation (Lagos, 23 March 2015) www.thenation onlineng.net/ifc-cbn-to-launch-collateral-registry/. 6 Central Bank of Nigeria, ‘Circular to Banks and Other Financial Institutions – Secured Transactions and National Collateral Registry for MSME Financing in Nigeria’ (29 June 2016) FPR/DIR/GEN/06/00. 7 See O Adetayo, ‘Osinbajo Signs Acts to Ease MSMEs’ Access to Credit’ Punch (Lagos, 31 May 2017) www.punchng.com/osinbajo-signs-acts-to-ease-msmes-access-to-credit/. 8 World Bank, ‘Nigeria's New Collateral Registry Aims to Increase Access to Finance for Small Business’ (3 August 2016) www.worldbank.org/en/news/feature/2016/08/03/nigerias-new-collateral-registry-aims-to-increase-accessto-finance-for-small-business. 9 The Equipment Leasing Act (ELA) was enacted in 2015 to provide businesses and individuals with easier ways to acquire assets that enhance living standards and business growth. ELAN Nigeria, ‘Equipment Leasing Act 2015 Official Gazette is Out’ Linkedin (7 January 2016) www.linkedin.com/pulse/equipment-leasing-act2015-official-gazette-out-elan-nigeria/.

140 Nigeria the 1980s.10 Its enactment has not been coordinated with the STMAA, and, among other features, the ELA contemplates the establishment of a registry for equipment leases, including those that functionally create security interests.11 So far, there is no evidence that the Equipment Leasing Registration Authority contemplated by the ELA has been established. As a result, the implementation status of the ELA is uncertain.

A. Economic Overview Nigeria’s economy is heavily dependent on oil for government revenues and foreign exchange.12 It is Africa’s biggest oil exporter and has the largest reserves of natural gas on the continent.13 Agriculture employs 70 per cent of the labour force,14 while the services sector is the largest contributor to GDP, comprising 60 per cent.15 Nigeria’s economy grew at an average rate of 5.7 per cent between 2006 and 2016.16 The steep decline in global oil prices, combined with a resumption in militancy in the oil-producing Niger Delta area and a scarcity of foreign exchange, caused Nigeria’s economy to contract, ushering in a period of economic recession from 2016 to 2017.17 The IMF forecasts that the economy will grow by 2.0 per cent in 2018.18 Nigeria’s economic growth has been stifled by an inadequate power supply, lack of infrastructure, delays in implementing legislative reforms, a slow and ineffective judicial system, pervasive corruption, insecurity, and an inconsistent regulatory environment.19 As a result, the government has sought to increase transparency, to diversify the economy away from oil and to improve fiscal management.20 It is estimated that over 90 per cent of Nigerian businesses are MSMEs,21 employing more than 80 per cent of the labour force.22 However, MSMEs face severe challenges 10 IO Olubor, ‘The Next Frontier for Finance Leasing in sub-Saharan Africa: Revisiting Secured Transactions Law in Nigeria’ (UNCITRAL, 2017) www.uncitral.org/pdf/english/congress/Papers_for_Congress/84OTAOR-OLUBOR-The_Next_Frontier_for_Finance_Leasing_in_sub-Saharan_Africa.pdf. 11 ELA, explanatory memorandum. 12 CIA, ‘Economy: Nigeria’ (The World Factbook, 23 April 2019) https://www.cia.gov/library/publications/ the-world-factbook/geos/ni.html. The petroleum sector provided 93 per cent of export earnings and 51 per cent of federal revenues in 2015. ‘Nigeria Country Commercial Guide: Exporting to Nigeria – Market Overview’ Export.gov (29 November 2018) www.export.gov/article?id=Nigeria-Market-Overview. 13 World Bank, ‘Nigeria: Overview’ (9 April 2019) www.worldbank.org/en/country/nigeria/overview. 14 See ‘Exporting to Nigeria – Market Overview’ (n 12). 15 See CIA (n 12). 16 See World Bank (n 13). 17 Y Kazeem, ‘Nigeria Has Now Been in Recession for Five Quarters – But There May be Light at the End of the Tunnel’ Quartz Africa (23 May 2017) www.qz.com/989805/nigerias-economy-has-been-in-recessionfor-the-fifth-consecutive-quarter/. 18 World Economic Outlook Update, ‘Brighter Prospects, Optimistic Markets, Challenges Ahead’ (IMF, 22 January 2018) www.imf.org/en/Publications/WEO/Issues/2018/01/11/world-economic-outlook-updatejanuary-2018. The economy is further projected to grow by 2.5 per cent in 2019. African Development Bank Group, ‘Nigeria Economic Outlook’, www.afdb.org/en/countries/west-africa/nigeria/nigeria-economic-outlook/. 19 See CIA (n 12). 20 ibid. 21 ‘MSMEs Represent 96 Per Cent of Businesses in Nigeria – DG SMEDAN’, Punch (Lagos, 2 March 2019) www.punchng.com/msmes-represent-96-per-cent-of-businesses-in-nigeria-dg-smedan/. 22 In December 2013, the Small & Medium Enterprises Development Agency of Nigeria (SMEDAN) announced that MSMEs employ 84.02 per cent of Nigeria’s labour force. SMEDAN, SMEDAN and National Bureau of Statistics Collaborative Survey: Selected Findings (2013) (SMEDAN, 2013) 6.

Pre-Reform Secured Transactions Framework 141 in their operations, especially a lack of access to finance. An overwhelming number of micro-enterprises (in excess of two-thirds) identify access to finance as a great challenge.23 Similarly, 30 per cent of all businesses identify access to finance as the main business obstacle, a figure that is higher than the sub-Saharan Africa average,24 whereas 70 per cent of all private enterprises have limited or no access to credit.25 Most businesses finance their operations through savings and retained earnings, or funds borrowed from friends and family.26 To increase the availability of credit, Nigeria embarked on reforms to its secured credit legal and institutional framework, enacting a new law and regulations on secured transactions and establishing a collateral registry.27 This led to an improvement in Nigeria’s ranking in the World Bank’s 2018 Doing Business report. Nigeria was placed sixth out of 190 economies in the Getting Credit indicator,28 a marked improvement from its ranking of 44 in the 2017 edition.29

B. Pre-Reform Secured Transactions Framework The recent reforms replaced a secured transactions system based on the English security devices. Previously, a security right in movable property could be created using a bill of sale, a pledge, a lien or a charge.30 In addition, movable property could secure an obligation under functional equivalents of the security devices, such as finance leases, sale agreements with the reservation of ownership, and hire-purchase agreements. As explained below, the reforms have not completely replaced these devices with the functional concept of security right contemplated by the Model Law. The STMAA purports to create a unitary and functional framework for secured transactions in Nigeria. Thus, a number of secured transactions, such as assignments of receivables and rights of owners under hire-purchase agreements, constitute security interests for the purposes of the STMAA. Nevertheless, some uncertainty persists, especially because the STMAA neither repeals the Hire-Purchase Act (HPA), nor expressly prevails over the Bills of Sale Acts of the respective states.31 Because the Bills of Sale Acts falls within the competence of the state governments, it is unclear to what extent the STMAA (which is a federal law) will prevail over the continued effectiveness of the Bills of Sale Acts.32 23 ibid 35. 24 Enterprise Surveys: Nigeria Country Profile 2014 (Washington, DC, World Bank, 2014) 4. 25 L Nwabughiogu, ‘Osinbajo Signs Law to Ease Access to Credit Facilities for MSMEs’ Vanguard (Lagos, 30 May 2017) www.vanguardngr.com/2017/05/osinbajo-signs-laws-ease-access-credit-facilities-msmes/. 26 See Enterprise Surveys, Nigeria Country Profile 2014 (n 24) 11. 27 ibid. 28 Doing Business 2018: Reforming to Create Jobs (Washington, DC, World Bank, 2018) 183. 29 Doing Business 2017: Equal Opportunity for All (Washington, DC, World Bank, 2017) 229. 30 Interlaw Banking and Finance Team, Worldwide Survey on Taking Security over Personal/Moveable Property and Collateral (Interlaw, 2015) 39–40. Under s 197 of Companies and Allied Matters Act (CAMA) Cap C20, Laws of the Federation of Nigeria 2004, a charge includes a mortgage. 31 See B1 below. 32 For a brief discussion of the relationship between the Bills of Sale Acts and the Secured Transactions in Movable Assets Act 2017 (STMAA) see ‘The Secured Transactions in Movable Assets Act: Key Highlights and Impact on the Existing Legal Landscape’ Olaniwun Ajayi (November 2017) 2, www.olaniwunajayi.net/wp-content/ uploads/2017/11/Understanding-the-Secured-Transactions-in-Movable-Assets-Act-6.pdf.

142 Nigeria This ambiguity is exacerbated by the fact that the regulation of secured transactions is not explicitly listed in the Exclusive Legislative List of the federal legislature under the 1999 Constitution.33 Thus, although the functional approach is reflected in the scope provision of section 2(1) and the definition of security interest,34 the assumption is that the HPA and the Bills of Sale Acts remain in force. As explained below, the STMAA affects the Companies and Allied Matters Act (CAMA), which provides for the registration of floating and fixed charges created by companies.35

1. Bill of Sale The nineteenth-century English law Bills of Sale Acts36 regulated the use of bills of sale, which were written documents whereby an interest in goods could be conveyed by the owner to another person without delivery of possession. The Bills of Sale Acts only covered bills of sale granted by individuals, such as consumers or unincorporated businesses. Although a bill of sale could result in an absolute transfer of title, the most common use was to create a mortgage of the goods to which it related, that is, a transfer of title by way of security.37 The Bills of Sale Acts regulated their use by imposing requirements of form (which had the effect of limiting the assets that could be the subject matter of a bill of sale) and a registration requirement (for security bills of sale).38 The English Bills of Sale Acts were made applicable in Nigeria by the Bills of Sale Act 1923.39 When Nigeria became a federation, the competence over bills of sale legislation was transferred to state governments.40 Although some Nigerian states adopted specific bills of sale laws,41 most continue to apply the Bills of Sale Act 1923.42 Regardless, there is no material difference between the Bills of Sale Act 1923 and the states’ versions.43 Since the provisions of the bills of sale legislation across Nigeria are similar, this subsection will be restricted to the provisions of the Lagos State Bills of Sale Law

33 See C1 below; See also Second Schedule, Part I (Exclusive Legislative List), of the Constitution of the Federal Republic of Nigeria 1999. 34 See D1 below. 35 See B4 below. 36 Bills of Sale Act 1878 and Bills of Sale Amendment Act 1882. See ch 1, E1a. 37 Note that the definition of ‘bill of sale’ in s 4 of the Bills of Sale Act 1878 is wide enough to include charges as well as mortgages. For an account of the history of the Bills of Sale Acts, see Law Commission of England and Wales, ‘Bills of Sale’ (2016) LC369, 2.6–2.9. 38 Note that the Law Commission of England and Wales has proposed the repeal of the Bills of Sale Acts and replacing them with a system of goods mortgages. These proposals are not being taken forward by the UK government at the moment. 39 The Nigerian Bill of Sale Act 1923 was essentially a copy of the English Bill of Sale Acts. See C Mwalimu, The Nigerian Legal System: Private Law, vol 2 (New York, Peter Lang, 2009). 40 The 1954 Lyttleton Constitution made Nigeria a federal state with three geopolitical regions; Northern Region, Eastern Region, and Western Region. See JS Coleman, Nigeria: Background to Nationalism (Berkeley, University of California Press, 1971) 371–73. Presently a federation of states, the legislative powers of the aforementioned regions now lies with the states. See the Second Schedule (Legislative Powers) to the Constitution of the Federal Republic of Nigeria 1999, Part II (Concurrent Legislative List), item 18. 41 The erstwhile Western Region parliament was particularly active in law-making and enacted the Bills of Sale Law 1958, which was similar to the Bills of Sale Act 1923. Now defunct, the states that emerged from that region retain use of the Bill of Sale Law 1958. Lagos state, Nigeria’s economic and commercial capital, is one such state. 42 IO Smith, Nigerian Law of Secured Credit (Lagos, Ecowatch, 2001) 152–53. 43 ibid 152.

Pre-Reform Secured Transactions Framework 143 (‘the BSL’).44 Despite its long history, the bill of sale has fallen into disuse in Nigeria and only a few states, such as Lagos, maintain a dedicated bills of sale registry.45 Unless a security bill of sale complies with the requirements of the BSL as to the form prescribed in the schedule, it will be void.46 The bill of sale must be duly stamped in accordance with the Stamp Duties Act.47 It must also contain, or have annexed to it, a schedule in which the goods covered by the bill of sale are described.48 A bill of sale is void in respect of any goods which the grantor did not own at the time of the bill’s execution, which means that a bill of sale cannot cover future property.49 The BSL mandates that a bill of sale be registered within seven days after execution,50 and an unregistered bill is void against all third parties.51 If, despite paying registration fees and submitting the bill of sale for registration within seven days, the registry fails to register it on time, the bill of sale will be deemed to have been properly registered.52

2. Pledge The pledge, a possessory security interest, has a long history in Nigeria, having been recognised in the customary law of certain tribes and used long before the introduction of the English security devices.53 The device of a pledge in Nigeria can also be found in statutory law (eg, the Pawnbrokers Act 1917).54 Nigeria’s Supreme Court held a pawn to be synonymous with a pledge, stating in Labode v Otubu55 that ‘pledge is interchangeable with pawn and both have been defined to mean a bailment of movable assets as security for some debt or engagement’.56 Pledges over ‘documentary intangibles’, such as bills of lading 44 Lagos State (Nigeria), The Laws of Lagos State of Nigeria: In Force on the 1st Day of February, 2003, Chapter B2, vol 1 (Lexis Nexis Butterworths, 2003). Lagos is the economic nerve centre of Nigeria. As such, the state represents Nigeria in the World Bank’s annual Doing Business reports. See Doing Business 2014: Nigeria (Washington, DC, World Bank, 2014). 45 IO Olubor, ‘Stuck in a Time Warp: Security Interests in Chattel Mortgages and the Bills of Sale Legislation in Nigeria’ (2015) 26(11) International Company and Commercial Law Review 345, 352–53. Olubor also notes the dysfunction wrecked by the existence of separate state bills of sale registries and the challenges it poses to unsuspecting assignees and grantees who may be unaware of a bill of sale registration in another state. 46 BSL, s 8. 47 Stamp Duties Act 2004, s 45. 48 BSL, s 9. 49 BSL, s 10 (s 5 in the 1882 Act). 50 BSL, s 14. APK Keesee, ‘Secured Lending in Nigeria’ (1981) 15(1) International Lawyer 25, 36. Adetoyin v Bank of the North Ltd [1976] 3 OYSHC. 51 BSL, s 14 (s 8 in the 1882 Act). See, eg, Adetoyin v Bank of the North Ltd (n 49). 52 International Bank for West Africa v Azubuike [1972] NCLR 37. See also Olubor (n 44) 349. 53 The concept of the pledge is common to the people of the defunct Eastern Region of Nigeria. These peoples now inhabit the states of Anambra, Abia, Enugu, Imo, Ebonyi, Rivers, Cross River, Bayelsa, Akwa Ibom and Delta. It is also known to other Nigerian tribes such as the Yoruba. See TO Nwogu, ‘Secured Transactions Laws of Nigeria and Cameroon through the Lens of Article 9 of the Uniform Commercial Code of the United States of America’ (unpublished SJD dissertation, Central European University, June 2011) 42. See also Ahaneku v Iheaturu [1995] 2 NWLR (Pt 380) 770, where a customary pledge was defined to include a security, agreement, pawn, undertaking, guarantee, etc. The Igbos (one of Nigeria’s largest tribes) have an indigenous system for securing credit through the pledge of land or personal property. Not entirely similar to the common law pledge, certain shared traits such as the requirement of possession are present. For an understanding of the concept of the pledge as it applies to traditional Igbo society see, SNC Obi, The Ibo Law of Property (London, Butterworths, 1963); Nwogu (n 52) 42–53. 54 Modelled on the Pawnbrokers Act 1877 of England. 55 Labode v Otubu [2001] 7 NWLR (Pt 712) 256. 56 ibid 276, citing Halsbury’s Laws of England, vol 36(1), 4th edn (London, Butterworths) para 101, 72.

144 Nigeria and warehouse receipts, have been used in the Nigerian banking system in the financing of commodities.57 No documentation is required for the creation of a pledge, which is created by the pledgee taking possession of the pledged assets,58 but as a matter of commercial practice, a letter of pledge evidencing the transaction and stating the rights of the pledgee to sell the goods upon default is commonly executed.59

3. Lien A lien has been defined by Nigerian courts as a ‘right to retain that which is in one's possession belonging to another till certain demands of the person in possession are satisfied’.60 In Co-operative Bank of Eastern Nigeria v Maria Eke, the court stated that a lien is ‘a right to retain property belonging to a debtor until he has discharged a debt’.61 A lien, unlike a pledge, arises where possession of goods is given to the lienholder for a reason other than the creation of a security interest.62 A number of Nigerian statutes contain provisions imposing non-consensual liens.63 Liens may also arise under the common law (originally by trade usage)64 or by contract. Contract can also extend a lien arising by operation of law, for example, by turning a particular lien (where the right to possess only secures the payment of sums relating to the goods possessed) into a general lien (where the right to possess secures the payment of all sums due from the owner to the lienholder).65 Although a lien is generally a right of retention with no accompanying power of sale, a contract creating a lien may grant the lien holder a power of sale, although such an arrangement may be recharacterised as a pledge.66 57 See B Olutola, ‘Trade Finance in Nigeria: Structured Commodity Financing as an Instrument for Mitigating Risk’ Banwo & Ighodalo (2011) 2, www.banwo-ighodalo.com/assets/resources/9ac537f61ae130f0544bd 49e6e8b10bd.pdf; O Anyanechi, ‘Bank Finance and Regulation: Multi-Jurisdictional Survey: Nigeria, Enforcement of Security Interests in Banking Transactions’ (International Bar Association, 2010) 2–3, www.ibanet.org/LPD/ Financial_Services_Section/Banking_Law/BankinglawSurveyApril2010.aspx. 58 Donald v Suckling [1866] LR 1 (QB) 585, 613. 59 Smith (n 41) 183–84; K Dosekun and O Senbore, ‘The International Comparative Legal Guide to Securitisation 2006 – Nigeria’ (Global Legal Group, 2006) 245. 60 Afrotec Technical Services (Nig) Ltd v MIA & Sons Ltd [2000] 15 NWLR (Pt 692) 730, 756; [2000] LPELR-SC132/ 1992; Smith (n 41) 188; Co-operative Bank of Eastern Nigeria v Maria Eke [1979] 2 FNLR 190; Hammond v Barclay [1802] 2 East 227, 235. 61 Co-operative Bank of Eastern Nigeria v Maria Eke (n 59) 196–97. 62 Re Cosslett (Contractors) Ltd [1997] EWCA Civ 2229. 63 See, eg, the Sale of Goods Act, the Nigerian Railway Corporation Act, the Customs and Excise Management Act, and the Merchant Shipping Act. See Smith (n 41) 207–09, 244–55. The English Law Sale of Goods Act of 1893 is applicable in Nigeria as a Statute of General Application. Statutes of General Application are laws that were enacted in England before 1 January 1900, which were extended to apply in Nigeria as received English law. See s 32(1) Interpretation Act, Cap I23, Laws of the Federation of Nigeria 2004; Nze Bernard Chigbu v Tonimas Nigeria Limited & Another [2006] 4 SCNJ 262. The Sale of Goods Act still applies in the majority of Nigerian states, either in its original form, or in an adapted form, without any material difference. 64 Tappenden v Artus [1964] 2 QB 185, 194–95. An example would be a warehouseman’s general lien, which, as a matter of usage and regular incorporation into warehousing contracts, has been held to be valid. K. Chellaram & Sons (London) Ltd v Butlers Warehousing & Distribution Ltd [1977] 2 Lloyds’s Rep 292. 65 H Beale et al, The Law of Security and Title-Based Financing, 3rd edn (Oxford, Oxford University Press, 2018) 5.85–5.87. 66 Anyanechi (n 56) 3; L Gullifer (ed), Goode and Gullifer on Legal Problems of Credit and Security, 6th edn (London, Sweet & Maxwell, 2017) 1–53. Chellaram & Sons (London) Ltd v Butlers Warehousing & Distribution Ltd [1978] 2 Lloyds’s Rep 412.

Pre-Reform Secured Transactions Framework 145

4. Charge The CAMA provides a statutory regime for securing obligations with charges created by companies.67 It was enacted in 1990 to replace the Companies Act of 1968, which was based on the English Companies Act of 1948, and was the product of the recommendations of a Law Reform Commission established in 1987.68 The CAMA’s provisions drew inspiration from English company law.69 Charges are commonly used to secure the fulfilment of obligations in debentures, which are instruments issued to acknowledge the indebtedness of the company.70 Nigerian courts defined a debenture as a security for a loan that creates a charge on the company’s property.71 The definition of ‘charge’ in the relevant part of CAMA includes a mortgage,72 and companies, according to the circumstances, can grant both charges and mortgages. For example, a mortgage may well be granted over contractual rights or receivables, taking the form of an assignment to the mortgagee who has an obligation to reassign only when the secured obligation is paid.73 Such a mortgage is likely to be an equitable mortgage, since this can be effected without notice being given to the account debtor.74 Although there is, in practice, very little difference between a fixed charge and a mortgage,75 it is important to realise what the differences between mortgages and charges are. Unlike a mortgage, a charge does not involve a transfer of title to the chargee. However, a charge is an encumbrance on the asset securing the debt. Charges are a creation of equity, so all charges (whether fixed or floating) are equitable. For a charge to be a fixed charge, the chargor must not be able to dispose of the charged assets without the consent of the chargee.76 A fixed or floating charge may be created over the assets of a company,77 and it is possible to take security over all the assets of a company through a combination of fixed and floating charges. A floating charge is defined by the CAMA as ‘an equitable charge over the whole or a specified part of the company’s undertakings and assets …, but so that the charge shall not preclude the company from dealing with such assets’.78 It entitles a chargor to use its 67 Cap C20, Laws of the Federation of Nigeria 2004. 68 A Bukola, ‘A Critical Appraisal of the Doctrine of Corporate Personality under the Nigerian Company Law’ NLII Working Paper Series 002, www.nlii.org/files/NLIIWPS002.pdf. See also A Adeniran, ‘A Mediation-Based Approach to Corporate Reorganizations in Nigeria’ (2003) 29(2) North Carolina Journal of International Law and Commercial Regulation 291, 291. 69 E Azinge and DCJ Dakas, NIALS’ Laws of Nigeria (Annotated): Companies and Allied Matters Act (Lagos, NIALS, 2013) vii. See also CW Iheme, Towards Reforming the Legal Framework for Secured Transactions in Nigeria: Perspectives from the United States and Canada (Cham, Springer, 2016) 40; MN Umenweke and AR Aladegbaiye, ‘Powers and Duties of the Corporate Affairs Commission as a Regulatory Body in Nigeria’ (2011) 2 Nnamdi Azikiwe University Journal of International Law and Jurisprudence 15, 15. 70 NC Ogbuanya, Essentials of Corporate Law Practice in Nigeria (Lagos, Novena Publishers, 2010) 502; Levy v Abercorris Slate and Slab Co [1888] 37 Ch D 260. 71 Nigerian Industrial Development Bank Ltd v Olalomi Industries Ltd [2002] 5 NWLR (Pt 761) 532. 72 CAMA, s 197(11). 73 Anyanechi (n 56) 3. 74 See Gullifer and Goode (n 65) 3-09. 75 See Beale et al (n 64) 6.52–6.65. 76 National Westminster Bank Plc v Spectrum Plus Limited [2005] UKHL 41. 77 United Bank of Nig Ltd v Tropic Foods Ltd [1992] 3 NWLR (Pt 228) 231; Intercontractors Nigeria Ltd v National Providence Fund Management Board [1988] 2 NWLR (Pt 76) 280, 292. 78 CAMA, s 178(1).

146 Nigeria assets in the ordinary course of business until the charge crystallises by enforcement by the chargee, the appointment of a receiver or manager by the court, or the company going into liquidation.79 The chargor may create other charges over the same collateral,80 but a later-in-time fixed charge may have priority over a floating charge, unless the fixed chargee had notice of a negative pledge clause contained in the floating charge.81 A judgment creditor will have priority to the extent that execution is completed before the floating charge crystallises,82 but not if execution is not complete at that date. Thus, in Mandilas & Karaberis Ltd v Anglo-Canadian Cement Company Ltd,83 the right of a judgment creditor was held to be subordinate to the right of a floating charge holder who appointed a receiver prior to the completion of execution by the judgment creditor. On the insolvency of the chargor, or enforcement by a receiver, the priority of a floating charge (but not a fixed charge) is subject to preferential claims.84 Because a bill of sale cannot be created over goods not owned by the grantor,85 only a company can grant a floating charge over goods. The qualities of a floating charge have the result that a company can grant security interests over all its assets, including those that circulate. The registration regime under the CAMA only applies to the categories of charges listed in section 197(2), replicating the list in the English law Companies Acts until the 2013 reforms, which made all charges registrable unless exempt.86 Failure to register the listed charges with the Corporate Affairs Commission (CAC) will render the charge void against the liquidator and any (secured) creditor of the company.87 The charge must be received for registration by the CAC within 90 days from the date of its creation.88 This means that there is, potentially, an ‘invisibility period’ of 90 days in which searchers of the register will not discover a charge which has been created and which, if registered within the 90-day period, will be valid against subsequent chargees in accordance with the priority rules. The priority rules between charges are complex.89

5. Other Security and Quasi-Security Devices A number of quasi-security devices have been made available in Nigeria to functionally take movable assets as collateral. Examples of such quasi-security devices include assignments of receivables,90 used in receivables financing, and hire-purchase transactions. Outright assignments of receivables were not required to be registered and the relevant priority rule

79 ibid. Intercontractors Nigeria Ltd v National Providence Fund Management Board (n 77). 80 CAMA, s 179. 81 ibid. 82 ‘Nigeria’ in Worldwide Survey on Taking Security over Personal/Moveable Property and Collateral (n 30) 41; Evans v Rival Granite Quarries Ltd [1910] 2 KB 979; Re Standard Manufacturing Co [1891] 1 ch 627, 640. 83 Mandilas & Karaberis Ltd v Anglo-Canadian Cement Company Ltd [1967] NCLR 42. 84 CAMA, ss 182 and 494(1). 85 B1 above. 86 See the UK’s Companies Act 2006 (Amendment of Part 25) Regulations 2013. 87 CAMA, s 197(1). 88 ibid. 89 For example, they depend to some extent on whether the security interest is legal or equitable, and over what type of assets it has been created. See Beale et al (n 64) s IV. 90 ibid 7.75–7.133; and F Oditah, Legal Aspects of Receivables Finance (London, Sweet & Maxwell, 1990).

Recent Reforms 147 was that in Dearle v Hall,91 namely that the first assignee to give notice to the account debtor had priority, despite not being first in time, unless the notice-giving assignee knew of the earlier assignment at the time it took its assignment. As in other common law jurisdictions, an assignment of a receivable in breach of an anti-assignment clause was invalid and without effect vis-a-vis the debtor, although once paid, the assignor would hold the proceeds on trust for the assignee.92 As in other jurisdictions, this can have the practical effect that many receivables financiers will not finance receivables containing anti-assignment clauses.93 Hire-purchase transactions, governed by the federal HPA,94 were mainly used to finance the acquisition of motor vehicles. The hirer has the right to return the goods at any time without being obliged to pay any further balance of rent that may accrue after return of the goods.95 The HPA applies to hire-purchase transactions as well as credit-sale transactions under which the total purchase price or the hire-purchase price does not exceed 2000 naira (roughly US$6).96 Since the HPA was enacted in 1965 and last amended in 1970, the effect of inflation has rendered it inapplicable to transactions of commercial significance.97 However, the HPA’s relevance is saved by section 1(b) which provides that it shall apply to all hire-purchase and credit-sale agreements in respect of motor vehicles, irrespective of the hire-purchase price or total purchase price. The reference to motor vehicles has been held by Nigeria’s Supreme Court to include specialised vehicles such as rigs and agricultural scrapers.98 Where one-half of the purchase price has been paid, or three-fifths in the case of motor vehicles, an owner has no right to recover possession of the goods from the hirer otherwise than pursuant to a court action.99

C. Recent Reforms Nigeria embarked on a flurry of reforms to its secured transactions legal framework between 2014 and 2017. These reforms led to the enactment of legislation on leasing and secured transactions, namely the Equipment Leasing Act (ELA), the Secured Transactions in Movable Assets Act, and the Central Bank of Nigeria Registration of Security Interests in Movables Regulations. An electronic notice-based collateral registry was established under the Regulations. 91 Dearle v Hall [1828] 38 ER 475. 92 Gullifer and Goode (n 65) 3–38 to 3–45. Helstan Securities Ltd v Hertfordshire CC [1978] 3 All ER 262. Any stipulation in the contract that consent of the debtor be obtained before any assignment must be complied with. S Harwood, Shipping Finance, 3rd edn (London, Euromoney, 2006) 228; T Owolabi et al, ‘Lending and Taking Security in Nigeria: Overview’ in Practical Law Multi-Jurisdictional Guide 2014/2015 Finance (2014) 3; Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85. 93 See ch 1, E4a. 94 Hire-Purchase Act, Cap H4, Laws of the Federation of Nigeria 2004. Modelled on the English Hire-Purchase Act 1938. 95 Samuel Aro v Joe Allen & Co Limited [1979] 2 FNR 292, 295. See also Kasali A Raimi v Moshudi Funso Ogundana & 2 ors [1986] 3 NWLR (Pt 26) 97; Afrotec Technical Services (Nig) Ltd v MIA & Sons Ltd & Anor [2000] 15 NWLR (Pt 692) 730. 96 HPA, s 1(a). 97 See N Ofo, ‘Hire-Purchase Legislation in Nigeria: Making a Case for Reform’ (2007) 2(1) Ebonyi State University Law Journal 159. 98 CDC (Nigeria) Ltd v SCOA (Nigeria) Plc [2007] 6 NWLR (Pt 1030) 300. 99 See HPA, s 9(1) and (4). CDC (Nigeria) Ltd v SCOA (Nigeria) Plc (n 98).

148 Nigeria

1. Leasing The 2015 ELA provides the framework for leasing transactions in Nigeria.100 Before the ELA, equipment leases were governed by the common law, the HPA and the agreement of the parties.101 Efforts to reform the legal framework for leasing go back as far as 1976 when the Okigbo Report on the Nigerian Financial System recommended the revision of certain laws in order to promote the growth of leasing.102 The effort to enact the ELA was not coordinated with the development of the STMAA, even though both laws were being developed at the same time. An example of the lack of coordination between the ELA and the STMAA is that the ELA provides for the establishment of a lease registry where registration of lease agreements affects the rights of third parties. As discussed below,103 finance leases fall within the definition of ‘security interest’ in the STMAA104 and the right of a finance lessor is a purchase money security interest.105 Therefore, arguably, the legal effect of finance leases that qualify as security interests should be governed by the STMAA, and notices with respect to those leases should be registered in the collateral registry only. Those equipment leases that fall outside the scope of the STMAA (such as operating leases, which do not secure payment of an obligation) should be registered in the lease registry. However, while efforts have been made to establish a registration entity for equipment leases (as contemplated by the ELA),106 there is no evidence that a leases registry has been established. For the time being, therefore, it appears that the ELA has not been put into operation. Furthermore, the STMAA requires that any public registry created pursuant to federal law for the registration of transactions in movable assets must be operated in a manner that provides an accessible automatic interface between it and the collateral registry.107 Therefore, it is expected that any registry eventually established under the ELA would be accessible through the collateral registry. The nature of the required automated interface remains unclear. The ELA defines a lease agreement as a written agreement between a lessor and lessee for the lessee’s use of a lessor’s own equipment (either already acquired or to be acquired from a supplier) in consideration for the payment of an agreed rental over a specified period.108 It defines ‘equipment’ to include movable and immovable equipment, but there is no definition of immovable equipment.109

100 ELA, s 1. The ELA recognises the following as types of equipment leases: finance leases, operating leases, syndicated leases, sale and lease back, cross-border lease, leverage lease, and any other equipment lease approved from time to time by the minister. ELA, s 3. 101 Oserogho & Associates, ‘Legal Alert – August 2016 – Equipment Leasing Laws and Taxation’ (7 September 2016) www.oseroghoassociates.com/articles/198-legal-alert-august-2016-equipment-leasing-laws-and-taxation. 102 CU Uche, ‘Equipment Leasing in Nigeria: Problems and Prospects’ (2002) 3 Journal of International Banking Regulations 381. 103 See D1 below. 104 STMAA, s 63(1). 105 ibid. 106 A Akuki, ‘ELAN To Implement Equipment Leasing Act 2015, Inaugurates ELRA’ Independent (Lagos, 22 April 2017) www.independent.ng/elan-implement-equipment-leasing-act-2015-inaugurates-elra/. See also ELA, s 12. 107 STMAA, s 2(1)(c). 108 ELA, s 44. 109 ibid.

Recent Reforms 149 The ELA stipulates that the lease agreement must be registered with the Equipment Leasing Registration Authority (ELRA)110 (a body to be created pursuant to the ELA to register lease agreements and leasing businesses) not later than 14 days after the agreement’s execution.111 An unregistered lease agreement is invalid between the lessor and lessee, and unenforceable against a third party acting in good faith for value without notice of the lease agreement who may take free of the interest of the lessor.112 Registration confers priority on the lessor, as the lessor’s interest in the leased equipment will prevail against all third party claims, including claims by the lessee’s creditors.113 Affixing the leased equipment to the land or building will not affect the lessor’s right.114 A lessee is barred from creating any rights, including sublease, mortgage, charge, etc, in favour of a third party unless the lease agreement expressly authorises such transactions.115 In the event that the lessee defaults, the lessor has to serve a default notice on the lessee giving the lessee not less than 14 days to remedy the default,116 failing which the lease agreement may be terminated by the lessor.117 If the lessee does not return the leased equipment to the lessor, the lessor may bring an action for repossession.118

2. Central Bank of Nigeria Registration of Security Interests in Movables Regulations In February 2015, Nigeria’s Central Bank issued the Registration of Security Interests in Movable Property by Banks and other Financial Institutions in Nigeria Regulations (Regulations No. 1 of 2015) (‘the CBNR’).119 The CBNR provides a framework for accessing secured credit and creating, perfecting, and realising security rights in movable assets.120 It also provided the legal basis for the establishment of a collateral registry and rules for the collateral registry’s operations.121 The CBNR has been criticised by a number of commentators for failing to provide a unitary and functional approach for secured 110 The ELRA is established under ELA, s 7 and is required to maintain a register of equipment leases (ELA, s 9(1) (a) and 9(2)(e)). 111 ELA, s 13. Registration is to be effected using Form A of the First Schedule to the ELA, together with other documents prescribed by the ELRA. ELA, s 14. 112 ELA, s 17. Note that the phrase ‘invalid between the parties’ is very odd: if the lease is invalid, it is not clear how it can be enforceable against a third party who is not a good faith purchaser without notice. 113 ELA, s 21. 114 ELA, s 19. Note that parties cannot agree to exclude the application of this provision of the ELA, as any such excluding provision will be null and void to the extent of its inconsistency with the ELA. ELA, s 19(4). 115 ELA, s 20(1) and (2). 116 ELA, s 37. Default notice should be in the manner prescribed in Form B of the first schedule to the ELA. ibid. 117 ELA, s 38. 118 By ex parte motion to the Federal High Court. ELA, s 39(1). This is without prejudice to other remedies that may be claimed by the lessor under law. ELA, s 39 (3). 119 IO Olubor, ‘Reforming the Law of Secured Transactions: Bridging the Gap between the Company Charge and CBN Regulations Security Interests’ (2016) Journal of Corporate Law Studies 18; WC Iheme and SU Mba, ‘Towards Reforming Nigeria's Secured Transactions Law: The Central Bank of Nigeria's Attempt through the Back Door’ (2017) 61 Journal of African Law 131. 120 See Circular to Banks and other Financial (n 6). 121 ibid. The registry became operational in May 2016. World Bank, ‘Nigeria's New Collateral Registry Aims to Increase Access to Finance for Small Business’ (3 August 2016) www.worldbank.org/en/news/feature/2016/08/03/ nigerias-new-collateral-registry-aims-to-increase-access-to-finance-for-small-business.

150 Nigeria transactions in Nigeria.122 For example, critics highlight the fact that the CBNR excludes from its scope charges created under section 197(2) of the CAMA.123 This criticism fails to take into account the underlying considerations involved in the reform process, as explained in the following paragraphs.124 The CBNR provided the basis for the establishment of a collateral registry, anticipating the passage of the Secured Transactions in Movable Assets Bill that was already under development at that time. Thus, the motivation behind the adoption of the CBNR was to expedite the operationalization of a future secured transactions regime. Furthermore, taking into account the risk that the enactment of the STMAA would be delayed, the policymakers sought to establish a notice-based registration system, albeit only for a narrow set of secured transactions. This step was taken to allow Nigerian lenders to familiarise themselves with the new registration mechanism, and to incentivise lending secured with movable assets to which the CBNR would be applicable. The CBNR are limited in scope because, as regulations issued by the Central Bank, they cannot supersede or override the CAMA or any other statute. The Central Bank has the power to make secondary legislation (such as the regulations) for the promotion of a sound financial system,125 and for the regulation of financial institutions under the Banks and Other Financial Institutions Act (BOFIA).126 Accordingly, the power of the Central Bank to make secondary legislation is limited to the scope of its enabling Act127 and the BOFIA. It is in this light that the CBNR was drafted to apply solely to regulated financial institutions subject to the supervision of the Central Bank and to those security devices not governed by a statute, such as company charges registrable under the CAMA.128 The CBNR are structured as a modern secured transactions law covering all elements of a security interest, as they are based on the then available draft of the Model Law. They served as the basis for the drafting of the STMAA, so a discussion of their provisions here would duplicate those of the STMAA in the following section. It is unclear whether the STMAA fully superseded the CBNR.129 The website of the Collateral Registry lists the STMAA and the CBNR, which may imply that the CBNR became the Registry Regulations for the STMAA.130 Some of the CBNR’s general rules might have been superseded by the STMAA, but those relating to the operations of the Collateral Registry would seem to remain applicable. In summary, the CBNR incorporated a number of modern principles,131 including that a security right may be taken in after acquired assets,132 that the effect of anti-assignment clauses on the validity of a transfer and a security interest in

122 See Olubor (n 118). 123 ibid. 124 One of the authors of this work served as the IFC consultant on secured transactions reform in Nigeria. 125 Central Bank of Nigeria Act 2007, s 2(d). 126 Banks and Other Financial Institutions Act (BOFIA), s 57; Laws of the Federation of Nigeria 2004, Cap B3. 127 Central Bank of Nigeria Act 2007. 128 CBNR, reg 3(3)(b). See B4 for a discussion of the CAMA registration regime. 129 For transition, see D8 below. 130 The Nigerian Collateral Registry website includes the CBNR as the governing regulations, see ‘National Collateral Registry’, www.ncr.gov.ng/. 131 See ch 1, A2. 132 Provided the after-acquired goods fall under the collateral description in the security agreement. CBNR, reg 4(2).

Secured Transactions Law – The Reform 151 receivables is overridden,133 and that a security right can be perfected by registration of a financing statement in a notice-based electronic collateral registry.134 Although modern secured transactions registration regimes typically apply equally to security interests and outright transfers of receivables, there is no overt indication in the CBNR that its registration regime applies to such outright transfers.135 The collateral registry is maintained by the Central Bank.136 There are some differences in terminology between the STMAA and CBNR frameworks, such as the use of the terms ‘security right’ and ‘security interest’ or ‘grantor’ and ‘debtor’. This chapter uses the terminology of the relevant framework.

3. Secured Transactions in Movable Assets Act of 2017 The STMAA was enacted by the National Assembly and entered into force on 30 May 2017. Based on the CBNR, it reflects a number of modern principles, but at the same time, departs from the approaches of the Model Law, especially with respect to the creation of a unitary secured transactions framework. It also prescribes some novel approaches, which the drafters thought were necessary for its successful implementation, particularly in the area of enforcing security interests. The following section examines the STMAA in detail.

D. Secured Transactions Law – The Reform 1. Scope The STMAA implements the functional approach, defining a security interest as a property right in collateral that is created by agreement and secures payment or other performance of an obligation, regardless of whether the parties have denominated it as a security interest but it does not include a personal right against a guarantor or other person liable for the performance of the secured obligation.137

This broad definition, which is very similar to the definition of ‘security right’ in the Model Law,138 clearly includes devices based on retention of title such as conditional sales, hirepurchase agreements and finance leases. Moreover, rights of finance lessors are included in the definition of purchase money security interests.139 Finance leases, therefore, are within the scope of the STMAA. Curiously, the STMAA includes a definition of ‘operating lease’, 133 CBNR, reg 5(2). Under Nigerian common law an assignment in breach of an anti-assignment clause does not bind the account debtor. See C5 above. The Regulations thus create a statutory exception to the anti-assignment rule in favour of security rights created, or any transfer of receivables under the CBNR. For discussion of the wording of this provision, see D2 below. 134 A financing statement refers to the prescribed forms on which information is provided to effect, amend, cancel or continue a registration under the CBNR. See ‘financing statement’ in CBNR, reg 2(1). 135 The definition of the term ‘security interest’ includes no mention of outright transfers of receivables. 136 CBNR, reg 9(1). 137 STMAA, s 63(1). 138 ML, Art 2(kk). 139 STMAA, s 63(1). Finance lease is defined as ‘a lease which transfers ownership of the asset to the lessee at the end of the lease term’.

152 Nigeria and section 39(2) provides that operating lessors may enforce their rights under other law or the lease agreement. However, as a whole, the STMAA does not provide how it applies to operating leases, as is commonly done in the PPSAs. Since operating leases do not functionally create security interests, a secured transactions law must specifically prescribe that it applies to operating leases, which the STMAA has not done. Similarly, nothing in the STMAA expressly subjects outright transfers of receivables to its scope. This is puzzling, especially because a number of provisions are drafted as if the STMAA applied to such transfers. For instance, section 4(2) provides ‘[a] transfer of account receivable or security interest in an account receivable by a grantor is effective’. The scope of the STMAA is limited in that it does not preclude the creation of charges under the CAMA. The STMAA did not repeal the provisions of the CAMA on the registration of companies’ charges,140 and section 2(3) of the STMAA provides that: ‘Nothing in this Act shall prevent the creation of security interest [sic] in the form of charges by companies registered under the Companies and Allied Matters Act.’ The meaning of this provision, and the lack of repeal, is unclear. It is very unclear whether the phrase ‘registered under the Companies and Allied Matters Act’ refers to the word ‘charges’ or the word ‘companies’. If it referred to ‘companies’, the provision could be read to mean that a security interest in the form of a charge can still be created, but would be a ‘security interest’ within section 2(1) of the STMAA. In that case, any security interest that is labelled ‘charge’ could still be registered in the National Collateral Registry (NCR), although the combination of section 2(3) and the lack of repeal of the provisions of CAMA probably would lead to the conclusion that the ‘charge’ could also (or alternatively, at the choice of the secured creditor) be registered with the CAC. This interpretation is problematic because, if effectively both regimes apply to such a ‘charge’, lack of registration in either registry could lead to invalidity against third parties and in insolvency. Therefore, on this interpretation, double registration is effectively required. The other problem with this interpretation is that it has the result that a ‘security interest’ is also a charge, which is either fixed or floating, and, if the latter, falls within the sections of the CAMA giving priority to certain preferential claims.141 Alternatively, if the phrase referred to ‘charges’, then the provision appears to mean that a company seeking to use its movable asset as collateral has a choice between creating a security interest over that asset under the STMAA (registered in the NCR) or creating a charge over the asset under the CAMA (registered in the CAC). On this view, as long as the interest is registered in the relevant registry, it will be valid against third parties and in insolvency, but additional problems arise, which are discussed in the following paragraphs. On either interpretation, section 2(3) coupled with the lack of repeal of the CAMA, creates a dual system, which increases the burden of searchers, who must consult both registries to assess the status of the company’s assets, giving rise to considerable uncertainty for lenders and corporate borrowers as to what security interests can be created. Furthermore, there is an increase in legal complexity in relation to priority. On the first interpretation, all interests will be registered in both registries. However, the priority rules in relation to each regime are completely different (the rule under the STMAA is basically first to file and under the CAMA is first to be created), and so the priority result is entirely unclear. 140 CAMA has been the subject of some amendment efforts in 2017 and 2018, but none of them relate to company charges. See Companies and Allied Matters Act 1990 (CAP C20, LFN 2004) Repeal and Re-enactment Bill, 2018. 141 See B4 above.

Secured Transactions Law – The Reform 153 On the second interpretation, the assets of a company may be encumbered under the STMAA in favour of one creditor and under the CAMA in favour of another creditor. This results in a conflict for which neither the STMAA nor the CAMA provides a priority rule. Arguably, a ‘first in time’ rule should determine the outcome of this priority conflict, but this approach is problematic. If it means that the first to register wins, then this is flatly contradictory to the regime under the CAMA and would, arguably, require legislation to bring it about. If it means that the first to create a security interest wins, then a later secured creditor may lose to an earlier security interest of which it had no notice because of the 90-day invisibility period under the CAMA.142 Moreover, a secured creditor taking advantage of the ability under the STMAA to register a financing statement in advance of creation of a security interest to secure priority143 would not necessarily achieve priority if a charge entered into under the CAMA was created in the period between the STMAA registration and the creation of the security interest. Even more problematic is the fact that a charge under the CAMA would be equitable,144 and so, under the common law, an earlier equitable charge would cede priority to a later legal interest taken by a bona fide purchaser without notice of the earlier charge. Although the distinction between law and equity does not apply in relation to a reformed secured transactions law regime, such as the STMAA, it applies in the rest of Nigerian law. The approach likely to be followed is that of the Supreme Court of Canada, which has held that a security interest under a PPSA regime is a statutory interest and therefore a legal interest.145 On this approach, a later STMAA interest taken by a creditor who had no notice of an earlier CAMA charge (because of the invisibility period)146 would prevail over the CAMA charge.147 Another effect of the failure to harmonise the regime is that it will hamper the prompt adoption of the STMAA regime by creditors, because finance companies are likely to prefer to continue extending credit under the familiar rules of the CAMA rather than the new regime. One example, pointed out by local commentators,148 is that the holder of a fixed or floating charge under the CAMA has the power to appoint a receiver if the charge becomes enforceable,149 and this is seen as desirable and therefore an incentive to create a CAMA charge. However, the holder of a security interest under the STMAA can also appoint a receiver, although the exact meaning of the provisions is far from clear. Section 39(1)(a) provides that, on default, in addition to exercising its rights under the STMAA, a creditor may exercise any rights under the security agreement. These rights can therefore include a right to appoint a receiver. Confusingly though, section 39(5) provides that ‘[t]he remedies available under this Part are in addition to those available under 142 See B4 above. 143 See D3. 144 See B4 above. 145 Innovation Credit Union v Bank of Montreal [2010] SCC 47; Bank of Montreal v I Trade Finance Inc [2011] SCC 26. 146 Or, perhaps, because the STMAA creditor did not search the CAC and had no constructive notice of the CAMA interest. 147 The discussion in this paragraph owes a great deal to Gregory Esangbedo, whose MPhil thesis on STMAA and the funding of MSMEs discusses the priority problems in much more depth. 148 Email from Chima Williams Iheme, to Marek Dubovec, Executive Director, NatLaw, 21 June 2017 (on file with authors). 149 This can be done by the court (CAMA, s 180) or under a power in the charge instruments (see CAMA, ss 178(a) and 206, which refer to such a power).

154 Nigeria [the CAMA], including the right to appoint a receiver’, which might be thought to indicate that the CAMA right still remains, even if not included in the security agreement. Accordingly, the CAMA does not appear to provide for a more favourable enforcement regime than the STMAA; despite this, a combination of preference for a familiar regime and the unclear drafting in section 39 of the STMAA is likely to lead many creditors to use CAMA charges rather than STMAA security interests.150 There are many other differences that prospective creditors may take into account when deciding whether to take a security right under one or the other regime (on the basis that the correct interpretation is that there is a choice). For instance, the STMAA does not require the payment of stamp duties, while the CAMA does.151 However, the Stamp Duties Act stipulates that the failure to stamp any document will render it inadmissible in evidence in civil proceedings.152 Thus, although the STMAA disposes of the requirement to pay stamp duties for the validity of the agreement or registration of a financing statement, the civil procedural requirements, practical realities and the custom of stamping documents in Nigeria would seem to defeat the STMAA’s good intentions. However, it may be argued that under the STMAA an agreement could be stamped later in time where a court dispute between the parties arises. Another factor which might point in the opposite direction is that, on the taking or possession of a company’s assets by a receiver or on the winding up of a company, preferential debts (eg unpaid tax and unpaid wages) are to be paid out of any assets subject to a floating charge in priority to the floating chargee.153 There appear to be no equivalent provisions in relation to a security interest created under the STMAA. However, the STMAA does not include the equivalent of Article 36 of the Model Law where the enacting State would list the preferential claims, so it remains unclear whether the security interest is subordinate to any preferential claims. If the first interpretation of section 2(3) discussed above is correct, then at least some security interests created by companies would also be floating charges and would be subject to the provisions in the CAMA about preferential claims. If, however, the second interpretation is correct, there is then no provision for preferential claims in relation to STMAA security interests. Although this is, perhaps, an incentive for creditors to use the STMAA regime rather than the CAMA, it is also an unfortunate omission. All common law regimes that have reformed their secured transactions law have had to struggle with finding an equivalent to the floating charge to act as a trigger for the priority of preferential creditors on insolvency.154 The alternative is for the state to decide to abolish the priority given to preferential creditors, but this should not be done without a policy debate about the consequences of such abolition.155

150 This is on the basis that there is a choice, but, as mentioned above, it may be that the correct analysis is that a security interest can just be registered in both registries. 151 STMAA, s 54 provides ‘[t]he provisions of the Stamp Duties Act shall not apply to any secured transactions subject under this Act’. 152 See ‘The Secured Transactions in Movable Assets Act: Key Highlights and Impact on the Existing Legal Landscape’ (n 31) 6. 153 CAMA, ss 182 and 494. 154 See Kenya C9 and Conclusion C6bii. 155 For example, the UK took the conscious decision in 2002 to abolish the preferential status of the tax authorities, see Enterprise Act 2002, s 251.

Secured Transactions Law – The Reform 155 As mentioned earlier,156 the STMAA provides that: [E]very public registry established by any Act of the National Assembly to coordinate or warehouse or oversee transactions in movable assets in Nigeria shall be operated in a manner that creates automated interface between such a registry and the National Collateral Registry, with a purpose to ensuring and guaranteeing that the registry is made accessible through, by, and from the National Collateral Registry.

The effect of this provision is uncertain, but the intention seems to require more than a simple weblink to the CAC on the website of the NCR. Aside from the CAC, it is unclear which other registries provide an interface with the NCR.157 Furthermore, the STMAA does not spell out the legal effect of the failure to provide such an interface.

2. Creation and Perfection A security interest must be created by a security agreement, which must meet certain requirements.158 Strangely, and unlike under the Model Law,159 there is no requirement that the agreement be in writing or signed by the grantor.160 The agreement may provide for the creation of a security interest in both present and future assets. Unlike under the previous law in relation to pledges,161 a security interest cannot be created by delivery of possession to the secured creditor, and possession by the secured creditor is also insufficient to perfect a security interest, as will be further explained below. The anti-assignment override introduced in the CBNR162 is included in STMAA section 4(2)(b). However, the drafting of this subsection is defective in that, first, it overrides any agreement limiting the ‘account debtor’s’ right to transfer or create a security interest. Instead of ‘account debtor’, the reference should have been to the ‘grantor’ who is the person that has the right to transfer the receivable. Secondly, the subsection makes a transfer of, or security interest in, an account receivable effective against the account debtor notwithstanding ‘any agreement’ limiting the grantor’s right to transfer or create a security interest. The phrase in inverted commas should have been qualified by the words ‘between the account debtor and the grantor’,163 otherwise, the provision could have unintentional effects on negative pledge clauses in the loan agreement between the grantor and the secured creditor. In any event, the anti-assignment clause override in the STMAA may be an incentive for creditors to choose the STMAA regime over the CAMA, which does not include a similar override of anti-assignment clauses in relation to the creation of a charge. Section 4(1) provides that a security interest is only created to the extent of the rights the grantor has in the collateral. This is a rather unfortunate provision, because the priority 156 See C1 above. 157 Any registry established under the ELA would presumably need to provide an interface. 158 STMAA, s 5. 159 ML, Art 6(3). 160 The definition of ‘security agreement’ in s 63(1) provides that it can be ‘in any form’ and ‘howsoever entitled’. The requirements in STMAA, ss 3–6 do not include writing or a signature. 161 B2 above. 162 See C2 above. 163 A more complicated version of this qualification appears in ML, Art 13(2).

156 Nigeria rules can sometimes act as an exception to the nemo dat rule, in that a later interest has priority over an earlier-created interest. In such situations, the grantor may create a security interest beyond the extent of its rights in the collateral. It is for this reason that the equivalent provision in the Model Law and in many PPSAs is drafted as a threshold provision: a security interest can only be created (or attach) if (and when) the grantor has rights in the collateral or the power to transfer rights in the collateral.164 The problem derives from the use of the words ‘to the extent of ’ in the Nigerian provision, which seems to exclude the operation of any exception to the nemo dat rule, and is, of course, inconsistent with the later priority provisions. A security agreement must contain information that mirrors that of Article 6(3) of the Model Law, namely, identification of the grantor and the secured creditor, and an adequate description of the collateral, which can be specific or generic, including by the use of the phrase ‘all the present and future assets of the grantor’.165 The security agreement must also include a description of the secured obligation by the maximum amount for which the security interest is enforceable, which is an optional feature of the Model Law.166 In addition, section 5 of the STMAA requires an indication of the tenor of the secured obligation, which is customary in Nigerian banking transactions, as well as a confirmation by the parties to submit their disputes to arbitration ‘as first recourse’. This provision appears to reflect a clause in Nigerian agreements where parties have expressly agreed that no right of action shall accrue in respect of any disputes between them until such disputes have been adjudicated via arbitration. The Nigerian courts have held that such a clause operates as a bar to commencing any action prior to arbitration.167 The requirement that the parties include such a clause in a security agreement is a highly unusual provision for a modern secured transactions law, effectively forcing the parties to agree to resort to arbitration as a condition for the validity of the security agreement. This provision was motivated by the desire to lessen the burden on the courts. Registration is the sole method of perfection for all security interests in any type of asset. Section 8(2) confirms that taking possession of the collateral by the secured creditor does not perfect a security interest. The STMAA does not contemplate any additional perfection methods, such as control or automatic perfection, provided for in the Model Law. This approach to perfection is consistent with a number of recently enacted secured transactions laws in Africa, including Kenya and Sierra Leone.168

3. Registration Part III of the STMAA provides for the establishment of the NCR, which was already established pursuant to the CBNR. Consequently, unlike in a typical reform, the entry into force of the STMAA was not suspended for a long period to allow for the establishment of 164 See ML, Art 6(1), Australian PPSA, s 19(2), UCC Art 9-203(b)(2), Saskatchewan PPSA, s 12(1)(b). cf New Zealand PPSA, s 40 (this omits any reference to a power). See also discussion in ch 2, C2. 165 STMAA, s 6(1)(b), which reflects ML, Arts 8 and 9. 166 ML, Art 3(d). 167 See SCOA Nig PLC v Sterling Bank PLC, Appeal No CA/L/170/2013, citing AIDC v Nigeria LNG Ltd [2000] 4 NWLR (Pt 653) 494 SC; [2000] 2 iLAW/SC.174/95. 168 See conclusion D4, and Kenya C3, Sierra Leone C3.

Secured Transactions Law – The Reform 157 a registry. The NCR is operated by the Central Bank, the Governor of which is vested with the authority to appoint a Registrar.169 The Registrar has the power to determine the fees for various NCR services through regulations.170 In addition, the Registrar may issue rules and guidelines for the proper administration of the NCR.171 At present, the regulations governing the registration process appear to be the CBNR,172 and their provisions are therefore referred to in the following discussion. A financing statement may be registered (only) electronically upon obtaining the grantor’s consent, which may be given prior to the execution of a security agreement or by (the grantor’s) entering into the security agreement.173 The secured creditor must properly complete the financing statement, providing the following information: (i) an identification of the grantor, which, for individuals, is a unique identification number from approved biometric-based identification174 and, for organisations, a number assigned by an appropriate agency, such as the CAC; (ii) the name and address of the secured creditor or its representative; (iii) a description of the collateral;175 (iv) the maximum amount for which the secured obligation may be enforced; and (v) the period of time for which the registration is to be effective.176 By virtue of section 14(2), the financing statement may also identify borrowers who have not granted a security interest but received a loan secured by the security interest granted by the grantor in the security agreement,177 which is a common practice in the Nigerian banking industry. A confirmation that the parties agreed to mediation and arbitration must also be indicated in the financing statement.178 Unlike the Model Law, but consistent with many other African secured transactions reforms,179 the STMAA contemplates serial-number indexing for certain types of assets, unless they are held as inventory.180 Serial-numbered goods are defined and limited to motor vehicles, airplanes and boats.181 Errors in the grantor’s unique identification number or a serial number that cause the financing statement not to be retrieved in a search will render the financing statement wholly (in the former case) or partially (in the latter case – that is, with respect to the asset identified by the incorrect serial number or not identified by a serial number at all when such identification is required) ineffective.182 Section 17 deals with the legal effect of errors 169 STMAA, s 10(2). 170 STMAA, s 57. 171 STMAA, s 61. 172 See C2 above and n 129. 173 STMAA, s 13. 174 If the individual is not a citizen or resident of Nigeria, the number appearing on the individual’s foreign government-issued passport will be the unique identification number, or any other document as the Central Bank of Nigeria may prescribe. CBNR, reg 12(4). 175 The description in the financing statement must reasonably identify the collateral such as by item, kind, type or category, or a statement that a security right is taken in all of the debtor’s present property and property acquired after the security agreement is signed, or in any other way that reasonably identifies the collateral. CBNR, reg 13(1) and (2). If the collateral is a serial-numbered good, the financing statement must contain the serial number, unless the serial-numbered goods are held as inventory. CBNR, reg 13(3). 176 CBNR, reg 12(2). 177 See the definition of borrower in STMAA, s 63(1), which, though not entirely clear, appears to support the sentence in the text. 178 STMAA, s 14(5). 179 For example, Kenya C4; Sierra Leone C3; Zimbabwe C2; Nigeria E3; Zambia D2. See also conclusion D5c. 180 STMAA, s 15(3)–(5). 181 STMAA, s 63(1). 182 STMAA, s 16.

158 Nigeria in the other information in the financing statement, setting out two distinct standards: first, errors in the collateral description that seriously mislead the searcher will render the financing statement ineffective; and, secondly, errors in all the other information, however seriously misleading, will not have an impact on the effectiveness of the financing statement. Although departing from the Model Law approach,183 it would be extremely unlikely for those other items of information, such as an error in the address of the secured creditor, to seriously mislead a searcher. The section 17 approach establishes a hard rule designed to reduce wasteful litigation over relatively immaterial erroneous information.

4. Priority The STMAA provides a comprehensive set of priority rules, which, compared to the Model Law, are less complicated due to the absence of alternative methods of perfection. Section 23 prescribes the fundamental priority rule, which relies on the time of registration. Unlike Article 29 of the Model Law, it was not necessary to elaborate on this rule because the STMAA recognises registration as the sole method of perfection. Although unnecessary, section 33(3) expressly confirms that the time of notification of the account debtor is immaterial for the purposes of establishing priority between competing assignees of an account receivable. The intention of the drafters was to send a clear signal that this approach breaks from the past, that is, the application of the rule in Dearle v Hall.184 The priority of a security interest in the original collateral extends to any proceeds, which are defined widely as including proceeds of sale or other dispositions, collections, natural fruits and proceeds of insurance and other claims relating to the collateral.185 Section 24(2) provides that the priority of a security interest also applies to all obligations and advances, present and future, thus abolishing any rule against tacking.186 Although the language of that subsection appears to extend the priority to any future advances, it should be read in conjunction with the other provisions that require an indication of the maximum amount in a security agreement and financing statement.187 These do not just affect priority but prevent enforcement of the security interest in relation to advances which result in the amount outstanding exceeding the maximum amount. Section 27 provides an exception for purchase money security interests from the general priority rule of ‘first to register’. That provision partially implements Option B of Article 38 of the Model Law, which does not distinguish among the different types of assets, so that the same requirements apply to equipment and inventory. However, unlike Option B, the purchase money secured creditor is not provided with a grace period in which to register a financing statement after the grantor takes possession of the collateral. In other words, the purchase money security interest must be registered before the grantor acquires possession, which may delay delivery of the asset until registration is complete. This approach has been followed in some other African jurisdictions that were concerned about the ‘invisibility 183 See ch 2, E. 184 Dearle v Hall [1828] 38 ER 475. See B5 above. 185 STMAA, s 63(1). 186 Such a rule generally applies, with exceptions, in common law jurisdictions, see Gullifer and Goode (n 65) 5–10, 5–17–5–24. See ch 2. B4. 187 STMAA, ss 5(c) and 14(1)(f).

Secured Transactions Law – The Reform 159 period’ during which a person may be in possession of an asset subject to a security right that may be made effective against third parties and acquire super-priority if a notice is registered within a grace period, and felt that access to an electronic collateral registry would allow secured creditors to complete registrations without affecting delivery of goods to their buyers.188 The STMAA contains a number of ‘taking-free’ rules under which transferees of funds, cash, negotiable instruments and title documents, as well as buyers and lessees of goods, may take free of the security interest. Although these rules generally follow the Model Law, the STMAA incorporates some notable deviations. For instance, under Article 34(4) of the Model Law, a buyer in the ordinary course of business takes free of any security interests, whether created by the seller or its predecessor in title. In contrast, under section 32(2) of the STMAA, the protection for buyers in the ordinary course of business is limited only to the security interest created by the seller. The STMAA approach is aligned with the UCC Article 9 approach,189 rather than the Model Law. The STMAA’s priority rule for transferees of cash and funds is also inspired by the UCC Article 9 collusion standard.190 Unlike under the Model Law where the transferee does not take free of a security right if it takes with knowledge that such transfer violates the rights of the secured creditor,191 section 29(2) of the STMAA prescribes a higher standard, so that the transferee only takes subject to the security interest if it acts in collusion with the transferor. Section 34 also addresses a conflict with a judgment creditor who may obtain priority over a holder of a competing security interest if, prior to the perfection of the security interest, the judgment creditor registers a financing statement, seizes the collateral or serves a notice on the party holding the relevant asset of the debtor. Accordingly, the STMAA gives a judgment creditor who has not completed execution192 an option of notice-based registration in order to protect its rights, and provides a ‘first to register’ priority rule accordingly.

5. Rights and Duties of the Parties As in Chapter VI of the Model Law, the STMAA provides for the rights and duties of the parties to a security agreement as well as some third parties, but in a much more limited way, leaving the other aspects to be addressed contractually by the parties. Section 35(1) imposes an obligation to discharge any duties and rights arising under the STMAA in a commercially reasonable manner. The failure to do so entitles the aggrieved party to claim damages.193 While the imposition of this obligation mirrors Article 4 of the Model Law, the latter does not expressly provide for a right to damages, although it does provide for a right to relief from a court for breach of any obligation imposed by the chapter on enforcement.194 The other sections of Chapter VI impose a duty on the secured creditor 188 See Kenya C9. 189 The same approach is taken by the Australian PPSA, s 46(1), the New Zealand PPSA, s 53(1) and the Saskatchewan PPSA, s 30(2). 190 UCC, Art 9-332. 191 ML, Art 47(6). 192 By seizing the asset or giving notice to the third-party holder. 193 STMAA, s 36. 194 ML, ch VII.

160 Nigeria to provide certain information to the grantor (a confirmation statement with respect to a registration)195 and to the grantor or other third parties (eg statements about the amount of indebtedness and the status of the collateral), but only upon request of the grantor or the borrower.196 Article 56 of the Model Law limits the right to request and receive information to the grantor, whereas section 38 also entitles the borrower to make a request, and provides that the information shall be provided to any stipulated person.

6. Enforcement of Security Interests Part VII provides a comprehensive set of extrajudicial remedies, which may be complemented by any additional remedies that the parties have agreed on in a security agreement. In contrast to Article 72(1) of the Model Law, section 39 does not limit the available remedies to those compatible with the STMAA, although the overarching duty to act in a commercially reasonable manner will limit the types of contractual remedies to which the parties can agree. Article 72(1)(b) of the Model Law states that parties may exercise any remedies provided by some other law, except to the extent that it is inconsistent with the provisions of the Model Law. Instead of this general principle, section 39 is more detailed, stating that operating lessors, holders of purchase money security interests, outright transferees of receivables, and secured creditors who also hold charges under the CAMA may enforce any remedies available under other laws (or, in the case of charges, remedies available under the CAMA) governing their respective transactions, such as, in the case of charges, the appointment of a receiver. It is not clear why operating lessors are even mentioned since operating leases do not fall within the scope of the STMAA. It is understandable that an outright transferee should be able to enforce its rights under the general law, as they are effectively owners. On the other hand, the holders of purchase money security interests are secured creditors whose rights and remedies should be fully subject to the STMAA. The preservation of the remedies under the CAMA in section 39(5) is confusing197 and potentially produces uncertainty. It seems to suggest that, if the correct interpretation of section 2(3) is that the creditor has a choice whether to create a ‘security interest’ under the STMAA or a charge under the CAMA,198 the remedies under the CAMA are available if the creditor chooses the first option. However, it is not clear whether this relates only to a security interest which could have been created as a charge under the CAMA or whether it applies to all possible security interests a creditor might take. The repossession remedy in section 40 is modelled on Article 77 of the Model Law, requiring the secured creditor to send a notice of default and its intention to repossess the collateral to the grantor (and borrower, if different). After sending the notice, the secured creditor must wait 10 days before taking any action.199 As a condition for extrajudicial repossession of the collateral, the grantor’s consent must be obtained in the security agreement. Section 40 provides one very powerful remedy that is not contemplated by the Model Law, namely, empowering the secured creditor to seek assistance from the Nigerian police to effectuate

195 STMAA,

s 37. s 38. 197 See D1 above. 198 For discussion of different interpretations of s 2(3), see E1 above. 199 STMAA, s 40. 196 STMAA,

Secured Transactions Law – The Reform 161 peaceable repossession. The police are required to lend such assistance upon presentation of a copy of the security agreement and a confirmation statement from the NCR of the registration of the relevant financing statement. This remedy may go a long way towards facilitating expeditious extrajudicial enforcement of security interests in a peaceable manner. It is also a strong incentive for secured creditors to register financing statements, as a confirmation of the registration is one of the conditions for seeking the assistance of the police. The Model Law generally encourages resolution of disputes through alternative dispute resolution mechanisms, including those that may relate to some aspects of enforcing security interests.200 Section 41 of the STMAA takes this one step further in that it expressly provides for the establishment of a Mediation and Dispute Resolution Panel to resolve any disputes that may arise under the STMAA. It is uncertain whether such a panel has been established. After repossession or rendering the collateral unusable,201 the secured creditor may dispose of it by sale, lease, licence or other form in an auction, public tender, private sale or other type of disposal provided for in a security agreement. Section 44(3) imposes on the secured creditor a duty to obtain a reasonably fair price at the time of disposal. Arguably, this provision is not necessary in light of the overarching duty to proceed in a c ommercially reasonable manner, but the rationale for including it was that it would give additional comfort to secured creditors and protect them against any unjustified arguments of grantors about a low selling price. Like the Model Law, the secured creditor must provide a notice of disposal to any affected parties, but a notice may be excused in certain circumstances, such as where the collateral is perishable. The rules on the discharge of security interests after disposal, distribution of proceeds and redemption of the collateral mirror those provided in the Model Law.202 Nevertheless, in contrast to the Model Law, section 50 provides for reinstatement of a security agreement by remedying any default. This right of reinstatement is characteristic of the PPSA regimes.203

7. Conflict of Laws The STMAA includes a set of rules that determine the law applicable to various aspects of a security interest. Unlike the Model Law rules, which inform the user about the law that generally applies to a particular aspect of a transaction even if that law is not the law of the enacting State, the STMAA’s conflict rules are drafted to determine when the STMAA applies. For instance, Article 85(1) of the Model Law provides: ‘[T]he law applicable to the creation, third-party effectiveness, and priority of a security right in a tangible asset is the law of the State in which the asset is located.’ In contrast, section 51(1) of the STMAA provides: ‘The law applicable to the creation, perfection and priority of a security interest in tangible property when the tangible asset is located in Nigeria is this Act.’ Section 51 provides general conflict of laws rules along the lines of Articles 85–90 of the Model Law. Since the STMAA recognises only registration as the method of perfection for security interests in any asset, there was no reason to incorporate the asset specific conflicts rules (from the Model Law)

200 See

ML, Art 3(3). remedy is provided for in STMAA, s 42. 202 STMAA, ss 46, 48, 49. 203 See, eg, Saskatchewan PPSA, s 62. 201 This

162 Nigeria on security rights in bank accounts and non-intermediated securities.204 This Part of the STMAA also contains a provision that deals with the payment of stamp duties (section 54), a matter that has nothing to do with conflict of laws and thus appears to be misplaced.

8. Miscellaneous This Part of the STMAA addresses various aspects of the secured transactions system, some of which were already highlighted above (eg the determination of fees for registry services). Section 58 provides a powerful deterrent to the misuse of the NCR, imposing steep fines and imprisonment for registering false or misleading information.205 In addition, a grantor who disposes of the collateral without disclosing to the third party, such as a buyer, that it is subject to a security interest commits an offence punishable by a fine or imprisonment. Even though this approach may reduce the incentive of the third party to search the NCR, it should still do so because, by virtue of section 20, an unauthorised transfer does not extinguish the security interest. The most important aspects dealt with in this Part are transitional matters and definitions. Section 62 addresses the transition of two types of security interests, namely those created under the CBNR, and those created under other laws. The former continue to be valid and effective against third parties for 180 days after the entry into force of the STMAA, unless a financing statement is registered in the NCR that complies with the STMAA. Since the registration requirements under the STMAA correspond to those under the CBNR and the NCR is the collateral registry established under the CBNR, financing statements relating to security interests created under the CBNR should remain effective until they expire. For any other security interests, they will remain valid and effective on the terms agreed to between the parties. Consequently, the STMAA does not contemplate a transition of prior security interests in such a manner that after a period of time the NCR would become the main source of information about security interests in movable assets. As a result, a searcher of the NCR should be wary of the possible existence of pre-STMAA security interests that are effective against third parties. Factoring in the possibility to create charges under the CAMA, the NCR does not provide as much comfort to the searcher as a registry established in a state that implements the Model Law. These approaches run counter to the elimination of alternative methods of perfection, which was done pursuant to a desire to strengthen the role of the collateral registry.

E. Final Remarks Nigeria has taken important steps towards modernising its secured transactions framework. These efforts were undertaken to tackle the persisting challenge of access to credit for MSMEs. Unlike the wholesale reform contemplated by the Model Law, Nigeria 204 See ML, Arts 97 and 100. 205 STMAA, s 58(3) defines false and misleading information as ‘the entry of information of facts required to assure the public of the truth and accuracy of the information by a person who knew that the entry is false in any material particular or is recklessly neglectful of the duty to ensure the truth or accuracy of the information supplied by that person’.

Final Remarks 163 first adopted a set of regulations with the goal of establishing a collateral registry that would be ready for deployment when a substantive law was brought into force. This approach faced a number of challenges, especially with respect to defining the powers and limitations of the Central Bank to regulate certain security devices. It also gave some basis to commentators to criticise the reform process as not providing a comprehensive and functional regime. Although the enactment of the STMAA addressed such criticism, it also raised a new set of issues caused primarily by its unique approaches and ambiguous drafting, which deviate from the Model Law. The STMAA implements the functional approach, but fails to establish a unitary framework for all secured transactions in movable assets. First, it does not seem to apply to outright transfer of receivables, despite a number of references to both transfers and security interests in receivables. Similarly, the STMAA refers to operating leases without any explicit provision as to whether it applies to operating leases and how. Secondly, it introduces a great deal of ambiguity with respect to the reference to charges that may be created under CAMA. This chapter considers two plausible interpretations of that clause. The first is that it provides for the creditor to have a choice to create a charge either under the CAMA or a security interest under the STMAA. The second is that a security interest created in the form of a charge would need to be registered both in the CAC and the Collateral Registry. The confusion and complexity are compounded by the lack of any priority rules that would resolve the conflicts resulting from this provision. Thirdly, the coherence of the secured transactions framework was further eroded by the enactment of the ELA, which contemplates the registration of lease agreements, including those that functionally create security interests. Although it contemplates the establishment of a registry, no such registry has yet been established, further compounding the uncertainty. Fourthly, the STMAA did not include any express repeal and amendment provisions, as contemplated in Article 101 of the Model Law. The extent to which it superseded some relevant state legislation, especially the HPA and Bills of Sale Acts, is uncertain. Some other approaches that deviate from the Model Law appear to have been influenced by UCC Article 9. The prime examples are the priority rules under which buyers in the ordinary course of business take free of only the interest created by the seller and transferees of money and funds take free unless they acted in collusion with the grantor. These alternative approaches have no negative effect on the internal structure of the priority rules. Other provisions have been drawn from the PPSA, such as the right of the grantor to reinstate a security agreement. As with the UCC-inspired provisions, this departure from the Model Law has no negative effect on the internal structure of the priority rules. Nigeria falls into the group of African reformers that designated registration in a collateral registry as the sole method of perfection, thus rejecting the alternatives, such as possession and control. The reliability of the Collateral Registry is undermined by the transitional rules under which some prior security interests will remain effective according to the terms and conditions agreed to between the parties. The STMAA also provides for unique approaches with respect to enforcing security interests, particularly the power to seek assistance of the police, as well as the resolution of any disputes through non-judicial mechanisms. The implementation of these approaches in practice should be closely monitored, as they may become models for other reforming jurisdictions. Considering the relatively recent implementation of Nigeria’s reform, there has been little opportunity to assess the efficiency of the STMAA’s unique approaches as well as the

164 Nigeria overall application and economic impact of the statute. The number of MSMEs against which registrations in the Collateral Registry were made, suggesting they obtained a secured loan, rose from 100,049 in 2017 to 154,827 as at 19 December 2018, an increase of over 54 per cent.206 The report issued by the Collateral Registry also showed that 22,251 of the MSMEs were female entrepreneurs, 146,777 of the borrowers were individuals, 3,416 were microbusinesses, 2,169 were medium businesses, 1,777 were small businesses and 687 were large businesses. Regarding the types of collateral identified in registrations, a significant proportion covered documents of title/negotiable instruments (17,105), motor vehicles (11,725), and consumer/household goods (9,029). Only 2,573 covered intangibles, 2,531 inventory, 1,580 farm products and 537 equipment indicating a lack of commercial lending. At the time of preparing this chapter, the STMAA had not generated any actual complications, particularly with respect to the troublesome intersection between the STMAA and the CAMA. This may be partially attributable to the STMAA’s facilitation of secured loans to borrowers who would not be companies falling under the CAMA, as evidenced in the statistics above. However, creditors should remain cautious and appreciate the complexities discussed herein that could have been avoided by a more faithful implementation of the Model Law. Finally, in light of the many deficiencies, it is surprising that Nigeria ranks sixth in the world in the Doing Business, Getting Credit rankings 2018 and 12th in the 2019 rankings.207

206 See B Komolafe, ‘Lending to MSME’s Under National Collateral Registry to Hit N2trn’ Vanguard (Lagos, 8 February 2019) www.vanguardngr.com/2019/02/lending-to-msmes-under-national-collateral-registry-to-hit-n2trn/. 207 See World Bank, ‘Rankings & Ease of Doing Business Score’ (May 2018 and 2019) www.doingbusiness.org/en/ rankings.

8 Zambia In April 2016, Zambia enacted the Movable Property (Security Interest) Act (MPSIA).1 The MPSIA is primarily modelled on the Canadian, Malawian and New Zealand PPSA, but also draws inspiration from the recommendations of the Legislative Guide. In 2012, one of the authors was retained by the IFC of the World Bank Group to assist in the reform process that led to the MPSIA.2 Some statements and observations thus reflect this author’s actual experiences in the design and implementation of the reform.

A. Introduction Zambia is a landlocked, lower-middle-income economy in southern Africa.3 Agriculture is the main source of employment,4 while copper mining and processing are the mainstay of the economy as well as the chief source of foreign exchange.5 Zambia’s dependence on copper has made the economy vulnerable to fluctuations in commodity prices.6 After averaging annual growth in GDP of 7.4 per cent between 2004 and 2014, its economy grew by just 2.8 per cent in 2015 and 3.3 per cent in 2016.7 In 2016, agriculture, industry and services, accounted for 9.2, 29.2, and 61.7 per cent respectively of Zambia’s GDP.8 Zambia’s 1 No 3 of 2016. 2 IFC, ‘IFC Helps Zambia Expand Access to Finance with New Secured Transactions Law’ (1 June 2016) https:// ifcextapps.ifc.org/ifcext%5Cpressroom%5Cifcpressroom.nsf%5C0%5C0BEC2FDC2620E09885257FC5003A4F50. 3 World Bank Group, ‘World Bank Country and Lending Groups: Lower-Middle-Income Economies ($1,026 to $4,035)’ www.datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lendinggroups; Zambia Tourism, ‘Zambia: Geography’, www.zambiatourism.com/about-zambia/geography; CIA, ‘Geography: Zambia’ (The World Factbook, 30 April 2019) https://www.cia.gov/library/publications/the-worldfactbook/geos/za.html. 4 Agriculture employs 85 per cent of the labour force. CIA (n 3). See also C Sebatta, M Wamulume and C Mwansakilwa, ‘Determinants of Smallholder Farmers’ Access to Agricultural Finance in Zambia’ (2014) 6(11) Journal of Agricultural Science 63. 5 GJ Williams, ‘Zambia: Physical and Social Geography’ in Regional Surveys of the World: Africa South of the Sahara, 33rd edn (London, Europa Publications, 2004) 1195. Copper accounts for as much as 80 per cent of all foreign exchange earned from exports. G Chelwa, ‘The Charts Behind Zambia’s Struggling Economy and a Controversial IMF Loan’ Quartz Africa (23 November 2015) www.qz.com/557335/the-chartsbehind-zambias-flailing-economy-and-a-controversial-imf-loan/. 6 Zambia is Africa’s second largest producer of copper. S Tshabalala, ‘Zambia’s Economy is in Trouble – And the Government Is Lashing Out at a Rating Agency’ Quartz Africa (28 September 2015) www.qz.com/africa/512120/ zambias-economy-is-in-trouble-and-the-government-is-lashing-out-at-a-rating-agency/. 7 World Bank, ‘Overview: Zambia’, www.worldbank.org/en/country/zambia/overview. 8 CIA (n 3). Main agriculture products include, inter alia, corn, tobacco, flowers, sorghum, rice, cotton and livestock. The top industries are copper mining and processing, construction, horticulture, foodstuffs, fertiliser and emerald mining. ibid. The services sector is comprised of communications, transportation, tourism and financial services. International Trade Centre, ‘Services Snapshot: Zambia’, www.intracen.org/uploadedFiles/intracenorg/ Content/Redesign/Sector/Global_Services_Network_blog/Zambia.pdf.

166 Zambia top exports are copper, cobalt, tobacco, electricity, flowers and cotton.9 Its economy has been adversely affected by continuous budget deficits,10 which increased the government’s external debt to US$6.7 billion, representing 35 per cent of Zambia’s GDP.11 Zambia has 19 commercial banks,12 8 leasing and finance companies, and 36 microfinance institutions.13 The vast majority of businesses in Zambia are MSMEs. It is estimated that there are over 1.05 million MSMEs in Zambia, with 1.02 million operating in the informal economy.14 Collectively, MSMEs employ 88 per cent of the labour force.15 The major sectors of operation for MSMEs are agriculture and wholesale/retail trade.16 Historically, Zambian businesses faced challenges accessing credit. For instance, a 2007 survey by the World Bank found that only 16 per cent of SMEs surveyed reported having a loan or line of credit from a financial institution compared to 23 per cent in the region and 35 per cent for all countries included in that survey.17 In a speech in 2012, the Governor of the Bank of Zambia (BoZ) acknowledged that access to credit is the principal challenge for MSMEs.18 Before the implementation of the MPSIA, commercial bank lending was predominantly secured by immovable collateral.19 When MSMEs were able to access finance without immovable collateral, they were charged high interest rates, which banks attributed to the high cost of loan monitoring.20 To curtail the high interest rates, in January 2013 the BoZ introduced interest rate caps. Initially, the maximum rate was set at 18.25 per cent for banks,

9 CIA (n 3); ‘Zambia Facts’ National Geographic, www.travel.nationalgeographic.com/travel/countries/ zambia-facts/. Zambia’s top export partners are Switzerland, China, Singapore, South Africa and the Democratic Republic of Congo. In turn, Zambia imports machinery, transportation equipment, petroleum products, electricity, fertiliser, foodstuffs and clothing. Its major import partners are South Africa, the Democratic Republic of Congo, China, Mauritius, Kenya and India. CIA (n 3). 10 Zambia’s external and domestic debt has been rising since 2011. IMF, ‘Zambia: Staff Report for the 2015 Art IV Consultation–Debt Sustainability Analysis’ (5 May 2015) 2, www.imf.org/external/pubs/ft/dsa/pdf/2015/ dsacr15152.pdf. 11 C Mfula, ‘Zambia to Cut Budget Deficit, Sees Higher 2017 GDP Growth’ Reuters (11 November 2016) www.reuters.com/article/zambia-budget-idUSL8N1DC3N5. The public debt has been reported to be as high as 57.2 per cent of GDP. CIA (n 3). 12 ‘Registered Banks’ (BoZ, March 2017) www.boz.zm/REGISTEREDBANKSMARCH2017.pdf. 13 BoZ, ‘Final List of NBFIs’ (September 2017) www.boz.zm/FinalListofNBFIsSeptember2017.pdf. 14 MK Shah, ‘The Informal Sector in Zambia: Can it Disappear? Should it Disappear?’ International Growth Centre Working Paper (June 2012) 8, www.theigc.org/wp-content/uploads/2014/09/Kedia-Shah-2012-WorkingPaper.pdf. In 2016, taxes constituted only 16.6 per cent of GDP. CIA (n 3). For Zambia to meet its development goals, the IMF and World Bank recommend that this figure be raised to 20 per cent. WN Nhekairo, ‘The Taxation System in Zambia: Final Report’ (JCTR) 11, www.taxjustice-and-poverty.org/fileadmin/Dateien/Taxjustice_and_ Poverty/Zambia/JCTR/JCTR_2014_taxstudy.pdf. 15 World Bank, ‘Zambia Business Survey: The Profile and Productivity of Zambian Businesses’ (June 2010) 3, siteresources.worldbank.org/INTZAMBIA/Resources/The_Profile_and_Productivity_of_Zambian_Businesses.pdf. 16 R Mukwena and F Sumaili (eds) Zambia at Fifty Years: What Went Right, What Went Wrong and Wither To? A Treatise of the Country’s Socio-Economic and Political Developments Since Independence (Partridge Africa, 2016) 4.2.1. 17 ‘Resolutions and Recommendations of the Bank of Zambia Conference on Increasing Access to Finance to MSMEs’ (14 June 2012) (on file with the authors). 18 M Gondwe, ‘Opening Remarks at the Launch of the Soweto Branch of Stanbic Bank Zambia Limited: Importance of the SME Sector in Zambia’ BIS (31 August 2012) 1, www.bis.org/review/r120911e.pdf. 19 CE Deressa, ‘MSMEs Access for Finance in Zambia’ (2014) 5(10) Research Journal of Finance and Accounting 77. 20 ibid 77–79. Interest rates are as high as 22 per cent. A Gray, ‘Access to Finance for Inclusive Businesses in Zambia’ Inclusive Business Action Network (1 November 2012) www.inclusivebusinesshub.org/access-tofinance-for-inclusive-businesses-in-zambia/.

Background to the MPSIA 167 30 per cent for non-banks and 42 per cent for microfinance institutions.21 However, the policy was not successful, and in November 2015, the BoZ withdrew it.22 Access to credit was further muted by the lack of sufficient understanding and expertise, on the part of Zambia’s commercial banks, to provide credit secured with movable property.23 An IMF study found that in 2013, lending to the private sector stood at 12 per cent of GDP, a very low figure compared to the sub-Saharan Africa average of 21 per cent.24 The implementation of the MPSIA was heralded in Zambia because of its potential to revolutionise commercial lending in the country.25

B. Background to the MPSIA The two main drivers of the reform project were the BoZ, through its Non-Bank Financial Institution Supervision Department, and the Patent and Company Registration Agency (PACRA). Representatives from both entities chaired the working group, which included other stakeholders such as commercial banks. The involvement of the BoZ and PACRA was critical to the successful completion of the project. PACRA was experienced in managing the Companies Registry and was expert in companies and insolvency legislation, whereas the BoZ provided insights into the practices and needs of financial institutions. The working group was advised by a legislative drafter whose skills and understanding of the subject matter ensured the passage of the Bill with minimal amendments and within a reasonable timeframe. Accordingly, the Bill was submitted to the government in a format that complied with the standard drafting style guidelines set by the Ministry of Justice.26 This saved time as otherwise it would have taken two or three years before the legislative drafting arm of the Ministry of Justice could review the Bill and revise it in compliance with the statutory drafting guidelines. Engaging an experienced and respected legislative drafter early in the process is therefore an important lesson for similar projects. Curiously, in 2012, Zambia’s legal framework – then based on English law security devices – was ranked globally at number eight in the Getting Credit Indicator of the World Bank’s Doing Business Report, whereas Canada, whose legal framework formed the basis of the Model Law, was ranked 24.27 Zambia was therefore not one of the e conomies 21 IMF, ‘Enhancing Financial Inclusion in Zambia’ (2014) 6. 22 Bank of Zambia, ‘Removal of Interest Rate Caps and Consumer Protection Measures’, Circular No 19/2015 (4 November 2015) www.boz.zm/Circular2015-19RemovalofInterestRateCapsandConsumerProtectionMeasures. pdf. For a succinct discussion of the frequent use of interest rate caps to combat predatory lending in Africa, see DM Mbengue, ‘The Worrying Trend of Interest Rate Caps in Africa’ CGAP (11 November 2013) www.cgap.org/ blog/worrying-trend-interest-rate-caps-africa. 23 Deressa (n19) 77-78. 24 IMF, ‘Enhancing Financial Inclusion in Zambia’ (2014). 25 See, eg, E Mseteka, ‘IFC Instrumental in Moveable Collateral Law, Says World Bank’ Zambia Daily Mail (Lusaka, 7 June 2016) www.daily-mail.co.zm/ifc-instrumental-in-moveable-collateral-law-says-world-bank/; J Wei, ‘Microcapital Brief: Zambia Passes Secured Transactions Law to Boost Lending to SMEs Based on Movable Collateral’ Micro Capital (19 June 2016) www.microcapital.org/microcapital-brief-zambia-passes-secured-transactions-law-to-boost-lending-to-smes-based-on-movable-collateral/. 26 Often, legislative drafting departments of Ministries of Justice are overloaded with Bills, so presenting a Bill that already complies with the style guidelines ensures quicker review. 27 See Doing Business 2012: Doing Business in a More Transparent World (Washington, DC, World Bank, 2012) 86 and 137.

168 Zambia in the world for which improving a low ranking in the Doing Business Report was the main motivating factor for undertaking reforms. Its decision to reform the secured transactions framework was based on empirical evidence of the difficulties local businesses face in accessing secured finance.28 The enactment of the MPSIA, promulgation of the Registry Regulations, and the launch of the Collateral Registry elevated Zambia to number two in the world in the Strength of Legal Rights Index of the World Bank’s Doing Business Report for 2018.29 Also, largely due to the secured transactions reform, Zambia was listed as one of the top 10 global reformers in 2017.30 The path to achieving that enviable ranking was long and tortuous. The working group had to grapple with many challenges, all of which were successfully overcome. The first of these challenges was the identification of the right approach for the reform. In the early years of the project, PACRA argued for an approach whereby the system of registration of company charges existing at that time would remain unaffected and the MPSIA would apply only to security interests created by non-companies (ie unincorporated debtors). Under this approach, PACRA would also maintain two registries: one for the registration of company charges and a collateral registry for security interests of non-companies. This approach raised a plethora of legal issues and risks that the group discussed in a series of meetings. For instance, under PACRA’s suggested approach, the new MPSIA would supersede the existing legislation, such as the laws under which a seller could retain ownership until the full satisfaction of the debt, for example, the Hire-Purchase Act (HPA) and the Trades Charges Act, but only with respect to unincorporated debtors. This approach would have led to further fragmentation of the legal framework, the complexity of which had stifled secured lending. Many situations were examined with respect to the prospective interaction of the Companies Act and the MPSIA, such as the third-party effectiveness and priority of a security interest created by a non-company in an asset that is transferred to a company which has already created a floating charge under the Companies Act over all its assets or that class of assets (assuming that the transfer would not extinguish the MPSIA security interest), and the effect of the incorporation of a previously unincorporated debtor on an MPSIA security interest. Many questions were also thoughtfully considered in relation to the impact of both approaches on the insolvency framework, which was being reviewed at the time. Coordination of the secured transactions reform with the concurrent reforms of the Companies Act and insolvency legislation was challenging because independent teams worked on the reform of each framework; at that time, the latter two projects were not sponsored by the IFC. Eventually, PACRA realised the challenges raised by its initial suggestion and agreed to pursue a complete overhaul of the framework resulting in a single body of law and a single registry. On reflection, all of these discussions expounded many policy and practical questions. Although the MPSIA was enacted and entered into force in April 2016, it could not be effectively implemented until January 2017 when the Collateral Registry was launched. Zambia is thus another African example of a secured transactions law that entered into force 28 The high ranking raises a separate question regarding the accuracy of the Doing Business Reports, and the precise measurement of the impact of secured transactions reforms modelled after the Model Law. 29 Doing Business 2018: Reforming to Create Jobs (Washington, DC, World Bank, 2018) 205. 30 See https://www.worldbank.org/en/news/press-release/2017/10/31/sub-saharan-african-economies-set-newrecord-in-reforming-their-business-climate-doing-business.

Pre-MPSIA Framework 169 before the Collateral Registry became fully functional.31 This challenge, examined below,32 was caused by the Ministry of Justice’s policy of expediting the entry into force of newly enacted legislation. Despite this significant transitory problem, expectations remain high. Estimates suggest that the reform would ‘generate $2 billion in financing and benefit 6,500 SMEs, 1,200 micro enterprises, and 1,200 women entrepreneurs’.33

C. Pre-MPSIA Framework The MPSIA replaced a patchwork of statutes governing lending secured on movable property. Section 6 of the MPSIA, entitled ‘Supremacy of Act’, provides that the MPSIA prevails over any inconsistent provisions of any other written law. The generality of this clause is supplemented by paragraph (2), which specifically provides that the MPSIA takes precedence over any written law that provides for the creation of a security interest in movable property, including under (i) an agreement to sell subject to retention of ownership, (ii) a hire-purchase agreement in accordance with the HPA, (iii) an outright transfer of accounts receivable, (iv) an operating lease, (v) a commercial consignment that does not secure an obligation, and (vi) any other interest that secures an obligation. Finally, paragraph (3) provides that the principles of common law, equity and the law of merchants continue to apply, but only to the extent they are not inconsistent with the MPSIA. The enactment of the MPSIA was accompanied by a consequential amendment to the Agricultural Credits Act (ACA) and a repeal of the Trades Charges Act (TCA). This approach corresponds to Article 101 of the Model Law, which invites enacting States to identify the relevant laws to be amended or repealed. The legislative drafter advised the working group that specific amendments were required only with respect to the ACA and the TCA. In contrast, the HPA, the Co-operative Societies Act and the Companies Act were indirectly (consequentially) amended on the MPSIA’s entry into force.34

1. The Companies Act Part 5.2 of the Companies Act 1994 deals with the registration of company charges in the company’s register and the Companies Registry, based on provisions in the UK Companies Act 1948. The rules on priority of charges in Zambia before the Companies Act was enacted were those of English law. Although these rules are based primarily on the order of creation, there are some quite complex exceptions, which depend on the nature of the security interest, including whether a charge is fixed or floating.35 Section 101 of the Companies Act 1994 was designed to simplify the Zambian priority by providing that the first charge to be 31 See Conclusion C2a and Liberia C5, Sierra Leone C6 and Kenya C. 32 See E. 33 IFC, ‘Zambia Expands Access to Finance with Secured Transactions Law and IFC Help’ (1 June 2016) www.ifc.org/wps/wcm/connect/news_ext_content/ifc_external_corporate_site/news+and+events/news/ za_ifc_zambia_secured_transaction_law_wcm. 34 For discussion of different methods of repeal and amendment, see Conclusion C6a. 35 See H Beale et al, The Law of Security and Title-Based Financing, 3rd edn (Oxford, Oxford University Press, 2018) ch 15.

170 Zambia lodged with the registrar had priority. This applied regardless of the nature of the charges (ie whether they were fixed or floating)36 provided that they fell within the scope of the registration requirements in section 99(1) of the Act.37 The provisions of the Companies Act (specifically Part V)38 on the registration of charges over movable assets of companies have been superseded by the MPSIA. PACRA, the entity empowered to establish and operate the Collateral Registry, also operates the Companies Registry and has been instrumental in the establishment of the new Collateral Registry.39 The MPSIA does not affect the provisions of the Companies Act on the registration of charges with respect to immovable property. The Companies Act is presently under review, with an eye to enacting a new law in the near future.

2. Agricultural Credits Act The ACA was enacted in 2010. In the same way as under previous versions of the ACA,40 it allowed farmers to use their farming stock and other agricultural assets as collateral for loans under floating or fixed charges.41 Farming stock is defined to include livestock, poultry, fish stock, standing and cut timber, fertiliser, but also agricultural vehicles and machinery.42 Under section 133, the floating charge had the same effect as a charge created by a company under the Companies Act. Under the ACA, the priority of charges was determined solely by the time of registration, irrespective of their nature, whether fixed or floating.43 This priority rule is bolstered by section 15(3), which provides that a fixed charge has no effect as regards property comprised in the floating charge as long the floating charge remained in effect. Once the floating charge has been discharged by payment of the secured obligation or by enforcement, any surplus value in the asset falls under the fixed charge.44 All agricultural charges were required to be registered within 30 days of their creation.45 The ACA provided for the establishment of a registry for the registration of agricultural charges and appointed the Registrar of the Lands and Deeds Registry as the Registrar of agricultural charges.46 In addition to charges of agricultural assets, the ACA sets out the rules for the warehousing of agricultural commodities, including the licensing of warehouses, and the issuance and transfer of warehouse receipts, as well as their use as collateral for loans. It authorises the issuance and transfer of warehouse receipts in electronic form, and the establishment of an

36 Re Sapco Fibreboard & Wood Products Ltd (in receivership) – An Application for Directions under s 113 of Cap 388 2001/HPC/0225. 37 Zambia National Commercial Bank Ltd v Mwila (SCZ appeal no 94, 11 November 1999 and 6 April 2000) per Ngulube CJ. See also AM Ngenda, ‘Registration and Ranking of Company Charges in Zambia’ (2009) 53(2) Journal of African Law 276. 38 Companies Act of 1994. 39 See ‘Patents and Companies Registration Agency’ www.pacra.org.zm/#/ and B above. 40 Agricultural Credits Act (ACA) 1962 and 1995, clearly based on the UK Agricultural Credits Act 1928. 41 ACA, s 2(1) defines agricultural charge as ‘a charge, lien or assignment.’ 42 ibid s 2(1). 43 ibid s 15(2), based on the UK Agricultural Credits Act 1928, s 8. 44 Beale et al (n 34) 69. 45 ACA, s 16(1). The failure to register would render the charge void, but an application to the Registrar for leave to register out of time was possible (s 16(2)). 46 See the definition of ‘registry’ and ‘registrar’ in ACA, s 2(1).

Pre-MPSIA Framework 171 exchange for their trading.47 A warehouse receipts system was established by the Zambian Commodity Exchange (ZAMACE) in 2015.48 Warehouse receipts issued by ZAMACE can be pledged ‘as collateral to access funds from a ZAMACE participating Bank’.49 Sections 65–71 confer on the warehouse operator a lien on agricultural commodities stored in the warehouse for any unpaid charges and fees. The MPSIA was accompanied by a consequential amendment to the ACA, providing that the MPSIA applied to any security interest over farming stock and agricultural commodities. The amending legislation50 replaced the term ‘agricultural charge’ wherever it appeared in the ACA with the term ‘security interest’51 and introduced a definition of ‘security interest’ in the ACA that was identical to that in the MPSIA52 as well as a section replacing the registration requirements and priority rules in the ACA with those in the MPSIA.53 A section stating that the MPSIA applied to any security interest over farming stock and agricultural commodities was also introduced.54 Thus the MPSIA applies to what were agricultural fixed and floating charges under the ACA. As a result, farmers are able to encumber farming stock and other agricultural assets under a security interest governed by the MPSIA and registrable in the Collateral Registry. The ACA provisions on warehouse receipts were unaffected by the consequential amendment, as they supplement the MPSIA’s provisions on security interests in negotiable documents, which was defined to include warehouse receipts.55

3. Trades Charges Act The MPSIA resulted in the repeal of the TCA, enacted in 1973.56 The main purpose of the TCA was to address the shortcomings of the Bills of Sale Act governing security interests in goods created by individuals and unincorporated businesses.57 Due to the strict formalities and severe penalties for the failure to strictly satisfy the bills of sale requirements, which included required wording for the creation of a bill of sale, Zambian banks were reluctant to accept bills of sale.58 Also, the Bills of Sale Act did not permit a creditor to take a floating charge over the assets of the borrower. The TCA authorised persons licensed under the Trades Licensing Act to encumber their property as security for loans given under the TCA by banks, financial institutions and parastatal bodies. Accordingly, the TCA’s scope 47 ibid s 37(3). ZAMACE was tasked with the establishment of a commodity exchange. See www.zamace.co.zm/ content/about-us-0. 48 USAID, ‘Warehouse Receipts System in Zambia to be Launched for 2015 Harvest Season’, www.satradehub. org/food-safety-and-production/192-warehouse-receipts-system-in-zambia-to-be-launched-for-2015-harvestseason-warehouse-receipts-system-in-zambia-to-be-launched-for-2015-harvest-season. 49 ZAMACE, ‘The Smallholder Farmer Participation Warehouse Receipt System’, 3–4, www.zamace.co.zm/ download/file/fid/166. 50 An Act to Amend the ACA 2010. 51 ibid s 3(a). 52 ibid s 2 amending ACA, s 2. 53 ibid ss 5 and 6 amending ACA, ss 16 and 17. 54 ibid s 3(b) introducing a new s 2(A) into the ACA. 55 MPSIA, s 2(1). 56 An Act to Repeal the Trades Charges Act 1973. 57 SV Mubako, ‘The Trades Charges Act 1973’ (1973) 5 Zambia Law Journal 163. See ch 1, E1a. 58 ibid 165.

172 Zambia was confined to qualified debtors and secured creditors.59 Under section 3(1) a qualified secured creditor could take a floating charge on the stock-in-trade and other movable property of the trader, whether presently owned or acquired thereafter,60 and under section 3(2) the trader could create a fixed charge over property that it owned at the time of creation of the charge. The TCA also created a Register of Trade Charges. However, the Office of the Registrar of Trade Charges, under the Ministry of Commerce, was not functional.61 As a result, trade charges could not be registered. By virtue of section 5(4) of the TCA, a registered charge has priority over all subsequent rights acquired in the charged property except for sales of inventory in the ordinary course of business or where there is a written subordination agreement in favour of the junior secured creditor; in relation to consecutive security interests this would result in the first to register having priority, as under section 101 of the Companies Act. Section 4(5) of the TCA provided that registration was actual notice of the charge ‘for all purposes connected with the secured property’. On the MPSIA’s entry into force, the TCA was repealed.62

4. Hire-Purchase Act The HPA of 1957 applied to hire-purchase agreements, that is, instalment sale on retention of title terms, hire contracts including an option to purchase or to continue hiring at a nominal rate, and any other contract with similar effect instalments’.63 The fact that the HPA was designed to apply to the acquisition of consumer goods and low-value equipment influenced its underlying approach, particularly in the context of remedies.64 The HPA did not apply to hire-purchase agreements where the purchase price exceeded 3,000 kwacha (US$232), an amount that has not been adjusted to reflect inflation.65 Under the HPA, the ownership rights of sellers under hire-purchase agreements were fully effective against the purchaser and third parties without any form of registration.66 Reflecting its consumer protection purpose, the HPA included various terms which had to be included in the hire-purchase agreement,67 including details of the price and the interest charged, and 59 The TCA scope became uncertain after the repeal of the Trades Licensing Act in 2011. 60 cf the position under the Bills of Sale Acts, see ch 1, E1a. 61 C Mapani and L Musonda, ‘The Zambian Movable Property (Security Interest) Act No 3 of 2016: A Stepby-Step Explanatory Guide to the Act’ (PACRA) 1, www.pacra.org.zm/crs2/docs/Guide%20to%20the%20 Movable%20Property%20(Security%20Interest)%20Act%20No.%203%20of%202016.pdf. 62 Trade Charges (Repeal) Act 2016. 63 The HP Act, ch 399 of the Laws of Zambia, s 3(1)(a). 64 Despite this focus, the HPA did not appear to be limited to unincorporated debtors. 65 HPA, s 4 of the provides: ‘Except for the provisions of sections five, twenty-three and twenty-four, which shall apply to every agreement or, as the case may be, to the parties to every agreement, the provisions of this Part shall not apply to an agreement under which the purchase price exceeds the sum of three thousand kwacha.’ The part it refers to is Part II of the HPA titled ‘General Provisions Relating to Agreements’. Part II contains 20 of the 29 sections of the HPA, including, inter alia, provisions on remedies, disposal of the goods, passage of ownership, rights of the parties, and conditions and warranties. 66 Under HPA, s 17 ownership is passed to the purchaser upon payment of all sums payable under the agreement. This appears to be irrespective of the detailed terms of the agreement, so long as it fell within the definition of ‘hire-purchase’ agreement in s 3(1), which included agreements which did not provide for the passing of ownership (s 3(1)(b)). 67 cf UK Consumer Credit Act 1974, s 60 and Consumer Credit (Agreements) Regulations 2010/1014, sch 1. Other consumer protection measures are included in HPA, ss 8 and 9.

Structure of the MPSIA 173 also a description of the asset in a manner sufficient to identify it – a description standard similar to modern secured transactions laws.68 Again, reflecting its consumer protection purpose, the HPA circumscribed the remedies of the seller in various aspects. For example, it invalidated clauses in hire-purchase agreements empowering the seller to enter onto the purchaser’s premises to repossess the asset, it allowed the purchaser to cure the default and reinstate the agreement within 21 days after the seller repossessed the asset, it stipulated that the seller could not repossess the asset where the purchaser has paid more than 50 per cent of the purchase price (the seller had to request the court to appoint a third party to sell the asset), and it rendered the purchaser’s waiver of any right ineffective.69 The HPA was superseded by the MPSIA.70

5. Other Relevant Legislation Prior to the MPSIA’s entry into force, several other statutes governed aspects of secured lending in Zambia. One such statute was the Co-operative Societies Act of 1998, which provided, in its Part XVI, for the creation of charges by co-operative societies. Such charges had to be registered with the Registrar of Co-operative Societies within 30 days of their creation.71 The provisions on the creation and registration of charges, as well as any matters incidental thereto, have been superseded by the MPSIA.72 Stamp duties are inapplicable to secured transactions in Zambia,73 but they may be required for various documents and transactions at nominal or ad valorem rates.74

D. Structure of the MPSIA The MPSIA repealed or superseded the secured transactions framework examined in the preceding section and contains all the elements of a modern secured transactions law, which are included in the Model Law.75 Although the Bill was drafted to follow the organisation of the Malawi and New Zealand PPSAs, its structure was changed because the Zambian drafting style standards dictate that the rules governing the body tasked with the implementation of the law (in this case, PACRA) must be included in the first sections of the legislation. The MPSIA is divided into seven Parts that in total contain 101 sections. Unlike the Model Law, it does not include a standalone Part on rights and obligations of the parties and third-party 68 HPA, s 7(1)(b). 69 See ss 8(1)(a) (cf UK Consumer Credit Act 1974, s 92), 15(1), 20(1) (cf UK Consumer Credit Act 1974, s 90), and 22 (cf UK Consumer Credit Act 1974, s 173). 70 MPSIA, s 6(2). 71 Co-operative Societies Act, CAP 397, s 129(1). 72 MPSIA, s 6(1)–(2). 73 See PWC, ‘Zambia: Corporate – Other Taxes’ (30 January 2019) www.taxsummaries.pwc.com/ID/ Zambia-Corporate-Other-taxes; B Chanda and D Tembo, ‘Special Report on Investment in Africa 2018: Country Chapter for Zambia’ African Law & Business (2018) www.africanlawbusiness.com/specialreport/zambia. 74 Africa Tax and Legal Services Division, Guide to Fiscal Information: Key Economies in Africa 2018 (Johannesburg, Deloitte, 2018) 359; C Adams, P Collier and M Gondwe (eds), Zambia: Building Prosperity from Resource Wealth (Oxford, Oxford University Press, 2014) 174 and 176. 75 See ch 2, A2.

174 Zambia obligors; rather, a number of provisions governing rights and obligations of the parties are scattered across the various Parts.76 Part II, which deals with registration, is supplemented by the Registry Regulations. Sections 11(5), 32(1) and 100 provide three separate bases for the Minister to issue regulations.77 The authorisation given to the Minister under section 100 includes the power to set the fees for various registry services. The following subsections analyse the fundamental provisions of the seven Parts and identify those that deviate in substance from the Model Law rules.

1. Preliminary Provisions The first Part of the MPSIA, titled ‘Preliminary Provisions’, includes six sections that cover a variety of topics, such as definitions, scope, conflict of laws rules, supremacy of the MPSIA78 and the effect of security interests in bankruptcy. The definitions follow the PPSA terminology (eg ‘account debtor’ and ‘debtor’) rather than the Model Law nomenclature (eg ‘debtor of the receivable’ and ‘grantor’),79 but without any material difference. The MPSIA also incorporates some definitions from Article 2 of the Model Law (eg ‘competing claimant’ and ‘financial contract’) and the Legislative Guide (eg ‘financial lease’).80 The MPSIA applies to security interests in movable property.81 It defines ‘security interest’ as a property right in movable property that is created by agreement and that secures performance of an obligation, including any type of charge over movable property, chattel mortgage, consensual lien, retention of title in movable property, right under a financial or operating lease, right of a transferee of accounts receivables, and right of a commercial consignor.82

The MPSIA does not apply to every operating lease, as this would have captured transactions such as short-term rentals of equipment and motor vehicles and forced compliance with the MPSIA registration rules without providing much of a benefit to these transactions. An operating lease is defined to cover only leasing agreements with a duration exceeding one year, in line with the PPSA models.83 Regarding financial leases, reflecting the corresponding definition from the Legislative Guide, the MPSIA applies irrespective of the duration of the agreement, and includes hire-purchase agreements. Following its PPSA forbearers, the MPSIA’s scope of application is broader than the Model Law, extending to transactions that do not secure obligations, such as operating 76 For instance, MPSIA, s 66 deals with the rights of account debtors against assignees. 77 See D3 below. MPSIA, s 11(5) requires the Minister to issue regulations that are, however, limited to the storage and indexing of information in the Collateral Registry. S 32(1) provides for the power to issue regulations for, or with respect to any matter under Part II, and s 100 provides the power to issue regulations with respect to anything under the Act. 78 MPSIA, s 6. This section provides that the provisions of the MPSIA prevail over any other inconsistent written law relating to security interests in movable property or the creation and maintenance of a registry for such interests. It is for this reason that it is stated above that the previous legislation has been superseded by the MPSIA. 79 See ch 1, D. 80 LG, Introduction para 20 (where it is qualified as a ‘term used only in the context of the non-unitary approach to acquisition financing’). 81 MPSIA, s 3. 82 MPSIA, s 2(1). 83 ibid s 2(1). Compare with the definition of ‘lease for a term of more than one year’ in Saskatchewan PPSA, s 2(1)(y) and New Zealand PPSA, s 16(1).

Structure of the MPSIA 175 leases, commercial consignments and outright assignment of receivables. Section 3(1)(a) builds on the definition of security interest, and expressly provides that the MPSIA also applies to interests which would previously have been seen as security interests, such as a charge, lien, pledge, security trust deed and trust receipt, as well as those which would not have been considered security interests, such as financial leases and hire-purchase agreements. Arguably, these latter types of transactions would fall under the MPSIA even if they were not explicitly mentioned, since its functional definition of security interest is broad enough to cover these functional equivalents. However, the working group felt that for the avoidance of doubt and to reduce the potential risk of litigation, it would be useful to explicitly list the types of transactions to which the MPSIA applies. The PPSAs adopt the same drafting technique.84 The MPSIA carved out interests in immovable property, ship mortgages, interests in aircraft objects and pledges of intermediated securities from the scope of its application, in a way similar to Article 1(3) of the Model Law.85 However, it does apply to investment securities, which are defined by a reference to the definition in the Securities Act, which is limited to non-intermediated securities.86 As in Article 1(5) of the Model Law, the MPSIA does not affect the consumer rights provided under the Competition and Consumer Protection Act of 2010.87 In addition to consensual security interests,88 the MPSIA applies to interests created by judgments.89 Only the registration, perfection and priority rules of the MPSIA apply to interests created by judgments, since their creation and enforcement remains governed by other applicable laws, such as the Civil Courts (Attachment of Debts) Act. In contrast, the registration and third-party effectiveness rules of the Model Law do not apply to such interests; the Model Law only covers the priority aspect of a competition between ordinary/ acquisition security rights and rights of judgment creditors.90 The enacting State is required to complete Article 37(1) by setting out what steps must be taken under its law by a judgment creditor for it to acquire rights in encumbered assets, such as seizure of the grantor’s assets or service of a garnishment order on a person against whom the grantor has a claim for payment of money, or the registration of a financing statement.91 Under the MPSIA,92 the steps that must be taken are seizure of the asset by or on behalf of an execution creditor, the registration of a financing statement by a judgment creditor, or the service of the court order on the person holding property for, or on behalf of, the debtor. Contemplating the registration of notices with respect to such rights has become a standard feature of newly reformed secured transactions laws in Africa. This approach provides more certainty and transparency in environments previously fraught with the risk that a judgment creditor may unexpectedly seize collateral. 84 Compare with Saskatchewan PPSA, s 3(1). 85 ibid s 3(3). 86 This is narrower than the scope of the PPSAs and the US UCC Art 9 which cover both intermediated and nonintermediated securities, but is the same as the ML. Intermediated securities were excluded because holding and transfer of these is governed by a specific regulation issued by BoZ. 87 MPSIA, s 3(1)(c). 88 See definition of ‘security interest’ in s 2(1) and also s 3(1). 89 ibid s 3(2). 90 ML, Arts 37 and 40. 91 Guide to Enactment para 317. 92 MPSIA, s 69(2).

176 Zambia Section 4 provides conflict of laws rules, but they are less comprehensive than those contained in Chapter VIII of the Model Law. Like the Model Law, it defers to the contractual freedom of the parties to designate the law governing their mutual rights and obligations, but prescribes mandatory rules that determine the law applicable to the creation, perfection and priority of a security interest. Following the approach of the Model Law, it selects the location of tangible assets and the location of the debtor (for intangible assets) as the connecting factors for its application.93 Likewise, following Article 94 of the Model Law, the commencement of bankruptcy or insolvency proceedings does not override the application of the rules that determine the law applicable to the creation, perfection and priority of security interests.

2. Collateral Registry Office and Collateral Registry Part II of the MPSIA provides comprehensive rules on the establishment and operation of the Collateral Registry housed in the Collateral Registry Office managed by PACRA. PACRA is a statutory body under the Ministry of Commerce, Trade and Industry that possesses statutory powers prescribed in the Patents and Companies Registry Authority Act.94 Zambia’s statutory framework required the formation of a designated office under which the registry would operate. The Collateral Registry is an electronic notice-registration system where the registry staff are statutorily barred from providing legal advice to users, altering or removing any information from registered financing statements, or conducting any scrutiny of the information provided in financing statements.95 With respect to the information to be provided in a financing statement, section 13 follows Article 8 of the Model Registry Provisions, requiring identification of the debtor and secured creditor by their names and identification numbers, their addresses, a description of the collateral, as well as an indication of a statement of the maximum amount for which the security interest may be enforced.96 Under the Model Registry Provisions, only the name of the grantor is the searchable criterion, whereas under the MPSIA it is the name plus a unique number, both of which must be entered correctly.97 Concerning the duration of a financing statement, the MPSIA allows registrants to select an appropriate period, which must not exceed five years.98 This approach corresponds to Option C of Article 14 of the Model Registry Provisions. Under section 15, a financing statement becomes effective from the time it is entered into the Collateral Registry, whereas Article 13(1) of the Model Registry Provisions stipulates that the registration of a notice is effective from the time it becomes searchable. Although the formulation of the provisions differ, their effect is the same given the design 93 MPSIA, s 4(1). However, with respect to security interests in bank accounts the connecting factor is the place of business of the bank or other financial institution that maintains the relevant bank account. 94 See Act No 15 of 2010. See B above. 95 MPSIA, ss 8(3) and 12(5). 96 This is an optional element that enacting States may implement or omit from their enactment of the ML. See MRP, Art 8(5). 97 MPSIA, s 17(1)(a). 98 Validity of the financing statement may be extended, for a term not to exceed five years, before its expiration by the registration of an amendment statement. See MPSIA, s 19.

Structure of the MPSIA 177 and operation of the Collateral Registry, which makes financing statements searchable on their entry into the database. Therefore, section 15 does not make financing statements effective on their lodgement and prior to their public searchability, as remains the case in the filing systems under UCC Article 9. Section 17 addresses the effect of errors in the information in registered financing statements, adopting the ‘seriously misleading’ test of Article 24(4) of the Model Registry Provisions. However, section 18 adds that the validity of a registration is not affected unless the omission, error, irregularity or defect is of a ‘material nature’ as well as seriously misleading (as specified in section 17). The reference to ‘material nature’ seems to impose an additional test that an omission, error, irregularity or defect must satisfy before it renders the registration invalid. The models that Zambia followed do not provide a similar two-prong test, and it is difficult to predict how the Zambian courts might apply this test.99 Essentially, it is difficult to identify the difference between the tests of ‘seriously misleading’ and ‘material nature’, and, despite the apparent cumulative nature of the two requirements, the Zambian courts should interpret them as a single test. Under sections 17 and 18, the registration of a financing statement is invalid (the MPSIA refers to the notion of ‘(in)validity’ rather than ‘(in)effectiveness’ in the Model Registry Provisions, but the effect is the same) if the financing statement contains an error or omission in the name or identification number of the debtor or the serial number of the collateral.100 The Model Registry Provisions do not contemplate indexing and searching according to serial numbers of encumbered assets, but the MPSIA follows the approach of its PPSA models with respect to serial number indexing, in that registration in relation to the particular collateral is invalid if there is an error in the serial number.101 Furthermore, both the name and identification number constitute the debtor identifier, and an error in either of them would render the financing statement unsearchable and thus invalid with respect to that debtor.102 In addition to an error (Article 24 of the Model Registry Provisions refers only to ‘errors’) section 17 refers to an ‘omission’, ‘defect’ and ‘irregularity’. These references were inserted from the PPSA models, but it would be difficult to envisage a situation where the notion of error would not cover omissions, defects and irregularities.103 Section 17(5) identifies the information in relation to which omissions, errors, defects or irregularities will have no effect on the validity of the registration, whether or not the information is seriously misleading, including the name, identification number and address of the secured creditor, address of the debtor, and any other information voluntarily entered by the secured creditor. An incorrect description of collateral that is part of other, adequately described, collateral will also not render the registration invalid. Subsection 17(6) contains an unusual provision that prevents the registrant from recouping the registration fee if the registration is invalid. Although nothing in the MPSIA would entitle the registrant to a fee refund under these circumstances, the working group chose to make it explicit to avoid having to deal with claims of unsophisticated registry users.

99 See Saskatchewan PPSA, s 43(6), which adopts the seriously misleading test. 100 The Saskatchewan PPSA, s 43 also refers to ‘(in)validity’ rather than ‘(in)effectiveness’. 101 MPSIA, ss 17(1) and (2). 102 ibid. 103 Email from Professor Roderick J Wood to Marek Dubovec, 31 October 2017, stating that there is no Canadian case-law drawing a difference among these four notions.

178 Zambia Regarding the discharge of registrations, the MPSIA follows the approach of the PPSAs in that it distinguishes between registrations of financing statements relating to consumer and non-consumer collateral.104 With respect to the former, the secured creditor must register a discharge within 14 days of the full satisfaction of the secured obligation,105 whereas for non-consumer collateral the debtor must request a discharge, which triggers the secured creditor’s obligation to do so within 14 days after delivery of that request.106 Fees for the various services of the Collateral Registry are regulated in the Movable Property (Security Interest) (Fees) Regulations.107 The fee for the registration of a financing statement or its amendment and discharge is approximately US$10, and the fee for searches is about US$2.108 The Fees Regulations were issued separately from the general regulations that address other aspects of the Collateral Registry.

3. Registry Regulations The Movable Property (Security Interest) (General) Regulations were gazetted on 4 November 2016.109 Given the premature entry into force of the MPSIA, the promulgation of the Registry Regulations was expedited in order to allow a fast-tracked design and implementation of the Collateral Registry. In addition to the usual operational details found in regulations of this nature (eg allocation of a registration number to every financing statement), the Regulations contain two sets of forms. The First Schedule includes the physical reproduction of fields to be completed when filing a financing statement electronically. Although such physical reproduction is not strictly necessary, given the exclusively electronic nature of accessing the Collateral Registry and the inability to register financing statements in physical form, it must be remembered that the Regulations were promulgated before the Collateral Registry ‘went live’. Local secured creditors could therefore use these forms to capture relevant information related to pre-MPSIA security interests to facilitate their transition into the MPSIA. The registration forms were based on the Registry Forms included in Annex II of the UNCITRAL Guide on the Implementation of a Security Rights Registry.110 The First Schedule also contains forms (not found in Annex II) that facilitate the performance of various registry functions. For instance, Form III entitled ‘Demand to Discharge or Amend Registered Financing Statement’ facilitates the procedure for compulsory amendment or discharge contemplated in section 23 of the MPSIA. Although this form is not to be communicated to the Collateral Registry, the drafters believed that standardising the practice would reduce the risk of disputes in connection with exercising the rights and performing the duties under this section. The First Schedule also includes forms to effect a notification of disposition and proposals to retain collateral in full satisfaction of 104 MRP, Art 20 does not draw a similar distinction. 105 MPSIA, s 22. 106 ibid s 23. 107 Statutory Instrument No 73 2016. 108 See The Movable Property (Security Interest) (Fees) Regulations, 2016, Statutory Instrument No 73 of 2016. 109 The Movable Property (Security Interest) (General) Regulations, 2016, Statutory Instrument No 77 of 2016 (Registry Regulations). 110 See ‘UNCITRAL Guide on the Implementation of a Security Rights Registry’ (Vienna, UNCITRAL, 2013).

Structure of the MPSIA 179 the secured obligation.111 Overall, the First Schedule contains three sets of forms to be communicated: (i) to the Collateral Registry; (ii) between the debtor and secured creditor in relation to a registration; and (iii) to the debtor and third parties in relation to matters other than registration. The Second Schedule contains a model security agreement. The use of this form has no legal effect since it does not constitute a safe harbour and is not prescribed by the MPSIA. Instead of being included in the Regulations, it would have been sufficient to post it on the Collateral Registry’s website as a service to potential users. The Zambian drafters were inspired by the Regulations issued under the Malawi PPSA, which also contain a model security agreement.112 The Regulations address a number of important aspects that determine the legal effect of registrations. For instance, the MPSIA requires serial number descriptions and indexing for certain goods, which are defined in Regulation 2 as ‘motor vehicle, trailer, or locomotive’. However, it does not provide for the exception from this description requirement when these types of assets are held as inventory, such as motor vehicles for sale by a car dealership. An exception of this kind is common to the PPSA regimes. The inventory financing for these types of assets is thus more cumbersome because of the requirement to identify all motor vehicles individually and to then regularly amend the registration to delete the vehicles that have been sold and add new ones acquired by the dealership. Regulations 7–9 determine the appropriate identification numbers for various types of debtors, such as domestic and foreign natural persons, domestic and foreign corporate bodies, and unincorporated bodies, implementing section 13 of the MPSIA. The Regulations also contain certain provisions that appear to be inconsistent with the MPSIA. For instance, Regulation 4(2) requires every financing statement to be electronically stamped. The stamp shall include the time and date of filing, as well as the name of the person accepting the filing. This last requirement is inconsistent with the electronic nature of the registration process since there are no Registry staff to accept individual filings. Furthermore, Regulation 4(4) states that a financing statement that does not bear an electronic stamp shall be deemed incomplete. The meaning of ‘incomplete’ is uncertain in light of the MPSIA concept of the ‘validity’ of a financing statement.113 The Regulations included forms of which the electronic equivalents are used to effectuate various registrations. The inclusion of ‘paper forms’ in regulations, also done elsewhere in Africa,114 is unnecessary and potentially confusing. Under section 80 of the MPSIA, extinguishment of a security interest triggers the duty of the secured creditor to register the extinguishment in the prescribed form. Generally, the effectiveness of a registration may be discharged by registering a discharge form under section 21 of the MPSIA, which does not specify the circumstances under which the registration may or must be discharged (this could occur on full satisfaction of the secured obligation or on disposal of the collateral upon default). This is satisfactory since, from the perspective of the Collateral Registry and searchers, the underlying reason for the discharge is immaterial. 111 Registry Regulations, Forms IX and X. 112 See M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in L Gullifer and O Akseli (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016) 186. 113 MPSIA, ss 17 and 18. 114 See Kenya C4.

180 Zambia However, rather than providing for a single discharge form, the Registry Regulations introduced a new registration function whereby the secured creditor must register a form to indicate extinguishment of a security interest: a notice of extinguishment of a security interest (Form XI). The wording of Form XI suggests that this notice is to be registered by the enforcing secured creditor who must also identify the registration numbers of any financing statements relating to subordinate security interests that have been discharged. This registration may have the effect of discharging not only the registered notice of the enforcing creditor, but also any other notices perfecting the discharged security interests. The ability of a person to affect registered notices submitted by others presents a number of risks, especially when the registered notices related to subordinate security interests also cover some collateral that has not been disposed of in an enforcement process undertaken by the senior secured creditor. Other aspects of the Regulations would also benefit from more clarity, including Recommendation 17(2), which provides for searches against various search criteria including ‘generation of statistical information’. It is difficult even to envisage how such a search function might be designed.

4. Creation The general approach of Part III is to enable the creation of a security interest by any person, whether a natural or legal entity, to secure obligations of any nature.115 The security agreement that creates a security interest must identify the parties, reasonably identify the collateral, and describe the secured obligation, including by a reference to the maximum amount for which the security interest is enforceable.116 The collateral may be described by item, kind, type, category or a statement that a security interest is taken in all present and after-acquired movable property.117 However, unlike the Model Law, section 35(3) requires collateral that is consumer goods to be described specifically.118 Furthermore, section 37 requires the further written consent of the debtor for the security interest to extend to after-acquired consumer goods. Although the MPSIA does not expressly refer to ‘attachment’, it effectively reflects this concept in section 36, which makes the effectiveness of a security interest conditional on the satisfaction of three criteria: that the debtor has rights in the collateral; that the security agreement has been concluded; and that the secured creditor has given value. The relationship between creation and effectiveness of a security interest is not clear. If the effectiveness of a security interest as between the debtor and secured creditor is conditional on the satisfaction of the aforementioned three requirements, then there does not appear to be a need to provide for the creation of a security interest.119 However, section 52 in providing that the priority between unperfected security interests is determined by the time of their 115 MPSIA, s 33. 116 ibid s 34(1). Compare with ML, Art 6(3)(d) under which the maximum amount is an optional element that enacting States may incorporate into their domestic law. See ch 2, C1. 117 MPSIA, s 35(2). 118 Compare with Saskatchewan PPSA, s 10(3). 119 The time of creation is provided for in MPSIA, s 33(4) which requires the debtor to have rights or the power to create a security interest (in an asset) before a security interest can be created in that asset. There seems an overlap in function between ss 33 and 36.

Structure of the MPSIA 181 creation questions the effect of section 36, and perhaps renders the concept of ‘effectiveness’ in section 36 redundant since it can only refer to effectiveness between the parties to the security agreement. Effectiveness against third parties (perfection, in the terminology of the MPSIA) is not achieved until the criteria in sections 44(1) and 44(2) are fulfilled.120 Section 39 contains provisions relating to the sufficiency of the rights of a debtor to create a security interest in specific circumstances. If the debtor is an outright transferor of receivables, it has sufficient rights to create a security interest until the transferee perfects its interest. If the debtor is a buyer or lessee under a financial or operating lease, it has sufficient rights to create a security interest whether or not the interest of the seller or lessor has been perfected. This approach differs from the Model Law under which the grantor no longer possesses any rights to the transferred receivable or may not have acquired rights in an asset subject to a financial lease, but retains/acquires the power to create a security interest that, as a matter of priority, may trump the ownership interest of the receivables transferee or financial lessor. Section 42 overrides the effect of anti-assignment clauses.121 Unlike Article 13 of the Model Law, it is not limited to so-called ‘trade receivables’, but applies to any receivable. As provided by Article 15 of the Model Law, a security interest in a bank account is effective despite any agreement between the bank and the debtor barring the latter from creating a security interest. However, despite the creation of a security interest, the bank shall not owe any duties to the secured creditor.122

5. Perfection Section 44 recognises three methods of perfection: registration of a financing statement, possession, and control of the collateral by the secured creditor. The last of these only applies if the collateral is a bank account.123 Section 44(2) also contains a drafting error. It says that ‘subsection (1) applies regardless of the order in which the steps referred to in paragraph (b) of subsection (1) have occurred’. It is clearly directed at the point that the order in which creation (referred to in (1)(a)) and the relevant one of the three methods (listed in 1(b)) have occurred, which is an important point. Registration (for example) can occur before creation, as can either of the other methods. However, the reference in section 44(2) to ‘paragraph (b) of subsection (1)’ should have been a reference to ‘subsection (1)’. Unlike Article 24 of the Model Law, there is no exception for acquisition security interests in consumer goods: like all other security interests, these must be perfected by registration or possession.124 The rules on possession and control generally follow the Model Law.125 Control of a bank account under the MPSIA arises either where the parties enter into a control agreement, defined in section 2(1) or automatically (upon the creation of a security interest) if the secured creditor is the bank that maintains the bank account. The working 120 For further discussion about ‘creation’ under the ML and ‘attachment’ under the PPSAs, see ch 2, C. 121 See ch 1, E4a. 122 MPSIA, s 42(3). 123 Compare with ML, Art 27 that provides for control with respect to security rights in non-intermediated securities, and the PPSAs some of which also include provisions for control of intermediated securities. 124 See ch 2, D. 125 ibid.

182 Zambia group was not aware of any practice of creditors becoming account holders on bank accounts of their debtors held with a third-party bank and did not see the utility of recognising this method as an additional form of control that would perfect a security interest.126 The MPSIA extends the perfection of a security interest to one arising in proceeds127 without any further action if the proceeds are money, accounts receivable, negotiable instruments, securities or funds credited to a bank account, or if the proceeds are assets covered by the description of the collateral in a registration.128 A security interest which arises in other types of proceeds is temporarily perfected for 21 days from when the proceeds arose, and the secured creditor can continue this perfection of its security interest if it perfects within that period.129 The MPSIA incorporates provisions on continuous perfection in the event of a change of the method of perfection130 or transfer of the collateral without the authorisation of the secured creditor,131 and extends the perfection of a security interest in a negotiable document to the goods covered by that negotiable document, in accordance with the Model Law.132 Section 51 provides a special rule (not found in the Model Law) applicable to the perfection of security interests in farm products. Section 51(1) does not squarely fit within the perfection Part of the MPSIA because it is a rule of creation, which provides that a security interest may be created in farm products irrespective of whether it is growing, grown, harvested, stored or kept on any land or premises. This subsection does not add much to what is already stated in the definition of farm product, but emphasises that a security interest may be created and continues to exist irrespective of the physical association of the farm product to the land. A rule of this nature may be useful in those legal systems where growing crops, and even livestock, are considered part of the land to clearly signal a break from the past. Section 51(2) provides that any transfer of the land with which the farm product is physically associated does not affect the perfection of a security interest.

6. Priority Part V of the MPSIA sets out a comprehensive schema of priority rules that resolve a variety of conflicts according to policies and approaches consistent with Chapter V of the Model Law. The fundamental priority rule of ‘first-to-perfect or first-to-file’133 is stated in section 52(b). Section 52 also sets out the residual priority rules for conflicts between a perfected and unperfected security interest. Further, it stipulates that in conflicts between two unperfected security interests, the time of creation determines priority. Sections 53–56 and 61 contain rules on (i) the priority of security interests in proceeds; (ii) the effect of a transfer of a security interest; (iii) agreed subordination of priority; (iv) priority with respect to future advances; and (v) the effect (on priority) of commingling goods subject to 126 Compare with ML, Art 25(c). 127 A security interest in an asset automatically continues in the proceeds of that asset, whether or not the security agreement contains a description of the proceeds, see MPSIA, s 38. 128 MPSIA, s 46. 129 MPSIA, s 47. 130 MPSIA, s 45. 131 MPSIA, s 48. 132 MPSIA, s 49. 133 See ch 2, F1.

Structure of the MPSIA 183 a security interest in a mass or product. Other than some drafting differences, the policies and substance of these rules match those of the comparable provisions of the Model Law. These temporal priority rules are supplemented by a set of exceptions, which, in certain circumstances, prioritise a later-in-time perfected security interest; for instance, purchase money security interests (PMSIs) in goods and security interests in bank accounts perfected by control. Under section 62, a bank’s security interest perfected automatically by control134 has priority over the security interest of a third party perfected by a control agreement. Both types of control would trump a security interest in a bank account perfected by registration, irrespective of the respective times of perfection.135 Similarly, PMSIs may attain priority over earlier-perfected security interests which extend to after-acquired collateral.136 The working group did not feel that it would be commercially practicable to recognise PMSIs in intellectual property rights or licences thereof, as contemplated in the Model Law.137 The MPSIA distinguishes between the requirements for the priority of a PMSI in inventory,138 as opposed to other types of goods, imitating Article 38 of the Model Law. For the priority of a PMSI in inventory, section 57(2) implements Option A of Article 38 of the Model Law, which requires the satisfaction of two conditions prior to the debtor’s receipt of the inventory: first, third-party effectiveness of the PMSI before the debtor takes possession of the collateral; and, second, notification of the holders of competing security interests in the inventory. However, unlike the Model Law, section 57(2) does not require the MPSI creditor to inform a competing secured creditor prior to delivery of the inventory to the debtor, which defeats the purpose of requiring such notifications to alert the competing secured creditor before it makes an advance on the inventory subject to a PMSI. For a PMSI in non-inventory collateral, the MPSIA departs from the Model Law under which the PMSI holder is given a grace period (typically 20 days) to make its PMSI effective against third parties after delivery of the collateral to the debtor. Section 57(1) of the MPSIA does not accord any such grace period, so that, in relation to non-inventory collateral, a secured creditor only obtains PMSI status if it perfects before the debtor receives the collateral. This may not be a concern if the collateral is a large piece of equipment sold or financed by a relatively sophisticated party, but it may be a concern for credit sales of low-value consumer goods by ordinary retail stores, especially in the absence of automatic perfection for such PMSIs. However, the working group determined that such secured financing did not exist in Zambia, and if it emerged, the design of the Collateral Registry (24-hour electronic access) would facilitate a seamless flow of registrations relating to PMSIs in consumer goods, equipment and farm products before the debtor takes delivery. Overall, the working group was concerned about prospective debtors being in possession of assets for which discovery of security interests might be difficult because they were (as yet) unperfected. Even though this departure from the Model Law may be viewed as causing an inconvenience to certain secured creditors, the counter-argument that this approach eliminates the risk of third parties dealing with a debtor in possession of encumbered goods (with respect to which no public notice exists) prevailed. 134 That is, by the bank that maintains the account being the secured creditor, see MPSIA, s 44(3)(a). 135 See discussion in ch 2, D. 136 See ch 2, F2. 137 For the definition of acquisition security right see ML, Art 2(b). 138 ‘Inventory’ is defined as tangible assets which are held for sale or lease in the ordinary course of business and raw materials or work in progress: MPSIA, s 2(1).

184 Zambia Section 59 provides a specific rule for a PMSI in fixtures. As explained elsewhere, the Model Law did not incorporate any recommendations of the Legislative Guide with respect to attachments to immovable property.139 The MPSIA priority rule for a PMSI in fixtures is a purely temporal rule that determines, based on the time of registration, the conflict between the holder of a PMSI in fixtures and the holder of a right in an immovable to which the fixture is physically attached. Accordingly, the PMSI in a fixture will have priority if it is perfected (by registration) before a third party acquires rights in the immovable. This rule thus forces the PMSI creditor to conduct a search of the land registry to ensure its registration in the Collateral Registry is first in time. The priority of holders of non-PMSIs in fixtures as against third parties with rights in the immovable is not covered by the MPSIA, as a result of which such secured creditors must look to land law to determine their priority. The MPSIA should have provided more certainty for creditors that finance fixtures, especially against competing claimants who acquire their rights under the land law.140 The remainder of Part VI covers the rights of competing claimants other than secured creditors, especially buyers, transferees and holders of non-consensual liens. Sections 64 and 65 protect transferees of money, funds, negotiable instruments, investment securities and negotiable documents. Transferees of money or funds take free of a security interest unless they acted in collusion with the debtor.141 This standard, reflecting the PPSAs and UCC Article 9 approaches, is different from the ‘knowledge of violation of the rights of a secured creditor’ standard under Articles 47(6) (for funds) and 48 (for money) of the Model Law. The requirements for transferees of negotiable instruments, investment securities and negotiable documents to take free of security interests, namely the giving of value, acting without knowledge that the acquisition violates the rights of the secured creditor, and taking possession, comport with the Model Law.142 Section 67 addresses the rights of buyers and lessees. In contrast to Article 34 of the Model Law, the MPSIA does not deal with the rights of licensees. Under subsection (1), a buyer or lessee takes free or is unaffected by an unperfected security interest if it takes possession of the goods and provides value. The rule applicable to buyers and lessees of goods subject to perfected security interests is found in section 48 in Part IV (perfection) which provides that the perfection continues in ‘the hands of a transferee’ unless the secured creditor authorised such transfer free of the security interest. This provision should have been formulated as a priority rule and included in Part V (priority). However, its substance would lead to the same result as Article 34(1) and (2) of the Model Law, namely that the transferee takes subject to a security interest that is effective against third parties unless authorised otherwise by the secured creditor. Subsection (2) of section 67 addresses sales and leases in the ordinary course of business in a manner that is substantively different from Article 34(4)–(5) of the Model Law. Although the knowledge standard is the same (the buyer/lessee is disqualified if it takes with knowledge that the sale or lease constitutes a breach of the security agreement), the buyer/lessee takes free of a security interest created only by the seller or lessor, as contrasted with Article 34(4)–(5) of the Model Law, which protects the buyer/lessee against any security interests in the goods, including those created

139 Conclusion

C1b. Kenya C5; Zimbabwe C1, C3, C3d; Rwanda C4; Liberia C1. 141 MPSIA, s 64. 142 For instance, for negotiable instruments, see ML, Art 46(2)(b). 140 See

Structure of the MPSIA 185 by the predecessor in title of the seller or lessor. This limited protection rule in the MPSIA is based on the PPSAs and UCC Article 9.143 In transactions covered by the MPSIA, buyers and lessees must be aware of the possibility that they acquire their rights subject to a security interest created by the seller or lessor’s predecessor in title. As a practical matter, this risk is limited to sales or leases of used goods held as inventory. With respect to the priority of security interests against two types of non-consensual liens held by judgment creditors and providers of services and materials, sections 68 and 69 follow the policies and approaches of the Model Law.144 As mentioned above, judgment creditors may ‘perfect’ their rights by registration in the Collateral Registry, the time of which will determine their priority. However, the priority of a first-in-time security interest as against a judgment lien is limited to any advances made within 30 days after the secured creditor received a notification of the judgment lien or indefinitely in a situation where the secured creditor had irrevocably undertaken to extend credit to the debtor. Both of these limitations mirror those of Article 37(2) of the Model Law.145

7. Enforcement Part VI of the MPSIA provides a comprehensive regime for the enforcement of security interests. It does not apply to rights of outright transferees of receivables and operating lessors.146 This limitation of the scope of application of this Part follows the Model Law147 and the PPSAs.148 In a manner similar to Article 72 of the Model Law, section 70 of the MPSIA authorises the use of any remedy provided for in the security agreement or granted under other law, including the appointment of a receiver, receiver and manager, or official receiver. However, the relationship of the MPSIA with these ‘other remedies’ differs from the Model Law under which non-Model Law remedies may be resorted to only if they are consistent with the Model Law. The MPSIA does not incorporate a similar qualification, so, theoretically, the secured creditor may use remedies that would otherwise be inconsistent with Part VI. This may reflect the importance of the ‘other remedies’ specified in Zambia, where long experience of the floating charge will have led to familiarity with the appointment of a receiver or receiver and manager, and appreciation of the advantages of this approach in some circumstances.149 For the MPSIA remedies, section 87(1) curbs the secured creditor’s latitude by imposing a duty to discharge any rights in accordance with reasonable commercial standards. Any remedy may be enforced extrajudicially or pursuant to a judicial process.150 Prior to the commencement of enforcement, the secured creditor must register an enforcement notice in the Collateral Registry.151 The working group felt that such registration would 143 See Saskatchewan PPSA, s 30(2) and UCC, Art 9-320(a). 144 ML, Arts 36 (enacting State to stipulate what claims had priority over a security right and up to what amount) and 37 (judgment creditors). 145 See ch 2, F4. 146 See ch 2, B2c. 147 ML, Art 1(2) (in relation to outright transfers of receivables). 148 Saskatchewan PPSA, s 55(2); Australian PPSA, s 109; New Zealand PPSA, s 105. 149 See further conclusion D6. 150 MPSIA, s 72(3). Compare with ML, Art 73(1). 151 MPSIA, s 71.

186 Zambia publicise the default of the debtor to searchers who contemplate dealing with the debtor, saving them the time and cost of due diligence beyond searching the Collateral Registry. Furthermore, it appears to be a substitute for the notice that many laws, including the Model Law, require to be given to the debtor prior to repossession of the collateral,152 since section 72(5) expressly provides that ‘a prior notice to a debtor is not required for the secured creditor to repossess the collateral’. The MPSIA does not provide for any legal effect resulting from a failure to register an enforcement notice. Given its limited function, the court should not suspend enforcement if the secured creditor has failed to register an enforcement notice or the notice contains seriously misleading errors. Any remedy given to an affected searcher should be limited to damages for any loss or damage that was reasonably foreseeable, in accordance with section 89(1). The secured creditor may exercise its remedies only if the debtor is in default or the collateral is at risk.153 Default is defined (in a way similar to the Model Law) as any event under the security agreement that gives the secured creditor the right to enforce its security interest. Section 72(2) specifies the situations where the collateral is considered to be at risk, including when the secured creditor believes that the collateral has been or will be destroyed, damaged, endangered, disassembled, removed, concealed, sold or otherwise disposed of contrary to the security agreement. A properly drafted clause in a security agreement should, in any event, identify these events as constituting default. The MPSIA expressly authorises the secured creditor to render the collateral unusable or dispose of it on the debtor’s premises.154 Although the Model Law does not expressly provide for such remedies, they may be available if the parties stipulate so in their security agreement. Section 74 of the MPSIA provides for the collection of accounts receivable, money and negotiable instruments, as well as the enforcement of security interests in bank accounts perfected by control. The policies and approaches mirror those underlying Article 82 of the Model Law, including the ability of the secured creditor to collect accounts prior to default, and the right to enforce a security interest in a bank account extrajudicially against a bank that maintains it, if the security interest has been made effective against third parties by a control agreement. The actual disposal of the collateral may be effected in the form of a sale, lease, licence, auction, public tender or other method provided for in a security agreement.155 The secured creditor may not dispose of the collateral unless it gives a notice, at least 14 days before the intended disposal, to the debtor, the secured creditors who have registered a financing statement against the same collateral, and other persons who have given the secured creditor a notice of their interest in the collateral.156 Under Article 78(8) of the Model Law, giving notice is not required if the collateral is perishable, may decline in value speedily or may be sold on a recognised market. Section 79(2) of the MPSIA provides two additional exceptions to the notice requirement. The first is where the cost of care and storage of the collateral is disproportionately high in relation to its value, and the second is where the collateral consists of inventory or farm products. Both exceptions were modelled on the PPSAs.157

152 ML,

Art 77(2)(b). s 72(1). 154 MPSIA, s 73. 155 MPSIA, s 78. 156 MPSIA, s 79. Compare with the ML, Art 78. 157 See Saskatchewan PPSA, s 59(16)(c) and New Zealand PPSA, s 114(2)(e). 153 MPSIA,

Structure of the MPSIA 187 The effect of the disposal is the extinguishment of the security interest of the enforcing creditor and any subordinate security interest, along with other subordinate liens in the collateral.158 Section 82, in prescribing the manner and order in which the proceeds must be applied, is consistent with Article 79 of the Model Law; the cost of enforcement is satisfied first, and then the security interests and other liens in their order of priority, with any surplus going to the debtor. A security interest senior to that of the enforcing secured creditor survives the disposition and remains effective against the person who acquired rights in the collateral at the disposition.159 In addition to disposal, the secured creditor may retain the collateral in satisfaction of the secured obligation. As compared to the equivalent provision of the Model Law (Article 80), section 83 of the MPSIA imposes two limitations. The first is that the remedy cannot be used with respect to consumer goods, and the second is that the secured creditor can retain the collateral only in full (rather than partial) satisfaction of the secured obligation. Parties entitled to receive a notice of disposition must be notified of the proposed retention of the collateral and be given a period of 14 days to object. On the expiration of the 14-day period, the secured creditor automatically retains the collateral in full satisfaction of the secured obligation. Sections 85 and 86 respectively provide for the right to redeem the collateral and the right to reinstate the security agreement. The right to redeem the collateral is available only to the debtor and other secured creditors, whereas article 75 of the Model Law confers the right to terminate enforcement on any person with a right in the encumbered asset, as well as the person who owes the secured obligation. The Model Law does not provide for the right to reinstate the secured obligation, which is a standard feature of the PPSAs.160

8. General Provisions and Offences This Part of the MPSIA addresses various aspects incidental to security interests, some of which have been already examined above (eg the duty to act in accordance with reasonable commercial standards).161 Equally, as earlier mentioned and unlike the Model Law, the MPSIA does not include a separate part on the rights and obligations of the parties to a security agreement and third-party obligors. Some sections of this Part address those aspects, including the right of the debtor to request information from the secured creditor with respect to the status of the indebtedness or collateral,162 following Article 56 of the Model Law. Sections 93–99 provide a set of sanctions for the fraudulent registration of financing statements, including fines and imprisonment. These sections do not apply to unauthorised registrations, a concept broader than those effectuated with a fraudulent intent. The Model Law leaves the question of sanctions for the registration of fraudulent and unauthorised notices to other laws. Sections 93–99 were modelled on the scheme of sanctions

158 MPSIA, 159 ibid.

s 80.

160 Saskatchewan

PPSA, s 62 and New Zealand PPSA, s 133. reflects the duty set out in ML, Art 4. 162 MPSIA, s 91. 161 This

188 Zambia applicable to the lodgement of documents with the Companies Registry. In addition to fines and imprisonment, the Registrar possesses the power to impose administrative penalties.163 The final section of the MPSIA refers to Schedule 1, which sets out savings and transitional provisions. Due to statutory drafting conventions, such provisions must be set out in a schedule, rather than constitute a part of the statute. The three sections of Schedule 1 address the transition in accordance with the Model Law. For example, section 2(2) designates a period of six months after the MPSIA’s entry into force, within which prior security interests must be brought into compliance with the MPSIA for their perfection and priority to continue.164

E. Implementation Challenges The MPSIA came into force on the date it was published in the Gazette, 6 April 2016, in accordance with section 10 of the Acts of Parliament Act of 1964. Zambian statutes become operative on the day of their publication in the Gazette, unless they indicate a specific commencement date or an alternative mechanism for entering into force. The secured transactions bill drafted by the working group contained a commencement clause, stating in section 1: ‘This Act may be cited as the Movable Property (Security Interests) Act, 2015 and shall come into operation on such date as the Minister may, by statutory instrument, prescribe.’ The intention was for the Minister to issue the statutory instrument of commencement once the Collateral Registry was fully functional. Without notifying the working group, this clause was removed from the bill prior to its adoption. The removal was the result of a new policy of the Ministry of Justice to remove clauses of this nature from recent bills because a high number of Acts containing such clauses are not operationalised swiftly. A blanket application of this new policy adversely affected the MPSIA, since deferral of its commencement until the Collateral Registry became fully functional was essential. Under section 44 of the MPSIA, a security interest may be perfected by registration, possession or control. The last two methods of perfection do not depend on the establishment of a registry system, and could be deployed in practice from the date of commencement of the MPSIA. Although security interests in movables excluded from the scope of the MPSIA (eg pledges of intermediated securities) or for which possession and control were applicable (and practically possible) remained unaffected, registration, which is by far the predominant method of perfection, could not be used at the time of commencement of the Act to perfect security interests. The commencement of the MPSIA also triggered the application of an important provision (in the First Schedule) inserted to facilitate a seamless transition of prior security interests, especially those perfected by registration under a prior law: the six-month window for prior security interests to be registered in the Collateral Registry or otherwise perfected.165 Consequently, not only was the Collateral Registry non-functional when the MPSIA entered into force, but PACRA and the BoZ were under 163 MPSIA, s 98(1). 164 See NL, Art 105. MPSIA, sch 1 s 2(5) defines prior security interest as ‘an interest created or provided for by an agreement or other transaction that was made or entered into before the commencement of this Act’. 165 MPSIA, sch 1 s 2(2). See D8 above.

Final Remarks 189 extreme time pressure to launch it prior to the expiration of the six-month transitional period. On 21 September 2016, PACRA issued a press release announcing the launch of the Collateral Registry, more than five months after the MPSIA’s entry into force and a mere two weeks before the expiration of the transitional period.166 During this process, PACRA worked directly with banks and lenders, collecting registration information on Excel sheets to record it in the Collateral Registry, which was not fully functional at that point. In addition to collecting the relevant registration information, PACRA had to address a number of queries from lenders as to whether certain security devices (eg cessions of rental income and fixed deposits) fall within the MPSIA and whether they should be transitioned accordingly. Ordinarily, such questions should not be answered by the registry as the answers would be tantamount to providing legal advice, from which the registry is statutorily barred from providing. However, given the short window to register financing statements relating to prior security interests, as well as the general novelty of the MPSIA, PACRA did provide general advice on the transitioning of security devices. Despite the myriad of legal issues and disputes that did and could have arisen, by working diligently with the affected stakeholders PACRA managed to ensure a transition that did not generate any litigation or loss of any security interests.

F. Final Remarks The process leading to the enactment of the MPSIA and its implementation provides a number of lessons for reformers in Africa, but also elsewhere. First, Zambian stakeholders were motivated to undertake reform by empirical evidence showing that MSMEs faced severe difficulties in accessing credit, as opposed to merely seeking an increase in the ranking under the World Bank’s Doing Business Report. Secondly, PACRA and the BoZ diligently considered the various reform approaches, including the possibility of only applying the new framework to individuals and unincorporated businesses. Although these considerations prolonged the completion of the project, they allowed the stakeholders to think through the important issues and ultimately reinforced their confidence in the Model Law/PPSA approach. Thirdly, early engagement of an experienced and respected statutory drafter ensured expeditious passage of the bill by the government and the parliament. Finally, Zambia should be a lesson as to the unnecessary risks that can arise from attempting to operationalise a reformed framework too quickly. Thanks to the efforts and hard work of those involved in this reform process, Zambia has overcome these challenges and was rewarded with the number two ranking (globally) in the Strength of Legal Rights Index of the World Bank’s Doing Business Report for 2018. The MPSIA was modelled on the Canadian, Malawian and New Zealand PPSAs, but also drew inspiration from the Legislative Guide, the Registry Guide and drafts of the Model Law available at the time of its finalisation. Although in some aspects it departs from the Model Law (eg the MPSIA applies to long-term operating leases), its fundamental approaches, including the functional approach, notice registration, and extrajudicial remedies, are consistent with the Model Law. The stakeholders also made an effort to facilitate

166 The

press release is on file with the authors.

190 Zambia the practical application of the MPSIA by including a number of forms and model agreements in the Registry Regulations. These efforts align with, and should further benefit from, the new project of UNCITRAL’s Working Group VI to develop a Practice Guide.167 In 2017, the IFC facilitated intensive trainings on asset-based lending products delivered by experienced members of the Commercial Finance Association. Presently, the BoZ is looking to review its prudential framework and guidelines for financial institutions to stimulate lending secured with movables even further.

167 See ‘Draft Practice Guide to the UNCITRAL Model Law on Secured Transactions: Annotated List of Contents’, A/CN.9/WG.VI/WP.75 (UNCITRAL, 3 October 2017).

9 Sierra Leone In November 2014, Sierra Leone enacted the Borrowers and Lenders Act (B&L Act) in order to reform its law of secured finance. In June 2019, a new Borrowers and Lenders Act (‘the new B&L Act’) entered into force. This chapter examines the law relating to secured transactions which is contained in the 2014 B&L Act, the effect of its adoption, and, in particular, the new challenges that it introduced. It concludes with a brief examination of the new B&L Act, which was enacted too late for a full discussion to be possible in this book. The focus of this chapter is therefore on the reform approach of the 2014 B&L Act and the results which flowed from it. The chapter draws lessons from that reform approach which resulted in further disintegration of the legal framework governing secured transactions in Sierra Leone. The chapter suggests ways to address the deficiencies of that approach. Unlike many other reforms discussed in this book, the 2014 B&L Act provides an additional tool to secure obligations as opposed to providing a single and integrated framework for movables financing.

A. Economic and Financing Overview Sierra Leone is a small economy with a population of 6.3 million that relies on exports of minerals as the principal source of foreign exchange. Agriculture contributes 60.7 per cent of GDP, industries contribute 6.5 per cent and services 32.9 per cent.1 All of these economic sectors suffered significant setbacks during the Ebola crisis and because of the fall in commodity prices when the economy contracted by over 20 per cent.2 The Ebola crisis also hindered implementation of legal reforms, including the B&L Act. The most common type of movable property used as collateral in Sierra Leone is motor vehicles. In practice, a lender arranges for the title document to indicate their name as a co-owner of the vehicle.3 In addition, the lender retains the title document until the loan is paid in full. The motor vehicle registry is not publicly searchable online, and interested persons must present a search request at the office of the Sierra Leone Road Safety Authority, which will conduct a search on their behalf. The government is hoping 1 CIA, ‘Africa: Sierra Leone’ (The World Factbook, 22 January 2019) www.cia.gov/library/publications/ the-world-factbook/geos/sl.html. 2 S Gbandia, ‘Sierra Leone Economy to Contract 22% on Iron Ore, Ebola’ Bloomberg Business (14 September 2015) www.bloomberg.com/news/articles/2015-09-15/sierra-leone-economy-to-contract-22-percent-on-iron-oreebola; see in general World Bank, ‘The Economic Impact of the 2014 Ebola Epidemic’ (2014). 3 Interview with the Sierra Leone Road Safety Authority (Freetown, Sierra Leone, 20 October 2016); see also Sierra Leone Road Safety Authority, ‘Vehicle Registration & Renewal’ (2019) www.slrsa.gov.sl/ vehicle-registration-renewal/.

192 Sierra Leone to encourage equipment finance with the adoption of a new leasing law. On the other side of the spectrum is the use of intangible property as collateral; such use is rare in Sierra Leone. Securities trading is conducted manually and limited to two brokerage houses that trade government bonds. The only shares that have been floated on the Stock Exchange are those issued by the Rokel Commercial Bank.4 Thus, securities are hardly ever given as collateral in Sierra Leone. Inventory and receivables financing also remains scarce. Very few businesses generate receivables, and lenders do not have sufficient expertise to provide facilities secured with inventory.

B. Legal Framework Affecting Secured Transactions The pre-B&L Act legal framework governing secured transactions consisted of a number of statutes, including the Companies Act of 2009, the Registration of Business Act of 2007, as well as subsidiary legislation for financial leasing. None of these pieces of legislation was repealed or consequently amended by the B&L Act; thus, they remained legal sources for the creation of rights in movable property. They were repealed or superseded by the new B&L Act. The relevant parts of these other statutes will now be discussed.

1. The Companies Act The Companies Act provides for the registration of companies that may either be public or private.5 Part VIII governs the issuance of debentures that may be unsecured or secured with a charge on some types of property.6 The charge may be either fixed or floating.7 A floating charge may be taken over the whole or a specified part of a company’s undertaking and assets, both present and future. On crystallisation, the floating charge becomes fixed.8 Section 151 provides for the general priority of a fixed charge over a floating charge, unless the former prohibits the creation of any later charge and the holder of the fixed charge took with actual knowledge of that prohibition. Such a prohibition can be made public on the register, by the registrant ticking a box for that purpose.9 Furthermore, the priority of a floating charge is subject to preferential claims enumerated in Part XVI,10 which in winding-up include taxes, social security deductions, and all wages and salaries due by the company 12 months before the date of the winding-up order.11

4 Interview with the Sierra Leone Stock Exchange (Freetown, Sierra Leone, 19 October 2016). 5 The Companies Act 2009, s 18. 6 ibid s 145(1). 7 ibid s 145(2). See ch 1, E1b. 8 ibid s 150. 9 Corporate Affairs Commission Sierra Leone, ‘Particulars of Charge: Form 170’ item 6. Form 170 is available at www.cac.gov.sl/PARTICULARS%20OF%20CHARGE.pdf. 10 The Companies Act 2009, s 154(1). 11 ibid s 438(1). See ch 1, E1b.

Legal Framework Affecting Secured Transactions 193 Section 170 requires every charge that is created by a company and specifically listed in this section to be registered with the Corporate Affairs Commission.12 The registration regime is very similar to the pre-2013 registration regime under English law.13 The failure to deliver the prescribed particulars with the charge instrument within 21 days after its creation renders the charge void as against the liquidator and other creditors of the company.14 The list of registrable charges includes a charge for the purposes of securing any issue of debentures, a charge on book debts, a fixed charge over tangible property,15 a floating charge on the undertaking or property of the company as well as a charge on intellectual property rights.16 The Commission has a duty to indicate in the register the date of creation, the amount secured by the charge,17 brief particulars of the property and the identity of a charge holder.18 Although the company that created a charge has an obligation to submit the particulars and the charge instrument for registration (in common with all regimes based originally on English law),19 any other person interested therein may also do so.20 The company and every officer responsible for the registration of charges will be guilty of an offence and liable to a fine for the failure to register a charge with the Commission within the prescribed period of time.21 Furthermore, the failure of an officer to knowingly or wilfully register a charge in the company’s own register of charges will be an offence that is also punishable by a fine.22 The Regulations to the B&L Act prescribe sanctions for the failure to perform obligations under the Regulations.23 These sanctions may be imposed primarily on lenders who fail to amend or discharge a registered notice where the borrower is entitled to such an amendment or discharge. This regime contrasts with that under the Companies Act where it is the borrower who will be liable if it fails to ensure the registration of a charge. Section 181 of the Companies Act confers statutory remedies on charge holders which may be exercised in addition to those that may have been provided for in the charge instrument. Registration documents may be delivered to the Commission either physically or by email.24

12 ibid s 70(1); see Corporate Affairs Commission Sierra Leone, ‘Our Services: General Registration & Incorporation’ (2015) www.cac.gov.sl/gen-incorporation.html. 13 See L Gullifer and M Raczynska, ‘The English Law of Personal Property Security: Under-Reformed?’ in L Gullifer and O Akseli (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016) ch 12. 14 The Companies Act 2009, s 170(1); see also Corporate Affairs Commission Sierra Leone, ‘Particulars of Charge: Form 170’ (n 9). 15 This is referred to in s 170(2)(c) as ‘a charge created or evidenced by an instrument which if executed by an individual would require registration as a bill of sale’. See ch 1, E1c. 16 The full list is included in The Companies Act 2009, s 170(2). 17 Any charge in excess of the stated amount is void. The Companies Act 2009, s 175. 18 ibid s 171(1)(b). 19 See Kenya B4 and ch 1, E1b. 20 The Companies Act 2009, s 172(1). 21 ibid s 172(3). 22 ibid s 164(2). 23 The Borrowers and Lenders (Collateral Registry) Regulations 2016, reg 28 provides that: ‘Any person who fails to meet his obligations under these Regulations commits an offence and shall be liable to a fine not exceeding ten million Leones.’ 24 Corporate Affairs Commission Sierra Leone, ‘Our Services: General Registration & Incorporation’ (n 12).

194 Sierra Leone

2. The Registration of Business Act of 2007 The Companies Act is not the only statute under which businesses may be formally registered.25 The Registration of Business Act provides the framework for the registration of proprietors26 in the Office of the Administrator & Registrar General (OARG).27 The registration process is manual, requiring submission of a physical form. Upon receipt, it is signed by a registrar, then submitted to another unit which creates an entry in the index, and finally passed on to a unit which scans the application documents and stores them electronically. Registration forms include a miscellaneous field in which parties often indicate that a charge has been created over some property of the business, movable or immovable.28 Section 13 of the Registration of Business Act prescribes the information that must be provided in an application but does not list charges or other encumbrances among them. Since it does not provide for the registration of any encumbrances over property, registrations that identify charges in the miscellaneous field do not seem to have any legal effect. It is noteworthy that the World Bank Group in one of its reports indicates that a leasing agreement is not enforceable unless it has been registered with the OARG.29

3. Leasing Guidelines Leasing is governed by the Operating Guidelines for Finance Lease Institutions issued by the Banking Supervision Department of the Bank of Sierra Leone in 2011, under section 53 of the Other Financial Services Act of 2001.30 However, the Guidelines only set out the regulatory framework for the establishment and operation of a leasing company, including the minimum capital and other prudential requirements, and do not cover other legal aspects of leasing. Commercial banks may not provide leasing directly, but only through subsidiaries.31 Section 15 of the 2001 Act prescribes the information that must be provided in a leasing agreement, including the identification of the lessor and lessee, a description of the asset, the estimated price of the asset, the total rentals to be paid by the lessee, the residual value and any option to purchase the asset. The effect of leases on third parties is governed by the B&L Act, under which they constitute security interests.32

25 The website of the Office of the Administrator & Registrar General (OARG) indicates that interested persons may register three different forms of businesses, including: (i) sole proprietorship; (ii) partnership; or (iii) company. It explains the difference among the three types, particularly in terms of their size, with sole proprietorships being suitable for individuals involved in micro-business activities, whereas companies are suitable for large businesses. See OARG, ‘What Form of Business Is Best for Me?’, www.oarg.gov.sl/Best%20Business%20for%20me.html. 26 The Act requires registration of ‘proprietors’, defined as an individual carrying on a business of which he is the sole owner and includes a person in private professional practice. See http://www.oarg.gov.sl/Gratuity%20Registry.html. 27 For information about the OARG, see www.oarg.gov.sl/. The OARG also administers the registration of intellectual property rights under the Trademarks Act, the Patent Act, and the UK Designs (Protection) Act. See OARG, ‘Intellectual Property’, www.oarg.gov.sl/Intellectual%20Property%20Registry.html. 28 Interview with the OARG (Freetown, Sierra Leone, 18 October 2016). 29 World Bank Group, Leasing in Sierra Leone (2016) 33. 30 Other Financial Services Act 2001, s 2 includes a definition of leasing. 31 World Bank Group (n 26) 39. 32 See C2 below.

Borrowers and Lenders Act of 2014 195

4. Road Traffic Act of 2007 The Road Traffic Act of 2007 provides for the registration of ownership of motor vehicles. It defines ‘owner’ in section 1 to include a person in possession of the vehicle subject to an agreement of hire, hire purchase or loan. However, the Act does not provide for the registration of encumbrances over motor vehicles. While indicating the name of a lender as a co-owner does not render any encumbrance effective against third parties, such an indication in practice prevents the borrower from selling the vehicle.

C. Borrowers and Lenders Act of 2014 The efforts of Sierra Leone to modernise its secured transactions legal framework culminated in the enactment of the B&L Act in November 2014.33 The Act is based on a similar statute enacted in Ghana in 2008. As well as adopting an incoherent approach to reforming secured transactions laws, the Ghanaian statute34 is fraught with gaps, inconsistencies and drafting errors that made it one of the worst possible models for Sierra Leone to follow. One of the factors that contributed to the approach taken in Sierra Leone was promotion of Ghana by the donor agencies as a success story and a model to follow.35 Sierra Leone’s approach is illustrative of the risks inherent in transplanting legislation without assessing its effect on the legal framework and practice in the host jurisdiction and its suitability for the transplanting economy.

1. Approach and models The IFC conducted an assessment of Sierra Leone’s legal and credit environment in 2013, which included recommendations for reform along the lines of the best international standards. In spite of those recommendations, Sierra Leone chose not to follow such standards but to emulate Ghana’s framework. One of the authors of this book provided comments on a draft of the Borrowers and Lenders Bill after it had been submitted to the Parliament, but, at that late stage, no substantial modifications could be made to the Bill.36 Unfortunately for Sierra Leone, this reform process resulted in a legal framework that fails to meet the test of the best international standards, leaves many critical questions unanswered, contains significant gaps and contradictions, and includes sections that are highly unusual for legislation of this kind. It presents a number of lessons for reformers in other jurisdictions, which are discussed below. The B&L Act is divided into seven Parts, which cover not only the usual issues addressed in a secured transactions law, such as the effectiveness of a security interest, but also issues 33 See www.sierra-leone.org/Laws/2014-13.pdf. 34 See Ghana D. 35 See IFC, ‘Secured Transactions and Collateral Registry’, 2, www.ifc.org/wps/wcm/connect/793e79804ac10fff9 ea69e4220e715ad/Secured+Transactions+and+Collateral+Registries+Brochure-English.pdf?MOD=AJPERES. 36 The comments and email communication with Frances Gadzekpo from the IFC Sierra Leone office are on file with the authors.

196 Sierra Leone that clearly belong to another form of legislation, such as those prescribing the regulatory powers of the Bank of Sierra Leone (BSL) in relation to the protection of borrowers from unfair credit practices. Part II, entitled ‘Supervision and Enforcement’, among other aspects, empowers the BSL to conduct investigations of violations of the Act, including by taking possession of books, making copies of any documents and requesting other information from lenders.37 The only violations to which these powers appear to apply (except for the offences of failing to assist in, or impeding, investigations)38 are to be found in Part VI, which is entitled ‘Borrowers Rights, Obligations and Disclosure of Information’. This Part prohibits discrimination in the making of credit decisions,39 and prohibits a lender from entering into a credit agreement without making certain precontractual disclosures.40 Moreover, Part II also imposes a number of requirements on the BSL to report to the Ministry of Finance on the administration of the B&L Act.41 All of these provisions are highly unusual for a secured transactions law because they deal with the regulatory powers of the central bank and the protection of borrowers from unfair credit practices. Both Parts were copied in toto from Ghana’s B&L Act.42

2. Issues As mentioned above, the B&L Act is a minefield that presents many challenges for lenders and others affected by security interests in movable assets. First, based on the definition of ‘lender’ in section 1, the B&L Act applies only to commercial banks and other licensed financial institutions.43 Institutions such as insurance companies and brokers regulated by the Sierra Leone Insurance Commission do not fall under the definition of lender.44 As a result, it does not create an integrated and unified legal framework that would apply to all providers of secured credit. The narrow scope of this provision seems to be a reflection of the BSL’s regulatory powers that may be exercised only over certain financial institutions or the fact that only financial institutions provide credit in Sierra Leone today. It is, of course, the case that a central bank might want to provide a legal framework only for those lenders that it directly regulates. However, taking such a narrow view preserves the existing risk or indirectly exposes the regulated financial institutions to many risks that were sought to be eliminated. Take the example of a farmer who acquires a plough from a neighbour under reservation of ownership. Since the neighbour is not a financial institution, this transaction falls outside the scope of the B&L Act, and under the general sales law 37 See Borrowers and Lenders Act 2014 (B&L Act 2014), ss 4–8. 38 ibid s 5(3). 39 ibid s 32(1). 40 ibid s 36(1). These disclosures include the amount lent, the interest rate, any other charges, the repayment schedule and any penalty for breach. Only this section overtly states that contravention can trigger enforcement measures by the BSL. 41 ibid ss 4 and 8. 42 See ch 3, D for further discussion of the inclusion of these types of provisions in the Ghanaian Act. 43 The list of licensed financial institutions is available at Bank of Sierra Leone, ‘Directory of Financial Institutions’ (2016) www.bsl.gov.sl/Directory_of_Financial_Institutions.html. 44 A list of licensed and registered insurance brokers is available at Nexus Commonwealth Network, ‘Sierra Leone: Find Insurance and Reinsurance Expertise in Sierra Leone’ (2019) www.commonwealthofnations.org/ sectors-sierra_leone/business/insurance_and_reinsurance/.

Borrowers and Lenders Act of 2014 197 the neighbour’s ownership is effective without any form of registration. If the farmer subsequently applied for a loan, the bank would face difficulties discovering whether the farmer owns the plough outright or whether it is subject to a right of a third party, as was the case before the enactment of the B&L Act. Some risks will be exacerbated as the financial institution would need to determine that the prospective borrower has not acquired the asset from a company subject to a registered fixed charge. Furthermore, the definition of lender includes a regulated financial institution that sells goods as part of its business. It is counter-intuitive to define ‘lender’ as limited only to regulated financial institutions, but in the same definition to state that it includes an institution that sells goods. On the other hand, those who do regularly sell goods on credit on reservation of ownership terms do not fall within the scope of the Act, and such transactions are ‘invisible’ as discussed above. It should also be noted that the definition of security interest is a typical functional one found in the Model Law and other national laws, so that reservation of title devices and finance leases are included, as are outright assignments of receivables.45 However, the functional approach is limited by the scope provisions that exclude non-financial institutions from the definition of lender. Operating leases, even lengthy ones, are not included.46 Many errors in the B&L Act stem from the failure to appreciate the need for drafting precision in order to bring about the desired legal result, and also from the drafters’ inattentiveness to detail. The following provisions are illustrative of the latter type of problem. Section 10(2) provides: ‘Where a security interest is created by a lender…’. By definition, security interests are created by borrowers who are persons that used their assets as collateral for loans. Sections 38(6) and (7) provide: (6) This Part shall apply to all security interests with the exception of – (a) an outright assignment of an account receivable; and (b) an operating lease that does not secure payment or performance of an obligation. (7) Where there is a conflict between this Act and other law in relation to the enforcement of a lender’s rights, this Act shall prevail.

It can be seen from the context, and from the reference to ‘security interests’, that this subsection should have been included in Part VII (which deals with enforcement of security interests) and not in section 38, which deals with repayment under the credit agreement and not with security interests at all. It appears that the ‘Part VII’ heading was just put in the wrong place, and should come above section 38(6).47 Even more troublesome is the reference in this subsection to operating leases because the definition of security interest48 encompasses only a finance lease and does not mention an operating lease. Section 38(6) therefore creates doubt as to whether the B&L Act applies to operating leases, and, if so, which Parts apply. These and many other challenges are confirmed by the low usage of the Collateral Registry (CR) operated by the Bank of Sierra Leone. In 2018, only 137 registrations were made. On the positive side, inventory was covered by 126 of those registrations and 45 Part VII on Enforcement does not seem to apply to outright assignments of receivables, though see C(v) below. 46 Unlike, for example, the Canadian PPSAs and the Australian PPSA, but like UCC, Art 9 and the ML. 47 This mistake may not be troublesome at the moment because of the scarcity of receivables financing mentioned earlier, but it may become a problem in the future if such financing becomes more common, which is likely under the reformed system. 48 In B&L Act 2014, s 1.

198 Sierra Leone achinery was designated as the collateral type in 89. Furthermore, 149 search reports were m issued by the CR, exceeding the number of registrations, which has not been seen elsewhere in Africa.49 Both the positive ratio of searches to registrations and a significant majority of registrations covering commercial collateral are promising trends for Sierra Leone.

3. Creation and Perfection Part III of the B&L Act covers both creation and third-party effectiveness (perfection) of security interests. The organisation of this Part does not clearly distinguish between creation and perfection rules. For instance, section 9 provides for the creation of a security interest by execution of a credit agreement, section 10 governs perfection by registration, and section 11 provides the standard for describing collateral in credit agreements. Creation of a security interest requires execution of a credit agreement in writing or by electronic means that evidences the parties’ intent to create a security interest.50 Curiously, section 11(1) requires ‘specific description’ of the collateral, whereas subsection 2 states that descriptions such as ‘all inventory’ and ‘all movable property’ are sufficient for the purposes of subsection 1, although such descriptions clearly are not ‘specific’ in the literal sense. The final creation rule is included in section 11(3), which provides for an automatic extension of a security interest to any identifiable proceeds. The reference to ‘identifiable’ may also be problematic because section 1 defines proceeds as any ‘identifiable or traceable p roperty’. Accordingly, an argument can be made that section 11(3) extends the security interest only to identifiable proceeds, leaving the status of proceeds that cease to be identifiable but are still traceable, especially funds deposited in a bank account where they have been commingled with other funds of the borrower, uncertain. Under sections 12 and 13, registration is the only method to perfect a security interest. In contrast to the Model Law,51 possession and control are not recognised as methods of perfection. Given that the B&L Act only applies to regulated financial institutions, possession might not be practicable because banks and other regulated financial institutions do not generally perfect their security interests by taking possession of the collateral. Section 21 includes yet another drafting error in that it provides that priority shall be determined by the order of possession or registration. As a result, one is left wondering as to the effect of taking possession of the collateral under the B&L Act. Section 15 specifically authorises the BSL to establish and maintain a CR as a unit or department, and to appoint a Registrar. However, the details of the CR’s operations are mainly contained in the Borrowers and Lenders Act Regulations of 19 May 2016 (‘the Regulations’).52 The CR operates as an electronic notice system, which went live in December 2016. The CR is administered by the Banking Supervision Department, rather than as a stand-alone department/unit within the Central Bank’s structure, as in Ghana. Unlike in Ghana, the Sierra Leone Central Bank has not taken any proactive role that would 49 On file with the authors. 50 ibid s 9(1). 51 Arts 18(2), 25, and 27. The B&L Act 2014 also does not recognise automatic perfection of acquisition security rights in consumer goods, as contemplated in ML, Art 24. 52 The Borrowers and Lenders (Collateral Registry) Regulations 2016.

Borrowers and Lenders Act of 2014 199 be inconsistent with the operation of such registries, such as authorising enforcement. The Central Bank was also leading the reform leading to the implementation of the Model Law through the new B&L Act. During the workshops conducted by one of the authors in Freetown (Sierra Leone), the stakeholders raised concerns about the notice function of the CR, arguing that in the absence of any verification of the information submitted to the Registry, its records will be unreliable. These arguments were reflective of the mind-set accustomed to the traditional registries where the actual instrument creating a security right is submitted for registration and verified by the registry staff.53 A notice filing system, on the other hand, operates on the basis that registration gives notice to third parties that a security interest might have been created covering the identified collateral.54 The risk of inaccuracy is taken by the party registering (usually the creditor), who is only protected to the extent of the information disclosed in the notice. Under the B&L Act, the notice to be submitted to the Registry must state the identity of the borrower and lender, a description of the collateral, the maximum amount secured and any other information required by the Registrar.55 The term ‘identity’ is defined in regulation 12 of the Regulations in relation to different kinds of borrowers, including individuals and entities, whether domestic or foreign.56 For instance, the identity of individuals who are citizens of Sierra Leone is established according to section 99 of the Goods and Services Tax Act,57 which provides for the issuance of a unique Taxpayer Identification Number (TIN) that identifies a person when dealing with the National Revenue Authority. Section 16(1)(e) of the B&L Act requires inclusion in the registration of ‘any other information required by the Registrar’. This appears to give a power to the Registrar to require ‘other information’, a power which, because it is implicit, does not require the issuance of Regulations and which could raise concerns as to the broad, and largely unchecked, authority to require additional information, and if that happens, the risk that the additional information may be onerous enough so as to discourage registrations. As in most African countries, in fact, in Sierra Leone, the Registrar decided only to require registrants to identify the industry in which the borrower is involved and to provide other information about the borrower’s business, such as whether the business is majority owned by a woman, whether the borrower is a new customer of the lender, etc. Accordingly, the Registrar exercised its power in a reasonable manner. This kind of information is commonly required by the funding and donor agency to be better able to track the impact of the reform, and is unobjectionable.58 The potential for overreaching, however, remains. Under section 17(3) of the B&L Act, the registration may be ineffective in case of an error in the identity of the borrower or serial number of the collateral that is required to be identified by a serial number. The search logic is strict, so any error, no matter how minor, would not retrieve the registration and thus render it ineffective. An error in some other information, including a description of the collateral where that collateral is not required to be identified by a serial number, would render the registration ineffective only if the error is 53 See ch 1, E1c. 54 See ch 2, E1. 55 ibid s 16(2). 56 The title of Part IV of the Regulations, under which Regulation 12 is included, erroneously refers to ‘registrant’ even though such a term does not appear in this Part at all. 57 Companies Act 2009. 58 See conclusion C7.

200 Sierra Leone misleading.59 This test is different from the ‘seriously misleading’ test included in the Model Law and could render more registrations ineffective than the standard.60 During meetings, the stakeholders regularly inquired whether the CR would be connected to the motor vehicles registry so that information could be exchanged, and so that it could be verified whether a motor vehicle which is to be taken as collateral has in fact been properly registered and remains in good standing. Stakeholders also requested that additional integration with the systems that issue and maintain related records, such as the TINs, should be explored. Unfortunately, such interfaces may not be possible to establish if the other registries and databases are paper-based or their records organised in a manner that makes it impossible to facilitate any information exchanges.

4. Priority Rules Part IV of the B&L Act contains the general priority rules. Aside from the lack of clarity created by the inclusion of possession as a priority point when it is not a mode of perfection, discussed above,61 Part IV is a reasonably straightforward part, which includes a number of sections addressing conflicts among competing claimants, including PMSIs,62 in a way that is similar to the Model Law and other national models. Other conflicts involving rights of buyers, claims of holders of judgment liens, and other competing claimants are included in Part V, which provides special priority rules. In Part IV, section 22 extends the priority of a security interest into proceeds of the original collateral, and section 24 provides that a security interest shall have the same priority in relation to all extensions of credit up to the maximum amount secured indicated in the registration pertaining to that security interest. Section 23 allows lenders to modify their respective priorities in a subordination agreement. Section 25 provides for the priority of PMSIs over non-PMSIs in the same collateral, distinguishing between inventory and other tangibles. For the former, the PMSI holder must complete registration and inform the lender before the borrower receives possession of inventory, whereas for the latter it is sufficient that a registration is completed 10 days after delivery of an asset, such as equipment. These priority rules correspond to Option A of Article 38 in the Model Law. Section 26(1) provides a specific priority rule for security interests in accessions. The Model Law has not incorporated the relevant recommendations of the Legislative Guide on attachments to a movable asset (accessions). Part V governs the rights of third parties that deal with the collateral, including buyers of goods, purchasers of instruments and transferees of money. The basic rule is that such third parties take free unless they have some sort of knowledge, although the subject matter and details of the knowledge required varies considerably. For example, section 28 contains priority rules applicable to three different categories of transferees and, curiously, applies three different knowledge standards. Under subsection 1, a transferee takes money63 free

59 B&L Act 2014, s 17(2). 60 See MRP, Art 24. 61 See C3. 62 Called acquisition security rights in the Model Law, see ch 2, F. 63 Money is not defined in the B&L Act 2014 or the Bank of Sierra Leone Act 2000, but from the context of the B&L Act 2014 provisions, it is presumed that it means physical notes and coins.

Borrowers and Lenders Act of 2014 201 of a security interest unless it acted in collusion with the borrower to violate the rights of the lender. Under subsection 2, a transferee of funds from a bank account takes free of a security interest, whether or not the creditor had knowledge of the security interest. Finally, under subsection 3, a purchaser of a negotiable document, instrument or security certificate takes free of the security interest if it gave value and took possession without knowledge that the transfer was in violation of the credit agreement.64 The logic for applying the three different standards is difficult to discern. Section 29 allows buyers in the ordinary course of business to take free of a security interest, but only that created by the seller. The reach of this rule contrasts with that of the Model Law, under which the buyer takes free of a security right created not only by the immediate seller, but also a security right that might have been created by any predecessor in title.65 The B&L Act is more aligned with some modern secured transactions laws, including the US UCC Article 9.66 Under section 31, a judgment creditor may obtain priority over a security interest by completing execution in relation to collateral before the security interest is perfected.67 Execution is completed by seizure, by the service of a judicial order on the person holding the collateral, or by registration in the CR. Accordingly, it is important for the registry system design to take into account the possibility of non-consensual creditors needing access to the CR.

5. Enforcement Part VII covers enforcement of security interests. As mentioned before, this Part does not seem to apply to outright assignment of receivables, even though the subsection excluding them is in the wrong Part.68 The rules in this Part, as with those in the Model Law and other PPSA models, are designed to enable the creditor to enforce reasonably speedily and effectively, while still protecting the interests of the borrower.69 As a condition of exercising remedies, the lender must deliver a notice to the borrower, who can cure the default within the time stipulated in the notice or, if no period is stipulated, within 30 days. Section 39(3) sets out what amounts to ‘delivery’: this varies according to the method used, but generally requires receipt by the borrower. Ghana’s experience has shown that borrowers often attempt to stifle enforcement by avoiding delivery of a notice of default, which, as opposed to the mere sending of such a notice, remains a statutory condition for enforcement under the Ghanaian Act.70 It is unfortunate that Sierra Leone had not learned from the Ghanaian experience and did not make any provisions in the B&L Act for deliberate avoidance of delivery, for example, that the notice of default is effective when sent rather than received by the borrower. Section 40 authorises the lender to realise the security interest in the collateral or sue the borrower. However, the availability of these remedies does not seem to be conditioned 64 B&L Act 2014, s 28(3) also refers to chattel paper, a term that covers a special type of asset commonly utilised as collateral in the United States, but nowhere defines it or provides other rules that may be required to recognise it as a distinct category of collateral. 65 See ML, Art 34(4). 66 UCC, §9-320(a). 67 See ML, Art 37. 68 See B&L Act 2014, s 38(6) and above C. 69 See ch 2, H. 70 S 32(3) of the Ghana Act is identical to s 39(3) of the Sierra Leone Act.

202 Sierra Leone on the expiration of the grace period within which the borrower may cure the default. Section 40 should be read in conjunction with section 39, authorising the lender to exercise remedies only on the expiration of grace period to cure default indicated in the notice of default. It appears to be implicit in section 41(2) (which enables a lender to use the services of the police or bailiffs to enforce a right to repossess providing that it has a warrant from the court, and that the lender is unable to enforce the right in a peaceable manner) that repossession of tangible collateral must be completed in a peaceable manner. However, in the absence of a specific provision on what constitutes a ‘peaceable manner’, it remains unclear what actions will satisfy this standard in Sierra Leone. In a workshop with judges in Freetown (Sierra Leone), one of the authors discussed some of the cases from other jurisdictions where the courts decided whether the secured creditor’s actions breached the peace.71 For example, while repossessing a car parked in a public place at night does not breach the peace in the United States, the judges concluded that this action would certainly breach the peace in Sierra Leone. These discussions indicated that only the borrower’s voluntary surrender of the collateral might be considered peaceable. Sections 39–41 empower any lenders, including those with junior priority, to enforce their security interests against the collateral. However, section 42(1) authorises only a lender with the first priority to enforce its security interest by collecting and applying accounts receivable, money, securities, or negotiable instruments. Although Ghana’s B&L Act does not include similar restrictions, they may be found in secured transactions laws elsewhere, including in New Zealand.72 Section 43 allows the lender to dispose of the collateral by public auction, private sale or any other method provided for in the credit agreement. The lender is required to provide a notice to the borrower and interested parties, such as other lenders, at least 10 days before the disposal unless one of the exceptions listed in section 43(3) applies, including that the collateral may perish within 10 working days. The lender is then obliged to distribute the proceeds of sale in accordance with the scheme prescribed in section 46, which takes account of the priority of other lenders and claimants, with any surplus being paid to the borrower. Section 47 provides that a lender can, by providing a notice to the borrower and interested parties, propose retention of the collateral in satisfaction of the obligation. However, this remedy is available only to the lender with the first priority.73 The lender would be required to dispose of the collateral if it received a timely objection from an interested party.74 Section 50 also authorises the lender to appoint a receiver or manager or apply to a court for the appointment. A notice of appointment must be registered in the CR, a function unusual for registries of security interests. Lenders may also enforce their rights through the court system. In addition to the general courts, in 2010, the Fast Track Commercial Court (FTCC) was established to provide expeditious relief in commercial disputes.75 According to its Manual of Operations, 71 The workshop took place in Freetown 20 October 2016. 72 New Zealand PPSA 1999, s 108. 73 cf ibid s 120. 74 B&L Act 2014, s 47(2) identifies the affected parties entitled to receive the proposal from the lender. These parties may also exercise the right to redeem the collateral under s 49(1) which incorrectly cross-refers to s 49(2) rather than s 47(2). 75 UK Foreign and Commonwealth Office & Government of Sierra Leone, ‘Sierra Leone: An Investor’s Guide’ (Herbert Smith Freehills, July 2015) 14.

Borrowers and Lenders Act of 2014 203 the objective of the FTCC is ‘to expedite the efficient disposal of claims in trade and commerce in order to promote trade, investments and other economic activities in the country’.76 A judge must hold a ‘pre-trial settlement conference’, a variation of mediation, within three days after receipt of a response from the defendant. On average, a conference takes place within 14 days after the filing of an initial petition.77 The FTCC became more efficient in 2015 when the first resident judges were appointed. In October 2016, it had three resident judges with two more acting in an ad hoc capacity. For the FTCC to assume jurisdiction, the amount in dispute must exceed a certain threshold.78 So far, examples of disputes that the FTCC has dealt with include conflicts of priority between two secured creditors, one of whom had not conducted proper due diligence and took an asset as collateral that had previously been encumbered, and disputes related to the theft and hiding of the collateral.79 There is a degree of uncertainty as to how the general courts may approach resolving secured transactions disputes. During a workshop in Freetown, some judges discussed the possibility of swift judicial remedies that could be made available ex parte.80 While some judges expressed an opinion favouring such remedies, arguing that honest debts should be satisfied, the others felt sympathetic with borrowers, arguing that they should be given a last opportunity to cure the default, which would also allow the judge to ensure that the lender has done everything to prevent the enforcement.

6. Transition Part VIII provides for transitional rules which essentially require lenders to ensure that their prior security interests are perfected under the B&L Act within a period of six months after the effective date.81 The Act does not provide when it takes effect or that its taking into effect is conditional on the establishment of the CR. Under section 12 of the Interpretation Act,82 every Act takes effect upon its publication in the Gazette or upon such date as is provided in or under the Act, or in any other enactment’. Since the B&L Act does not state or provide for a date of its effectiveness, it took effect on 30 December 2014 when it was published in the Gazette. Accordingly, the transitional period to allow lenders to reperfect their prior security interests had expired long before the CR was actually established. This Part also contains section 51, one of the most problematic provisions in the entire B&L Act. It provides: ‘Security interest registered in another registry other than the Registry established under this Act shall not render security interest effective against third parties.’ During all workshops that one of the authors conducted in Freetown, this provision was debated by the stakeholders at length, and three different interpretations emerged. One is that the intent of the B&L Act was to provide for a semi-unitary framework for security

76 On

file with the authors. with one of the resident judges (Freetown, Sierra Leone, 20 October 2016). 78 ibid. At the time of the interview, the threshold was about US$2,500. 79 ibid. 80 The workshop took place in Freetown 20 October 2016. 81 B&L Act 2014, s 52(2)(b). 82 Interpretation Act No 8 of 1971. 77 Interview

204 Sierra Leone interests in movables with a single registry, which means that floating and fixed charges, as well as other security interests taken by regulated financing institutions, whether registrable or not, must now be registered in the CR.83 This interpretation was objected to by the Corporate Affairs Commission, which argued that the provisions of the Companies Act remain unaffected in the absence of a specific consequential amendment. Accordingly, the second interpretation is that the B&L Act imposes a double-registration requirement for certain security interests, such as floating charges and many fixed charges created by companies, which would now need to be registered both with the Corporate Affairs Commission84 and in the CR in order to be effective against third parties. Third, one could also argue that a charge85 created by a company and registered with the Corporate Affairs Commission has its own third-party effectiveness and priority, so the failure to register it in the CR has no negative effect on the charge. All interpretations are problematic. The second and third interpretations could lead to additional legal and practical problems, such as when charge A is created on 1 January and registered with the Corporate Affairs Commission on 20 January, and charge B is created on 10 January and registered with the Corporate Affairs Commission on 25 January;86 however, a registration with respect to charge B is completed with the CR on 10 January, while the holder of charge A registers with the CR on 20 January. In such a case, charge A has priority under the Companies Act whereas charge B has priority under the B&L Act, thereby creating considerable complications and uncertainty.87 One can only hope that parties would react to this state of affairs by making sure that they had subordination agreements in place. This will, however, add to the costs of financing and will also reduce the availability of finance for some types of borrowers. The first interpretation has its own vices too because it provides for a semi-unitary framework which does not apply to security interests taken by creditors who do not fall under the definition of lender in the B&L Act.88 There will be also difficulties applying the respective priority rules when the borrower changes its nature, such as when it initially creates a security interest as an individual under the B&L Act, which is properly perfected by registration in the CR, but subsequently incorporates and creates a floating charge in favour of another bank under the Companies Act that is properly perfected by registration in the Corporate Affairs Commission. It is unclear what the effect of this conversion on the security interest, including its priority status, would be. There is also the potential for problems in the case that a conflict were to arise between a secured creditor who qualifies as lender under the B&L Act and perfects by registration in the CR, and a secured creditor who is not a lender and perfects its security interest under a different law.

83 For the registration of floating and fixed charges see B1 above and, generally ch 1, E1b. 84 ibid. 85 This could be a floating charge or a fixed charge falling within one of the categories in s 170(2) of the Companies Act. 86 The charges could be floating or fixed. 87 To add to the complexity, if charges A and B are both fixed, or both floating, the priority rule under the Companies Act regime is not ‘first to register’ but ‘first in time’, so that provided the charges were registered within the 21-day period, priority would be governed by the date of creation. Priority between a fixed and a floating charge is even more complex, and is governed by s 151 of the Companies Act, see B1 above. 88 See C2 above.

The 2019 Borrowers and Lenders Act 205

7. Regulations The implementation of the B&L Act restarted after the Ebola crisis waned in the autumn of 2015, with the IFC providing support to the BSL in this regard. As one of the first steps, the implementation team drafted the Regulations. In this phase, the parties involved had already realised the magnitude of the problems created by the B&L Act and took the opportunity to clarify certain grey areas and fill gaps where possible. As a result, some provisions of the Regulations are unlike any other regulations that are typically adopted to implement a secured transactions law. Some Regulations were expected to be drafted to implement specific sections of the B&L Act. For instance, the reference to the identity of the borrower in section 16(2) of the B&L Act was specified in Regulation 12. This is the kind of regulation that is common for subsidiary legislation of this nature. However, Part VII of the Regulations, which governs miscellaneous items, covers issues of great importance that belong in the substantive law. For instance, Regulation 23 deals with post-registration transfer of the collateral. This provision corresponds with Option C of Article 26 in the Model Registry Provisions of the Model Law, under which a transfer of the collateral does not affect the security right.89 In other words, the initial transferee and any subsequent transferees or lenders acquire their rights subject to a security interest, whether or not the original lender registers an amendment identifying the transferee as an additional borrower. Regulation 24(2)–(3) provides a priority and enforcement rule for security interests in attachments to an immovable (fixtures). It is not only inappropriate to govern such aspects of security interests through regulations, but their legal effect is also questionable in the light of the absence of any authority in the B&L Act to provide such rules in the Regulations. The implementation team hoped that this stop-gap measure could at least provide some guidance to the users and judges if an issue covered by these Regulations comes up before the B&L Act is revised. The Schedule to the Regulations sets out the fees applicable for various Registry services. The CR charges fees for the registration of initial notices, amendment notices as well as searches. Concerning is the level of fees, which is about twice as high as in neighbouring Liberia and higher than in Ghana. There is a commitment to reconsider the level of fees in the near future.

D. The 2019 Borrowers and Lenders Act On 20 May 2019, the Parliament passed the Borrowers and Lenders Act No 50 (‘the new B&L Act’), which was gazetted on 20 June 2019, and thus has already entered into force. However, a number of its provisions contemplate that the details will be set out in the Regulations, including the standards for describing immovable collateral in a notice. The CR must also be modified to reflect the changes introduced in the new B&L Act properly; therefore, the actual implementation may be delayed. As with the entry into force of the B&L Act, some legal uncertainty will exist until the Regulations are revised and the CR modified. The new B&L Act repealed the B&L Act and the relevant provisions of the Companies Act affecting

89 See

ch 2, E1.

206 Sierra Leone the registration of floating and fixed charges.90 This addresses the issue of several possible interpretations of section 51 of the B&L Act. Furthermore, the new B&L Act supersedes the registration function for conveyances, deeds and instruments that purport to create security interests in the OARG.91 Consequently, the CR will be the only source of information regarding security interests created by legal entities and individuals with respect to movable and immovable property.92 The new B&L Act implements the Model Law. It does not deviate from the approaches or drafting of the Model Law, except that it extends the scope of application to security interests in intermediated securities and immovable property. It is based on the functional approach. One consequence of this approach is that the Act expressly applies to, inter alia, a ‘lease intended as a security interest’,93 which is defined in section 1 of the new B&L Act along the lines of the recommended definition of a financial lease from the UNCITRAL Legislative Guide on Secured Transactions.94 The new B&L Act thus clarifies the position with respect to its application to leases that was muddied by the references to operating leases in the B&L Act.95 Moreover, unlike its predecessor, it applies to any type of lender, whether or not a financial institution.96 It also affects the priority of non-consensual creditors, which are defined to include not only judgment creditors, but also any tax authority.97 To obtain priority over subsequently registered security rights, non-consensual creditors need to perfect their interests by registration in the CR.98 With one exception, the new B&L Act also follows the structure of the Model Law. Unlike the Model Law, the conflicts of laws provisions are set out in section 2 as a definition of ‘law applicable’, though their content reflects the Model Law. Since the B&L Act also applies to immovable property, it designates the law of the State where the immovable property is located as the law applicable to the creation, perfection and priority of a security interest.99 Section 2 also includes all the transitional rules under the heading of ‘Matters that are the subject of proceedings commenced before the entry into force of this Act’: these rules could have been set out more clearly in a separate part. The transitional period of six months for prior security interests in movable property and a period of 12 months for security interests in immovable property commenced upon the gazetting of the new B&L Act, which is problematic in the absence of the new framework’s full implementation.100 As for the structure, unlike its predecessor, the new B&L Act does not cover the supervisory powers of the BSL and the protection of borrowers.101 The B&L Act incorporates the requirements for creation of a security interest from the Model Law, but adds that a description standard for immovable property in a security 90 S 77(1), which repeals ss 150–51, 175 and 180–81 of the Companies Act. 91 S 77(2). See B2 above. 92 The new B&L Act does not apply to security interests over limited categories of movable assets, such as ships and aircraft. See s 3(2). 93 S 3(1)(iv). 94 See ‘Legislative Guide on Secured Transactions’ (Vienna, United Nations, 2010) (LG). 95 See C2 above. 96 See C2 above. 97 S 2. 98 See s 48. 99 S 2(b)(iv). 100 See ch 19 C2a. 101 See C1 above.

The 2019 Borrowers and Lenders Act 207 agreement and a notice will be prescribed in the Regulations.102 Many other aspects are also left to be dealt with in the Regulations. For instance, the security agreement must state the identity of the lender and the borrower,103 and a notice must provide an identifier of the borrower and the lender, or its representative.104 The new B&L Act does not define either ‘identity’ or ‘identifier’. It is expected that the Regulations issued under the B&L Act, which already defined these notions, would serve as the basis for the new Regulations.105 Under section 78(1), any regulations issued pursuant to a repealed Act remain in force until they are expressly repealed. Preserving the effect of the Regulations issued under the B&L Act that also define the identifier for different types of borrowers ensures that the CR is capable of processing notices with the effect under the new B&L Act, though limited to movable collateral. Section 15 of the new B&L Act recognises registration as the only method of perfection for security interests in any movable or immovable property. Accordingly, control is not a perfection method even for security interests in deposit accounts and securities, as recommended in international models. Section 18 empowers the BSL to establish and maintain the CR. Since the BSL has already established the CR under the B&L Act, this registry will be modified, including to facilitate registration of notices relating to security interests in immovable property. The unclear relationship between the B&L Act and the CR, on the one hand, and the Companies Act and the registry maintained thereunder by the Corporate Affairs Commission, exemplified in section 51 of the B&L Act, has been clarified through the scope provision. Section 3(b)(vi) of the new B&L Act provides that floating and fixed charges are security interests. However, the distinction between these two types of charges has not been entirely abolished. The priority of preferential creditors over a floating (but not fixed) charge holder when a receiver takes possession of assets subject to a floating charge106 or on a liquidation107 has not been abolished as the relevant sections of the Companies Act have not been repealed or replaced.108 Furthermore, section 18(3) of the new B&L Act states that the requirements of other laws, other than for perfection and priority of security interests, remain applicable. In other words, only a registration in the CR will perfect and establish the priority for a security interest. Paragraph (b) of section 18(3) makes it clear that, in the context of the Companies Act, a company borrower will still need to notify the Commission of the creation of any security interest, including a charge, and will also need to maintain its own register of charges. The Commission will have the ability to verify the satisfaction of those requirements since the Registrar of the CR must provide to the Commission records of all notices registered therein which name a company as the borrower.109 Section 20 of the new B&L Act restates the approach of section 18(3). Searches of the CR can be conducted according to the identifier of the borrower or a serial number of the serial-numbered asset, as defined in section 2. Serial-numbered

102 S

12(4) and 26(3). 6(2)(b). 104 S 19(a)–(b). 105 See C7 above. 106 Companies Act s 154. 107 Companies Act s 438(4)(b). 108 See also B1 above. 109 S 18(4). 103 S

208 Sierra Leone assets are limited to motor vehicles and trailers.110 Any error in a serial number renders the registration ineffective against buyers and lessees only.111 Accordingly, an error in the serial number does not have any impact on the effectiveness of the notice as against other competing claimants, including other lenders and trustee in bankruptcy. Finally, when motor vehicles or trailers are held as inventory the notice is effective, regardless of whether it provides serial numbers in the relevant field.112 The priority scheme is based on the time of registration in the CR, subject to the usual take free rules protecting buyers, lessees and licensees in the ordinary course of business,113 transferees of money and funds, and purchasers of negotiable instruments, negotiable documents of title, security certificates and intermediated securities.114 The new B&L Act also recognises a PMSI that differs from the Model Law notion of acquisition security right in name only; the definition as well as the priority approach of the PMSI reflect the Model Law.115 The enforcement part tracks the remedies provided for in the Model Law. In a similar way to its predecessor, the new B&L Act empowers the lender to appoint a receiver or a receiver and manager, or request a court to appoint one.116 A notice of appointment must be registered in the CR. The borrower is entitled to a notice of default that may indicate a period within which the default must be cured; if none is stated, it must be cured within 30 days.117 If the collateral is an immovable property and the borrower is an individual, the 30-day cure period cannot be modified by the parties.118 For security interests in both movable and immovable property, section 66 provides that a security interest in both types of property can be enforced under the law governing enforcement of rights in immovable property. Movable collateral may only be repossessed in a ‘peaceable manner’; a notion inspired by the US UCC, but similarly left undefined.119

E. Final Remarks and Future Steps This chapter provides a clear illustration of a misguided approach to reforming secured transactions laws. At the same time, it shows that the implementers did have some understanding of the issues involved and that they desired to correct the mistakes in the 2014 reform by enacting the new B&L Act. These mistakes range from the serious ones, such as the limited scope and the uncertain relationship with the Companies Act to a number of drafting issues, whether placement of certain provisions under an improper Part of the B&L Act or incoherent knowledge standards for transferees of different types of collateral to take free of security interests. The Regulations are not a proper legislative

110 See

the definitions of ‘serial number’ and ‘serial-numbered asset’ in s 1. 35(5). 112 S 26(2)(b). 113 S 45(4)–(6). 114 S 46. 115 For priority of the PMSI, see s 49. 116 S 63. 117 S 64(1). 118 S 64(3). 119 UCC, §9-609. 111 S

Final Remarks and Future Steps 209 tool to address those mistakes. Even before the CR was launched, the implementation team drafted a set of amendments to the B&L Act that were enacted in 2019. Many stakeholders, particularly the BSL and the Corporate Affairs Commission, were motivated to clarify the uncertainties caused by the approach of copying the Ghana statute, but reforming laws of this nature is always a formidable task. An additional motivating factor was the World Bank’s Doing Business Report120 which could not assign higher marks to Sierra Leone for this reform, as many local stakeholders had hoped. A coherent secured transactions framework not only required a reform of the B&L Act, but also a close coordination of the reform effort with the project to introduce new leasing legislation and the consequential changes to the related laws, especially the Companies Act. The new B&L Act drew inspiration from the Model Law to both correct the many mistakes in approach and drafting in the 2014 reforms and to fill the gaps.

120 See

ch 1, B2.

210

part iii Civil Law

212

10 OHADA Uniform Act on Securities A. Introduction In 1993, 14 African nations formed a regional organisation with the purpose of harmonising their business laws; this organisation became known under its French acronym OHADA (Organisation pour l’Harmonisation en Afrique du Droit des Affaires, or the Organization for the Harmonization of Business Law in Africa).1 The underlying objective for business law harmonisation is to provide investors with certain and reliable legal rules that can be applied predictably.2 Since its formation, OHADA has adopted several uniform acts relating to insolvency, contracts for the carriage of goods by road, business entities and economic interest groups, recovery procedures, and arbitration.3 Article 10 of the OHADA Treaty provides that: ‘Uniform Acts shall be directly applicable to and binding on the State Parties notwithstanding any previous or subsequent conflicting provisions of the national law.’ Therefore, the Uniform Acts are unification measures.4 The OHADA Uniform Act on Securities (‘the former Uniform Act’) was adopted on 1 January 1998 and published in the OHADA Official Journal on 1 July 1998.5 It was subsequently revised in December 2010. The revised Uniform Act (‘the Uniform Act’), which became effective on 16 May 2011, was based mainly on the 2006 reform of the French law while also attempting to adhere to some of the approaches of the Legislative Guide.6 The Uniform Act specifies that security rights granted before 16 May 2011 will continue to be governed by the former Uniform Act. Unlike the Model Law, it does not provide for the transition into the 2011 regime of security rights created under the former Uniform Act. 1 OHADA now has 17 members: Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Comoros, Congo, the Democratic Republic of the Congo, Equatorial Guinea, Gabon, Guinea, Guinea Bissau, Ivory Coast, Mali, Niger, Senegal and Togo. See: Organization for the Harmonization of Business Law in Africa, ‘History of OHADA’ (2017) www.ohada.org/index.php/en/ohada-in-a-nutshell/history. 2 P Winship, ‘Law and Development in West and Central Africa (OHADA)’ Southern Methodist University Dedman School of Law LS Research Paper No 272 (2015) 1, 17. 3 The complete list of uniform acts is available (under the ‘OHADA Laws: Uniform Acts’ tab) at: Organization for the Harmonization of Business Law in Africa, ‘History of OHADA’ (2017) www.ohada.org/index.php/en/ organisation-des-suretes-en/auds-presentation-et-innovations-en. 4 J Issa-Sayegh, ‘Quelques aspects techniques de l‘intégration juridique: l’exemple des actes uniformes de l’OHADA’ [Some Technical Aspects of Legal Integration: The Example of the Uniform Acts of the OHADA] (1999) 6 Revue de droit uniforme 5; OHADA, ‘Uniform Act Organising Securities: Compilation of Treaties and Uniform Acts: Official Translation’ (2010) www.droit-afrique.com/uploads/OHADA-Uniform-Act-2010-securities.pdf (Uniform Act). 5 P Crocq et al, ‘Le nouvel acte uniforme portant organisation des sûretés: La réforme du droit des sûretés de l’OHADA’ [The New Uniform Act on the Organisation of Securities: OHADA’s Securities Law Reform] (Collection Lamy Axe droit, 6 September 2012). 6 JF Riffard, ‘The Still Uncompleted Evolution of the French Law on Secured Transactions Towards Modernity’ in L Gullifer and O Akseli (eds), Secured Transactions Law Reform: Principles, Policies and Practice

214 OHADA Uniform Act on Securities Accordingly, the former Uniform Act may still govern a number of secured transactions. The absence of any transitional rules forces prospective creditors to conduct due diligence beyond the Uniform Act, thus increasing the cost of secured transactions. This chapter examines only the 2011 edition of the Uniform Act. OHADA assumed responsibility for facilitating the uniform interpretation of its Acts by establishing the Common Court of Justice and Arbitration (CCJA).7 The seat of the Court is in Abidjan (Ivory Coast), but it may hold hearings in other Member States.8 Article 14 of the OHADA Treaty specifically provides: ‘The Common Court of Justice and Arbitration, shall, in the Member States, ensure uniform interpretation and application of the Treaty, regulations laid down for its application, as well as the Uniform Acts.’ Despite this objective of ensuring uniformity of laws, it has been reported that national courts do not consistently apply OHADA Uniform Acts, either because they are unaware of their existence or simply due to a preference for domestic law.9 The Uniform Act is accompanied by a commentary to facilitate the understanding of its provisions. The International Institute of French Law of Expression and Inspiration published annotations on the former Uniform Act,10 while the Uniform Act and Treaties, also known as ‘Code Vert’ for the publication’s green cover, contains commentaries and annotations on the 2011 version.11 The Uniform Act contains 228 articles governing personal securities, such as suretyship (cautionnement) and independent guarantees (garanties autonomes),12 as well as security rights in both movable and immovable property.13 Article 4 distinguishes between personal securities and real security interests, the former being defined as an undertaking of a person to fulfil an obligation of the debtor either in the event of default or on demand. Accordingly, the scope of the Uniform Act is much broader than that of the Model Law, as the Uniform Act also applies to security rights in immovable property and guarantees that do not create any proprietary interest in the guarantor’s assets. The 2011 revision introduced a number of changes to modernise the regime for taking security rights in movable property.14 However, in a number of aspects, especially with respect to the asset or transaction-specific security devices contained in the Uniform Act (eg the security transfer (assignment) of receivables, (Oxford, Hart Publishing, 2016); see UNCITRAL, ‘UNCITRAL Model Law on Secured Transactions: Guide to Enactment’ (United Nations, Vienna, 2017). 7 P Meyer, La sécurité juridique et judiciaire dans l’espace OHADA [Legal and Judicial Security in the OHADA Area], 855 Université de Franche-Comté (UFR) de sciences juridique et politique 151. 8 See Organization for the Harmonization of Business Law in Africa, ‘Court commune de justice et d’arbitrage: Abidjan, Cote d’Ivoire Présentation de la CCJA’ [‘Common Court of Justice and Arbitration: Abidjan, Ivory Coast: Presentation of the CCJA’] (2017) www.ohada.org/index.php/fr/cour-commune-de-justice-et-d-arbitrage-ccja/ ccja-en-bref. 9 Winship (n 2). 10 Institut international de Droit d’Expression et d’inspiration Françaises, ‘Code IDEF annoté de l’OHADA’ [International Institute of French Law of Expression and Inspiration, ‘Annotated IDEF Code of the OHADA’] (2019) www.institut-idef.org/-Code-IDEF-annote-de-l-OHADA-.html. 11 Juriscope (ed), Traités et actes uniformes commentés et annotés [Uniform Act and Treaties with Commentary and Annotations] (Code Vert) (University of Poitiers, France, 19 April 2016). 12 Uniform Act (n 4) Arts 12–49. 13 ibid Arts 50–189. 14 One such innovation is the ability to appoint a security agent, a device particularly useful in syndicated loans, structured finance and international project financing. See Uniform Act (n 4) Art 5; HD Amboulou, Le droit des sûretés dans l’espace OHADA [Security Law in the OHADA Area] (Paris, France, Études Africaines, L’Harmattan, 2014) 9.

Scope: Types of Security Rights 215 the fiduciary transfer of money, the pledge of a bank account and the pledge of intellectual property rights), it departs from the unitary and functional approach of the Model Law.15

B. Scope: Types of Security Rights The Uniform Act defines a security as ‘the allocation of an asset or property in favour of a creditor to guarantee the discharge of an obligation or obligations, whatever their legal nature, provided that such obligation or obligations is existing, prospective, ascertained or ascertainable, conditional or unconditional and of a fixed or changing amount’.16 Article 4(4) stipulates that security rights in assets subject to waterway, sea, and air laws may be governed by specific legislation. In a similar way, Article 1(3)(e) of the Model Law allows enacting States to designate assets to which the enactment of the Model Law shall not apply. However, a State enacting the Model Law is expected to specify the laws that govern the special types of assets, whereas Article 4(4) of the Uniform Act is too vague to determine with certainty which assets are excluded from the scope of the Uniform Act. For a uniform act that directly applies in multiple jurisdictions, it would have been arduous to identify specifically the relevant laws of every State governing security rights in assets excluded from its scope. Other than that, the Uniform Act does not exclude any type of movable assets that may be subject to a security right.17 Unlike the Model Law, however, the Uniform Act does not mention finance leases that may be created under other laws, in following with the classical civil-law approach of numerus clausus that limits the scope of secured transactions laws to specific types of security devices. Finance leases may be registered in the same Register as pledges and other security devices governed by the Uniform Act, but the Uniform Act neither requires their registration nor provides consequences for failure to register.18 The Uniform Act was adopted in the general context of facilitating secured transactions for businesses. While it generally applies to all types of movable assets, whether held by consumers or businesses, the Uniform Act contains some special rules that apply in the context of obligations incurred by professional debtors in the performance of professional duties or activities. Although the Model Law also places a special emphasis

15 Amboulou (n 14) 10. 16 Uniform Act (n 4) Art 1; Permanent Secretariat of the Organization for the Harmonization of Business Law in Africa (OHADA), ‘Uniform Act Organising Securities’ (15 December 2010) Art 1 in OHADA, Compilation of Treaties and Uniform Acts: Official Translation (Official Bulletin, 24 November 2016) 309. The Uniform Act is also available in the original French at: OHADA, ‘Acte uniforme revise portant organisation des duretes’ (15 December 2010) www.ohada.com/actes-uniformes/938/955/chapitre-2-droit-de-retention.html. The OHADA website provides an automatic English translation tool (activated by clicking on the English flag on the top right side of the page). Minor differences exist between the official and unofficial versions. For instance, Art 1 is officially translated in the text to this note; whereas the unofficial version of the same Article reads as follows: ‘[T]he pledge, for the benefit of a creditor of an asset, a set of properties or assets to secure performance of an obligation or a set of obligations, irrespective of their legal nature whether existing or future, determined or determinable, conditional or unconditional, and whether their amount is fixed or fluctuating.’ 17 cf ML, Art 1(3), which, inter alia, excludes from the scope of its application security rights in intermediated securities. 18 See discussion of the lack of clarity regarding the requirement to register, below at nn 100–02 and text (to end of paragraph).

216 OHADA Uniform Act on Securities on facilitating commercial transactions, its general rules apply equally to security rights in consumer goods, unless stated otherwise.19 Unlike the Uniform Act, the Model Law does not make any distinction based on the nature or purpose of the secured obligation (as opposed to the nature of the encumbered asset). Article 50 of the Uniform Act provides that movable property may be encumbered under: (i) a right of retention, (ii) the reservation of ownership as security, (iii) an assignment of ownership as security (security transfer of ownership), (iv) a pledge over tangible assets, (v) a pledge over intangible assets; and (vi) liens. Title 2, Chapters 2–6, prescribes the rules that apply to each type of security right.20 Accordingly, the Uniform Act has not embraced the functional approach, under which all legal devices that function to secure the performance of an obligation are treated as security rights and to which common rules apply. The functional approach is the bedrock of the Model Law.21 Furthermore, the Uniform Act does not recognise a general security right that can encumber all assets of the grantor. Instead, the secured creditor must deploy the individual security devices, subject to their own specific creation and third-party effectiveness requirements, to encumber different types of assets or the totality of assets belonging to the grantor. Any type of security right can be granted by the borrower or a third party.22 A security right can be granted either to the actual secured creditor or to its agent who may act on behalf of multiple creditors.23 Whenever a security agent acts on behalf of the actual secured creditors, it must clearly identify itself as a security agent.24 The following subsections of this chapter examine the application of the individual security devices governed by the Uniform Act to specific transactions, deferring the discussion of third-party effectiveness and enforcement to later in this chapter.25

1. Right of Retention A right of retention is the first type of security right enumerated in Title 2. Article 67 of the Uniform Act provides that the creditor in lawful possession of an asset may continue to hold it until full payment of the debt.26 For the right of retention to be lawful, the holder must satisfy the requirements of article 68, including that the claim must be for 19 eg, as in ML, Art 24. See ch 2, B2a and F3. 20 Uniform Act (n 4) Arts 67–189. 21 See ML, Arts 1(1) and 2(kk) and ch 2, B2. 22 Uniform Act (n 4) Art 4. 23 ibid ch 2 (Arts 5–11). See also KO Coffi Dogué, ‘Regard critique et pratique sur l’agent des sûretés de l’OHADA’ [Critical and Practical Look at the OHADA Surety Agent] OHADA Revue de l’Ersuma (September 2014) revue. ersuma.org/no-4-septembre-2014/doctrine/article/regard-critique-et-pratique-sur-l; GA Likillimba, ‘L’agent des sûretés en OHADA’ [The OHADA Surety Agent] OHADA Revue de l’Ersuma (November–December 2011) revue. ersuma.org/numero-special-novembre-decembre/etudes-et-pratiques/L-agent-des-suretes-OHADA. 24 Uniform Act (n 4) Art 7. 25 See discussion of third-party effectiveness, at C below and discussion of enforcement at G below. 26 Acte uniforme (n 16) Art 67 provides: ‘Le créancier qui détient légitimement un bien mobilier de son débiteur peut le retenir jusqu’au complet paiement de ce qui lui est dû, indépendamment de toute autre sûreté, sous réserve de l’application de l’article 107 alinéa 2, du présent Acte uniforme’ [‘A creditor in lawful possession of an asset belonging to the debtor may, regardless of any other security, continue to hold the said asset pending payment in full of the debt due, without prejudice to the provisions of Article 107 (2) of this Uniform Act’].

Scope: Types of Security Rights 217 a due and certain amount, and arise in relation to the retained asset. An annotation to the Uniform Act identifies a situation where a car seller who has not received payment of the purchase price retains the vehicle as an example of lawful possession that gives rise to the right of retention.27 In one case, the CCJA held against a creditor who asserted its right of retention against the debtor’s canoes even though the debt was not payable until two months later.28 Under the former Uniform Act, the right of retention was assimilated in the possessory pledge, which enabled its holder to dispose of the asset. However, under the revised Uniform Act, a right of retention merely allows the holder to retain the asset until full satisfaction of the obligation.29 The creditor is permitted to sell the asset with the authorisation of the court, but only if it is perishable.30

2. Reservation of Ownership A clause retaining ownership has the effect of suspending the transfer of ownership to the buyer until full satisfaction of the obligation it secures.31 It must be evidenced by a written agreement executed prior to delivery of the asset to the buyer.32 A single agreement may effectively reserve ownership with respect to future transactions between the parties.33 However, it is ineffective against third parties before it has been registered in the Register of Commerce and Securities.34 As will be explained below, this approach is similar to the Legislative Guide’s non-unitary approach to acquisition security rights, which does not characterise rights of sellers and financial lessors who retain ownership as security rights but, rather, treats the rights of sellers and lessors as functionally similar to security rights.35 The actual effect of this treatment seems quite limited. As mentioned, these rights are required to be registered to be effective against third parties (in a way similar to ‘true’ security rights). The retention of ownership has the effect that such rights are not affected by a prior security right over the buyer’s after-acquired property; thus, the position is very similar to that of an acquisition security right under the Model Law.36 The enforcement rights of the retention of title creditor are set out in Article 77 of the Uniform Act, under which the creditor can claim restitution of the asset and can apply its value to payment of the secured debt. Any surplus value is payable by the creditor to the debtor. 27 Institut international de Droit d’Expression et d’inspiration Françaises, ‘Acte uniforme du 17 Avril 1997 portant organisation des suretes’ [International Institute of French Law of Expression and Inspiration, ‘Uniform Act of 17 April 1997 on the Organisation of Securities’] (1997) www.institut-idef.org/Article-41.html, Art 41. 28 Code Vert (n 11) 913; court decision available online, to members, at Judgment No 16/2002, Maregel Company v Serigne Moustapha Mbacke (Common Court of Justice and Arbitration (CCJA), Ohadata J-02-165, 27 June 2002) www.ohada.com/jurisprudence/ohadata/J-02-165.html. 29 Z Njutapvoui, ‘Le Droit de rétention dans le nouvel acte Uniforme portant organisation des sûretés: sûreté active ou passive?’ [‘Right of Retention in the New Uniform Act on the Organization of Security Rights: Active or Passive Security?’] (OHADA, January 2016) www.ohada.org/index.php/en/ohada-doctrine/studies/445-le-droitde-retention-dans-le-nouvel-acte-uniforme-portant-organisation-des-suretes-surete-active-ou-passive. 30 Uniform Act (n 4) Art 70. 31 ibid Art 72. 32 ibid Art 73. 33 ibid. 34 ibid Arts 74 and 51–61 governing registration of security rights. 35 See ch 2, B2b. 36 See ibid and ML, Art 38.

218 OHADA Uniform Act on Securities Thus, the main characteristics of a security right are preserved; that is, that the debtor can obtain full rights to the asset on payment of the secured obligation, that the secured creditor is obliged to account for any surplus to the debtor on enforcement, and that the debtor continues to owe the balance of the obligation due after taking into account the value generated by enforcement.37 The lack of recharacterisation changes very little of substance but means that the conceptual analysis of the transaction (as a retention of ownership rather than a grant of a security right) is very different.

3. Security Transfer of Receivable Article 79 provides the general rule that ownership of an asset may be transferred as security for the satisfaction of a present or future debt. The Uniform Act sets out specific rules for such transfers in relation to two types of assets: debts (receivables) and funds. Under Article 80, any debt may be assigned as security to a national or foreign legal entity conducting banking or credit operations in the regular course of business for its own account.38 As such, not every type of creditor may benefit from this type of security device. The second paragraph of Article 80 provides that any contractual anti-assignment clause is ineffective against the assignee, as long as the debt of which the clause prohibits assignment is a business debt. As with Article 13 of the Model Law, the purpose of Article 80 is to override the effect of anti-assignment clauses restricting the transfer of trade receivables.39 However, the application of the override in Article 80 of the Uniform Act is not limited to a particular subset of trade receivables, such as those listed in paragraph 3 of Article 13 of the Model Law. Furthermore, only a national or foreign legal entity conducting banking or credit operations in the regular course of business for its own account may benefit from this override of anti-assignment clauses, a limitation not included in the Model Law. A security transfer of a receivable must be evidenced in a writing that includes: (i) the name or trade name of the assignor and the assignee; (ii) the date of assignment; and (iii) a description of the secured obligation(s) and the assigned receivables.40 The agreement may also provide for an assignment of future debts as security.41 However, in this case, the written agreement must identify the future debt(s) by means such as the name of the person owing the debt (receivables debtor), the place of payment, the amount and, where applicable, the due date. The recognition of the security assignment of future debts was motivated by the objective of facilitating access to credit.42 However, as compared to Article 9 of the Model Law, the requirement to describe future receivables in a security 37 See, eg, the English case: Re George Inglefield Ltd (1933) 1 Ch 1 (CA). 38 Uniform Act (n 4) Art 80 para 1. 39 For discussion of the policy reasons for this kind of provision, see ch 1, E4a. 40 Uniform Act (n 4) Art 81 para 1. 41 ibid Art 81 para 2. 42 GK Douajni, ‘Les nouvelles sûretés introduites dans l’Acte uniforme sur les sûretés adopté le 15 décembre 2010’ [‘The New Security Rights Introduced in the Uniform Act on Security Interests Adopted on 15 December 2010’] OHADA Revue de l’Ersuma (November–December 2011) www.revue.ersuma.org/numerospecial-novembre-decembre/legislation/Les-nouvelles-suretes-introduites.

Scope: Types of Security Rights 219 agreement, especially the place of payment and the amount, is onerous and does not facilitate this type of secured transaction, as it requires the parties to revise the security agreement regularly as new receivables arise. A notification of the receivables debtor is not required to render the assignment effective between the assignor and assignee, but in the absence of a notification the receivables debtor may discharge its obligation by paying the assignor.43 This corresponds to the approach of the Model Law in Articles 58 and 59. The security nature of the assignment is made clear by the provision that sums paid by the receivables debtor to the assignee reduce the secured debt (if overdue), and any surplus is to be returned to the assignor.44 The parties cannot contract out of this provision. This particularly shows that the arrangement is a security interest and not an outright assignment; it is analogous to Article 72(3) of the Model Law which prohibits the grantor or debtor from contracting out of its rights under the enforcement chapter, including the right to the surplus in Article 79(2)(b).

4. The Fiduciary Transfer of Funds Under Article 87, a grantor may transfer funds (ie money) as security for the performance of an obligation. The transferred funds must be deposited in a blocked account maintained in the name of the secured creditor by a credit institution authorised to receive deposits.45 An agreement to create a fiduciary transfer of funds must identify: (i) the secured obligation, (ii) the amount of funds transferred in security, and (iii) the blocked deposit account. If it does not, the agreement is a nullity.46 A transfer is not effective against third parties until the credit institution receives a notification and credits the amount to the identified blocked deposit account. The Uniform Act does not recognise any alternatives to this form of third-party effectiveness such as registration or a control agreement between the grantor, the secured creditor and the financial institution that maintains the blocked account.47 Furthermore, since the funds are deposited in a blocked account, this means that the grantor is stripped of access to those funds. In contrast, the Model Law not only recognises control agreements as a method to achieve third-party effectiveness but also allows the grantor to retain access to the funds. Under the Uniform Act, this security mechanism is not useful for businesses, especially those involved in trading activities that require constant access to funds.48

43 Uniform Act (n 4) Art 84. 44 ibid Art 86. 45 ibid Art 87 para 2. 46 ibid Art 88. 47 cf ML, Art 25, which provides not only for a control agreement as an alternative form of third-party effectiveness but also for such effectiveness when the deposit-taking institution is the secured creditor. See also ch 2, D. 48 cf the English jurisprudence on the creation of a fixed charge over receivables which requires that they are paid into a blocked account. See Re Spectrum Plus Ltd [2005] UKHL 41, [2005] 2 AC 680; see also H Beale, M Bridge, L Gullifer and E Lomnicka, The Law of Security and Title-Based Financing (3rd edn, Oxford, Oxford University Press, 8 May 2018) ch 6.

220 OHADA Uniform Act on Securities

5. Pledge of Tangible Assets The Uniform Act distinguishes between pledges of tangible49 and intangible assets.50 The rules governing the pledge of tangible assets (gage de meubles corporels) provide that dispossession is not required to create the pledge or to make it effective against third parties.51 Under Article 92, a pledge of tangible movable assets is created by a contract between the grantor and the secured creditor. The pledge can be over existing or future assets52 and can secure existing or future obligations, provided they are determined or determinable.53 A pledge agreement must identify the pledged object by quantity, specification or nature, which seems to suggest that a detailed description is required54 and which, despite the express mention of future assets in Article 92, would seem to make it difficult to give a pledge over anything other than very specific future assets. The pledge may be made effective against third parties either by registration (which would be the case if the pledge was non-possessory) or delivery of the asset to the secured creditor.55 The Chapter of the Uniform Act on pledges of tangible assets contains two subsections that provide additional rules specific to two types of pledges: the pledge of professional equipment and motor vehicles and the pledge of inventory. Professional equipment and motor vehicles may be pledged as stand-alone assets or encumbered with other elements of an enterprise (fonds de commerce), in accordance with Articles 162–65 of the Uniform Act.56 For motor vehicles, a reference to the pledge must be included on the registration (certificate of title) document. However, failure to include such a reference does not affect the third-party effectiveness of the pledge.57 Articles 120–24 contain special rules for pledges of inventory, such as raw materials, agricultural and industrial products, and other goods (presumably those manufactured and/or held for sale).58 Although the former Uniform Act defined inventory, that definition was not carried over to the Uniform Act, so the exact limits of this subsection are unclear. Pledgors may sell the pledged inventory in the ordinary course of business.59

6. Pledge of Intangible Assets Chapter 5 of the Uniform Act governs the pledge of intangible assets listed in Article 126: receivables, bank accounts, partnership interests, securities and financial instruments, enterprise (fonds de commerce) and intellectual property rights.60 Under Article 125, a pledge of intangible assets (nantissement de meubles incorporels) is created by 49 Uniform Act (n 4) Arts 92–124. 50 ibid Arts 125–78. 51 This is clear from the provisions of ibid Art 96, which provide for delivery of the pledged asset to the pledgee, and Arts 99–102, which refer to possessory and non-possessory pledges; see also Amboulou (n 14). 52 Uniform Act (n 4) Art 92. 53 ibid Art 93. 54 cf ML, Art 9, which recognises generic and ‘all assets’ descriptions. 55 ibid Art 96. 56 Uniform Act (n 4) Arts 118–19. 57 ibid Art 119. 58 ibid Art 120. 59 ibid Art 124. 60 ibid Art 126.

Scope: Types of Security Rights 221 an agreement. The pledge may be over existing and/or future assets, and may secure one or more present or future obligations provided that they are determined or determinable. For the pledge of future receivables, the security agreement must describe them so as to enable them to be identified individually. Article 127 provides the matter which can be used to achieve this, such as the identification of the receivables debtor, the place of payment, the amount payable and/or the maturity date. The Model Law does not impose similarly onerous requirements as to the description of encumbered assets, and the criticism expressed above in connection with the equivalent requirement for the security assignment of receivables applies equally here to the pledge of receivables.61 Articles 130–35 provide rules for the operation of a pledge of receivables, dealing with such matters as the effect of notification to the account debtor (the person owing the receivable),62 and the effect of collection of the receivable by the secured creditor,63 both of which are very similar to the position in relation to a security transfer of a receivable discussed above.64 Chapter 5 includes rules specific to the following types of pledges of different types of intangible assets: (i) pledge of a bank account; (ii) pledge of partnership interests, securities and financial instruments accounts; (iii) pledge of intellectual property rights; and (iv) pledge of an enterprise and an enterprise seller’s lien. These pledges may be created not only contractually but also by a judicial order.65

a. Pledge of Bank Accounts The Uniform Act treats the pledge of bank accounts like a pledge of receivables.66 The rules governing the latter, outlined in the last paragraph, are applicable to the pledge of bank accounts, unless otherwise provided by Articles 137–39.67 Under Article 138, the parties can agree on conditions for allowing the pledgor to dispose of the funds deposited in the pledged account. The Model Law distinguishes between receivables and rights to payment of funds credited to bank accounts, providing distinct sets of rules on the third-party effectiveness, priority, enforcement and conflict of laws rules.68 This makes a great deal of sense because the two types of assets, though similar in that each are choses in action relating to money, are different in many other ways, especially the nature of the third party who owes an obligation to the pledgor, such as a financial institution in the case of a bank account and, typically, a trading partner of the pledgor in the case of a receivable.

b. Pledge of Partnership Rights, Securities and Financial Instruments These pledges were introduced in the 2011 revision of the Uniform Act. Generally, the pledge of partnership rights and securities is synonymous with the Model Law’s security right in non-intermediated securities, whereas the pledge of financial instruments

61 See

above B3. Act (n 4) Arts 132–33. 63 ibid Arts 134–35. 64 See above B3. 65 See, eg, Uniform Act (n 4) Art 142 on the court-ordered pledge of rights of partners and securities. 66 ibid Art 136. 67 ibid. 68 See ML, Art 2(dd) for the definition of ‘receivable’. 62 Uniform

222 OHADA Uniform Act on Securities is synonymous with the security right in intermediated securities excluded from the scope of the Model Law.69 Article 140 determines the type of partnership rights and securities that can be pledged. It provides: ‘Partnership rights, securities of commercial companies and assignable securities of other legal entities subject to registration in the Register of Commerce and Securities may be subject to a contractual or judicial pledge.’ In addition to the general requirements (eg a description of the pledged object), the pledge agreement must also identify the registered office and the registration number of the issuing entity, as stated in the Register of Commerce and Securities, as well as the quantity or the means to identify the quantity of the pledged rights or securities.70 These requirements go beyond those prescribed in Article 9 of the Model Law, which requires only sufficient information to identify the pledged assets, without specifying the particular means. In particular, it is sufficient under that article to describe the assets as ‘all the grantor’s movable assets’. That clearly would not be sufficient to comply with Article 141 of the Uniform Act. The former Uniform Act required a notification of the pledge to the entity that issued the partnership rights or securities, a requirement that was eliminated in the 2011 revision. Article 146 of the Uniform Act defines the pledge of financial instruments credited to accounts as an agreement whereby a person pledges financial instruments deposited in an account to secure the performance of an obligation. The financial instruments must be negotiable or transferable and may include shares, bonds and units in mutual investment organisations.71 The creation of this pledge requires the issuance of a form (déclaration) signed and dated by the pledgor who is the holder of the securities account.72 The déclaration must identify the pledged financial instruments by their number and nature and must also identify the special pledge account73 to which the securities will be credited.74 As with the pledge of a bank account discussed above, the secured creditor and the holder of the pledged account can prescribe the conditions under which the holder may dispose of the financial instruments credited to the account.75 Once the intermediary that maintains the securities account is informed of and accepts the conditions under which the pledgor is permitted to dispose of the financial instruments, it will have a duty to ensure the satisfaction of those conditions.76 This type of pledge is one of the two security rights dealt with in the Uniform Act which cannot be made effective against third parties by registration.77 Thus, the requirements just described take the place of a requirement to achieve third-party effectiveness. They can be contrasted with the provisions related to third-party 69 ML, Art 1(3)(c). 70 Uniform Act (n 4) Art 141. 71 AI Bah, ‘Le nantissement de compte de titres financiers’ [‘Accountability of Financial Securities’] (14 January 2017) www.coursdedroit.wordpress.com/2017/01/14/le-nantissement-de-compte-de-titres-financiers/; Actualités du droit OHADA, ‘Droit des sûretés: Nantissement: Le nantissement de meubles incorporels’ [‘Collateral Law: Pledge: Pledging Intangible Property’] (21 October 2015) www.actualitesdroitohada.com/droit-des-suretes/ nantissement/le-nantissement-de-meubles-incorporels/le-nantissement-de-compte-de-titres-financiers. 72 Uniform Act (n 4) Art 147 para 1. 73 ibid Art 149 mandates that the pledged account must be a special account opened in the name of the holder (the pledgor) and maintained either by the issuing legal entity or an intermediary. 74 ibid Art 147. 75 ibid Art 151 para 1; see also O Fille-Lambie and A Marceau-Cotte, ‘Les sûretés sur les meubles incorporels: Le nouveau nantissement de l’Acte Uniforme sure les sûretés’ [‘Collateral on Intangible Property: The New Pledge of the Uniform Security Act’] (2010) 77 Revue droit et patrimoine 197. 76 Uniform Act (n 4) Art 151 para 2. 77 This is evident from the fact that there is no mention of registration in ibid ch 5 (Arts 86–90).

Scope: Types of Security Rights 223 effectiveness under the UNIDROIT Convention on Substantive Rules for Intermediated Securities (‘the Geneva Securities Convention’).78 Article 12 of that Convention gives three possible means of granting, and making effective against third parties, a security interest in intermediated securities: (a) by the security taker being the intermediary with whom the securities account is held, (b) by a designating entry in favour of the security taker, or (c) by a control agreement between the security taker, the security provider and the intermediary. The conditions specified in paragraphs (b) and (c) are both means whereby the intermediary is not permitted to comply with instructions given by the security provider without the consent of the security taker, and/or the intermediary must follow the instructions of the security taker, but before that point the intermediary can follow the instructions of the security provider.

c. Pledge of Intellectual Property Rights Under the former Uniform Act, intellectual property rights could be pledged only as part of the enterprise.79 The 2011 revision introduced the pledge of intellectual property rights as a stand-alone security device.80 The pledge may cover any patents, trademarks, designs and models, and can be created contractually or by a judicial order.81 Security rights in intellectual property rights can be created under the Model Law but only to the extent that the provisions of the security agreement are not inconsistent with the law relating to intellectual property of the enacting State.82

d. Pledge of the Enterprise and Lien of the Seller of an Enterprise Section 5 of Chapter 5 deals with the pledge of an enterprise and the lien of the seller of an enterprise. The Model Law does not contemplate similar security devices, although it does permit a security right to be granted over all the assets of a business, as well as over more limited classes of assets. It is thus possible to grant a security right over the same types of assets as can be included in the pledge of an enterprise.83 The enterprise (fonds de commerce) is defined in Article 135 of the Uniform Act on General Commercial Law (UAGCL) as ‘a set of resources that allow the merchant to attract and retain customers’.84 The UAGCL also provides an illustrative list of tangible and intangible assets that may constitute an enterprise, including, but not limited to, installations and fixtures, equipment, furniture, inventory, leasehold to the immovable property, and

78 See UNIDROIT, ‘UNIDROIT Convention on Substantive Rules for Intermediated Securities’ (9 October 2009). The Model Law does not cover intermediated securities. 79 Fille-Lambie and Marceau-Cotte (n 73). 80 Uniform Act (n 4) Arts 156–61. 81 ibid Art 156. 82 ML, Art 1(3)(b). 83 See ch 2. 84 OHADA, ‘Acte uniforme revise portant sur le droit commercial general’ [‘Uniform Act on General Commercial Law’] (UAGCL) (15 December 2010), www.ohada.com/actes-uniformes/940/1041/titre-2-fonds-de-commerce. html; Art 135 provides: ‘Le fonds de commerce est constitué par un ensemble de moyens qui permettent au commerçant d’attirer et de conserver une clientèle’ [‘The enterprise shall consist of a set of resources that allow the merchant to attract and retain customers’]. For an official English translation of the UAGCL, see OHADA (n 4).

224 OHADA Uniform Act on Securities intellectual property rights associated with the business.85 Accordingly, an enterprise may be composed of immovable and movable assets, which may be both tangible and intangible. Article 162 of the Uniform Act, which provides for the enterprise pledge, is much narrower in terms of the components of the enterprise that it extends to. As a general rule, the enterprise pledge extends only to the list of customers and the trade name of the enterprise.86 Only if the parties agree, and specifically provide so in a security agreement, will the enterprise pledge extend to the other intangible components of the enterprise, such as intellectual property rights and operating licences.87 Article 162 also permits the extension of the enterprise pledge to equipment. However, in order to have a security right over the inventory or receivables of the enterprise, the secured creditor must take a separate pledge of inventory or receivables under the Uniform Act. The enterprise pledge may also encumber the enterprise of the pledgor’s subsidiaries, in which case the subsidiaries must be specifically identified in the security agreement.88 A lien of the seller of the enterprise secures the payment of the purchase price with respect to the sale of an enterprise that the seller financed. It must be registered to become effective against third parties.89 Notably, the request for registration must be submitted by the purchaser of the business.

7. Liens Chapter 6 of the Uniform Act governs liens, which are preferential rights granted by law, as opposed to consensual security rights.90 The Uniform Act recognises general and special liens.91 The Uniform Act contemplates that liens may arise under other law, which may also determine their priority. In the absence of a specific priority rule, the liens arising under other law would rank last in the priority order prescribed by Article 180 of the Uniform Act.92 The Model Law does not contemplate a similar treatment of the priority of non-consensual interests (liens). General liens may extend to any movable and immovable assets and are effective against third parties without registration.93 Article 180 sets out a list of six liens that must be satisfied in the order they appear in the list: (i) burial expenses and any medical expenses of the 85 ibid Art 137 provides: ‘Le fonds de commerce peut comprendre différents éléments mobiliers, corporels et incorporels, notamment les éléments suivants: les installations; les aménagements et agencements; le m atériel; le mobilier; les marchandises en stock; le droit au bail; les licences d’exploitation; les brevets d’inventions, marques de fabrique et de commerce, dessins et modèles, et tout autre droit de propriété intellectuelle nécessaires à l’exploitation’ [‘The enterprise can consist of different movable, tangible and intangible elements, including the following: facilities; facilities and fixtures; equipment; furniture; goods in stock; right to lease; operating licences; patents, brand names and trademarks, drawings and models, and any other intellectual property right necessary to the operation’]. 86 ibid. 87 ibid Art 162 para 2. 88 ibid Art 162 para 5. 89 Uniform Act (n 4) Art 166. 90 Y Kalieu Elongo, Droit et pratique des sûretés réelles OHADA [The Law and Practice of Real Security Interests] (Yaoundé, Cameroun, Presses Universitaires d’Afrique, 2010) 153. 91 Uniform Act (n 4) Arts 179–89. See discussion of both types of liens, see G below. 92 Uniform Act (n 4) Art 179. 93 ibid Art 180.

Third-Party Effectiveness 225 grantor prior to the seizure of assets; (ii) minimum living expenses of the grantor in the last year; (iii) wages owed to employees during the last year; (iv) debts owed to authors of intellectual, literary and artistic works for the last three years; (v) debts owed to social security and welfare agencies; and (vi) debts owed to tax and customs authorities. In addition, liens of welfare, tax and customs agencies for debts owed beyond the amounts prescribed in court decisions are effective against third parties only if they have been registered pursuant to Article 181. With regard to regulating the priority of liens, the Uniform Act goes beyond Article 36 of the Model Law, which recommends that enacting States set out in the enacted law the various preferential claims and their amounts.94 Special liens are given to the seller of a movable asset, the lessor of an immovable as against the furniture installed in the leased premises, the land carrier, contractors, suppliers of services, brokers, and any person who incurs expenses or provides services to preserve an asset. The lien is over the movable asset in question and relates just to the amount owing in relation to the relevant transaction. Thus, for example, the lien of a seller of a movable asset is over that asset for the unpaid price,95 and the lien of a land carrier is over the property transported for all that is due to him in relation to that property.96

C. Third-Party Effectiveness Unlike Chapter III of the Model Law, the Uniform Act does not include a stand-alone chapter on third-party effectiveness. The provisions that regulate the individual pledges and the various ‘ownership-security’ devices specify how these encumbrances may be made effective against third parties. With the exception of the pledge of financial instruments credited to accounts and fiduciary transfer of funds,97 all pledges, reservation of ownership and security transfer of ownership may (and in some cases must) be made effective against third parties by registration. For instance, Article 74 of the Uniform Act expressly provides that ‘reservation of ownership is effective against third parties only if it has been properly registered in accordance with articles 51–61’. A fiduciary transfer of funds becomes effective against third parties when the funds are credited to a pledge bank account, and pledge of financial instruments is effective when the instruments are credited to a pledge account.98 Pledges of tangible assets can also be made effective against third parties by transferring possession to the secured creditor.99 The Uniform Act does not provide for automatic third-party effectiveness for any type of pledges, differing from Article 24 of the Model Law that contemplates automatic third-party effectiveness for acquisition security rights in consumer goods. Registration and possession are therefore the general m ethods of third-party effectiveness under the Uniform Act. Control is available only for the pledge of financial instruments and the fiduciary transfer of funds, both of which require their

94 See also conclusion D1b. 95 Uniform Act (n 4) 183. 96 ibid Art 185. 97 See discussion of fiduciary transfer of funds, B4 above, discussion of pledging financial instruments credited to accounts, B6b above. 98 See discussion of pledging partnership rights, securities, and financial instruments, B6b above. 99 Uniform Act (n 4) Art 97.

226 OHADA Uniform Act on Securities credit to a pledge account. Unlike Article 25 of the Model Law, which sets out three possible methods of third-party effectiveness as an alternative to registration,100 the only method of third-party effectiveness for the pledge of bank accounts is registration. The Uniform Act eliminated certain requirements necessary for third-party effectiveness under the former Uniform Act. For instance, the general requirement to register the pledge agreement with the tax authority no longer applies.101 However, since taxation falls outside OHADA’s legislative powers, the Uniform Act may not preclude Member States from levying taxes on registrations relating to security interests.102 Furthermore, notification of the receivables debtor is no longer a condition for creating and making a pledge over those receivables effective against third parties, but rather a requirement to make the pledge effective against the receivables debtor who may be otherwise discharged by paying the pledgor.103 Article 63 of the Model Law takes the same approach. However, certain requirements have been preserved, but without any or with uncertain legal effect. For motor vehicles, the Uniform Act requires the pledge to be indicated on the administrative title (registration document), in addition to registering the pledge in the Register of Commerce and Securities.104 However, the absence of the pledge reference on the administrative title has no impact on the effectiveness of a pledge that has been properly registered in the Register of Commerce and Securities, and whether it has any other effect is unclear.105 For the pledge of intellectual property rights, registration in the Register of Commerce and Securities, as well as in any other register applicable to intellectual property, is required.106 The effect of a failure to register the pledge in an intellectual property register is unclear: this should be addressed in the relevant law relating to intellectual property. Article 143, which governs the pledge of partnership rights and securities, provides that the 100 ML, Art 25; for detail see ch 2D. 101 O Fille-Lambie and A Marceau-Cotte, ‘Les sûretés sur les meubles incorporels: Le nouveau nantissement de l’Acte Uniforme sure les sûretés’ [‘Securities in Intangible Property: The New Pledge of the Uniform Security Act’] [1 November 2010] Droit & Patrimoine 197. 102 IFC & OHADA Permanent Secretariat, ‘An Impact Assessment of OHADA Reforms: Uniform Acts on Commercial, Company, Secured Transactions, and Insolvency’ (IFC’s OHADA Investment Climate Program (2007–2017) 2018) 25. 103 Uniform Act (n 4) Art 132(2). 104 ibid Art 119: ‘En ce qui concerne les véhicules automobiles assujettis à une déclaration de mise en circulation et à immatriculation administrative, le gage doit être mentionné sur le titre administratif portant autorisation de circuler et immatriculation. L’absence de cette mention ne remet pas en cause la validité ou l’opposabilité du gage dûment inscrit au Registre du Commerce et du Crédit Mobilier.’ [‘In the case of motor vehicles subject to a declaration of registration and administrative registration, the pledge must be mentioned on the administrative title bearing the authorization to travel and to register. The absence of this declaration does not call into question the validity or opposability of pledges duly registered in the Trade and Credit Register.’] 105 ibid. 106 ibid Art 160: ‘Le nantissement conventionnel ou judiciaire n’est opposable aux tiers dans la mesure et selon les conditions prévues par les articles 51 à 66 du présent Acte uniforme que s’il est inscrit au Registre du Commerce et du Crédit Mobilier. L’inscription provisoire et l’inscription définitive doivent être prises, respectivement, après la décision autorisant le nantissement et la décision de validation passée en force de chose jugée. Si le nantissement a pour objet un droit inscrit sur l’un des registres régis par la réglementation applicable en matière de propriété intellectuelle, il doit, en outre, être satisfait aux règles de publicité prévues par cette règlementation.’ [‘A conventional or judicial pledge shall not be binding on third parties to the extent and under the conditions provided for in Articles 51 to 66 of this Uniform Act unless it is registered in the Trade and Personal Property Credit Register. The provisional and definitive registrations must be taken, in that order, after the decision authorizing the pledge and the decision of validation become res judicata. If the pledge relates to a right registered in one of the registers governed by the applicable rules on intellectual property, it must, in addition, follow the notice rules provided for by this regulation.’]

Registration of Security Rights in the Register of Commerce and Securities 227 pledge may be notified to the company that issued the partnership rights and securities. As with the preceding types of pledges, the Uniform Act does not specify any consequences for the failure to effect such a notification, which in this case appears to be optional.

D. Registration of Security Rights in the Register of Commerce and Securities 1. General Overview The Uniform Act contemplates the registration of pledges and other security devices in the Register of Commerce and Securities (Le Registre du Commerce et du Crédit Mobilier, RCCM).107 Articles 51–66, which deal with registration, are much less comprehensive than the Model Registry Provisions of the Model Law. Many aspects of the registration process are governed by the UAGCL, specifically its Book II.108 As noted above, the rules regulating the individual security devices of the Uniform Act also include provisions relating to registration of those devices. In addition, Articles 174–78 contain some common provisions on various effects of registration, which go beyond the Model Registry Provisions of the Model Law and do not appear to have any or only a remote relation to registration. Article 174 deals with the power of unsecured creditors to petition the court to satisfy their debts in the case of sale or liquidation of a business which is not related at all to registration. Article 176 provides for the duty of the lessor of the building in which the business subjected to a registered security right operates to notify the secured creditor before terminating the lease. Accordingly, the potential registrant needs to review Book II of the UAGCL, Articles 51–66 and 174–78 of the Uniform Act as well as the relevant articles relating to the type of security device in order to find out information about registration and its effects. Such presentation of the registration provisions is cumbersome compared to that of the Model Law. The Register is not a typical registry of security rights as envisaged by the Model Law; rather, it is a registry of businesses that, inter alia, also accepts registrations relating to security rights.109 Other than organisations, any natural person that carries on some business activity or any merchant may be registered in the Register.110 The Register also accepts registrations of pledges, other security devices and financial leases granted by natural persons who are not merchants.111 Under Article 34 of the UAGCL, the Register must process registrations relating to finance leases and security rights created under the Uniform Act or other legislation, and make them available to the public.112 The UAGCL does not provide any rules on finance leases, other than those relating to their registration. Any of the secured creditor, trustee or grantor of a registrable security device can request registration.113 107 ibid Arts 51–66. 108 UAGCL (n 81) Arts 34–100. 109 For more details, see Eversheds, Business Law in Africa: OHADA and the Harmonization Process (London, Kogan Page, 2002) 33–34. In this regard, the Register is similar to the Register of Company Charges in common law jurisdictions, see UK Companies Act, s 859A. 110 UAGCL (n 81) Art 35. 111 ibid Art 38(3). 112 ibid Art 34. 113 Uniform Act (n 4) Art 51.

228 OHADA Uniform Act on Securities The Register is maintained by the clerk of the competent court, or by any other competent body designated by the Member State.114 Each OHADA Member State is required to establish its own national register, which would be supervised by a judge or some other competent authority.115 Registration systems in some OHADA Member States have not been established, primarily due to the lack of resources, and in others they are limited to the registration of companies. A State may have multiple national registers, but is required to centralise the registered information in a single database. For instance, Cameroon has 120 registers spread across different localities for registering security rights, none of which provides for electronic registration.116 The information from the national registers of Member States must then be centralised in the ‘OHADA Register’ maintained by the Common Court of Justice and Arbitration.117 The former Uniform Act provided for the same structure of the registration system.118

2. Requirements for the Initial Registration In addition to the Uniform Act, a number of requirements regarding the registration process are set out in the UAGCL. Those requirements apply to the registration of finance leases and other security rights not governed by the Uniform Act. Article 40 of the UAGCL provides that the application ‘shall be made on the form provided for that purpose by the clerk or the competent body in the Member State’. Although permitted, electronic registers are not yet the norm in the OHADA Member States.119 While some attempts at providing electronic functionality have been made in Ivory Coast and Senegal, the maintenance of ‘paper records’ complicates the centralisation of information in the respective national registers and the ‘OHADA Register’. An application for registration must be signed by the grantor and the secured creditor.120 Under Article 53 of the Uniform Act, the registration form must indicate the following: (i) the full names of the grantor and the secured creditor; (ii) the trade names of the grantor and the secured creditor; (iii) the domicile or registered office; (iv) the registration number or the business activity of the secured creditor or its security agent; (v) the grantor, and the debtor 114 UAGCL (n 81) Art 36 para 1. 115 ibid. 116 IFC & OHADA (n 98) 25. 117 UAGCL (n 81) Art 20 paras 2–3: ‘Un Fichier National centralise les renseignements consignés dans chaque Registre du Commerce et du Crédit Mobilier. Un Fichier Régional, tenu auprès de la Cour Commune de Justice et d’Arbitrage, centralise les renseignements consignés dans chaque Fichier National.’ [‘Information entered in each Trade and Personal Property Rights Registry shall be centralised in a National Register. Information entered in each National Register shall be centralised in a Regional Register kept at the Common Court of Justice and Arbitration.’] 118 PR Wood, Comparative Law of Security Interests and Title Finance, The Law and Practice of International Finance Series vol 2 (London, Thomson, Sweet & Maxwell, 2007) 23–94. 119 There is no official website for the Commercial Register for Securities in Movables (Registre du Commerce et du Crédit Mobilier, RCCM) in the OHADA region. In 2016, the OHADA launched an international tender notice in order to set up the electronic database. Permanent Secretariat of OHADA, ‘IDA Grant No H7990: International Tender Notice No 01/SP-OHADA/PACI/2016/AOI: Supply of Computer Tools for the Deployment of Integrated Software for the Management of National Trade and Securities Register (RCCM) Files in OHADA Countries’ (2016) www.ohada.org/index.php/en/news/ latest-news/1560-international-tender-notice-supply-of-computer-tools-for-the-deployment-of-rccm-ohada. 120 UAGCL (n 81) Art 51 para 1.

Registration of Security Rights in the Register of Commerce and Securities 229 if different from the grantor; (vi) the duration of the registration; (vii) the maximum amount of the secured obligation including the principal, interest, and other fees; and (viii) the due date of the secured obligation. The collateral must be identified by ‘its nature, location, and where applicable, its brand or serial number’. These requirements are entirely inconsistent with the Model Law,121 unnecessarily increasing the parties’ burden when effecting an initial registration. Some of the requirements are also ambiguous. For instance, it is not clear when the description by serial number (eg relating to a motor vehicle) is applicable, and what the third-party effectiveness and priority consequences of the failure to include a serial number are. If the encumbered assets include receivables, they must be described in a manner that allows their identification by the name of the receivables debtor, the place of payment, the amount and the due date.122 Such requirements make receivables financing very cumbersome and costly, particularly when the secured creditor takes a pledge over a pool of receivables or when the receivables regularly fluctuate, both instances being common in asset-based loans. With respect to a pledge over partners’ rights and securities, the form must also contain the registration number of the company whose partners’ rights and securities are the subject of the pledge,123 notwithstanding that the information is arguably immaterial to a searcher interested in taking such rights and securities as collateral. The Register clerk must ensure that the application is complete and verify the veracity of the statements against the submitted supporting documents.124 Both this verification function and the requirement to submit supporting documents along with the registration form are inconsistent with notice registries of the kind envisaged in the Model Law as well as with the criteria measured in the World Bank’s Doing Business Report.125 If a registration is rejected, the decision must be communicated not only to the person who submitted the form, but also to the grantor.126 A decision of the Register clerk to reject a registration application may be appealed within eight days of the date of the notification. A decision of the first-instance court or other competent authority may be appealed within 15 days of the date of notification of the decision.127

121 MRP, Art 8. 122 Uniform Act (n 4) Art 53 para 2: ‘Lorsque la sûreté a pour objet une créance ou un ensemble de créances, actuelles ou futures, la désignation du ou des biens grevés requiert l’indication des éléments de nature à permettre l’individualisation de cette ou de ces créances, tels que l’indication du débiteur, le lieu de paiement, le montant des créances ou leur évaluation et leur échéance.’ [‘Where the object of the security is a debt or a series of actual or future debts, the designation of the encumbered asset or assets shall indicate the elements naturally leading to the identification of the said debt or debts such as, the indication as to the debtor, the place of settlement, the amount of the debts or their valuation and date of settlement.’] 123 ibid Art 53 para 3: ‘En cas de nantissement des droits d’associés et valeurs mobilières d’une société commerciale et ceux cessibles de toute autre personne morale, le formulaire porte, en outre, mention du numéro d’immatriculation de la société dont les droits d’associés et valeurs mobilières font l’objet de ce nantissement.’ [‘In the case of a pledge of partnership rights and transferable securities of a trading company and those which are transferable from any other legal person, the form shall in addition to the aforementioned information, contain the registration number of the firm whose partnership rights and transferable securities are pledged.’] 124 UAGCL (n 81) Art 66 para 1. 125 See ‘Strength of Legal Rights’ at: World Bank, ‘Getting Credit Methodology’ (2019) www.doingbusiness.org/ Methodology/Getting-Credit, explaining how state Registry evaluations are handled under the Doing Business Report, which allots additional points to countries with a secured transactions law that provides for a notice-based collateral registry, ie a registry that files only a notice of the existence of a security interest (not the underlying documents) and does not perform a legal review of the transaction. See also ch 2, B2bii and E1. 126 Uniform Act (n 4) Art 54; cf Model Registry Provision (n 117) Art 6(4). 127 ibid.

230 OHADA Uniform Act on Securities A security right is effective against third parties from the date of its registration, provided the creation requirements have been satisfied.128 Advance registration that precedes the execution of a security agreement or creation of a security right is not possible under the Uniform Act given the requirement to submit supporting documentation for verification by the Register clerk. Article 57 sets out the priority rule between two security rights registered on the same day, which shall rank equally irrespective of the actual time of registration. This rule presupposes that local registers do not keep accurate records of the actual time of registration. There is an exception to the rule for a reservation of ownership or security transfer of ownership, which is accorded priority over a pledge registered on the same day.129 The Model Law does not contain such rules, since it contemplates an electronic registry that records the time of registration by the second, as is customary in modern electronic registry design. The parties may agree on the duration of the registration in the security agreement, but it must not exceed 10 years.130 This approach corresponds to one of the three options under the Model Law.131 The registration of liens of the treasury, customs administration, and social security institutions is effective for three years.132 The effectiveness of the registration may be extended before its expiration.

3. Amendments, Discharges and Searches Any modification to the registered information may be effected by an amendment entered on the margins of the initial registration.133 In addition to voluntary amendments, an affected person, such as the grantor, may at any time request a competent court or other authority to order the discharge or amendment of the registration.134 A registration may also be consensually discharged upon delivery to the Register of a discharge (cancellation) form, accompanied by a notarised deed or a privately signed deed evidencing the consent of the secured creditor or its agent to the cancellation.135 Article 64 of the Uniform Act expressly requires the Register clerk to verify that the content of the cancellation form c orresponds

128 Uniform Act (n 4) Art 57 para 1: ‘L’inscription régulièrement prise d’une sûreté mobilière soumise à publicité est opposable aux tiers à la date de son inscription au registre chronologique des dépôts du Registre du Commerce et du Crédit Mobilier.’ [‘Any duly registered transferable security subjected to publicity shall be binding on any third party from the date of entry in the chronological register of the Trade and Personal Property Right Register.’] 129 ibid Art 57 para 2: ‘Si les inscriptions de sûretés concurrentes grevant un même bien sont requises le même jour, celle qui est requise en vertu du titre dont la date est la plus ancienne est réputée avoir été inscrite en premier, quel que soit l’ordre du registre susvisé.’ [‘Where the registration of competing securities encumbering one and the same asset are due on the same date, the entry with the earliest date shall be deemed to have been registered first irrespective of the order of the register mentioned above.’] 130 ibid Art 58 para 2: ‘Pour les autres sûretés mobilières soumises à publicité, les parties peuvent convenir de la durée de validité de l’inscription au Registre du Commerce et du Crédit Mobilier dans l’acte constitutif de ladite sûreté mobilière sans que cette durée puisse dépasser dix années à compter de l’inscription.’ [‘As for the other transferable securities subject to publicity, the parties may in the deed establishing the said security agree on the validity period of registration in the Trade and Personal Property Right Register not exceeding ten years.’] 131 See MRP, Art 14 Option C. 132 Uniform Act (n 4) Art 58 para 1. 133 ibid Art 60 para 1. 134 ibid Art 61 para 1. 135 ibid Art 64 para 1.

Priority 231 to the deed. These requirements are cumbersome as compared to the Model Law, but minimise the risk of unauthorised cancellations.136 The response to any search request must be provided immediately or within two working days from the receipt of the request. The registry clerk must issue a certificate stating that no registration has been found against the name of the grantor, or a general statement of the existing entries, including any amendment entered on the margins, or one or several special statements when the request relates to an asset or a class of assets belonging to the grantor.137 Accordingly, one can submit a search request for all registrations against Company ABC or limit it only to registrations against Company ABC that cover inventory or some other type of encumbered asset. The Uniform Act does not provide for indexing and searching of registrations against serial numbers: they must be indicated in a registration form where applicable merely as a form of identification.138 Also, the Uniform Act does not include a rule on the effect of errors in the registered information. The lack of such a rule creates a legal risk as the question of whether a particular error or omission renders a registration ineffective can, therefore, only be determined by a court.139 One might argue that the risk of errors is lower under the Uniform Act, as compared to the Model Law, given the verification function of the Register clerk. However, the Model Law provides for a noticefiling system that requires much less information to be entered in a registration form than under the Uniform Act, which is also a feature that minimises such risks.

E. Priority The Uniform Act does not contain a stand-alone priority chapter similar to Chapter V of the Model Law. Under Article 57, registered security devices become effective against third parties from the date of their registration. Article 57 also provides some rules allocating the priority between security devices registered on the same day. The sections that deal with the individual types of pledges and ‘ownership security’ devices provide priority rules applicable to conflicts between those types of encumbrances. For instance, Article 107 provides that the order of priority of more than one non-possessory pledge over tangible assets is determined by the order of registration. An earlier ‘registered pledge’ also prevails over any later-in-time pledge made effective against third parties by the transfer of possession to the secured creditor, and vice versa.140 However, the Uniform Act does not provide specific priority rules for competing claims in an asset encumbered by different security devices. For instance, receivables may be pledged to Creditor A on day 1 and subsequently transferred as security to Creditor B on day 5. Although the Uniform Act does not provide a formula for resolving the ensuing priority conflict, the most reasonable interpretation is that the date of registration of the pledge and the security transfer of receivables should determine their respective priority. This is because on day 5 the pledgor could not have transferred to Creditor B a right in the receivables free of the pledge interest of Creditor A. The more

136 cf

MRP, Art 5(2). Act (n 4) Art 66 paras 1–2. 138 See discussion of applicable Uniform Act registration form requirements, D2 above. 139 cf MRP, Art 24. 140 Uniform Act (n 4) Art 107. 137 Uniform

232 OHADA Uniform Act on Securities difficult situation is a conflict where Creditor A takes a pledge of receivables on day 1 that the borrower collects on day 30 and deposits into a bank account. On day 35, the borrower enters into a fiduciary transfer of funds arrangement with Creditor B, and the funds are transferred into a blocked bank account. The Uniform Act does not provide a clear answer, but since the pledge of receivables does not extend to a bank account as proceeds, Creditor B will have priority.

F. Rights and Obligations of the Parties As in Chapter VI of the Model Law, the Uniform Act sets out some rights and obligations of parties to a secured transaction as well as of persons (eg debtors of receivables) affected by it. In many aspects, the content of the Uniform Act rules differs from the far more comprehensive Chapter VI of the Model Law. For instance, under Article 102 of the Uniform Act the secured creditor may, in a security agreement or separately, authorise the grantor to dispose of fungible goods, provided that the grantor replaces them with equivalent assets of the same quantity. The Model Law does not similarly limit the contractual freedom of the parties with regard to fungible goods, but permits the secured creditor to authorise a sale or other transfer of any encumbered asset free of the security right.141 In a possessory pledge, the Uniform Act does not allow the secured creditor to use the asset, unless otherwise provided by the agreement.142 The secured creditor or the third party in possession of the asset under a possessory pledge is required to keep and preserve the asset as a paid custodian143 although, unlike the Model Law,144 the Uniform Act does not provide a general duty for the parties to act in a commercially reasonable manner. If the secured creditor or the third party fails to perform the obligation to keep and preserve the asset, the grantor may demand the restitution of the pledged asset as well as damages.145 The grantor must reimburse the secured creditor or the third party for any necessary expenses incurred for the preservation of the encumbered asset.146 Article 55(1)(b) of the Model Law allows the secured creditor to use the encumbered asset reasonably and request reimbursement for any reasonable expenses incurred for its preservation. Under a non-possessory pledge, the grantor is responsible for preserving the asset and procuring insurance against risks of loss and damage.147 If the grantor fails to preserve the encumbered asset, the secured creditor can request immediate satisfaction of the secured obligation or provision of additional collateral.148 The Model Law leaves any other consequences of the grantor’s failure to preserve the encumbered asset to the agreement between the parties. The Uniform Act also makes provisions for certain rights and duties with respect to intangible assets. For instance, the pledgor/transferor of receivables cannot, without the

141 ML,

Art 34(2). Act (n 4) Art 103. 143 ibid Art 108 para 1. 144 ML, Art 4. 145 Uniform Act (n 4) Art 109 para 1. 146 ibid Art 113. 147 ibid Art 108 para 2. 148 ibid Art 109 para 2. 142 Uniform

Enforcement of Security Rights 233 consent of the pledgee/assignee, modify the terms and conditions of the receivables.149 A notification of the receivables debtor is not required to make the assignment effective against third parties but, in the absence of the notification, the receivables debtor may discharge its obligation by paying the pledgor/assignor.150 These provisions are consistent with Chapter VI of the Model Law.

G. Enforcement of Security Rights The Uniform Act provides enforcement remedies for the individual security devices. Accordingly, the secured creditor’s remedy may vary depending on the type of encumbrance it has taken. As an example, under a sale with reservation of ownership, if the debtor defaults in payment, the seller may terminate the agreement and request the return of the asset.151 Specific enforcement rules are also provided for non-consensual security devices. For instance, according to Article 67 of the Uniform Act, a creditor can retain possession of an asset until full payment of the debt. The Model Law’s unitary and functional approach results in a single body of enforcement rules applicable to security rights generally, whether possessory or non-possessory, of an acquisition or non-acquisition nature, and so on. Its simplicity contrasts with the security device-specific approach of the Uniform Act which requires familiarity with different types of security devices to apply the appropriate and permissible remedy.152 The Uniform Act contains several rules that correspond to the approach of the Model Law but, for the most part, the approach of the Uniform Act deviates from the Model Law. One example of the former is Article 107 paragraph 2, under which the right of the secured creditor to enforce its security may be suspended if a secured creditor that has priority in the same asset seeks to enforce its rights.153 This approach corresponds with Article 76 of the Model Law, which empowers a higher-ranking secured creditor to take over the enforcement process. The Uniform Act also contains some rules not found in the Model Law, such as the requirement that the pledgee notify the lessor of the immovable where the enterprise is operated.154 If properly notified, the lessor will have a duty to notify the secured creditor if the lease is to be terminated. The fundamental difference between the Uniform Act and the Model Law lies in the ability to enforce security rights extrajudicially. Generally, the Uniform Act requires enforcement to be effected pursuant to a court order, whereas the Model Law promotes efficient enforcement remedies,155 including by providing a comprehensive set of extrajudicial enforcement mechanisms, although judicial enforcement is also possible according to domestic law. However, the Uniform Act provides certain exemptions from the requirement to proceed judicially, such as for the fiduciary transfer of

149 ibid

Art 82. Art 132; see also ibid Art 84 (for security assignments of receivables). 151 ibid Art 77. 152 Conclusion B2aii. 153 Uniform Act (n 4) Art 107 para 2. 154 ibid Art 172. 155 See LG introduction (paras 16–18). 150 ibid

234 OHADA Uniform Act on Securities money.156 These exemptions are too few to satisfy the creditors’ expectations to be able to expeditiously enforce their rights in encumbered assets. As will be shown in the following paragraphs, the requirements to enforce rights via judicial means are onerous and costly. Article 104 of the Uniform Act provides that after default a pledgee who has obtained a writ of enforcement can proceed to sell the pledged asset.157 The sale must not occur before the expiration of eight days after sending a notice to the grantor, and the debtor if different from the grantor.158 An expert must determine the value of the asset, unless it is traded in a public market.159 An expert valuation is not a requirement to dispose of an encumbered asset under the Model Law, as such valuations tend to delay the completion of enforcement and increase its cost. In addition to disposal of the pledged asset, the parties can agree to transfer ownership to the secured creditor.160 This remedy is available only for those assets whose value may be determined through public market quotations.161 However, the secured creditor must obtain a court order to acquire the encumbered asset in satisfaction of the obligation.162 Article 80 of the Model Law also provides for the acquisition of an encumbered asset in full or partial satisfaction of the secured obligation, but the exercise of this remedy under the Model Law is not limited to those assets whose value may be established by public market quotations and, like the other remedies, it does not require an order of the court. Enforcement rules for intangible assets vary according to the type of asset. Regarding a pledge of receivables, if the receivables are payable before the secured obligation becomes due, the secured creditor must deposit the sums collected into a specific bank account and hold them as security until the maturity of the secured obligation.163 If the grantor does not perform its obligation, the sums collected may be applied to the secured obligation.164 If the grantor performs the secured obligation, the secured creditor must return the funds to the grantor.165 The Model Law does not impose a similar limitation on the collection rights of secured creditors and outright transferees.166 Such rules are not reflective of contemporary financing practices, where the grantor may collect the receivables and use the proceeds, or the sums collected by the secured creditor increase the availability of funds for drawing under the credit facility to the extent the grantor meets certain financial ratio requirements set out in the security or loan agreement. In a pledge of financial instruments, the secured creditor must notify the grantor, as well as the holder of the account to which the financial instruments are credited, of the debtor’s default.167 The notice must contain mandatory information, the omission of

156 Uniform 157 ibid

158 ibid. 159 ibid.

Act (n 4) Art 91. Art 104 para 1.

160 ibid

Art 104 para 3.

162 ibid

Art 104. Art 134 para 1.

161 ibid. 163 ibid

164 ibid. 165 ibid. 166 ML

Arts 82–83. Act (n 4) Art 152.

167 Uniform

Enforcement of Security Rights 235 which will render the notice invalid.168 The secured creditor must allow a grace period of eight days for the debtor to cure the default. This grace period is also generally applicable to the enforcement of other types of security rights.169 However, if the securities are traded on the stock exchange and the pledgee is an authorised credit or financial institution, the pledgee may dispose of them without any formality and without having to provide a grace period to the debtor to cure the default.170 When the holder of the account is the secured creditor, Article 154 authorises the secured creditor to appropriate the funds and exchange-traded financial instruments credited to the account,171 or dispose of the exchange-traded instruments in a public market.172 As noted above, the Model Law does not apply to security rights in intermediated securities. The remedies available to secured creditors under the Uniform Act reflect the best practice, as set out in Article 33 of the Geneva Securities Convention.173 The distribution of proceeds of enforcement, and the creditors’ entitlement to claim such proceeds, are governed by Articles 224 and 226.174 If there is only a single secured creditor, the proceeds are applied to the satisfaction of the obligation owed to that secured creditor, and the grantor is entitled to receive any surplus. According to Article 226, proceeds obtained from the enforcement of a security right shall be distributed in the following order: (1) court fees incurred during the enforcement process; (2) preservation costs with respect to the encumbered asset; (3) salaries of employees which correspond to the amount that may not be seized by judgment creditors; (4) obligations of secured creditors according to their respective priority; (5) special liens; (6) general liens not subject to registration according to the order established by Article 180 of the Uniform Act; and (7) unsecured creditors in possession of an enforceable title. The Model Law requires the proceeds to be distributed in accordance with the priority scheme of the Model Law, even if the secured creditor proceeded judicially.175 Although the Model Law does not accord employee claims any preferential rights with respect to any distribution under Article 79, it allows enacting States to designate certain preferential claims to have priority over the rights of secured creditors. As such, even under the Model Law, salaries of employees may be satisfied ahead of the obligations owed to secured creditors, depending on what is designated by the enacting State.176

168 ibid

169 ibid. 170 ibid

Art 153.

Art 145. Art 154 para 1. 172 ibid Art 154 para 2. 173 See further B6b above. 174 Uniform Act (n 4) Art 225 deals with the proceeds of enforcement against immovable property. 175 ML, Art 79(1). 176 See ch 2, F4. 171 ibid

236 OHADA Uniform Act on Securities

H. Conflict of Laws and Transition Provisions The Uniform Act does not contain a comprehensive chapter delineating conflict of laws rules, unlike Chapter VIII of the Model Law. The only rule which determines some aspects of the law applicable to third-party effectiveness and priority is Article 52, which determines the competent local Register for the registration of a security device by the location of the grantor. Under Article 52, the Register of Commerce and Securities in the jurisdiction where the grantor is registered or, if the grantor does not operate as a registered company, the jurisdiction in which the registered office or the domicile of the grantor is located is competent to register security rights.177 This rule does not apply to the registration of a pledge or assignment of receivables; in such cases, the location where the receivables debtor is registered or, if the receivables debtor is not subject to a registration requirement, where its registered office or domicile is located, determines the competent register. This requirement, set out in Article 52 of the Uniform Act, is incompatible with the best international practice and renders a number of security rights and outright transfers of receivables (especially those where receivables owed by multiple debtors are encumbered or transferred outright) practically inoperative. The cost associated with multiple searches and registrations in paper-based registries that remain prevalent in the OHADA Member States may be significant, or even prohibitive. With respect to a pledge of partners’ rights and securities of a commercial company or a legal entity subject to registration, the competent register is one located where the company or legal entity that has issued the partners’ rights or securities is registered. Unlike Article 100 of the Model Law, the Uniform Act does not provide separate conflict of laws rules for equity and debt securities. As for the pledge of the enterprise or the right of an unpaid seller with respect to an enterprise, the competent register is where the owner of the enterprise is registered. The Uniform Act provides additional rules for the continued third-party effectiveness of an enterprise pledge after the relocation of the grantor. The owner of the enterprise is required to notify the secured creditor at least 15 days before any relocation.178 A secured creditor can refuse the relocation request if in its opinion the move will reduce the value of the collateral.179 Upon relocation, the secured creditor will retain its security right if it registers an amendment on the margins of the initial registration indicating its consent to the relocation within 15 days.180 Arguably, such an amendment does not provide sufficient notice to searchers of the register in the new location of the grantor who should enquire about any relocations of the grantor and search the registers in the previous locations of the grantor. This problem is eliminated if the State centralises the records of its individual registers in a single database. If the enterprise is relocated to another State, the initial registration will be entered in the Register of Commerce and Securities of the new location of the enterprise at the request of the secured creditor.181 Accordingly, the registration file will be effectively transferred to the Register of the State that the grantor has relocated to. However, unlike for intra

177 Uniform 178 ibid. 179 ibid

Act (n 4) Art 52 para 2.

Art 175 para 3. Art 175 para 4. 181 ibid Art 175 para 5. 180 ibid

Final Remarks 237 State transfers, the Uniform Act does not provide a time period within which the secured creditor must effect an amendment or new registration for inter-State transfers. The drafters of the OHADA Uniform Act might have expected that the ‘OHADA Register’ would be established relatively quickly, centralising all registrations, which would have eliminated the need for the rules on relocation of grantors between OHADA Member States. However, the failure to establish such Register and the absence of adequate conflict of laws rules addressing the legal effect of the movement of grantors within a particular Member State as well as within the OHADA region creates significant obstacles for potential creditors and other searchers interested in dealing with the grantor’s assets. As for transitional and final provisions, Article 227 provides that: This Uniform Act which repeals the Uniform Act of 17 April 1997 on the organization of securities shall apply only to securities granted or constituted after its entry into force. Any security granted or constituted or created prior to this Uniform Act and in conformity with the laws in force at the time shall remain subjected to the said laws until its extinction.182

The Uniform Act was published in the Official Journal of OHADA within 60 days of its adoption.183 It entered into force on 15 May 2011, 90 days from the date of its publication in the Official Gazette of OHADA (on 15 February 2011).184 Proper transition of prior security rights into a new regime is important to balance the interests of creditors holding prior security rights and those who contemplate taking security rights after the entry into force of a new regime. This balance is reflected in Chapter IX of the Model Law, which requires States to designate a period of time within which prior security rights must be brought into compliance with the requirements of the new regime.185 The effect of this rule is that secured creditors must register a notice or take other steps to make their prior security rights effective against third parties to continue the third-party effectiveness and priority of prior security rights. After the expiration of the transitional period, prospective secured creditors and buyers of assets must concern themselves only with the effect of the new regime. That is not the case under the Uniform Act, which forces prospective secured creditors and buyers to conduct due diligence, not only with respect to the asset being potentially encumbered under the Uniform Act, but also under its predecessor. Such extensive due diligence is costly and raises a number of legal risks.

I. Final Remarks The 2011 revision of the Uniform Act on Securities introduced a number of improvements over its predecessor adopted in 1997. Some aspects have been brought in line with or closer to the best international standards, such as those enshrined in the Model Law. For instance, a security right may be held by a trustee, facilitating syndicated secured transactions. Interviews 182 ibid Art 227. 183 The Uniform Act was also published in the Official Journal of every Member State, pursuant to ibid Art 228. 184 OHADA, ‘Uniform Act on General Commercial Law’ (2010–17) www.ohada.org/index.php/en/component/ content/article?id=482:uniform-act-relating-to-general-commercial-law. 185 ML, Art 105(1)(b). See also ch 2, J.

238 OHADA Uniform Act on Securities with lenders in Cameroon and Ivory Coast confirm the substantial improvements.186 However, for the most part, the Uniform Act departs from the fundamental features of the Model Law, especially its functional approach, the simple notice-based registration requirements, the extrajudicial enforcement remedies and the predictable conflict of laws rules.187 The Uniform Act recognises a multitude of security rights, which may be created in the form of pledges, reservation of ownership, and security (fiduciary) transfers of ownership. However, it does not provide for an all-asset security right; thus, it leaves no option to prospective secured creditors but to use multiple security devices, increasing the cost of structuring such transactions.188 Furthermore, the Uniform Act does not apply to finance leases although they can be registered in the same Register as the security devices governed by the Uniform Act. The Uniform Act does not include a set of comprehensive priority rules, which is problematic particularly for competitions between secured creditors who acquired their rights in the collateral under different security devices.189 The registration procedures are cumbersome, requiring the Register Clerk to verify the veracity and legality of the information stated in a registration form against supporting documents. Furthermore, OHADA Member States have not established electronic registers, which significantly complicates the establishment of the ‘OHADA Register’ contemplated in the Uniform Act. Besides, multiple paper-based registries may have been established in individual States, as is the case in Cameroon, which has established 120 such registries. Extrajudicial enforcement procedures are available in very few instances, leaving secured creditors with no choice but to endure the inefficient judicial systems. Equally troubling for prospective secured creditors and buyers of assets is the absence of adequate conflict of laws rules, which may inhibit and increase the cost of cross-border transactions. The rules prescribed in the Uniform Act relating to the registration of pledges and security assignments of receivables are wholly impractical. An independent assessment also highlighted the failure of the prudential regulator, the Central Bank of West African States, to take into account the new mechanisms introduced in the reformed Act.190 Further, the assessment noted that the Côte d’Ivoire Banking Association identified prudential regulation as a disincentive to secured lending.191 This conclusion, however, leaves unclear what the Central Bank should do, especially in light of the significant deficiencies that remain in the secured transactions legal framework. The uncertainty is further exacerbated by preserving the applicability of the 1997 Uniform Act to security rights created prior to the entry into force of the 2011 Uniform Act. Moreover, one may identify a number of legal lacunae in the 2011 Uniform Act, such as the absence of rules on the identification of the grantor in registrations and the effect of errors in registered information, which further elevate the legal risk. OHADA’s approach also contrasts with the recent reforms undertaken by other African economies, which have embraced the bedrock principles and approaches of the Model Law. Given the evidence of early success and impact of the newly implemented secured transactions laws in Anglophone African economies, OHADA should consider yet again revising the Uniform Act on Securities along the lines of the Model Law.

186 IFC

& OHADA (n 98) 40. ch 2. 188 See discussion of the right of retention, see B1 above. 189 See E above. 190 IFC & OHADA (n 98) 25. 191 ibid 41. 187 See

11 Burundi A. Economic Overview Burundi is a small Central African country. After being a German colony until 1916, Burundi was part of the Belgian colonial empire until its independence in 1962. From this colonial past, Burundi retained the civil law system, based mainly on the Belgian civil and commercial codes, which were replicas of the French Napoleonic Civil and Commercial Codes of 1804 and 1807. Its economy is dominated by agriculture, which employs 90 per cent of the population,1 but suffers from slow growth and low productivity.2 Although economic growth in Burundi has ranged between 4 and 5 per cent per annum over the last decade,3 it has been hindered by the recent political crisis in the country.4 The financial sector remains under the strict control of the government, despite some recent liberalisation efforts.5 The 2012 National Financial Inclusion Survey revealed that the framework governing security rights was fragmented, registries were ineffective, the credit risk was high and the financial products offered by banks were limited.6 According to this Survey, in Bujumbura (the capital city) only 8.7 per cent of those applying for a loan in the previous 12 months were successful. In some provinces, only 1.1 per cent of loan applications were approved.7 The weakness of the overall secured transactions framework translated into a low ranking in the Getting Credit Indicator of the 2018 World Bank Doing Business Report, where Burundi ranked 177th, scoring only 2 points on the ‘strength of legal rights’ index.8 As a result of the challenging legal framework for secured transactions in Burundi, lenders very strongly prefer immovable collateral. Movable assets are taken as collateral mostly to support a mortgage over immovable property.9 However, even in those cases where movable assets are used as secondary collateral, it is merely symbolic because the procedures 1 World Bank, ‘Burundi Overview’, www.worldbank.org/en/country/burundi/overview. 2 JD Nkurunziza, L Ndikumana and P Nyamoya, ‘The Financial Sector in Burundi’, National Bureau of Economic Research Working Paper 18289 (August 2012) 10. 3 World Bank (n 1). 4 The situation in Burundi deteriorated following the re-election of the President, Pierre Nkurunziza, on 21 July 2015. See ‘Burundi Economic Studies and Country Risks’ Coface (February 2019) www.coface.com/ Economic-Studies-and-Country-Risks/Burundi. 5 Nkurunziza et al (n 2). 6 Banque de la République du Burundi, ‘National Inclusion Survey in Burundi’ (August 2012) 41, http://www. brb.bi/se/docs/enquete_inclusion_fin.pdf. 7 ibid. 8 See ch 1, B2 for a description of the strength of legal rights index. Also, Burundi is ranked 150 on the ‘enforcement of contracts’ index of the 2018 World Bank Doing Business Report. Doing Business 2018: Burundi (Washington, DC, World Bank, 2018) 33–34, 51. 9 Banque de la République du Burundi (n 6).

240 Burundi for enforcing rights in movable collateral are cumbersome.10 An ineffective judicial system further contributes to the hostile lending environment. The situation described in this paragraph should improve after the implementation of the recently enacted secured transactions and leasing laws. However, at the time of writing, no registry has been developed, and therefore the law has not yet been implemented. In 2017, the Burundian financial sector consisted of 10 commercial banks, 2 non-bank financial institutions, 6 insurance companies and 29 microfinance institutions (MFIs).11 The MFIs were established after the enactment of the law on micro-finance in 2006.12 MFIs play an important role, particularly in rural areas where formal banking services are largely unavailable.13 MFIs rely mostly on the salaries of borrowers as collateral for microloans, a practice governed by Article 98 of the Burundi Labour Code.14 The Labour Code and Article 250(8) of the Code of Civil Procedure15 limit the enforcement of security rights to a maximum of 75 per cent of the grantor’s salary. Apart from loans secured by salaries, the credit products offered by most MFIs are very limited and are not suited to the support of many business activities that, for instance, require lines of credit secured with inventory. Some MFIs located in rural areas accept crops as collateral, but only if they have been harvested and stored in a warehouse. The customary practice is for the MFI to take a pledge, which is made effective against third parties by taking possession of a warrant issued by a warehouseman under a 1923 law relating to warrants.16 Even though the economy relies heavily on agriculture, the volume of credit channelled to this sector has been negligible.17 Burundi’s MFIs also provide crédit solidaire, a classic microloan product, which is credit extended to a group of women who jointly guarantee its repayment. In 2013, the default rate on such microloans was 0.3 per cent.18 The Burundian banking system, including the Central Bank, holds very little monetary reserves and foreign currency. Most credit facilities are short term.19 The cost of credit varies according to the type of borrower: ‘outsiders’ (regular bank clients) who pay high interest rates, and ‘insiders’ (bank executives, employees and a small group of ‘special clients’) who pay no or very low interest rates.20

10 ibid. 11 The recent data in the official website of the Bank of the Republic of Burundi shows that the total number of microfinance institutions is 29. Microfinance is relatively new in the Burundian financial system. Apart from BNDE’s microfinance operations dating from the 1960s, genuine microfinance began with the creation of the savings and credit cooperatives (Coopératives d’Epargne et de Crédit, COOPECs) in 1985. However, starting from the mid-1990s, many institutions were created with diverse legal status, ranging from NGOs to cooperatives. In addition to COOPECs, as many as 19 organizations have been created since 2000, of which five were created in 2005 alone. Nkurunziza et al (n 2) 17. 12 Decree No 100/203 of July 22, 2006 related to the Regulation of Microfinance Activity. For a generic description of microfinance, see ch 1, C4a. 13 Nkurunziza et al (n 2) 17. 14 See Labour Code 1991. 15 Code of Civil Procedure 2004. 16 See Decree of March 20, 1923 relating to Warrants. 17 The agricultural sector mainly gets credit to finance coffee and tea chains, and to a lesser extent, the rice sector. Report on Indicators Performance of the Microfinance Institutions of RIM (Bujumbura, RIM, 2013) 15. RIM is an abbreviation for the French Réseau des Institutions de Microfinance (Network of Microfinance Institutions). 18 ibid. 19 Nkurunziza et al (n 2). 20 ibid 36. ‘Outsiders’ pay interest rates of between 18 and 23.5 percent.

Pre-Reform Legal Framework 241 Burundi’s economy is also plagued by a general mistrust of credit products by both individuals and businesses. Since the end of the genocide in 1993, Burundians have lost faith in, and respect for, authority.21 As part of this phenomenon, a number of borrowers, particularly in urban areas (mainly in the capital Bujumbura), do not feel the need to fulfil the obligations they have contractually undertaken, including repayment of debts. This indifference to the payment of debts is encouraged by the authorities, including the judiciary, which tends to overprotect debtors. Such an environment does not bode well for creditors who, due to the practical inability to collect debts, remain reluctant to extend credit. Until recently, the secured transactions regime had been primarily structured around a single security device recognised by the Civil Code,22 the possessory pledge.23 Burundi’s credit infrastructure was modernised in 2015 by the adoption of an amendment to the Code of Commerce.24 Even before modernisation, the Code of Commerce provided for nonpossessory pledges over specific types of property, namely professional equipment and motor vehicles, inventory, securities and shares, and the enterprise. The legal framework governing secured transactions was further strengthened by the enactment of a comprehensive law governing security rights in movable property (2016 Law on Security Rights) that provides rules for a non-possessory pledge and a retention of ownership device.25 Finally, the leasing law enacted in 2016 provides a framework for finance leases, revising a 1993 law (2016 Law on Leasing).26 The 2016 Law on Leasing complements the Law relating to the Regulation of Banks and Financial Institutions.27 Leasing is characterised as a credit transaction that can only be provided by banks and authorised financial institutions under the supervision of the Bank of the Republic of Burundi.28 Finance leases can also be provided by other financial institutions not specifically authorised to do so by the Central Bank as their primary activity, and by MFIs.29 Operating leases, such as short-term rentals, can be provided by anyone.30

B. Pre-Reform Legal Framework 1. Introduction Before the enactment of the 2016 Law on Security Rights, the main laws governing security rights in Burundi were the Civil Code and the Code of Commerce. The former dealt 21 ‘The lack of faith in the judiciary – as well as in the police – moved the population to take the law into their own hands.’ R Schweiger, ‘Late Justice for Burundi’ (2006) 55(3) ICLQ 653–71, 658. 22 See Civil Code (originally adopted by decree in 1888). 23 ibid Art 598. 24 Law No 1/01 of 16 January 2015. This new law promulgated a new version of the Commercial Code. The 2010 version of the Commercial Code is available in French at ‘Commerce Code’ (2010) www.assemblee.bi/IMG/pdf/ code%20du%20comerce.pdf; and the 2015 version is available in French at ‘Commerce Code’ (2015) www.droitafrique.com/upload/doc/burundi/Burundi-Code-2015-commerce.pdf. 25 Art 1 of the Law on Security Rights, Law No 10 of 12 August 2016. 26 See ‘Leasing Law’ (2016) www.droit-afrique.com/uploads/Burundi-Loi-2016-01-definition-credit-bail1.pdf. 27 No 1/017 of 23 October 2003. Art 1(1) of the 2016 Law on Leasing. 28 Arts 5, 8 and 14 of Law No 1/017 of 23 October 2003, related to the Regulation of Banks and Financial Institutions. 29 2016 Law on Leasing, Art 3. 30 Civil Code, Arts 373 et seq.

242 Burundi primarily with the possessory pledge, whereas the latter provided a regime for various specific types of non-possessory pledges31 and a system for their registration in the Movable Credit Register (Registre de Credit Mobilier) maintained by the Tribunal of Commerce.32 The only non-possessory pledges authorised by the Code of Commerce were: (i) the pledge over an enterprise, which is characteristic for legal regimes based on French law,33 (ii) the equipment and automobile pledge,34 (iii) the pledge of inventories,35 and (iv) the pledge of partners’ shares and securities.36

2. Pre-Reform Requirements for Creation and Effectiveness against Third Parties of Non-Possessory Pledges Non-possessory pledges had to be evidenced in writing, either in the form of a notarial deed37 or under a private signature,38 and they required registration to be effective against third parties. For each type of non-possessory pledge, the Code of Commerce provided that the pledge was only effective vis-a-vis third parties and the pledgor if it was registered in the Movable Credit Register.39 Unregistered pledges were therefore void, but the underlying obligation still stood as a personal obligation. Each type of non-possessory pledge was subject to specific rules. Thus, in the case of a pledge over an enterprise, only the elements of the enterprise enumerated by Article 81 of the Code (these elements do not include inventory) could be pledged.40 Furthermore, the agreement had to describe the assets covered by the enterprise pledge specifically. Where the agreement failed to do so, the pledge only extended to the trade name, leasehold property, any rights and information related to customers of the business, and the goodwill.41 The enterprise pledge had to be registered within 15 days from the date of the deed providing for the pledge.42 Regarding the pledge of professional equipment and motor vehicles,43 Article 138 stipulated that the pledge would not be effective until its registration in the Register. For vehicles, 31 See nn 25–28. 32 Also, the appellation ‘Book Archive’ is a translation for ‘Registre du Credit Mobilier’. 33 nantissement de fonds commerce, Commercial Code, Arts 113–34. 34 gage automobile, Commercial Code, Arts 135–43. 35 gage des stocks, Commercial Code, Arts 144–48. 36 nantissement des droits d’associés et valeurs mobilières, Commercial Code, Arts 149–54. 37 Notarial Deed is a translation of the French phrase Acte Authentique. 38 Private Signature is a translation for the French phrase Acte sous Seing Prive. These are used in documents that are written by the parties themselves or their agents (generally a lawyer). Generally, in such cases, all that is required is that the parties sign the document. It must be distinguished from the Notarial Deed (Acte Authentique) that requires much more formalities to be valid, such as being written by a public officer, ie a public notary. According to Art 115, the pledge of an enterprise (business) can be in the form of a notarial deed or under private signature. The same applies to a pledge of professional equipment and motor vehicles pursuant to Art 137, as well as to a pledge of inventory under Art 143. As for securities and shares owned by the partners of a company, Art 149 contains a similar provision. All these contracts must contain the required information to be valid. 39 Art 117 (enterprise pledge), Art 138 (equipment and motor vehicles pledge), Art 144 (pledge of inventory), Art 159 (pledge of shares, for which there were additional requirements, see below). 40 Art 114(1). The fonds de commerce includes elements such as customers, brand and trade name, right to lease, commercial movables, patents, trademarks etc., according to the Commercial Code, Art 81. 41 Art 114(2). 42 Art 117. 43 Category (ii) above.

Pre-Reform Legal Framework 243 the pledge must also be noted on the certificate of title.44 Equally, the pledge of inventory45 had to be registered in the Register.46 Unlike registrations of pledges of enterprise, equipment and motor vehicles, which were valid for five years, a registration relating to an inventory pledge was only valid for one year.47 As for securities such as shares, registration in the Register of a pledge over these assets was not sufficient, and the pledge took effect against third parties only upon notification of the company that issued the securities.48 The somewhat rigid formal requirements were ameliorated by a provision according to which generally an omission of, or inconsistency with, the formalities necessary for the creation or third-party effectiveness of these pledges would not automatically result in their invalidation unless the omission or error prejudiced a third party.49

3. Pre-Reform Priority of Non-Possessory Pledges The priority of competing pledges was determined by the date of registration. However, two pledges registered on the same day would have the same priority so that the pledgees are entitled to be satisfied on a pari passu basis.50 The Register was maintained in paper records in which pledge instruments were indexed manually and in the order that they were received. Under this indexing system, it was almost impossible for a third party to find a registration unless the searcher knew the exact date on which it was processed by the Registrar. As a result, prospective secured creditors and buyers of assets faced significant challenges in both discovering existing encumbrances over assets that a loan applicant may offer as collateral and ensuring that pledges were properly registered and had priority against competing claims.

4. Pre-Reform (and Post-Reform) Rules on Enforcement of possessory and Non-Possessory Pledges The Code of Commerce did not permit extrajudicial enforcement of pledges, leaving no other option for creditors but to seek relief from a competent court. Extrajudicial remedies were only provided for in the 2016 Law on Security Rights.51 The original rules for the judicial enforcement of debts secured with movable assets are detailed in the Code of Civil Procedure and the Code of Commerce, particularly in Title 13 (of the Code of Commerce), and the general rules of these Codes remain applicable after the enactment of the 2016 Law on Security Rights. The Code of Civil Procedure contains the general

44 Art

139.

45 Category

(ii) above. 144. 47 Art 169(2). 48 Art 152. 49 Art 154. 50 Art 157(2). 51 See C5 below. 46 Art

244 Burundi rules on enforcement of debts, whether secured with some property or unsecured, while the Code of Commerce contains some additional rules applying specifically to pledges of enterprise, inventory, equipment, automobiles and shares. For example, Articles 118–25 of the Code of Commerce govern the enforcement of debts secured with the pledge of an enterprise. For the enforcement of such debts, the pledgee must obtain a court order authorising sale of the enterprise.52 The pledgee must also provide a prior notification to the pledgor of its intention to initiate enforcement proceedings and give a grace period of eight days for the pledgor to cure the default.53 After the entry of judgment, the clerk must notify the pledgor and publicise an announcement about the auction in which the pledged asset will be sold.54 The auction must not be held earlier than 30 days after the publication.55 In the case of professional equipment and vehicles, the enforcement procedures are similar except that a court may authorise the creditor to accept the collateral in satisfaction of the obligation,56 although any rule to that effect in an agreement between the parties is unenforceable.57 Title 13 of the Code of Commerce, as supplemented by the Code of Civil Procedure, also provides rules for precautionary (‘provisional’) enforcement measures. Provisional measures are designed to protect creditors where the recovery of the loaned sum is put at risk, such as by the likelihood of dissipation of the collateral by the borrower. The Code of Commerce includes special provisions for tangible movables,58 receivables59 and securities.60 The Code of Civil Procedure contains general rules for the seizure of immovable and movable assets by secured and execution creditors. The rules in Articles 307–32 relate to immovable assets. Article 258 empowers a judgment creditor to seize the debtor’s movable assets. Other articles stipulate general rules for notification, the necessary information to be included in such notifications, and the procedure for the sale of the seized assets, whether movable or immovable. It is extremely important for the efficient enforcement of secured lending that disputes relating to such transactions are resolved quickly and efficiently. Article 352 of the Code of Commerce provides the basis for the establishment of commercial courts that aim to provide this kind of resolution for all commercial disputes. The Code includes short deadlines for the procedures of the commercial court.61 The appeal procedures are also subject to strict time limits. For instance, Article 356(3) provides that the clerk of the first-instance court must forward the dossier to the clerk of the court of appeals within 15 days of filing an appeal. The goal of the legislature was to have the commercial courts issue a final decision in less than one year.

52 Art 120. 53 Art 119. 54 Art 121(3). 55 The period can be extended by a court order to reach a maximum of 90 days. Art 122. 56 Art 141(2). 57 Art 141(3). 58 Art 396. 59 Art 401. 60 Art 407. 61 Art 354 provides that if the tribunal finds that the case is not ready and needs more inquiry, the judge will have a maximum of three months to issue her report to the tribunal. Art 355(2) provides that the tribunal will have two months to issue a decision if the case is ready.

The 2016 Law on Security Rights and the 2016 Leasing Law 245

C. The 2016 Law on Security Rights and the 2016 Leasing Law In 2016, Burundi enacted two significant pieces of legislation with the purpose of modernising its legal framework for secured lending and leasing. Initially, a bill on security rights was removed from the legislative agenda of the parliament because it was packaged together with a law on credit reporting that sought to establish a credit bureau.62 The latter was found to be too controversial, which adversely affected the bill on security rights. However, the law on Security Rights was enacted as a separate statute in 2016. It is based on the Model Law, but in relation to the treatment of retention of ownership it reflects the non-unitary approach to acquisition security rights which was included in the Legislative Guide, but not in the Model Law. Because of Burundi’s civil law tradition roots, some of its provisions are also inspired by the French law on secured transactions, as resulting from the reform of 23 March 2006 insofar as these provisions were compatible with the Model Law approach. Given the widespread use of devices based on retention of ownership and their relative efficiency in Burundi, it was felt that the recharacterisation of such a device as an acquisition security right, along the lines of the unitary approach of the Model Law, could have negative practical consequences, and perhaps even endanger its passage by the Parliament. It therefore appeared prudent not to question the retention of title in the context of the reform and to adopt the non-unitary approach advocated by the Legislative Guide. This entails the principle of functional equivalence, especially in terms of third-party effectiveness as a retention of title seller is required to register a notice.63 The 2016 Law on Security Rights governs two distinct types of security devices, namely pledges and retentions of ownership, reflecting the decision to merge all existing security rights, with the exception of retention of title, into a single concept. It was decided to call this new concept by the traditional term ‘pledge’ so as not to confuse lawyers and users. The traditional civil law distinction between the gage (relating to tangible assets) and the nantissement (relating to intangible assets) was, therefore, abandoned. However, despite its label and its association with the traditional concepts, the new pledge (which can be possessory or nonpossessory) has the features of a security right contemplated by the Model Law.64 The 2016 Law on Security Rights also provides for the establishment of the National Registry of Security Rights (‘the National Registry’). Although the law regulates all aspects of pledges and retentions of ownership, Article 17 also requires registration of functional equivalents to security, including the ‘vendor’s privilege’,65 the ‘sale with the option of redemption’ (a sale with the option to repurchase),66 outright assignments of receivables67 and finance leases.68 The creation and enforcement of these rights, however, remain outside the scope of the law. Similarly, a finance lease is not recharacterised as a pledge; rather, the ownership right of the lessor is effective against all third parties with one exception: an

62 During a parliamentary debate in August 2016, some deputies still questioned the Minister of Justice about the relation between the Credit Information Office and the National Register of Securities. 63 For discussion of the non-unitary approach, see ch 2, B2b. 64 See C1 and C2 below for the requirements of its creation and third-party effectiveness. 65 Le privilège du vendeur, which is similar to a non-possessory vendor’s lien. 66 La vente avec la faculté de rachat. 67 La cession de créance. 68 Le contrat de crédit-bail.

246 Burundi unregistered finance lease loses priority to a secured creditor who has registered its pledge.69 This approach is not the same as under the non-unitary approach of the Legislative Guide, which treats finance leases in the same manner as other retention of title devices. Under that system, if a lessor does not register its right, it loses its ownership and the ‘grantor’ (the lessee) becomes the owner.70 Finally, the law also provides priority rules that are applicable to non-consensual liens (privilège mobilier).71 Even though they arise and become effective against third parties outside the Law on Security Rights, their priority against a security right is determined under the Law on Security Rights, so that they must be registered for priority purposes. The 2016 Law on Leasing applies only to ‘finance leases’, defined in Article 2 as ‘any leasing operation of goods, equipment, materials, tools, or immovable property for professional use owned by lessors when the lessee is provided an option to acquire all or part of the leased property’. The only condition for its application is that the leased assets must be used in some business activity.72 An enterprise (fonds de commerce) can also be the subject of a lease.73 Under Article 7, the leasing agreement must provide that at the expiration of the lease, the lessee is required to exercise one of the three options: to purchase the asset, to renew the lease, or to return the leased asset to the lessor at the lessee’s expense.74 Article 3 of the 2016 Law on Security Rights sets out definitions of various terms. ‘Grantor’ (constituant) is defined as a person that creates a security right (sûreté mobilière) to secure either its own obligation or that of another person. ‘Control’ is defined as the power conferred on the secured creditor (créancier garanti) in respect of the right to payment of funds credited to a deposit account. Control is automatic when the secured creditor is the bank that maintains the deposit account in which the encumbered funds are credited. In all other cases, control must be constituted by an agreement between the grantor, the secured creditor and the depository bank. ‘Retention of ownership’ (la réserve de propriété) is defined as a provision in a contract of sale whereby the vendor remains the owner of the asset which is the subject of the contract until full payment of the purchase price.75 This approach to the scope, and the isolated treatment of retention of ownership, was inspired by the non-unitary approach of the Legislative Guide. However, this approach is narrower than that recommended in the Legislative Guide because it applies only to retention of ownership, whereas the Legislative Guide applies to both retention of ownership and finance leases. The underlying purpose of the non-unitary approach to acquisition financing is to provide for equal treatment of all providers of such financing.76

1. Creation Rules for the creation (création) and third-party effectiveness (opposabilité) of a pledge are provided in the 2016 Law on Security Rights. Some rules are generally applicable to all kinds

69 See

C2f below. ch 2, B2b, and C2e below 71 Law on Security Rights, Art 47. 72 2016 Law on Leasing, Art 2. 73 ibid. 74 2016 Law on Leasing, Art 7. 75 2016 Law on Security Rights, Art 81. 76 See the discussion in ch 2, B2b. 70 See

The 2016 Law on Security Rights and the 2016 Leasing Law 247 of pledges, whereas others only apply to pledges over receivables, bank accounts, vehicles and intellectual property rights. This structure resembles the Model Law, which also contains general and asset-specific rules. As mentioned above,77 since a ‘pledge’ under the law on Security Rights can be possessory or non-possessory, and has many of the features of a security right under the Model Law, for the purposes of comparison here the two will be equated. Article 4 provides that a pledge is created by an agreement by which a grantor specifically assigns, regardless of the form of the transaction or the terminology used by the parties, a movable asset or a set of movable assets for the benefit of the creditor in order to secure an obligation. The pledge may relate to future as well as existing assets of the pledgor, whether tangible or intangible, as is the case with a security right under the Model Law.78 Also, as with a security right under the Model Law,79 according to Article 6 the pledge may secure any obligation, existing or future, as long as it can be determined or is determinable. As for formalities, the agreement must be in writing evidenced by a notarial deed or a privately signed contract.80 The agreement must identify the secured debt specifically, by the amount and the due date, and provide a sufficient description of the encumbered asset.81 A generic description will suffice. Most of the requirements, therefore, mirror those of the Model Law.82 However, the requirement to indicate the amount due, rather than the maximum amount for which the pledge may be enforced as contemplated by the Model Law as a possible requirement which can be included by an enacting State,83 may appear to be troublesome, given that the amount of credit extended may not be known at the time of the agreement. However, the requirement to state the amount due merely entails that the amount must be determined or determinable, so a pledge agreement may state that the amount due is the maximum amount that the secured creditor commits to extend to the pledgor. Article 65 of the 2016 Law on Security Rights provides that, notwithstanding the general rules, if the pledge covers a specific type of asset or a group of assets it will be governed by the special provisions set out in Articles 66–80.84 For instance, under Article 66 a pledge of receivables (le gage sur créance) can cover a portion of a debt unless the debt is indivisible. This contrasts with the general rule under Article 8, which prevents the creation of a pledge over a fractional interest in an asset (eg 50 per cent of a car).85 If the collateral is subsequently divided, such division has no effect on the pledge that continues to encumber each part. Further, the law considers the pledge over bank accounts (le gage de compte bancaire) as a pledge of a receivable. Accordingly, the rules governing the pledge of receivables are applicable subject to the provisions of Articles 75–78.86 Under Article 77, for example, the pledgee may agree to grant the pledgor the power to dispose of the funds in the bank account. 77 C. 78 ML, Arts 6(2) and 8. 79 ML, Art 7. 80 2016 Law on Security Rights, Art 14(1). 81 ibid. 82 ML, Arts 9(1) and (2) 83 ibid, Art 6(3)(d). See ch 2, C1 for a discussion of the merits of this approach. 84 For instance, for the pledge of motor vehicles and intellectual property rights see the 2016 Law on Security Rights, Arts 79–80. 85 This is different from the approach in the Model Law where security rights over fractional interests are permitted, see Art 8(b). 86 2016 Law on Security Rights, Art 74.

248 Burundi Ownership of an asset may be retained as security by way of a retention of ownership clause in a contract of sale.87 This will have the effect of suspending the transfer of ownership until the full satisfaction of the purchase price.88 The retention of title clause must be agreed upon in writing and executed before delivery of the asset.89 These formality provisions are based on recommendation 189 of the Legislative Guide. Additionally, the parties may agree that the seller may retain ownership in their future transactions, so there is no need to execute a separate agreement containing a retention of ownership clause every time a sale is entered into between the two parties.90 Similarly, under the 2016 Law on Leasing the lessor’s ownership of the asset may not pass to the lessee until the expiration of the lease, and, even then, only passes if the parties agree that this will be the case.91

2. Third-Party Effectiveness A pledge may be made effective against third parties by two general methods: (i) registration in the National Registry of Security Rights (Registre National des Sûretés), or (ii) dispossession (dépossession).92 The naming of the principal methods of third-party effectiveness, supplemented by additional methods (such as control)93 that apply to specific assets, follows the structure of the Model Law.94

a. Registration The specifics of the registration procedures are prescribed in a separate decree.95 Article 15(3) mandates that the Registry be publicly accessible, allowing any person to submit a notice or search request. By virtue of Article 42, the time of third-party effectiveness determines the priority of the pledge (thus, an earlier registered pledge enjoys priority over a subsequently registered pledge). As mentioned earlier,96 Article 17 sets out the other rights with respect to which notices can be registered in order for those rights to become effective against a pledgee, such as vendor’s privilege, sales with the option of redemption, outright assignments of receivables, and finance leases. Articles 18–22 contain general provisions for the registration of a notice. The notice must identify the grantor, describe the encumbered assets in a manner that reasonably allows their identification, and set out the nature and the sum of the secured obligation, as well as its due date.97 In some ways, this is similar to the requirements for registering a

87 See

C above. Art 81. 89 ibid, Art 82. 90 ibid. 91 2016 Law on Leasing, Art 9. 92 2016 Law on Security Rights, Art 15(1). 93 See Art 76 and discussion below. 94 ML, Arts 18 and 25–27. 95 2016 Law on Security Rights, Art 15(2). 96 C above. 97 2016 Law on Security Rights, Art 18(1). 88 ibid,

The 2016 Law on Security Rights and the 2016 Leasing Law 249 notice under the Model Law,98 but there are two main differences. First, and very curiously, the law does not require that the notice identify the secured creditor. Second, there is a requirement to indicate a specific sum and nature of the secured obligation: the concern with this is the same as with the equivalent requirement applicable to security agreements,99 but Article 15 contemplates the issuance of regulations that may clarify the position with respect to the description of a secured obligation in a notice.100 The period of effectiveness of the registration is limited to five years.101 Extension is possible for another five years.102 A notice can be registered before or after the conclusion of a security agreement, but must be authorised by the grantor in writing.103 The authorisation can be granted before or after the registration.104 Unless otherwise agreed by the parties, authorisation for the registration of a notice is presumed by the grantor’s signature on the security agreement.105 These provisions are very similar to those in the Model Law.106 Generally, a pledge remains effective against a transferee who acquires rights in the encumbered asset after its registration, unless the transferee qualifies for priority as a buyer in the ordinary course of business.107 A system permitting registration in advance of the security agreement, and where registration is carried out unilaterally by the secured creditor, requires some protection for the grantor against registrations which are maliciously false or do not relate to potential transactions.108 This protection is provided here by a provision that the secured creditor must register a cancellation notice 15 days after receiving a written request from the grantor on the grounds that (a) no security agreement has been concluded, (b) the pledge has been extinguished for any reason or (c) the registration was not authorised by the grantor.109 Article 22 seems to contain a drafting error empowering the ‘secured creditor’ to request a court to cancel a registration. This power seems to be contingent on the secured creditor failing to register a cancellation notice, in which case it should be the grantor who has this power. This is the case under the Model Law110 and generally is included in similar registration provisions. A pledge also extends to the proceeds of any disposition, provided that the proceeds can be identified.111 When the identifiable proceeds are in the form of money, receivables, negotiable instruments or rights to payment of funds credited to a bank account, the pledge remains effective against third parties and retains its priority without any further action.112 If the proceeds of the pledged asset are a type of asset other than ‘cash proceeds’,

98 MRP, Art 8. 99 See C1 above. 100 See C1 above. 101 2016 Law on Security Rights, Art 21. 102 ibid. 103 See ibid Arts 18–19. 104 ibid. 105 ibid. 106 MRP, Arts 2 and 4. 107 2016 Law on Security Rights, Art 20 and Art 49 (protecting buyers in the ordinary course of business). cf ML, Art 34. 108 See ch 2, E1. 109 2016 Law on Security Rights, Art 22. cf ML, Art 20. 110 Model Registry Provisions, Art 20(6). 111 2016 Law on Security Rights, Art 40. 112 ibid Art 41(1).

250 Burundi the secured creditor must register an amendment notice identifying the new asset within 15 days. Otherwise, the pledge will be ineffective against third parties with respect to those proceeds.113 Again, these provisions are similar to those in the Model Law.114

b. Dispossession The second method for achieving third-party effectiveness is dispossession, which requires the secured creditor or a third party to take possession of the encumbered tangible asset.115 Instead of the actual possession of the tangible asset, the pledgee will have possession if it holds a negotiable document representing the asset, such as a bill of lading or a warehouse receipt.116

c. Attachment of a Sign Article 26 provides that the pledgee may attach any distinctive signs, such as a metal plate indicating where and when the pledge was registered, to the encumbered asset, but affixation of a plate on its own is insufficient to make the pledge effective against third parties. This mechanism is available when the collateral remains in the possession of the grantor, but does not apply to inventory.117 Even if the law does not attach any legal consequences on the third-party effectiveness of a pledge to this attachment of a sign, it appeared necessary to maintain this mechanism, which existed before the reform, in order to reinforce the confidence of the creditors. Indeed, in the particular context of Burundi, creditors tend to rely more on a material and tangible sign than on a simple electronic entry in a Register. In other words, the procedure may be seen as a good practice having a prophylactic effect of preventing conflicts between secured creditors and third parties who acquire rights in the encumbered asset without searching the registry. The 2016 Law on Security Rights imposes severe sanctions,118 detailed in Article 63, on anyone who destroys, removes or covers the distinctive signs affixed to the encumbered asset.119 In case of involuntary destruction of the signs, the grantor or the third party who possesses the collateral must notify the secured creditor within three days to enable it to attach new signs.120

d. Other Methods of Making a Pledge Effective against Third Parties Concerning the pledge over bank accounts (le gage de compte bancaire), the pledgee has two options: it can either register a notice in the National Registry or can obtain control.121 This approach is again consistent with the Model Law.122

113 ibid

Art 41(2). These provisions are similar to those in ML, Art 19. Art 19. 115 2016 Law on Security Rights, Art 23. 116 ibid Art 24. 117 ibid Art 26(1). 118 These are penal sanctions, including imprisonment and heavy fines. 119 ibid Art 26(2). 120 ibid Art 27. 121 ibid Art 76. 122 ML, Art 25. 114 ML,

The 2016 Law on Security Rights and the 2016 Leasing Law 251 The pledge over a motor vehicle (le gage automobile) is effective against third parties on registration of a notice in the National Registry.123 However, the pledge must also be indicated/registered on the ‘administrative title’ (ie the certificate of title) relating to the vehicle.124 Although both registrations are required, the absence of a reference in the administrative title will have no effect on the third-party effectiveness of the pledge duly registered in the National Registry. The 2016 Law on Security Rights does not specify the consequences of the failure to indicate/register a notice of the security right on the administrative title.

e. Third-Party Effectiveness for Retention of Ownership Sales According to Article 83, third-party effectiveness for a sale on retention of ownership terms is achieved by registration of a notice in the National Registry. The registration must occur within 30 days after the grantor has taken possession of the sold assets in order to have priority over all other competing security rights, even if these rights were made effective against third parties earlier.125 This priority comes from the fact that, if made effective against third parties as provided for in Article 83, ownership is retained by the seller and therefore the grantor cannot grant competing security rights in the asset. If registration is not effected within the 30-day period, ownership will be transferred to the buyer and the seller will be assimilated to a pledgee.126 The seller (now a pledgee) can register a notice of its interest after this has occurred, but will lose priority to any competing security right whose right was made effective against third parties before that registration took place. The scheme set out in Article 83 implements alternative B of recommendation 192 of the Legislative Guide,127 which, unlike alternative A of that recommendation, does not provide distinct requirements for the third-party effectiveness of acquisition security rights depending on whether the relevant asset is inventory or equipment. This consequence of the failure to register a notice within the 30-day window corresponds to recommendation 194 of the Legislative Guide.

f. Third-Party Effectiveness of Finance Leases Finance leases must be registered with the tax administration.128 The procedure for ‘tax authority registration’ is to be determined by a decree, but it requires the submission of a leasing agreement.129 The priority of the lessor’s right over a pledgee is established according to registration in the National Registry.130 An unregistered finance lease remains fully effective against other competing claimants, including an insolvency trustee as long as the lease agreement is stamped with a certain date. Article 28(2) of the 2016 Law on Leasing provides that registration in the National Registry will grant the lessor the same rights as a

123 2016 124 ibid. 125 ibid

Law on Security Rights, Art 79.

Art 83(1). Art 83(2). 127 Recommendations relating to the non-unitary approach. 128 2016 Law on Leasing, Art 26(1). 129 ibid Art 26(3). 130 ibid Art 28(1). See C above. 126 ibid

252 Burundi non-possessory pledgee under the 2016 Law on Security Rights,131 which would enable the lessor to enforce extrajudicially,132 even though it would also be obliged to account for any surplus value.133 For the purposes of the 2016 Law on Leasing, the lessor remains the owner of the asset.134

3. Rights and Obligations of the Parties Like the Model Law,135 the 2016 Law on Security Rights contains a section on mutual rights and obligations of the parties, which are primarily set out in Articles 28–37. Nevertheless, a major difference exists between these laws: the Burundi law imposes criminal penalties for the violation of these duties, whereas the Model Law does not deal with criminal sanctions at all. The 2016 Law on Leasing also sets out some rights and duties of parties, including that the lessee may not remove the movable asset to a location other than the one stated in the security agreement without the lessor’s prior written authorisation.136 Nothing in the 2016 Law on Security Rights would prevent the pledgee or the retention of ownership seller from including a similar limitation in the security agreement, but it is not expressly set out in the statutory lists of rights and duties. As a general rule, with respect to tangible encumbered assets under the 2016 Law on Security Rights, any person in possession of the pledged asset must exercise due care to prevent its destruction or a decline in its value.137 However, a decline in value due to the normal use of the asset (wear and tear) is permissible.138 Although these duties are similar, though not identical, to those provided for in the Model Law,139 and, under the Model Law, the parties cannot contract out of those duties,140 the Law on Security Rights is different in that any action taken in violation of the duty to exercise due care is punishable by the Penal Code.141 Furthermore, any person who with a fraudulent intent prevents the secured creditor from exercising his rights may be imprisoned for up to five years.142 The Law on Security Rights also provides that the secured creditor may inspect the grantor’s premises to check the condition of the assets and ensure that they are properly preserved.143 However, the secured creditor must notify the grantor of the date and hour of the inspection at least 48 hours prior to the visit.144 This advance notice requirement goes beyond Article 55(2) of the Model Law, which merely requires the secured creditor to act in a commercially reasonable manner. In some situations, this requirement may frustrate the purpose of the

131 ibid

Art 28(2). is not possible under the 2016 Law on Leasing; see C5c below. 133 See C5a below. 134 ibid. 135 ML, ch VI, s I. 136 2016 Law on Leasing, Art 14(2). 137 2016 Law on Security Rights, Art 62. 138 ibid. 139 ML, Art 53. 140 ibid Art 3. 141 2016 Law on Security Rights, Art 63(1). See n (123) above. 142 ibid Art 63(3). 143 ibid Art 29. 144 ibid. 132 This

The 2016 Law on Security Rights and the 2016 Leasing Law 253 inspection by allowing the grantor to take measures that conceal the actual status or value of the collateral. The grantor does not have the right to dispose of any pledged asset without the prior authorisation of the secured creditor.145 This authorisation can be granted in the security agreement creating the pledge or in a subsequent agreement.146 The secured creditor may revoke the authorisation at any time upon notification of the grantor.147 Article 63(2) subjects the person in possession of the encumbered assets who leases or sells it without prior authorisation and before the full satisfaction of the secured obligation to imprisonment.148 However, when the security right covers goods intended to be sold in the ordinary course of business, the secured creditor is presumed to have given its authorisation to the sale of those assets.149 Similarly, a lessee has the right to use the leased asset until the expiration of the lease,150 but cannot sell, mortgage or pledge the asset.151 A secured creditor who is in possession of the encumbered asset will be held responsible if, due to its negligence, the asset is lost or damaged.152 A secured creditor must also allow the grantor to verify the status of the encumbered asset. The grantor can perform inspections on the secured creditor’s premises to verify the condition of the assets and ensure that they are properly preserved.153 Notice of the date and hour of the inspection must be given to the secured creditor (by the grantor) at least 48 hours prior to the visit.154 The grantor must reimburse the secured creditor or any third party for any necessary expenses incurred for the preservation of the encumbered asset.155 The Model Law does not contain a similar provision because it is based on the expectation that in most secured transactions the grantor will remain in possession of the encumbered asset, although if the secured creditor is in possession, it must exercise reasonable care to preserve the asset.156 If the encumbered asset is a perishable good, the secured creditor or the third party in possession of such asset may sell it after notifying the grantor at least 48 hours before the sale. The pledge then extends to the proceeds of sale.157 The Model Law does not include a comparable provision in relation to the period before default.

4. Priority rules As a general rule, the priority of a security right is determined according to the time when it becomes effective against third parties.158 Article 29 of the Model Law provides a similar priority rule. An exception from this general rule, again corresponding to the priority

145 ibid

146 ibid. 147 ibid. 148 ibid

Art 30.

Art 63(2). Sanctions vary from two to five years imprisonment. Art 31. 150 2016 Law on Leasing, Art 13. 151 ibid Art 14(1). 152 2016 Law on Security Rights, Art 32(1). 153 ibid Art 32(2). 154 ibid. 155 ibid Art 32(3). 156 ML, Art 53. 157 2016 Law on Security Rights, Art 35. 158 ibid Art 42. 149 ibid

254 Burundi scheme of the Model Law, is a security right granted to finance the acquisition of movable assets. Even if later-in-time, such an acquisition security right has priority if it is registered within 30 days of the grantor taking possession of the assets.159 It will be seen that the effect of this rule is the same as where the acquisition of the asset is effected by a sale on retention of title terms,160 although the conceptual reasoning is, of course, different. This is because, in economic terms, these transactions are very similar, although the type of financier is different in each case.161 An acquisition security right of a lender in the purchased asset may be in competition with a retention of title seller. In this case, Article 50 gives priority to the retention of title seller. The priority of the acquisition security right also extends to any proceeds of the originally encumbered assets.162 Article 3 of the Law on Security Rights defines proceeds to include any products. Accordingly, an acquisition security right in raw materials will extend to manufactured products. The general applicability of the acquisition security right rule to all types of movable assets may be problematic where a creditor takes a security right over present and future inventory. In relation to any new acquisition of inventory, this creditor may not be able to determine for 30 days whether the new inventory acquired by the grantor is subject to an acquisition security right, if the holder of the acquisition security right did not make it effective until the end of the period of 30 days after delivery. Prudence would demand that the inventory financier delays making an advance against the new inventory until the grace period expires. In addition, since the acquisition secured creditor does not have any obligation to inform the inventory secured creditor of its intention to acquire a right in the inventory, the inventory secured creditor would need to conduct regular searches of the National Registry. These conclusions follow from the fact that the requirements for an acquisition security right follow option B of Article 38 of the Model Law, and do not differentiate between requirements for an acquisition security right over inventory and one over equipment.163 The priority scheme for transferees of encumbered assets follows Article 34 of the Model Law. First, if the grantor sold the encumbered asset with the authorisation of the secured creditor, the buyer will acquire its rights free of the security right.164 Second, a buyer of an asset sold in the ordinary course of the seller’s business acquires its rights free of any security right.165 The buyer acquires its rights free of any security right, whether created by the seller or its predecessor in title. The law also addresses the issue of conflict between the holder of a security right in an asset that has become immovable by destination and a mortgagee of the relevant immovable property. Article 3 of the 2016 Law on Security Rights defines ‘immovable property by destination’ as any tangible movable property that is physically and inseparably attached

159 ibid Art 50. For the reasoning behind this type of provision, see ch 2, F2. 160 See C2 above. 161 See ch 2, B2b and F3. 162 2016 Law on Security Rights, Art 51. 163 See ch 2, F3. The same approach is taken in relation to retention of title devices, as option B of recommendation 192 of the Legislative Guide is followed, see C2e above. 164 2016 Law on Security Rights, Art 48. 165 ibid Art 49.

The 2016 Law on Security Rights and the 2016 Leasing Law 255 to immovable property. It should be noted that this definition is narrower than the classic definition of real property by destination in civil law in general and the French law in particular. Indeed, civil law generally defines the phrase ‘immovable by destination’ as meaning not only any movable asset that is physically and inseparably attached to the immovable, but also movable property that is placed on the land so that it becomes necessary to the exploitation and service of that land.166 Where a security right in an immovable by destination competes with an interest of a mortgagee in the immovable to which the movable asset has been affixed,167 priority is determined by the order of two dates: the date of registration of the mortgage in the Land Register and the date of registration of the security right in the National Registry. If these two dates are exactly the same, the two secured creditors will share pari passu.168 However, because of the narrow definition of the notion of immovable by destination, this rule may have little practical value. A creditor holding a security right in a movable asset that has become immovable by destination will not be able to enforce its right under the 2016 Law on Security Rights. This is because the encumbered asset has become an immovable. Therefore, the creditor will either have to resort to the execution procedures relating to enforcement of rights in immovable property or to wait for the mortgagee to dispose of the immovable asset in order to assert its security right in the proceeds of the disposition. Outside insolvency proceedings, the normal priority rules apply where an asset is subject to a non-possessory pledge and a non-consensual lien.169 However, it is not clear whether the general Burundi laws give priority to non-consensual claims (privilèges généraux) of general nature, such as claims for owed wages. The same priority rule applies in collective insolvency proceedings, except as otherwise provided by other laws.170 Articles 38–41 of the 2016 Law on Security Rights cover the priority of security rights in assets that become commingled in a mass or transformed into a product, reflecting the approach of Article 33 of the Model Law.171 Article 85 deals with the matter of accessions, that is, when an asset subject to a retention of ownership is physically affixed to another movable asset.172 However, it is limited only to retentions of ownership, and the law does not provide a comparable rule for accessions subject to pledges. The Model Law did not incorporate recommendation 89 of the Legislative Guide, which sets out a priority rule for security rights in attachments to a movable asset. Under Article 85 of the 2016 Law on Security Rights, the incorporation of a movable asset subject to a retention of ownership into another movable asset does not affect the rights of the seller as long as the incorporated asset can be detached without any damage to the other asset.173 When the incorporated asset cannot be detached without damage, ownership of the whole asset will belong to the owner of the principal part. However, the owner of the whole asset must

166 E Steiner, French Law: A Comparative Approach (Oxford, Oxford University Press, 2018) 285. See also the French Civil Code, Arts 524–25. 167 This interest would be governed by the relevant immovable property law. 168 2016 Law on Security Rights, Art 46. 169 These privileges are listed in the Ordinance of the Governor General of 22 January, 1896 about Preferences added to the Civil Code and amended by other texts. See also C above. 170 2016 Law on Security Rights, Art 47. 171 See ch 2, C3. 172 See conclusion C1b 173 2016 Law on Security Rights, Art 85.

256 Burundi reimburse the seller for the value of the incorporated asset, pursuant to any conditions set out in the contract under which the retention of ownership arose.174

5. Enforcement a. Enforcement of Pledges In line with the Model Law, the enforcement chapter of the 2016 Law on Security Rights sets out general provisions applicable to the enforcement of pledges, which are then supplemented by specific rules that apply to certain types of encumbered assets, such as receivables and bank accounts. The enforcement chapter also contains provisions specific to sales on retention of ownership terms. The pledgee may repossess an encumbered tangible asset without application to the court if it provides a notice to the pledgor of its intention to obtain possession and the pledgor does not object at the time the pledgee attempts to obtain possession.175 Article 77(2) of the Model Law similarly makes the enforcement of the right to take possession conditional on the satisfaction of such requirements. However, unlike the Model Law,176 the 2016 Law on Security Rights does not recognise a general right to dispose of the encumbered asset extrajudicially. Generally, the pledgee must enforce its rights to dispose of the encumbered asset according to the judicial enforcement procedures prescribed in the applicable procedural law.177 However, the use of such disposal procedures is not necessary in two circumstances: first, when the encumbered asset is customarily sold on a recognised market,178 and second, when the pledgee is a bank or a financial institution and the security agreement provides for extrajudicial disposal.179 In these circumstances, to effect an extrajudicial disposal the pledgee must send a notice to the pledgor of its intention to sell the encumbered asset at least eight days before the date of the sale.180 The notification must contain a description of the collateral, an indication of the amount required to satisfy the secured obligation, the date after which the asset will be disposed of, and the manner of the intended disposition.181 The pledgee is under a duty to act in a commercially reasonable manner when disposing of the collateral in a private sale.182 Whether the assets are disposed of according to judicial procedures, or sold extrajudicially, the pledgor remains liable for any shortfall if the proceeds of the sale are not sufficient to satisfy the secured obligation fully.183 After the satisfaction of junior interests in the encumbered asset, the pledgor will be entitled to the surplus, if any.184

174 ibid. 175 ibid. 176 ML,

Art 78. Law on Security Rights, Art 54. 178 ibid Art 55(1). 179 ibid Art 55(2). 180 ibid Art 55(3). 181 ibid. 182 ibid Art 55(2). 183 ibid Art 56(1). 184 ibid Art 56(2). 177 2016

The 2016 Law on Security Rights and the 2016 Leasing Law 257 The pledgee is also entitled to propose acquisition of the pledged asset in satisfaction of the secured obligation. However, in this case, judicial involvement is usually required.185 Judicial involvement can be avoided if the pledge is between a bank or financial institution and a non-consumer grantor. In such cases, the parties can agree that upon default the pledgee may take the encumbered asset in satisfaction of the secured obligation.186 If the encumbered asset is not customarily sold on a recognised market, its value must be determined by a court appointed expert or pursuant to an agreement between the two parties.187 A pledge is extinguished when the secured obligation is fully satisfied.188 Moreover, a possessory pledge will cease to have any effect when the asset is voluntarily returned to the pledgor, lost, or when a court orders its return due to the pledgee’s failure to comply with the law,189 for example, where the pledgee fails to protect it against damage.190 The pledgor can fully perform the secured obligation at any time before enforcement is completed, and can redeem the encumbered asset, pursuant to Article 52 of the 2016 Law on Security Rights. Under a pledge of receivables, the security agreement (or later agreement) can provide either that the pledgor collects in the receivables or that they are to be collected by the pledgee before default. In the latter case, the pledgee must deposit the payments into a specific bank account and hold them as security until the maturity of the secured obligation, at which point the funds can either be applied in diminution of the secured obligation or returned to the pledgor.191 The Model Law does not contain a similar rule, although the parties could make an agreement to this effect. After default, the pledgee must notify the pledgor of its intention to collect the receivables or, in relation to payments it has already collected, enforce its rights against those deposited payments. Eight days after the notification, the pledgee may apply the deposited payments in partial or full satisfaction of the secured obligation.192 Similarly, the pledgee may apply the balance of a pledged deposit account in partial or full satisfaction of the secured obligation on default of the pledgor.193

b. Enforcement of Retention of Ownership Sales As mentioned above, the 2016 Law on Security Rights also provides enforcement rules specific to retention of ownership. As a general rule, if the buyer defaults in payment of the purchase price, the seller has the right to obtain, on request, the restitution of the asset subject to the retention of ownership.194 The rules also address a number of specific situations, including when inventory consisting of fungible assets is sold on retention of ownership terms and when the object in respect of which ownership has been retained is affixed to another movable asset. Regarding the former, the retention of ownership seller

185 ibid Art 57(1). cf ML, Art 80, which permits out-of-court acquisition subject to the fulfilment of certain conditions. 186 2016 Law on Security Rights, Art 58. 187 ibid Art 59. 188 ibid Art 60. 189 ibid Art 61. 190 See ibid Arts 32 and 62. 191 ibid Art 70(1). 192 ibid. 193 ibid Art 75. 194 ibid Art 86(1).

258 Burundi can enforce its right against all assets of the same kind and quality held by or on behalf of the grantor.195 In the case of an attachment to movable asset, the retention of ownership seller can enforce its rights and remove the object from the other asset if it can do so without damage.196 Since the retention of title, like any security rights, must not be a source of enrichment for the creditor, the Law provides that the value of the property taken back is applied towards the outstanding balance of the secured claim. Where the value of the property taken back exceeds the amount owed, the creditor owes the buyer an amount equal to the difference.197

c. Enforcement of Leases The 2016 Law on Leasing also provides a set of remedies. Under Article 10, if the lessee defaults, the lessor may terminate the lessee’s right to use the asset.198 As a condition of termination, the lessor must send a notice to the lessee and request compliance with the obligation within 30 days.199 However, the lessor may only recover the asset by applying to the court. The 2016 Law on Leasing does not provide for extrajudicial remedies, and lessors may not enforce their rights under the provision of the 2016 Law on Security Rights, which do not apply to leases. A court order may be granted by the President of the Commercial Tribunal or the President of the First Instance Court (Tribunal de Grande instance) of the place where the lessor’s business is located. The court will then appoint a bailiff who will supervise the recovery process.200 The court order can be appealed, but the appeal must be lodged within two days of the notification of that order to the lessee.201 The appellate court must then issue a decision within 30 days.202 These strict timelines ensure that the lessor is able to enforce its rights in an expeditious manner, which is likely to make credit more available.

6. Transitional and Conflict of Laws Rules By virtue of Article 88, the 2016 Law on Security Rights repeals all prior inconsistent laws.203 Similarly, Articles 34 and 36 of the 2016 Law on leasing provide that any inconsistent laws are repealed. These articles do not include a list of specific laws that were repealed or a list of articles in laws that are amended by the two new laws. The 2016 Law on Security Rights came into effect on the date of its promulgation, which was 12 August 2016.204 The 2016 Law on Leasing also became effective on the day it was enacted (5 January 2016).205 However, on

195 ibid

Art 84. Art 85. 197 This is consistent with the non-unitary approach, see ch 2, B2b. 198 2016 Law on Leasing, Art 10, para 1. 199 ibid para 2. 200 ibid. 201 ibid Art 10(3). 202 ibid. 203 2016 Law on Security Rights, Art 88. 204 ibid Art 89. 205 2016 Law on Leasing, Art 37. 196 ibid

The 2016 Law on Security Rights and the 2016 Leasing Law 259 this date the National Registry was not yet functional, and therefore the law was brought into force without its main supporting infrastructure, thus creating legal uncertainty. Bringing a secured transactions law into force before the establishment of a fully functional registry is not an uncommon development in Africa,206 but is very unfortunate. Under the transitional and final provisions of the 2016 Law on Security Rights, a prior security right that was effective against third parties under prior law continues its thirdparty effectiveness until the time it would have ceased to be effective.207 If the third-party effectiveness requirements of the new law are satisfied before the third-party effectiveness of a prior security right ceases in accordance with Article 87(1), the effectiveness and priority of the security right is preserved.208 Accordingly, all prior security interests, including pledges and retentions of ownership, will continue to remain effective against third parties until their effectiveness ceases under the prior law. This is a concern particularly with respect to long-term transactions entered into under the prior law, as a searcher of the National Registry established under the 2016 Law on Security Rights cannot proceed with confidence that certain assets are not encumbered under the prior law. The searcher would thus need to conduct due diligence beyond the 2016 regime for an indefinite period of time. In contrast, Article 105(1)(b) of the Model Law requires enacting States to stipulate an appropriate time period within which all prior security rights must be made effective against third parties under the newly implemented regime. Although seemingly a challenge for prospective secured creditors, long-term deals of this nature might be a rarity in Burundi. Furthermore, the predictable legal regime and efficient enforcement framework set out in the 2016 Law on Security Rights should be an incentive for prior secured creditors to comply with the requirements of the new law in relation to prior rights, or to create new rights under the new law. More troublesome, from a practical standpoint, is the absence of any conflict of laws rules relating specifically to secured transactions. Although, at present, cross-border transactions are a rarity in Burundi, as the economy grows it might be expected that it will attract the interest of foreign lenders or that Burundian companies may seek to expand their operations abroad. This would force local banks to determine the law applicable to security rights in assets located outside of Burundi or receivables owed by persons located in foreign countries. However, the omission of specific conflict of laws rules for secured transactions appears to be because the local stakeholders found them to be complicated and unnecessary within the local context. In the absence of specific conflict of laws rules, prospective creditors would need to examine the rules of the Civil Code. Under Article 6, agreements are governed by the law of the place where they are concluded, unless otherwise agreed by the parties. Accordingly, the parties may choose the law applicable to their security agreement. As for third-party effectiveness and priority, Article 3 provides that property rights in movable assets shall be governed by the law of the place where the asset is located. Although this rule provides some guidance with respect to tangible movable assets, it is difficult to apply it to intangible assets, which by definition do not have any physical location. It is thus uncertain whether,

206 See

conclusion C2a. Law on Security Rights, Art 87(1). 208 ibid art 87(2). This is similar to ML, Art 105(2). 207 2016

260 Burundi for the purpose of determining the governing law, the location of a receivable is where the pledgor (the receivables creditor) or the receivables debtor is located. Also, the Civil Code does not provide a predictable framework for many transactions involving tangible encumbered assets, such as transportation trucks that regularly cross the borders of neighbouring States. These uncertainties would have been avoided if Chapter VIII of the Model Law had been implemented in Burundi, as it sets out a comprehensive set of conflict of laws rules.

D. Final Remarks With the enactment of the laws on security rights and leasing in 2016, Burundi has taken significant steps towards modernising its outdated regime for secured transactions. Not only has it enacted new laws, but the 2016 Law on Security Rights is based on the Model Law. It has incorporated the fundamental elements of this model framework, including the creation of a security right on execution of a security agreement without any formalities such as notarisation,209 third-party effectiveness by registration of a notice, clear priority rules and efficient enforcement mechanisms. This is a huge step forward for a civil law country. At the same time, the 2016 Law on Security Rights deviates from the Model Law in some respects. Some of these deviations remain within the framework recommended by UNCITRAL, especially its Legislative Guide’s recommendations on the non-unitary approach to acquisition security rights that the 2016 Law on Security Rights chose to apply to retention of ownership sales. Others are deviations that take certain transactions beyond the UNCITRAL-recommended framework. For instance, under the enforcement chapter not all security rights may be enforced extrajudicially, whether by a sale or acceptance of the collateral in satisfaction of the obligation. Finally, other deviations are more significant, particularly the absence of a chapter on conflict of laws. Although the Model Law contains a comprehensive set of conflict of law rules, secured creditors applying the 2016 Law on Security Rights will need to resort to the Civil Code, which is not equipped to provide clear and predictable answers to the many questions that may arise, particularly when a secured transaction involves a foreign element. Likewise, the transitional rules do not provide any cut-off time period within which all prior security rights would need to be brought into compliance with the 2016 Law on Security Rights. Nonetheless, while seemingly a significant gap in the legislation, the practical effect of this omission may be mitigated by the current low overall use of movables in secured transactions and the absence of long-term deals of this nature. Even so, the unavailability of the National Registry at the time the 2016 Law on Security Rights came into force left the new secured transactions framework essentially inoperative. It is hoped that these reforms will increase access to secured finance in Burundi. However, the law and the future National Registry might not have a significant economic impact due to a number of factors. First, as described in the introduction to this chapter, local lenders have very little expertise in movables lending. Second, businesses and individuals are not accustomed to borrowing funds using their assets as collateral, and generally resist the repayment of debts. Third, the political environment remains unstable, presenting

209 Although

this is preserved as an alternative to a privately signed contract, see C1 above.

Final Remarks 261 a number of risks especially for mid- and long-term secured transactions. Once properly implemented, the 2016 Law on Security Rights is expected to significantly boost Burundi’s ranking in the World Bank’s Doing Business Report of 177 in the Getting Credit indicator. However, due to the factors identified in this paragraph the actual economic impact, especially on the agricultural sector, which employs 90 per cent of the population, may not be meaningful.

12 Ethiopia The current secured transactions system of Ethiopia exhibits a number of weaknesses, including the limited availability of non-possessory security rights for only a narrow set of transactions, the absence of a general registration system and cumbersome enforcement mechanisms. The inefficient legal framework coupled with the Ethiopian banks’ lack of expertise in SME finance explain why SME loans comprise only 7 per cent of the banks’ loan portfolios.1 One illustration of the lack of expertise in secured lending is the common practice of excessive overcollateralisation, which has resulted in collateral requirements averaging 234 per cent of the loaned amount.2 These deficiencies have been amply documented in studies which empirically show the negative economic impact caused by the lack of access to reasonably priced secured credit. As a result, policymakers have taken the decision to reform the secured transactions legal framework. This process has been proceeding on the basis of the UNCITRAL Model Law on Secured Transactions. Since a new law is yet to be enacted, this chapter examines and highlights only certain features of the draft, focusing particularly on those in which the drafters have departed from the Model Law.

A. Economic Overview Ethiopia ranks below the sub-Saharan Africa region average for private-sector credit as a percentage of GDP.3 Studies published by the World Bank Group (WBG) and others have amply documented the lack of access to credit for SMEs in Ethiopia, and diagnosed its causes. According to the WBG 2011 Enterprise Survey, 94.7 per cent of all firms (92.9 per cent of small firms, 95.5 per cent of medium firms and 99 per cent of large firms) in Ethiopia had a current or savings account.4 Despite such a high connection to the banking sector, 33 per cent of all firms identified access to credit as the biggest obstacle to their growth, and only 15.8 per cent reported having a bank loan or line of credit.5 A February 2015 WBG report titled ‘SME Finance in Ethiopia: Addressing the Missing Middle Challenge’ (‘Missing Middle Report’) provides further evidence of an environment inhospitable to

1 SME Finance in Ethiopia: Addressing the Missing Middle Challenge (Washington, DC, World Bank, 2015) 13 (hereinafter SME Finance in Ethiopia). 2 World Bank, ‘Enterprise Surveys: Ethiopia Country Profile 2011’ (2011) www.enterprisesurveys.org/~/media/ FPDKM/EnterpriseSurveys/Documents/Profiles/English/Ethiopia-2011. 3 See World Bank, ‘Domestic Credit Provided by Financial Sector (% of GDP)’ (2017) https://data.worldbank. org/indicator/FS.AST.DOMS.GD.ZS. 4 World Bank (n 2) 11. 5 ibid.

Economic Overview 263 credit transactions.6 Although the majority of micro and small enterprises have current and savings accounts, only 6 per cent of micro-enterprises, 1.9 per cent of small enterprises and 20.5 per cent of medium-sized businesses had a loan or a line of credit.7 Interviews conducted in Addis Ababa, Ethiopia, from 27 to 30 October 2015 (‘2015 interviews’), as part of an IFC-supported reform project, revealed that financial institutions are reluctant to extend credit against inventory and receivables unless the inventory is ‘locked in a warehouse’.8 The vast majority of credit remains secured with motor vehicles.9 The 2015 interviews revealed that Ethiopian lenders prefer to take security rights in machinery and motor vehicles for which registries have been established. In contrast, they are reluctant to take security rights in crops, accounts receivables or inventory because third-party effectiveness of security rights in those assets is not dependent on registration. The lack of adequate registration mechanisms results in uncertainty as to the existence of prior claims to the same asset. As a result, the financial institutions interviewed charge higher interest rates for loans secured with such assets. Furthermore, some lenders accept only those assets as collateral for which insurance may be procured, such as motor vehicles. The need for secured financing in the agricultural sector is urgent given that it accounts for 45 per cent of the country’s total output, employs nearly 80 per cent of the labour force, and represents more than 80 per cent of exports.10 Agriculture and services are the main growth industry, with manufacturing contributing only around 4 per cent of the GDP.11 Livestock farming in Ethiopia accounts for 12 per cent of the total GDP, 33 per cent of the agricultural GDP, and employs 65 per cent of the population.12 Ethiopia was ranked 175 out of 190 economies in the Getting Credit indicator of the 2019 World Bank Doing Business Report.13 Ethiopia’s credit infrastructure is weak when compared with other economies in sub-Saharan Africa under the Getting Credit indicator, ranking 43 out of the 48 economies in the region.14 Ethiopia’s financial sector is dominated by banks, which in 2014 accounted for more than 92.6 per cent of the total assets of the financial sector,15 with private banks comprising just 30 per cent of the total assets of the banking industry, compared to the 58 per cent share of the Commercial Bank of Ethiopia (CBE), a state-owned bank.16 MFIs represented 5.2 per cent and insurance companies 2.2 per cent of the total financial sector assets.17

6 SME Finance in Ethiopia (n 1). 7 ibid. 8 On file with the authors. 9 The Credit Reference Bureau reports indicate that 90 per cent of all movable collateral are motor vehicles. Statistics provided by the National Bank of Ethiopia are on file with the authors. 10 World Bank, ‘Ethiopia: World Bank to Finance Expanded Agriculture and Better Livelihood Opportunities for its Small Farmers’ (2015) www.worldbank.org/en/news/press-release/2015/03/31/ethiopiaworld-bank-to-finance-expanded-agriculture-and-better-livelihood-opportunities-for-its-small-farmers. 11 SME Finance in Ethiopia (n 1) 17. 12 A Solomon et al, ‘Livestock Marketing in Ethiopia: A Review of Structure, Performance and Development Initiatives’ International Livestock Research Institute Socio-economics and Policy Research Working Paper 52 (2003). 13 World Bank, ‘Rankings & Ease of Doing Business Score’, www.doingbusiness.org/rankings. 14 ibid; Doing Business 2019: Ethiopia (Washington, DC, World Bank, 2019) 28. 15 ‘Ethiopia: Banking Sector Liberalisation Can Lift Growth’ in ‘Middle Africa Insight Series’ Ecobank (30 December 2014). 16 H Belda, ‘Ethiopian Banking Sector Development’ (2017) 7 Research Journal of Finance and Accounting 25, 27. 17 ‘Ethiopia: Banking Sector Liberalisation Can Lift Growth’ (n 15).

264 Ethiopia In 2018, the Ethiopian financial sector consists of 3 government-owned banks,18 16 private banks,19 17 insurance companies,20 33 MFIs,21 and 5 capital goods finance companies.22 The top five banks were responsible for 77.7 per cent of loans disbursed between 2008 and 2014, making the Ethiopian banking sector highly concentrated.23 State-owned banks, which mainly finance large enterprises, constitute 70 per cent of total assets of the banking sector.24 Ethiopia has taken a more cautious approach to liberalising its banking industry than other African economies.25 These features distinguish the Ethiopian banking sector from the rest of sub-Saharan Africa. The National Bank of Ethiopia’s (NBE) policies affect commercial lending in various ways. One NBE Directive requires private banks to allocate 27 per cent of their loan portfolios to the purchase of government securities.26 The NBE also imposes comparatively high capital requirements on the formation of new banks, thereby fostering the dominance of the government-owned banks.27 The devaluation of the local currency in October 2017 resulted in a number of policies designed to manage the allocation of credit within the economy. For example, the NBE restricted banks from growing their loan portfolios beyond 16.5 per cent from the previous year’s balances with the exception of the export sector.28 The business models of the financial institutions are mostly inadequate to serve SMEs. According to the World Bank Group: They lack a dedicated and specialised SME unit or department within their organisational structure; loan appraisal techniques are still mostly based on traditional relationship lending rather than on transactional technologies such as credit scoring; products are highly standardised and there is very limited product innovation.29

B. Legal System Ethiopia’s legal framework is a civil law–common law hybrid system.30 Unlike many other African economies, Ethiopia was never colonised and was thus not exposed to outside culture and influences associated with colonisation.31 However, foreign influence featured 18 National Bank of Ethiopia, ‘Banks’, www.nbe.gov.et/financial/banks.html. 19 ibid. 20 National Bank of Ethiopia, ‘Insurers’, www.nbe.gov.et/financial/insurer.html. 21 National Bank of Ethiopia, ‘Microfinance Institutes’, www.nbe.gov.et/financial/microfinanceinstitutes3.html. 22 See A Olana, Lease Financing in Ethiopia: An Assessment of Five Leasing Companies (Lambert Academic Publishing, 2017). 23 Belda (n 16) 28–29. 24 ibid 28. 25 T Keatinge, ‘The Role of Public and Private Sector Banking in Ethiopia’s Future Economic Growth’ Global Centre on Cooperative Security (September 2014) www.globalcenter.org/wp-content/uploads/2014/09/14Sept26_ Public-Private_PB.pdf. 26 SME Finance in Ethiopia (n 1) 47. 27 ibid. 28 Muluken, ‘NBE Applies New Policies Following the 15% Birr Devaluation’ Capital Ethiopia Newspaper (Addis Ababa, 6 October 2017) available at http://ethiodailypost.com/2017/10/16/nbe-applies-new-policies-followingthe-15-birr-devaluation/. 29 SME Finance in Ethiopia (n 1). 30 ibid. 31 J Beckstrom, ‘Transplantation of Legal Systems: An Early Report on the Reception of Western Laws in Ethiopia’ (1973) 21 American Journal of Comparative Law 557, 557–58.

Legal System 265 prominently in the codifications of the 1950s and 1960s that modernised the ‘Law of Kings’ (Fetha Negast), which had been in force since the seventeenth century.32 The Fetha Negast had never been officially adopted by an Ethiopian king/emperor, and applied as a private compilation of laws.33 The codifications of the 1950s and 1960s resulted in the enactment of the Penal Code in 1957, Civil, Commercial and Maritime Codes in 1960, a Criminal Procedure Code in 1961 and a Civil Procedure Code in 1965.34 The process by which these codes had been drafted in French, and then translated into Amharic and English, resulted in a number of difficulties in properly conveying the concepts and terminology of the original text.35 The codes were intentionally drafted to address issues comprehensively and in detail in order to minimise the need for judicial interpretations because of the dearth of qualified judges and other legal professionals at that time in Ethiopia.36 Thus, the modernisation of Ethiopia’s legal system through codification differed from the rest of Africa (at that time) where modernisation was sought through the drafting of restatements of customary law.37 René David, one of the drafters of the Civil Code, noted that ‘the decision to codify was based on the idea that codification in itself is progress’.38 The Civil Code of Ethiopia follows various European models, including the civil codes of France, Portugal, Switzerland and Greece.39 However, in relation to its approach, style and substance, the French Civil Code was the predominant influence.40 The Commercial Code was also modelled on the French Commercial Code.41 It was criticised for taking inspiration from the French model, which was viewed at the time as outdated, having been designed largely for transactions of shopkeepers and artisans in the seventeenth century.42 Others criticised its highly technical nature and its approaches, which were designed for a society that did not exist in Ethiopia at the time of enactment, and predicted that many provisions would remain dormant for decades.43 David observed that the allocation of matters between the two Codes was not driven by any commercial criteria, and displays a degree of arbitrariness.44 Due to its detachment from the economic relationships and commercial transactions existing in Ethiopia, many of its provisions were ignored in practice. For instance, merchants continued to use the traditional methods of resolving insolvency rather than the bankruptcy law provisions set forth in the Commercial Code.45 As a result of the mixture of the outdated nature of the Commercial Code and a degree 32 P Brietzke, ‘Private Law in Ethiopia’ (1974) 18 Journal of African Law 149. 33 J Vanderlinden, ‘Civil Law and Common Law Influences on the Developing Law of Ethiopia’ (1966) 16 Buffalo Law Review 250, 250–51. 34 Beckstrom (n 31) 559. 35 Vanderlinden (n 33) 257. 36 ibid 260. 37 K Tuori, ‘Legal Pluralism and Modernization: American Law Professors in Ethiopia and the Downfall of the Restatements of African Customary Law’ (2010) 62 Journal of Legal Pluralism and Unofficial Law 43, 43–44. 38 R David, ‘A Civil Code for Ethiopia: Considerations on the Codification of the Civil Law in African Countries’ (1962) 37 Tulane Law Review 187, 188. 39 Civil Code of the Empire of Ethiopia Proclamation No 165/1960 (Civil Code). The drafters were also influenced by the codes of Egypt, Iran, Israel, Turkey, Quebec and the Philippines. M Ayalew, ‘Contracts’ in The International Encyclopaedia of Laws (Dordrecht, Kluwer Law International, 2010) ch 6 (Ethiopia). 40 Brietzke (n 32) 150. 41 Commercial Code of the Empire of Ethiopia Proclamation No 166/1960 (Commercial Code). 42 Brietzke (n 32) 152. 43 Beckstrom (n 31) 570. 44 David (n 38) 197. 45 Beckstrom (n 31) 570.

266 Ethiopia of high technicality in the financial industry, SMEs could only obtain short-term credit against postdated cheques. The repayment of such loans was ‘guaranteed’ by the sanctions imposed by the Penal Code for the writing of cheques without sufficient funds to cover.46 Furthermore, Book V of the Commercial Code provides bankruptcy rules that are based on the approach that the bankrupt is to be blamed for a business failure, and that bankruptcy is a form of punishment.47 With respect to banks, the Banking Business Proclamation 592/2008 provides additional rules on insolvency. Between these two regimes, there are many gaps that result in the unavailability of any insolvency regime for some types of debtors (eg consumers and sole proprietors).48 Whereas under the Commercial Code secured creditors stand outside of bankruptcy, and may enforce their security rights unaffected by it, the Cassation Division of the Supreme Court recently held that mortgagees are subject to bankruptcy proceedings. This decision, which contradicted the black-letter law in the Commercial Code, has led to some confusion in the market.49 The Civil and Commercial Codes, as well as other Ethiopian statutes, borrowed some common law principles from English, US and Indian law. In particular, the procedural codes were influenced by English law.50 The court structure reflected the English influence, and the appointment of English judges to serve on Ethiopian courts led to the introduction of the doctrine of precedent into Ethiopian legal thinking.51 English scholars were also involved in a project to compile a restatement of African customary law, including for Ethiopia.52 Common law influence was also instilled through education at the Addis Ababa law school where many professors came from US universities.53 After the enactment of the new Constitution in 1995, Ethiopia commenced work on modernising its laws.54 Over the past few years, several efforts have been made to revise the Commercial Code.55 An initiative to reform the secured transactions system has been part of these efforts since 2015. The reform initiative was delayed for a while due to political challenges, but was re-energised toward the end of 2017.

C. Overview of the Current Secured Transactions Regime The current legal framework governing secured lending against movable assets in Ethiopia is fragmented and found in several pieces of legislation. The Civil and Commercial Codes,

46 Brietzke (n 32) 164. 47 Meheret, ‘An Appraisal of the Ethiopian Bankruptcy Regime’ (2017) 50 De Jure Law Journal 111, 111–12. 48 ibid 130. 49 ibid 127. 50 Brietzke (n 32) 159. 51 Vanderlinden (n 33) 255. 52 Tuori (n 37) 46. The efforts to put together Restatements of customary laws eventually failed because they ceased to be of interest to the African leaders who were looking to modernise their legal frameworks, and did not associate collections of ancient laws with modernisation. ibid 61. 53 ibid 50. 54 Constitution of the Federal Democratic Republic of Ethiopia (1995). This Constitution replaced the former centralized unitary government with a federal democratic government. 55 Position of the Business Community on the Revision of the Commercial Code of Ethiopia (Addis Ababa, Addis Ababa Chamber of Commerce, 2008).

Overview of the Current Secured Transactions Regime 267 the Business Mortgage Proclamation,56 the Proclamation to Provide for a Warehouse Receipts System (‘the Warehouse Receipts Proclamation’)57 and the Capital Goods Leasing Business Proclamation58 provide different devices to secure obligations with movable assets. Proclamation is the primary type of legislation that would be equivalent to acts in other jurisdictions.59 Ethiopian legislation is adopted in Amharic although its text is often published in side-by-side Amharic and English versions. The Civil Code governs pledges of tangible assets and claims. It also provides special rules for pledges of warehouse receipts and securities, retentions of title, and assignment of rights. The Civil Code rules on pledges of warehouse receipts were modified by a Proclamation adopted in 2003.60 The Commercial Code regulates the mortgage of a business with further details included in the related Proclamation. The Capital Goods Leasing Business Proclamation governs financial and operating leases as well as hire-purchase agreements. These laws provide rules for creation, third-party effectiveness, registration, priority and enforcement of what would be security rights under the Model Law, which differ according to their nature, the type of movable asset, and the type of creditors and borrowers.61 Rights in many types of movable assets are not subject to any form of registration, exacerbating the risk of secret liens. Ethiopia has established several asset-specific registries in which agreements creating rights in those assets may be registered. These registries operate from physical offices located throughout the country. For instance, interests in vehicles are registered with the Transportation Authority of the corresponding locality where a vehicle’s plate number was issued. Ethiopia is a federal state divided into nine regions and two cities that retain certain powers to operate registries.62 In the absence of a unified legal framework and interconnection among the local registries, relocation of the vehicle to another region raises questions as to the effect of the registration in the original region. Financial institutions reported that some registry offices are more efficient than others.63 Some registries charge flat fees whereas others charge a fee based on a percentage of the asset value or the loan amount. Registration fees are charged in addition to stamp duties payable upon the execution of a security agreement.64 All registries are document-registration systems that require the submission of a registration/application form that must be accompanied by the agreement providing for the rights of the creditor. The records of the registry offices are neither accessible electronically nor connected. The challenges are further exacerbated by the power of local authorities to issue their own regulations governing the operations of the local registry offices, as well as to prescribe their fees. The effect of relocation of the mortgagor or encumbered assets to another locality on the rights of creditors is not addressed.

56 Business Mortgage Proclamation No 98/1998. 57 Warehouse Receipts Proclamation No 372/2003. 58 Capital Goods Leasing Business Proclamation No 103/1998. 59 B Fikre, ‘Ethiopia’s Legislative Drafting: A Bird’s Eye View’ (2016) 5 International Journal of Legal Discourse 31, 33. 60 Warehouse Receipts Proclamation No 372/2003. See below D7. 61 See the definition of security right in ML, Art 2. 62 Constitution of the Federal Democratic Republic of Ethiopia (1995) Art 47. 63 October 2015 interviews on file with the authors. 64 See Proclamation to Provide for the Payment of Stamp Duty, Proclamation No 110/1998, art 3.

268 Ethiopia Business mortgages are registered in the ‘Regional or City Bureau of Commerce and Industry or a Regional or City authority entrusted with the power to register mortgages’.65 Any concerned person may register a business mortgage upon completion of the necessary application forms.66 The law requires that the bureau or other authority entrusted with the registration of a business mortgage verify the information entered in the application form.67 For some assets, specific registries have not been established even though contemplated by the law. For instance, the 1999 Registration and Control of Construction Machinery Proclamation provides the basis for the registration of ownership rights to construction machinery with the Ministry of Urban Development and Construction. As stated in Article 3, one of its objectives is to ‘facilitate conditions so that construction machineries can be used for mortgage and other similar purposes’.68 Construction machinery is defined in Article 2(2) to exclude construction machinery registered pursuant to Proclamation No 14/1984 with regional Bureaux. The 1999 Proclamation does not include any specific language that would implement the objective to facilitate the mortgaging of construction machinery, so it remains unclear whether the registries indeed process such registrations. Some assets may be encumbered under different security devices registered in different registries. For instance, a business mortgage registered with the Bureau of Commerce and Industry might include a vehicle that could also be subject to a specific encumbrance registered with the Transport Authority. The Ethiopian legal framework does not provide a priority rule to resolve such conflicts. The current legal framework authorises extrajudicial remedies in the following transactions: 1.

The Civil Code Articles 2851(1)–2854 provide for the sale of the collateral, by public or private auction, upon an eight-day notice. 2. The Civil Code Article 2851(2) allows the parties to agree to the appropriation of the collateral, but only after default. 3. The Civil Code Article 2821 allows the pledge of a warehouse receipt to sell extrajudicially, but only after satisfying the Commercial Code requirements of protest that may be done before a notary; the Warehousing Proclamation authorises the pledgee to sell the goods covered by a warehouse receipt 15 days after the date indicated in a notice given to the pledgor. 4. Commercial Code Articles 329 and 342 allow the pledgee to sell the pledged shares by a public auction if the shareholder has not met the call of the company on the pledged shares upon company giving a 15-day notice to the shareholder. 5. Commercial Code Article 949 allows the bank to sell securities on which it made an advance. 6. Commercial Code Article 189 authorises attachment under a mortgage of a business only upon a court order and after a 30-day notice to the mortgagee; the Mortgage 65 See pt 1 and 2 of the Business Mortgage Proclamation (No 98/1998); see also A Gikay, ‘Rethinking Ethiopian Secured Transactions Law through Comparative Perspective: Lessons from the Uniform Commercial Code of the US’ (2017) 11 Mizan Law Review 153, 178. 66 Business Mortgage Proclamation, Art 5. 67 ibid Art 9. Incorrect information can be corrected in accordance with the instructions of the Bureau; Art 10. See further Art 12, for a similar requirement of verification with respect to cancellations. 68 Registration and Control of Construction Machinery Proclamation No 177/1999.

Current Security Devices 269 usiness Proclamation Article 13 authorises extrajudicial sale of the business or its B transfer to the mortgagee after the second auction. 7. Leasing Proclamation Article 6 authorises the lessor to rescind the agreement and repossess the goods after the expiration of a 30-day period to cure the default. The Enforcement Proclamation and the Business Mortgage Proclamation were enacted to specifically deal with enforcement of the rights of secured creditors that are banks.69 The Enforcement Proclamation allows creditors that are banks to agree with the debtors to repossess and sell the collateral extrajudicially, on a three-day notice, prior to default.70 The enforcement rights of creditors were strengthened particularly by empowering the Registrar to take some measures ‘for carrying out the sale by auction’, and thus assist the creditor in enforcing its rights without having to resort to the judicial system.71 As listed above, the Business Mortgage Proclamation authorises extrajudicial disposal of the encumbered business. The sale must be for the value of the business as estimated in the loan agreement (or, presumably, for a greater amount). If this amount cannot be realised in two public auctions, the mortgagee can take over the business in consideration of the estimated amount. In a similar way to the empowerment of the Registrar of relevant pledges under the Enforcement Proclamation, Article 18 empowers the Bureau in which the business mortgage has been registered to ‘take the necessary measures for carrying out the sale by auction’. Both the Registrar and the Bureau have the power to order the police force to carry out such measures.72 Relatedly, Article 17 of the Leasing Proclamation imposes a very unusual duty on the Ministry to ‘take appropriate measures to ensure return of the leased capital goods to the lessor’ if the lessee does not return them when obliged to do so. The Ministry also has the power to order the police to facilitate repossession. The inefficiency of the court system forced the drafters to provide for this unconventional assistance from the Ministry. The success of these procedures in practice led to their adaptation for the purposes of the draft Secured Transactions Proclamation, whereby the Collateral Registry will have the power to order the police to provide assistance to the secured creditor in repossessing the encumbered asset.

D. Current Security Devices This section examines the security devices currently available under the Ethiopian legal framework, providing background for the following section that illustrates how this framework will be affected by the adoption of the proposed Proclamation to reform the secured transactions system.73

69 Property Mortgaged or Pledged with Banks Proclamation No 97/1998 and Business Mortgage Proclamation No 98/ 1998. A bank is defined by the Banking Business Proclamation as a company licensed by the NBE to undertake banking business or a bank owned by the Government. 70 The Proclamation overrides Arts 2851 and 3060 of the Civil Code. It also provides for extrajudicial sale of mortgaged immovable property. 71 Enforcement Proclamation, Art 8. 72 ibid Art 8(2); and Business Mortgage Proclamation, Art 18(2). 73 See further, Gikay (n 65) 166–67.

270 Ethiopia

1. Pledge The Civil Code provides for the classical security device of possessory pledge in Book V (Special Contracts), Chapter 6 (Contracts of Pledge in General). Article 2825 defines a pledge as ‘a contract whereby a debtor undertakes to deliver a thing, called the pledge, to his creditor as security for the performance of an obligation’. The pledge may secure a future or conditional obligation.74 If the maximum amount exceeds 500 birr (approximately US$17 as of June 2019), the pledge agreement must be in writing and indicate the maximum amount of the debt secured.75 Any movable asset, including intangible assets, as well as ‘a totality of effects’ may be pledged, but only to the extent the asset is capable of being sold separately in a public auction.76 The creditor need not have actual possession of the goods pledged: possession of a document of title without which the goods cannot be disposed of will suffice,77 as will constructive possession where the goods are in the actual possession of a third party.78 As a general rule, any stipulation in a pledge agreement whereby the pledgor may remain in possession of the asset is void.79 While recognising only the possessory pledge, the Civil Code defers to other laws that may provide for a non-possessory pledge.80 In fact, legislation has been enacted for non-possessory security devices, particularly in the form of a business mortgage.81 Section 2 of Chapter 6 provides for pledges of claims and other intangible assets. Pledges of claims not embodied in a paper document require execution of a written contract that indicates the maximum amount of the secured obligation and ‘specifies the claim pledged’.82 For a pledge to have any effect as between the pledgee and pledgor, as well as against third parties, the person who owes the performance of the claim (eg debtor of the receivable) must be notified.83 This requirement is especially burdensome when the pledgor has multiple receivables owed by different debtors or seeks to encumber future receivables. Article 2872 appears to require the pledgee to collect payment on the pledged debt when it falls due. Presumably, the account debtor obtains a good discharge from paying the pledgee, and the pledgee owes an obligation to account for any surplus value as this arises from the nature of a pledge. On default, the pledged claim may be (i) collected, (ii) sold in accordance with the requirements of section 1 of Chapter 5 governing the enforcement of pledges in tangible assets, or (iii) accepted by the pledgee in settlement of the secured debt (satisfaction of the secured obligation).

2. Assignment of Rights The Civil Code governs the outright assignment of rights in Articles 1962–75. Under Article 1962, a right may be assigned to an assignee unless the assignment is prohibited by

74 Civil

Code, Art 2827. Art 2828(2). 76 ibid Art 2829. 77 ibid Art 2830(1). 78 ibid Art 2831 (1). 79 ibid Art 2832. 80 ibid. 81 See D5 below. 82 Civil Code, Arts 2864(1) and 2865. 83 ibid Art 2864(2). 75 ibid

Current Security Devices 271 law or an agreement of the parties. Accordingly, a right in the nature of a receivable subject to an anti-assignment clause cannot be assigned. If a right is assigned to multiple assignees, their priority is determined by the date of notification given to the debtor.84 Priority in relation to other third parties is not dealt with by the Civil Code, and nor is the position between multiple assignees where notification to the debtor has not been given.

3. Advances on and Pledges of Securities Sections 2 and 3 of Chapter 5 (Credit Transactions) of the Commercial Code deal with advances on securities and pledges of securities, respectively. Article 947 provides for an advance against securities by a banker. Any securities that are transferable may be pledged.85 The creditor may be in direct possession of the securities or hold them constructively through a third party that has ‘placed securities in a special account’.86 The Commercial Code does not specify what action would satisfy the requirement to place securities in a special account; the reference to placing the securities in an account may mean that the provisions on pledge extend to dematerialised securities as well as certificated ones. In any event, where securities are registered in the company register, the creditor must notify the issuer to register a ‘transfer by way of guarantee’ upon which the creditor would be deemed to be in possession.87

4. Retention of Title The Civil Code governs sale agreements with a retention of title clause that may be used to secure the payment of the purchase price.88 Under the retention of title clause, a seller can retain ownership to the goods until the buyer pays the purchase price in full. However, the retention of title clause has no effect against third parties unless it has been registered in the relevant registry.89 The Civil Code establishes that the registration must be in a public registry kept for this purpose at the place where a buyer resides.90 As noted above, registries have not been established for many types of assets, so although a retention of title arrangement may give the unpaid seller rights against the buyer, it remains unclear how it can protect itself against other persons who have a proprietary claim to the goods. The reference to the public registry in Article 2387 of the Civil Code must be read to mean the registry relevant to the type of asset. As a result, agreements that contain retention of title clauses with respect to motor vehicles must be registered with the Transportation Authority. The local offices of the Transportation Authority, which is part of the Ministry of Transport, register ownership over vehicles and issue title certificates. For the financing of 84 ibid Art 1967(2). The alternative to the notification of the receivables debtor is ‘a document with an authenticated date’. 85 Commercial Code, Art 950. 86 ibid Art 952(2). 87 Commercial Code, Art 952(3). 88 In Book V, Title XV, ch 1 (sale) s 4, para 5. 89 See Civil Code, Art 2387. 90 ibid.

272 Ethiopia motor vehicles, a creditor will request the title certificate from the borrower and file a letter with the Transportation Authority requesting it to register the interest and not to transfer the vehicle’s ownership without the creditor’s authorisation.91 The Civil Code also provides for letting or hiring agreements under which the possession of an asset is transferred to the lessee for a fixed period of time, after which the asset must be returned to the lessor.92 A letting or hiring agreement may confer a right on the lessee to acquire ownership to the object of the agreement after completing a number of payments,93 but, if so, it is characterised as a contract of sale, and the ‘lessor’ must satisfy the requirements applicable to the registration of agreements that contain a retention of title clause.94

5. Leasing The Capital Goods Leasing Business Proclamation, as amended in 2013,95 governs three types of leases that may be entered into only by institutions licensed for that purpose by the Ministry of Trade.96 The Leasing Proclamation limits the types of assets that can be leased to ‘capital goods’, which were defined in the amendment to the Proclamation as ‘any equipment or machine used to generate products or to provide services, and its accessories’.97 Leasing is defined as ‘a financing-in-kind … by which the lessor provides the lessee with the use of specified capital goods on financial or operating lease, or hirepurchase agreement without the requirement of collateral for a period of time’.98 The Leasing Proclamation governs and defines three types of leases. First, under the financial lease, a lessor provides a lessee with goods already acquired by the lessor or purchased by the lessor according to the lessee’s specifications. Under a financial lease, the lessor retains full ownership in the goods for the duration of the contract and the lessee may have99 an option to purchase at its expiration. The lessee has no ownership right whatsoever during the period of the lease. The second type of lease is hire purchase100 under which the lessee acquires proportional ownership corresponding to each payment it makes. Full ownership is automatically transferred to the lessee upon the last payment.101 The third type is the operating lease, by which a lessor provides a lessee the use of goods for a maximum period of two years with no option to purchase the goods.102

91 Creditors identify the corresponding office to register a security interest through the plate numbers, which contain a code linked to the region and office numbers. 92 In Book V, Title XVII, ch 2. 93 ibid Art 2728. 94 That is, the requirements in Art 2387. 95 The 2013 Amendment replaced Art 3 of the original Proclamation, and added a new Art 4. 96 Capital Goods Leasing Business Proclamation No 103/1998, amended by Proclamation No 807/2013, Arts 3 and 4 (as renumbered). 97 ibid Art 2(8). 98 ibid Art 2(1). 99 Whether there is an option to purchase depends on the agreement between the parties; ibid Art 2(3)(b). 100 See also Kenya B2 and Nigeria B5. 101 Capital Goods Leasing Business Proclamation No 103/1998, Art 2(4). Note that this is different from a common law hire-purchase. See, eg, Ghana B7; Zambia C4. 102 Capital Goods Leasing Business Proclamation No 103/1998, Art 2(5).

Current Security Devices 273 Under all three types of leases, the lessor remains the full owner of the goods with the exception of a hire purchase where the lessee incrementally acquires ownership with each payment. Article 6 reiterates that ‘the lessee shall hold the capital goods as a mere bailee … and shall not have any ownership right to the goods’. However, Article 12, entitled ‘Claim by Third Parties’, states that the ownership right of the lessor shall at all times prevail over claims by third parties, such as the creditors of a lessee. This priority rule appears to be redundant in light of Article 6 under which the lessee is a bailee, and Article 14, which renders any charge or encumbrances created by the lessee void. Logically, there cannot be any competition between the ownership right of the lessor and a right of a third party in the capital goods (the right of a third party would be void). However, this rule might have been designed to address a situation where a non-consensual creditor attempts to enforce its rights against the lessee’s right to possession. With respect to a hire purchase, Article 14 seems to limit unduly the ability of a lessee to encumber its proportional ownership right. This differs from the non-unitary approach to acquisition financing of the Legislative Guide which recognises the ownership right of the lessor, but under which the lessee can create a security right as if it were the owner of the asset. In such cases, the maximum value that a junior secured creditor or other third party can recover is limited to the surplus value in the goods over and above the amount due to the lessor.103 As will be explained below, the non-unitary approach is the basis of the draft Secured Transactions Proclamation. Under the Lease Proclamation, a lessee that acquires some goods subject to hire purchase may not encumber even its proportional ownership right. Article 17(1) contributes to the confusion as it requires the Ministry to establish a registry for the registration of lease agreements, but does not set out the legal effect of the failure to register a lease agreement. The likely effect is that the lessor’s ownership right may not be effective against buyers of the leased asset. One could read Article 11(2) (establishing the priority rule of the lessor’s ownership right against ‘claims by third parties’) as being limited to the claims of creditors of the lessee, but not those of buyers. Nonetheless, the legal effect of (non)registration of lease agreements on the ownership right of the lessor and rights/ claims of third parties remains very uncertain. The Lease Proclamation may have provided the framework to develop leasing products, but the ambiguities with respect to the effect of the lessor’s ownership right on the right of the lessee and third parties result in an unpredictable and uncertain legal environment. The proposed secured transaction law will regulate the rights of lessors (with the exception of operating lessors) vis-a-vis the claims of third parties, especially secured creditors of the lessee, in a more coherent way.

6. Mortgage of a Business The Commercial Code governs the mortgage of a business in Book I Title V (Businesses) Chapter 4,104 and is supplemented by the Business Mortgage Proclamation.105 According to the October 2015 interviews, this device is quite regularly used by Ethiopian lenders.

103 LG

recommendation 190. See also ch 2, B2b. Code, Arts 171 et seq. 105 Business Mortgage Proclamation No 98/1998. 104 Commercial

274 Ethiopia Article 124 of the Commercial Code defines business as ‘an incorporeal movable consisting of all movable property brought together and organised for the purpose of carrying out any commercial activity’.106 A ‘business’ is operated by a ‘trader’.107 Although it is not spelt out, it appears that a ‘trader’ can be a natural person or one of several types of legal person.108 ‘Business’ is therefore a collective name for all the value of the business, including goodwill and intellectual property, as well as any tangible assets such as equipment or goods.109 The inclusion of other intangible assets is less clear: the right to lease the business premises is expressly included, but not the other ‘assets and debts’ of the trader.110 The Commercial Code provides for two types of mortgage of a business: one created by a contract (contractual mortgage), and one given to the seller of a business to secure the payment of the purchase price that remains outstanding (legal mortgage).111 Only a business with respect to which the trader has been properly registered and issued an operating licence by the corresponding Bureau of Commerce and Industry can be the subject of a mortgage.112 Accordingly, this security device remains out of reach for most persons who operate Ethiopian businesses ‘informally’. Both types of mortgage must be registered in the corresponding Bureau of Commerce and Industry where the mortgagor is located/ registered within a month from the date of entering into a contract of sale (legal mortgage) or a mortgage deed (contractual mortgage).113 Registration is necessary for the mortgage to be effective in the insolvency of the owner or any other person.114 The registration process entails the delivery of two copies of the application form and the actual business mortgage agreement. Creditors usually include a detailed list of the items of movable property that are encumbered. The operations and procedures of each local registry office are governed by the regulations issued by that office. As mentioned above, the regulations, processing time, as well as the fees charged for various services vary. Article 178 of the Commercial Code prescribes the following information that must be entered in an application for the registration of a contractual mortgage:115 (i) the names and addresses of the mortgagor and mortgagee; (ii) the date and nature of the mortgage deed; (iii) the claim secured by the mortgage, the conditions on which it may become due and the interest rate; (iv) the objects and address of the business; (v) the scope of the mortgage; and (vi) the address of any branch or agency mortgaged with the principal business. Paragraph 2 of this article adds that ‘the mortgage shall apply only to such parts as are expressly specified in the entry’.116 The information requirements for the registration of a mortgage over a business are much more detailed and cumbersome than the 106 See Commercial Code, Art 124. 107 As owner, usufructuary or lessee; see Commercial Code, Art 125. 108 Various types of partnerships and companies are provided for in Book II Title 1 of the Commercial Code and are listed in Art 212. 109 See Commercial Code, Arts 127–28. 110 ibid Art 129(1). 111 Commercial Code, Art 171(2). The creditors of a bankrupt trader also have a legal mortgage over the business, which is governed by the Bankruptcy provisions in Book V of the Commercial Code. 112 See Business Mortgage Proclamation, Art 4. 113 See Commercial Code, Arts 173(2) (legal mortgage) and 177(2) (contractual mortgages). If a contractual mortgage encumbers branches or agencies of the business located elsewhere, additional registrations would need to be effected; see Art 179(3). 114 ibid Art 193. 115 Commercial Code, Art 175 prescribes the information for a legal mortgage. 116 There is an identical provision in the Commercial Code, Art 191(2).

Current Security Devices 275 corresponding requirements to register a notice under the Model Law.117 Furthermore, Article 178 calls for information that is financial in nature (eg the amount of the claim and the interest rate) and not related to the mortgaged assets. The Business Mortgage Proclamation sets out requirements that further distance the registration process from the notice-filing regime envisaged by the Model Law. Under Article 5, the registrant is required to attach the contract or other documentation to application forms. The registry must then verify the forms against the information from the contract. Article 5(4) does not clarify whether the verification is limited to ensuring that the required information has been stated or extends to additional verification of its accuracy and legality. However, given the powers conferred on the registry by Article 11 to investigate the status of the registered mortgage, including by on-site inspections of the actual business, it may be assumed that the registry also verifies the legality of the information. Given these powers, and the powers and duties to assist with an extrajudicial auction sale, the powers of the registry go way beyond those of not only notice-filing systems, but also the traditional document-registration systems in other economies in sub-Saharan Africa. We do not know of another regime that would vest the registry with the power to conduct physical inspections of an encumbered asset or get involved in carrying out auction sales of encumbered assets. A registry with such vast powers would not be supportive of a modern secured transactions regime along the lines of the Model Law under which the registry performs solely administrative functions.118 A mortgage of a business creates challenges with respect to its effectiveness as against assets that are subject to registration in different registries, such as vehicles and intellectual property rights. According to some interviewed stakeholders, registration of a business mortgage in the Bureau of Commerce and Industry should be sufficient to make the mortgage effective against third parties with respect to all assets comprised thereunder.119 Hence, these financial institutions do not register an interest in collateral in the corresponding registry for those assets, such as vehicles. Conversely, other financial institutions register both the business mortgage and interests in specific assets. In the absence of a thirdparty effectiveness and priority rule, the latter practice is more prudent. A judge may well decide that the priority rule is not first to register, but that specialised registration trumps general registration. For competing mortgages in the same assets of a business, the date of registration determines their priority.120 However, as between the legal mortgage and contractual mortgage, the former has priority irrespective of the time of registration.121

7. Pledges of Warehouse Receipts In 2003, Ethiopia enacted the Proclamation to Provide for a Warehouse Receipts System to supplement and replace some provisions of Chapter 5 of the Civil Code.122 The main

117 MRP,

Art 8. 2, A2bii. 119 October 2015 interviews. See above in this section for discussion of which assets are included. 120 Commercial Code, Art 192(2). 121 ibid Art 192(3). 122 Warehouse Receipts Proclamation No 372/2003. 118 Ch

276 Ethiopia bjective of the Warehouse Receipts Proclamation was ‘to enable any person store his o goods and in particular standardised agricultural products in warehouses and borrow money upon delivery of the warehouse receipt as collateral’.123 The content and concepts of the Warehouse Receipts Proclamation, such as ‘due negotiation’ and ‘security interest’ were clearly influenced by the US UCC’s Article 7 governing documents of title. One of the fundamental changes from the Civil Code was the provision for the issuance of a single warehouse receipt.124 Under the Civil Code, a warehouse had the power to issue a receipt and a voucher, but only the latter could be pledged. A holder of the voucher would have the same rights as a pledgee of the goods to which the voucher related.125 Under the Warehouse Receipts Proclamation, a warehouse receipt can either be negotiable or non-negotiable. A negotiable warehouse receipt can be transferred by way of outright transfer126 or by way of collateral.127 The remedy of extrajudicial sale is available to the collateral taker who, on default by the borrower, may sell the goods covered by a warehouse receipt 15 days after the date specified in a notice provided to the borrower.128 The Proclamation does not provide any exception from the notice requirement for perishable commodities.129 Article 24(6) of the Proclamation provides that the effect of offering a nonnegotiable warehouse receipt as collateral is governed by the rules on an ordinary pledge. In fact, to have a pledge over the goods in this situation, the potential pledgee would have to obtain an acknowledgement/attornment from the warehouse operator, because the delivery of the non-negotiable warehouse receipt will give the potential pledgee no rights over the goods. Article 18 of the Warehouse Receipts Proclamation provides for the warehouseman lien with respect to any unpaid charges, and Article 19 provides for extrajudicial enforcement by public auction or private sale. Ethiopia is one of the few economies in the world that operates a commodity exchange for warehouse receipts. The Ethiopia Commodity Exchange (ECX) began operations in 2008 trading coffee pursuant to a requirement of the Ethiopian government for the entire coffee crop (both for exports and domestic consumption) to be sold through ECX’s platform.130 Between 2015 and 2016, 715,000 metric tons of coffee and sesame were traded at the ECX.131 Originally serving as an all-in-one exchange, performing warehousing, quality control, trading, clearing and other services, in 2016 its warehousing, quality-control operations and central depository functions were moved to the Ethiopian Agricultural Commodities Warehousing Service Enterprise.132 The Ethiopian Commodity Exchange

123 ibid Art 3(1). 124 See ibid Art 8, which provides that there shall be only one warehouse receipt in respect of the same goods. 125 Civil Code, Art 2817. 126 Warehouse Receipts Proclamation, Art 20. 127 ibid Art 24. 128 ibid Art 24(3). 129 Compare with ML, Art 78(8). 130 USDA Foreign Agricultural Service, ‘Ethiopia Coffee Annual Report’ (15 May 2012) 5, http://gain.fas.usda. gov/Recent%20GAIN%20Publications/Coffee%20Annual_Addis%20Ababa_Ethiopia_4-25-2012.pdf. 131 A Haile, A Volk and T Rehermann, ‘Creating Agricultural Markets: How the Ethiopia Commodity Exchange Connects Farmers and Buyers through Partnership and Technology’ (IFC, April 2017) 2, www.ifc.org/wps/wcm/ connect/8e925b5a-94ff-476c-ba03-e5fdfb4b9c85/EMCompass+Note+37+Ethiopia+Exchange+FINAL+April+27. pdf?MOD=AJPERES. 132 ibid.

Proposed Reform 277 Authority oversees the ECX’s operations.133 Proclamation No 550/2007 to Provide for the Establishment of the Ethiopia Commodity Exchange establishes that the exchange may operate warehouses to carry out weighing and inventory management of agricultural commodities and issue exchange warehouse receipts for trading purposes.134 Depositors must be registered with ECX in order to deposit commodities at one of the licensed warehouses. Banks must be also registered as members with ECX. The collateral value of ECX receipts is strengthened by the ECX’s guarantee of quality and quantity of the commodities covered thereunder. Pledges of electronic warehouse receipts (EWRs) are governed by the Rules of the Ethiopian Commodities Exchange.135 After approving a loan, the bank must deliver an Exchange Electronic Pledge Request Form to request ECX to indicate a pledge on the EWR in its system.136 Once the pledge is indicated, the EWR is marked as pledged and can neither be traded nor commodities withdrawn until the pledge is released.137 Enforcement proceedings are governed by Article 9.5.2.4 of the Rules. Upon default, the pledgee must give the pledgor a five-day notice of its intention to dispose of the commodity covered by the pledged EWR. If the pledgor fails to cure the default within the stipulated time, the pledgee may sell the EWR 15 days thereafter.138 The pledgee must give a notice to ECX and then may sell the EWR through the ECX.139

E. Proposed Reform In 2015, Ethiopia decided to undertake a secured transactions reform. Assessments of the credit market and the legal framework led to the formulation of a preliminary draft Secured Transactions Proclamation,140 which is based on the Model Law.141 Like the Civil Code, the draft Proclamation is comprehensive, attempting to facilitate credit in transactions which are already common in the marketplace, as well as to provide a solid foundation for the development of new practices and credit products. The following subsections illustrate and explain some provisions from the draft Proclamation, and the approaches taken into consideration.

1. Scope The draft Proclamation sets out a broad scope narrowed down by a few exemptions along the lines of Article 1 of the Model Law. The specific parameters of its scope have been shaped 133 ibid. 134 Ethiopia Commodity Exchange Proclamation No 550/2007, Art 24. 135 See Art 9.5.2.1. On file with the authors. 136 ibid Art 9.5.2.1(c). 137 ibid Art 9.5.2.1(d). 138 ibid Art 9.5.2.4(a)(ii). 139 ibid Art 9.5.2.4(c). 140 The draft Proclamation is available at A Yohannes, ‘A Draft Proclamation to Provide for Movable Property Security Right’ Ethiopian Legal Brief (29 March 2019) www.chilot.me/2019/03/a-draft-proclamation-toprovide-for-movable-property-security-right/. 141 For an argument in favour of reforming the Ethiopian secured transactions law along the lines of the functional approach of the ML, see Gikay (n 65).

278 Ethiopia by two issues: first, whether the future Proclamation should apply to all types of creditors or limit itself to financial institutions; and, secondly, how to integrate capital goods leases. As for the former, Ethiopian legislation already limits the availability of certain security devices (eg mortgage of a business) to particular types of secured creditors.142 Furthermore, the NBE’s leading role in the reform process may have been another reason this approach was carefully considered, given the mandate of the NBE to regulate and supervise only banks and similar financial institutions.143 The exclusion of suppliers and other non-bank creditors would not make it impossible for such creditors to extend credit secured with a movable asset under other laws that would not be repealed on the entry into force of the future Proclamation. However, those other laws might render the right in the encumbered asset effective against third parties without any form of public notice, require registration in a registry system that is cumbersome or expensive to use, or provide for registration in a registry that has not been established, such as for retentions of title. As a result, the creation and third-party effectiveness of security rights of ‘non-bank’ creditors would be more cumbersome and expensive, particularly due to the continued requirement of notarisation. Another consequence of this exclusion would be that the priority rules of the Proclamation would not be capable of resolving conflicts between a security right of a ‘bank creditor’ and a claim of an ‘excluded creditor’. The exclusion would also increase due diligence of ‘bank creditors’ who would need to verify the existence of competing claims beyond a search of the Registry, effectively having to undertake complicated due diligence. The cost of verifying whether the inventory is subject to an acquisition security right made effective against third parties by registration is lower than the cost of verifying that the inventory has been sold under an agreement that contains a retention of title clause. The former entails an electronic search of the Registry, often provided free of charge, whereas the latter involves an inquiry with the grantor. Furthermore, the grantor might not provide accurate information, but the chances of getting such information from the Registry are much higher. Overall, this approach would perpetuate one of the current challenges, which is the absence of a predictable priority rule for competing rights in the same assets created under different laws. The second issue concerned the proper integration of leases governed by the Leasing Proclamation. The Ethiopian government’s policy to facilitate leasing transactions with the view of creating a vibrant leasing environment dictated the possible approaches considered for the integration of leases under the future Proclamation.144 The Model Law’s unitary approach would result in recharacterisation of financial leases and hire-purchase agreements as acquisition security rights where the lessee would be viewed as the owner of the leased goods and the lessor would have a security right. This approach was rejected because many considered that stripping lessors of ownership would be inconsistent with

142 See D6 above. 143 For a similarly limited approach, see Ghana’s Borrowers and Lenders Act of 2008 discussed in Ghana C1. 144 The Ethiopian government and the NBE have identified the development of the leasing sector as a major priority area. See H Teshome, ‘Leasing Regulatory Framework Analysis and Intervention Options for the PRIME Project’ (September 2014) www.prime-ethiopia.org/wp-content/uploads/2015/07/Leasing%20Regulatory%20 Framework%20Analysis%20and%20Intervention%20Options%20for%20the%20PRIME%20Project%20.docx. In fact, in 2016, the Development Bank of Ethiopia (DBE) adopted a lease financing policy for middle-level enterprises (MLEs) with the objective of providing substantial financial support for MLEs. M Teshome, ‘Applying Lease Financing to Support Industry’ Ethiopian Herald (Addis Ababa, 20 April 2016).

Proposed Reform 279 the policy objective of protecting lessors, or its effect would be misunderstood which would undermine the successful passage of the legislation in the Parliament. Although the Leasing Proclamation applies to three types of capital goods leases,145 the discussions centred on hire-purchase and financial leases, whilst an agreement was reached early to exclude operating leases from the scope of the draft Secured Transactions Proclamation, which is consistent with the Model Law.146 Nonetheless, the latest publicly available version of the draft Proclamation defines an operating lease, but does not otherwise state whether and how the Proclamation is to apply to it, which creates uncertainty.147 A decision was made to include hire-purchase and financial leases in the reformed scheme. For hire purchase, under the current law, the lessee acquires increments of ownership, but is unable to encumber this proportional right. In order to foster the policy of increasing access to credit, Article 4(2) of the draft Proclamation adopts the Legislative Guide’s non-unitary approach148 under which the lessee has sufficient rights in the leased object to create a security right, but the maximum realisable value is limited to the surplus value in the goods over and above the amount due to the lessor.149 For financial leases, the Leasing Proclamation vests full ownership rights in the lessor until the lessee makes all payments and exercises the option to acquire the leased object. The draft Proclamation defines a financial lease,150 but does not provide for any special treatment similarly to hire purchase. As a consequence of the functional definition of security right,151 an interest of a financial lessor shall be treated as a security right which will necessitate consequential amendments of the Leasing Proclamation. As regards the scope, the draft Proclamation excludes security rights in securities traded on exchanges, essentially ‘intermediated securities’.152 In relation to securities, Ethiopia has not established a securities exchange and the types of relationships based on securities accounts maintained by intermediaries do not yet exist. There are two options for the approach of the future Proclamation: it could include provisions facilitating the use of intermediated securities to secure obligations; or this matter could be addressed as part of a broader project to provide for a comprehensive framework on intermediated securities. The current draft Proclamation adopts the latter option. As a consequence, the draft Proclamation follows broadly the approach of the Model Law which excludes security rights in intermediated securities153 and excludes a subset of these, namely security rights in ‘securities traded on exchanges’.154 Regarding warehouse receipts, the drafters of the future Proclamation had to consider the ECX mechanisms for pledges of EWRs. Although those mechanisms operate efficiently, providing a transparent alternative to possession and registration, their integration under the future Proclamation would be advisable to address any

145 See D5 above. 146 Operating leases over a specified length are included in some schemes, eg the Australian PPSA 2009, the New Zealand PPSA 1999. 147 Draft Proclamation, Art 2(31). 148 See ch 2, B2a. 149 LG, recommendation 190. 150 Draft Proclamation, Art 2(17). 151 ibid Art 2(43). 152 ibid Art 3(2)(a). Compare with the ML, Art 2(w) definition of non-intermediated securities. 153 ML, Art 1(3)(c). 154 Draft Proclamation, Art 3(2)(a).

280 Ethiopia potential priority conflicts between creditors who finance crops and make their security rights effective against third parties by registration, and subsequent pledgees of EWRs.

2. Creation and Third-Party Effectiveness The draft Proclamation reinforces the objective to reduce the cost of credit by eliminating any formalisation requirements, such as notarisation, for creating a security right. This general rule is bolstered further by a specific provision exempting all transactions creating security rights in movable assets from stamp duties.155 Consistent with the Model Law, a security right will be created upon execution of a security agreement that identifies the grantor, secured creditor, encumbered assets and secured obligations.156 Furthermore, it incorporates the approach of the Model Law to override the effect of anti-assignment clauses, allowing the use of receivables, the transfer of which may be restricted, to secure obligations.157 Article 13 of the draft Proclamation identifies three methods for achieving third-party effectiveness of a security right: registration, possession and control. The latter two are limited to specific types of assets, whereas registration will be available for a security right in any type of an asset. Unlike under the modern principles,158 possession is limited only to money, negotiable instruments, negotiable documents and certificated securities. The draft Proclamation thus attempts to change the existing practice of lenders to take possession of the collateral even where it is not commercially reasonable, such as where the collateral is inventory. If possession were to have the same effect as registration, especially on the priority of a security right, it might perpetuate the existing cumbersome practice in inventory financing whereby the creditor takes constructive possession through an agent. Commercial realities dictate that possession should not be deployed with respect to certain types of tangible assets, especially inventory and equipment. For inventory, equipment and crops, registration will be the sole method of third-party effectiveness. Control is limited to deposit accounts and electronic securities, which is consistent with the modern principles.159

3. Registry The draft Proclamation provides the legal basis for the establishment of a national collateral registry.160 Although technologically and practically a superior solution, serious consideration was given to the sensitivities inherent in transitioning from regional registries to a nationwide system. Such centralisation of registration functions would have an effect on the 155 ibid Art 89. See also Proclamation to Provide for the Payment of Stamp Duties, which in Art 11 empowers the Ministry to grant exemptions from payment of stamp duties. 156 See Draft Proclamation, Art 4(1) and (5). The draft Proclamation has not incorporated the optional feature of the ML inviting enacting States to require the maximum amount for which the security right may be enforced as a mandatory element of a security agreement. 157 See ML, Art 9. 158 See ch 2, D. 159 ibid. 160 See Draft Proclamation, Art 20.

Proposed Reform 281 revenues of regional registries. However, the impact may be relatively small given the low volume of secured lending currently being registered in the regional registries. The greatest impact is expected to concern registrations of encumbrances over motor vehicles. Nonetheless, some guidance may be drawn from the 2016 Commercial Registration and Licensing Proclamation that empowered the Ministry of Trade ‘to establish and administer a central commercial register and trade name register which has nationwide application; and make open and accessible to the public at large’.161 It is unclear whether such a centralised registry has yet been established. Unlike in most countries in sub-Saharan Africa, Ethiopian citizens do not have a unique identifier according to which registrations could be indexed and searched. The NBE has already faced this obstacle during the implementation of the Credit Reference Bureau. Nevertheless, the draft Proclamation requires registrants to enter in a notice the name and the identifier of the grantor.162 Article 28 defines the identifier of a natural person and a legal person separately as the unique identification number as prescribed in the directive. A draft directive is not available to examine what those unique numbers might be, which could be complicated particularly for natural persons, given the absence of a national identification number. While both the name and the identifier must be entered in a notice, the registry records will be searchable against the identifier only, the accuracy of which also determines the legal effectiveness of the registration.163 The relevant identifiers of different types of grantors must be carefully considered before the directive is finalised. Depending on how quickly the implementation progresses, a new, reliable and unique identifier for individuals may be generated under another project (eg a digital identity project).

4. Priority In the priority chapter of the draft Proclamation, thoughtful consideration was given to the priority of ‘preferential claims’, particularly the claims of employees and tax authorities. The discussions were also shaped by the prospect of gaining points on the Getting Credit Index of the World Bank’s Doing Business Report.164 Such discussions occur in every single reform project. Nonetheless, the principles of certainty and predictability were the main guideposts. In that light, interesting discussions were held to ensure the same priority for preferential claims in both insolvency and non-insolvency situations. The insolvency priority aspect was critical, particularly in light of the application of Ethiopia’s outdated insolvency regime. Article 55 of the draft Proclamation provides for a priority regime applicable to non-consensual creditors defined in Article 2(29) as ‘a creditor that has obtained

161 Commercial Registration and Licensing Proclamation 980/2016, Art 4(1). Furthermore, Art 4(3) states that ‘regional organs administering commercial activities … may undertake commercial registration and trade name registration when delegated by the Ministry’. 162 See Draft Proclamation, Art 27(1)(a). 163 See Draft Proclamation, Arts 37 and 39. 164 Two priority-related questions are included in the Getting Credit Methodology. They ask: (i) whether secured creditors are paid first (eg before tax claims and employee claims) when a debtor defaults outside an insolvency procedure; and (ii) whether secured creditors are paid first (eg before tax claims and employee claims) when a business is liquidated. World Bank, ‘Doing Business Getting Credit Methodology’ www.doingbusiness.org/ Methodology/Getting-Credit.

282 Ethiopia a right in the collateral, on the basis of a court order or applicable law’. This broad definition would seem to encompass not only the traditional ‘judgment creditor’, but also a tax authority that acquires its interest in some assets of the debtor under the relevant law. The intention to apply this provision to wage claims is uncertain. It would seem that the plain language of the definition of non-consensual creditor supports such application, but the process set out in Article 55 for such creditors to obtain priority over a security right would seem to require the registration of individual wage claims, which does not seem practicable. Nonetheless, this approach would ensure compliance with the Getting Credit Index.

5. Enforcement The enforcement chapter provides for extrajudicial remedies that may be exercised in addition to the remedies allowed under any other law, including expeditious relief.165 Unlike the Model Law, Part VII of the draft Proclamation does not contemplate that additional remedies may be provided for in a security agreement. However, bearing in mind the grantor’s ability to direct the enforcement to the judicial system by objecting to extrajudicial procedures and the inefficiency of the judicial mechanisms, careful consideration was given to whether the policies and approaches of the Model Law would deliver outcomes facilitative of credit decisions. Having the latter overarching objective in mind, the Collateral Registry has been vested with powers to facilitate enforcement of security rights to specifically minimise the risk of abusive and delaying tactics. The Collateral Registry would ‘have the power and duties to summon the police force’, in keeping with the powers given to the present registries and the Ministry under the other Proclamations.166 The other example of addressing enforcement inefficiencies is embedded in the definition of ‘recognized market’. The Model Law exempts secured creditors from the requirement to provide a notice of enforcement to interested parties, provided the encumbered asset is of a kind that may be sold on a recognised market.167 In order to minimise the risk of delays with having to identify and properly give notice to all interested parties, the draft Proclamation defines recognised market very broadly.168 The definition is not limited to an officially organised market, such as a commodity exchange, but also extends to any market where prices are stated publicly and/or are presumed to be commercially reasonable.

6. Conflict of Laws Conflict of laws rules are expected to play an important role given the objective of the Ethiopian government to stimulate exports. Current legislation does not provide adequate guidance on the application of the relevant law when the transaction involves a foreign element. The original draft of the Civil Code contained conflict of laws rules, but they were 165 Draft Proclamation, pt VII. 166 ibid Art 81(2). 167 ML, Art 78(8). 168 Draft Proclamation, Art 2(37). One of the authors of this book saw one such market in Addis Ababa for the sale of goats.

Final Remarks 283 removed prior to its adoption.169 This unfortunate development might have influenced the thinking behind omitting those rules from the present draft Proclamation. Similar resistance against the inclusion of conflict of laws rules in secured transactions laws has been encountered elsewhere (eg Kenya).170 The absence of such rules leaves a gap that is expected to increase legal complexity and thus costs of certain secured transactions.

7. Transition Transitional rules will be critical to ensure a smooth transition from the prior regime into a system based on the Model Law. Previously, the functions and effect of transitional rules were not properly understood and resulted in difficulties transitioning into the Civil Code. In fact, the 84 articles of the draft Civil Code dealing with transition ‘were practically hollowed emptied of its content by the Parliament’.171 The draft Proclamation’s transitional chapter closely tracks the Model Law. For the length of the transitional period, the draft tentatively selected 12 months.172 Like the Civil Code, which repealed a number of laws, the Proclamation will result in consequential amendments to and repeals of other laws.173 In addition to overriding the provisions of the Codes governing the various types of pledges and mortgage of a business, the Proclamation will expressly repeal the Enforcement and Business Mortgage Proclamations.174 The enactment of the draft Proclamation would signal a clean break from the past.

F. Final Remarks Ethiopia may become the first civil-law jurisdiction in sub-Saharan Africa to comprehensively implement the Model Law. It is hoped that this implementation will become a model for others to follow. The draft Proclamation would overhaul the entire legal system for security rights, which is outdated, fragmented and relies on an inefficient network of paperbased registries. It would provide the basis for the establishment of a modern and electronic collateral registry. The secured transactions system of Ethiopia laid down in the Civil and Commercial Codes of the 1960s did not deliver the envisaged economic growth. Decades later, policymakers and legislators began eroding the credit-unfriendly nature of the Code system

169 David

(n 38) 196. Kenya C8. 171 David (n 38) 201. 172 Draft Proclamation, Art 82(1). 173 Civil Code, Art 3347 provides: 170 See

(1) Unless otherwise expressly provided, all rules whether written or customary previously in force concerning matters provided for in this Code shall be replaced by this Code and are hereby repealed. (2) In particular the following laws are hereby repealed: (a) The Law on prescription of 29th February, 1948; and (b) The Law on loans of 1923.

174 Draft

Proclamation, Art 93(1)–(2).

284 Ethiopia by introducing exceptions to the general prohibition on non-possessory security rights and extrajudicial remedies. Such piecemeal approach has not had an economic impact on Ethiopian SMEs, as documented, for instance, in the ‘Missing Middle’ study. The draft Proclamation, being based on the Model Law, incorporates some deviations and novel features. These deviations have been carefully thought through to integrate the new secured transactions law properly within the existing commercial legal framework (eg the approach to hire-purchase agreements)175 and to address some uneconomical and impractical tendencies (eg the unavailability of possession to render security rights effective against third parties for certain types of assets, such as inventory and equipment, which is a phenomenon common to a number of African economies).176 With the government’s focus on supporting the leasing industry, recharacterisation of the rights of owners under hire-purchase agreements into security rights in accordance with the Model Law’s unitary approach might attract political opposition. Instead, the non-unitary approach of the Legislative Guide was chosen to recognise the rights of owners under hire-purchase agreements.177 Accordingly, the Legislative Guide played an important role in shaping one of the most debated aspects of the draft Proclamation. However, some uncertainty remains with respect to the application of the future Proclamation to operating leases. Deviations have been also introduced to minimise the risk of delays and disputes (eg vesting the Collateral Registry with certain powers to facilitate the enforcement of security rights).178 Several challenges remain on the road to implementation. One such challenge is the critical function of the Amharic official text to properly channel the concepts and terminology introduced by the draft Proclamation, which were not previously known to the Ethiopian legal framework. This was a challenge for the codifications of the 1950s and 1960s.179 Ensuring that local experts understand the concepts and terminology properly has been a key concern of all those involved in the reform process. Ethiopia would also need to overcome some expected difficulties with transitioning certain registration functions from the regional registries into a central system maintained at the federal level. The policies of the NBE should be moulded to become more facilitative of SME credit, rather than requiring or incentivising the provision of credit to specific products or sectors, such as treasury securities and exporters. Last, but not least, the Ethiopian credit infrastructure needs a modern insolvency framework. As indicated above, the current framework is limited to bankruptcy rules designed to punish business failure.180 More robust protection of security rights in insolvency with the possibility of reorganisation would go a long way towards enhancing access to secured credit.

175 See

E1 above. E2 above. 177 See E1 above. 178 See E5 above. 179 Beckstrom (n 31) 563. 180 See B above. 176 See

13 Tunisia A. Introduction At present, the Tunisian law of secured transactions is inadequate for a modern economy. Not only is it complex and fragmented, but there are a number of specific shortcomings in relation to the taking of security over certain types of assets, and hence the registration system and the enforcement of security rights act as a disincentive to lenders and restrict the availability of finance. This chapter examines the inadequacies in the present system. Tunisia has enacted several statutes over the past decades governing security rights in movable assets. These are mainly of narrow application to specific types of assets, such as pledges of equipment and tools, or transactions, such as fiduciary assignments of receivables.1 The statutory requirements for the creation and third-party effectiveness of these security rights are onerous and out of step with 21st-century best practice. The framework does recognise a number of non-possessory security rights, known as pledges, but inventory cannot be encumbered in this way, despite it making no economic sense for a lender to take any form of possessory security over inventory. Structuring non-possessory security rights is fraught with many challenges under Tunisian law, including a requirement to describe the encumbered assets with specificity in both security agreements and registrations, and the impossibility of automatically extending a security right to future assets or proceeds of the collateral. Another obstacle is the requirement that secured obligations be identified in both security agreements and registrations by the specific amount actually lent to the borrower and not by a generic description or maximum sum for which the security right may be enforced.2 Tunisia has established a number of registries in which transactional documents creating security rights may be registered, including the Commerce Registry for pledges over fonds de commerce (meaning an ‘enterprise’ or ‘going concern’) and registries maintained by courts for the registration of pledges over tools and machinery. Many security devices, such as retention of title devices and assignments of receivables, are effective against third parties without any form of registration. Finally, the legal framework does not allow parties to agree on out-of-court

1 See, eg, European Bank for Reconstruction and Development (EBRD), ‘Commercial Laws of Tunisia’ (March 2013) 2, 48–49; loi no 2001-19 du 6 février 2001, relative au nantissement de l’outillage et du matériel d’équipement professionnel [Law No 2001-19 of 6 February 2001, Related to Pledges of Tools, Machinery and Professional Equipment]; loi no 2000-92 du 31 octobre 2000, relative aux actes de cession ou de nantissement de créances professionnelles et à la mobilisation des crédits rattachés [Law No 2000-92 of 31 October 2000, Related to Fiduciary Assignments of Accounts Receivable]. 2 See EBRD (n 1) 48.

286 Tunisia enforcement.3 The practical effect of the current framework for secured transactions is that Tunisian creditors rely primarily on immovable assets as collateral.

B. Economic Overview The Tunisian economy is reliant on services, which contribute 63.8 per cent of its GDP.4 GDP grew by 2.6 per cent in 2018, up from 1.9% in 2017; the growth was spurred on by agriculture, which grew by 8.7 per cent, and market services, which grew by 3.2 percent, year over year.5 However, Tunisia still suffers from a sluggish economy, thanks in large part to slow recovery in the tourism sector following the terrorist attacks of 2015 as well as low levels of investment, among other things.6 Lack of access to finance is the fourth largest obstacle for SMEs operating in Tunisia.7 In 2015, 97.8 per cent of all Tunisian firms were MSMEs,8 and 90 per cent of MSMEs were sole proprietors.9 Further, MSMEs accounted for 40 per cent of GDP and 56 per cent of all employment.10 A large number of these businesses have been in operation for years and can be considered mature.11 However, nearly 50 per cent continue to rely on internal finances to cover operating costs and investments. As of 2018, growth was projected to increase thanks to measures designed to simplify the business environment.12 Tunisia’s Ease of Doing Business (EODB) score fell by twotenths of a point from 2017 to 2018 (63.38 to 63.58) according to the World Bank’s Doing Business 2018 Report, but the EODB score increased from 2018 to 2019 (to 66.11).13 The banking sector dominates the Tunisian financial system; banks account for approximately 90 per cent of financing in the country.14 Such overreliance on bank financing is an obstacle to economic growth and job creation.15 Three Tunisian banks, representing a third

3 A Alvarez de la Campa, ‘Increasing Access to Credit Through Reforming Secured Transactions in the MENA Region’, World Bank Policy Research Working Paper 5613 (March 2011). 4 CIA, ‘Africa: Tunisia’ (The World Factbook, 4 February 2019) www.cia.gov/library/publications/the-worldfactbook/geos/ts.html. 5 African Development Bank (AFDB), ‘African Economic Outlook’ (2019) 1, 182. 6 US Department of State Bureau of Economic and Business Affairs (USDS), ‘Tunisia’, 2018 Investment Climate Statements (19 July 2018) www.state.gov/e/eb/rls/othr/ics/investmentclimatestatements/index.htm?dlid=281685& year=2018#wrapper. 7 S Stölting, ‘Neighbourhood SME Financing: Tunisia’ (European Investment Bank, October 2015) 15. 8 USDS (n 6). 9 Stölting (n 7) 14. 10 ibid. 11 World Bank, ‘Enterprise Surveys: Tunisia’ (2013) 4; A Sy, F Bennani, M Cherif et al, ‘International Bank for Reconstruction and Development Program Document for a Proposed Loan in the Amount of Euro 413.4 Million (Equivalent to US$500 Million) to the Republic of Tunisia for the Investment, Competitiveness and Inclusion Development Policy Financing’, World Bank, Report No 124855-TN (25 May 2018) 37. 12 ‘Developments in Individual OECD and Selected Non-Member Economies: Tunisia’ (OECD Economic Outlook 2018) 3. 13 World Bank Group, ‘Doing Business 2018: Reforming to Create Jobs’, 15th edn (1 June 2017) 4; World Bank Group, ‘Doing Business 2019: Training for Reform’, 16th edn (1 May 2018) 5. 14 Bureau of Economic and Business Affairs, ‘Investment Climate Statements for 2018: Tunisia’ (US Department of State, 19 July 2018) www.state.gov/e/eb/rls/othr/ics/investmentclimatestatements/index.htm?year=2018& dlid=281685. 15 ibid.

Economic Overview 287 of the entire banking industry’s assets, are state owned;16 the credit market is dominated by banks that lend primarily to the industrial, trade and tourism sectors;17 and banks focus on offering services tailored to large corporations.18 SMEs are left with limited access to banks because of high collateral requirements.19 In 2012, the IMF found many deficiencies in the practices of Tunisian banks, such as the constant overvaluation of collateral for provisioning purposes and other failures to comply with international loan classification standards, as well as the overconcentration of risks from loans given to connected borrowers.20 The high cost of bank finance was exacerbated by excessive requirements for collateral, which averaged 250 per cent of loan value.21 Further, the World Bank concluded that the current regulatory environment creates a burdensome system of rents and privileges which hampers the economy, restrains competition, obstructs entrepreneurship and is thus an obstacle to economic development.22 These are some of the indicators of the limited expertise of Tunisian lenders in asset-based lending. These challenges are symptomatic of the region. In its 2011 Enterprise Survey, the World Bank found that, in comparison with other regions, the Middle East and North Africa (MENA) had the lowest percentage of firms with access to credit.23 According to the Doing Business 2019 Report, however, MENA region countries undertook a record number of reforms in 2018 to improve the ease of doing business for their SMEs.24 Even so, the Report found that barriers against women starting a business remain firmly in place.25 Further, getting credit in the region is more difficult compared to the rest of the world due to insufficient insolvency laws and the inability of courts to provide effective and expeditious relief on enforcement.26 In its 2015 report, the European Investment Bank (EIB) identified inefficient secured transactions and insolvency frameworks as one of the contributing causes to the fragmentation of the uncompetitive banking sector.27 Improving the framework for access to finance for SMEs is one of the conditions of the IMF’s US$2.8 billion loan to Tunisia, agreed to in the spring of 2016.28 In the aftermath of making this loan, the Tunisian Ministry of Justice initiated the process of reviewing its legal framework with the objective of adopting a modern secured transactions law.29 The Borrower’s Council of Ministers, in May 2018, approved and submitted to Parliament a draft Law on Secured Transactions that is aligned with the Model Law.30 The Law was drafted by 16 IMF, ‘Financial Stability Assessment: Tunisia’ (2012) 13. 17 ibid. 18 Bureau of Economic and Business Affairs (n 15). 19 ibid. 20 IMF (n 18) 16. 21 ibid 11. 22 Sy, Bennani, Cherif et al (n 11) 60. 23 de la Campa (n 3) 3. 24 World Bank, ‘Doing Business Report: Economies of Middle East & North Africa Carry Out Record Reforms to Improve Business Climate’, press release (Washington, DC, 31 October 2018). 25 ibid. 26 ibid; IMF, ‘Tunisia’ (Country Report No 16/47, February 2016) 10. 27 Stölting (n 7) 10. 28 S Donnan, ‘IMF Agrees $2.8bn Bailout for Tunisia’ Financial Times (Washington, DC, 15 April 2016). 29 See ‘‫[ ’ةيلودلا' ةيمنتلل ةيكيرمألا ةلاكولا نع نيلثمم لبقتسي لدعلا ريزو‬The Minister of Justice Receives Representatives of the United States Agency for International Development (USAID)] e-justice (1 December 2016) www.e-justice.tn/index.php?id=2464. 30 Sy, Bennani, Cherif et al (n 11) 24.

288 Tunisia the Ministry of Justice with assistance from the World Bank and USAID, and the World Bank continues to provide assistance to authorities as they draft regulations and proceed with the next steps in the reform process.31

C. Secured Transactions Framework – An Overview and Common Transactions The most important sources of law governing secured transactions in Tunisia are the Code of Obligations and Contracts (COC),32 the Code of Real Rights (CRR),33 the Code of Commerce,34 the Commerce Registry Law35 and the Code of Civil and Commercial Procedures (CCCP).36 These codes are the result of ‘thematic codification’37 rather than an all-encompassing approach to codification, such as inspired by the comprehensive civil and commercial codes of continental Europe. Other statutes have been enacted to regulate security rights primarily related to specific industries or assets, including the Law Related to Pledges of Tools, Machinery and Professional Equipment,38 the Law Related to Finance Leasing,39 the Law Related to Fiduciary Assignments of Accounts Receivable40 and the

31 ibid 25. 32 Code des obligations et des contrats [Code of Obligations and Contracts (COC)] (as amended) 2005. The COC was established in 1906 and amended in 2005 by loi no 2005-87 du 15 août 2005 relative à approuver la réorganisation du le COC [Law No 2005-87 of 15 August 2005 relating to approving the reorganization of the COC]. French and Arabic versions of the COC are available at: Republic of Tunisia Presidency of the Government, ‘Obligations and Contracts Code’ (Legislation: National Portal of Legal Information, 2019) www.legislation.tn/en/ affich-code/Code-des-obligations-et-des-contrats__88. 33 Code des droits réels [Code of Real Rights (CRR)] 2011. The CRR was promulgated by loi no 65-5 du 12 février 1965 (11 chaoual 1384), portant promulgation du code des droits réels [Law No 65-5 of 12 February 1965 (11 Shawwal 1384) promulgating the Code of Real Rights]. A French version is available at: Republic of Tunisia Presidency of the Government, ‘Real Property Code’ (Legislation: National Portal of Legal Information, 2019) www.legislation.tn/en/affich-code/Code-des-droits-r%C3%A9els__81. Code des Droits Réels is translated as ‘Code of Real Rights’ or ‘Real Property Code’. However, it has to be noted that these translations are not accurate because the CRR deals with both real and personal property. 34 Code de commerce [Code of Commerce (CC)] (as amended) 2010. The CC was promulgated by loi no 59-129 du 5 octobre 1959 [Law No 59-129 of 5 October 1959]. French and Arabic versions of the CC are available at: Republic of Tunisia Presidency of the Government, ‘Code de commerce’ (Legislation: National Portal of Legal Information, 2019) www.legislation.tn/fr/affich-code/Code-de-commerce__74. 35 Registre de commerce [Commerce Registry Law] 1995. The Commerce Registry Law was promulgated by loi no 95-44 du 2 Mai 1995, relative au registre du commerce [Law No 95-44 of 5 May 1995, relating to the commercial register]. The French version, with an automatic translation tool, is available at: WIPO, ‘Tunisia Act No 95-44 of 2 May 1995 Concerning the Commercial Register’ (1995) wipolex.wipo.int/en/legislation/details/8176. 36 Code des procédures civiles et commerciales [Code of Civil and Commercial Procedures (CCCP)] (as amended) 2018. The CCCP was promulgated by loi no 59-130 du 5 octobre 1959 (2 rabia II 1379), portant promulgation du code de procédure civile et commerciale [Law No 59-130 of 5 October 1959 (2 springtime II 1379), promulgating the Code of Civil and Commercial Procedure]. The CCCP is available in French and Arabic at: Republic of Tunisia Presidency of the Government, ‘Code of Civil and Commercial Procedures’ (Legislation: National Portal of Legal Information, 2019) www.legislation.tn/en/affich-code/Code-de-proc%C3%A9dure-civile-et-commerciale__90. 37 DE Stigall, ‘Law and the Lodestar: Tunisian Civil Law and the Task of Ordering Plurality in the Aftermath of the Jasmine Revolution’ (2014) 7(1) Journal of Civil Law Studies 2, 16–23. 38 Law Related to Pledges of Tools, Machinery and Professional Equipment 2001 (n 1). 39 Loi no 94-89 du 26 juilet 1994, relative au leasing [Law No 94-89 of 26 July 1994, Related to Finance Leasing]. 40 Law Related to Fiduciary Assignments of Accounts Receivable 2000 (n 1).

Secured Transactions Framework – An Overview and Common Transactions 289 Law Related to the Dematerialisation of Securities.41 Assets such as large ocean vessels and aircraft which are commonly governed by special laws may be used as collateral under asset-specific legislation, including Articles 100–106 of the Code of Maritime Commerce.42 Although based on continental civil law, Tunisian law also incorporates elements of Islamic law, although the new Constitution does not contain any specific references to Islamic law.43 Some elements of Islamic law have been preserved, but primarily in legislation that does not affect secured transactions, such as the Personal Status Code.44

1. Multiplicity and Overlap of Laws The absence of an integrated legal framework forces potential secured creditors to examine, and potentially apply, multiple statutes when seeking to take security for obligations. For instance, with regard to security rights to be taken over shares, the following statutes may be applicable: (i) the Commercial Companies Code (CCC)45 that covers the general definition of securities and the authority of commercial companies to issue securities; (ii) the Law Related to Dematerialisation of Securities; (iii) the CRR, as to the creation and third-party effectiveness of a pledge; and (iv) the CCCP, in relation to the requirements for enforcing a pledge.46 Structuring a transaction secured with equipment can be even more complicated because security rights may be created under five different devices, namely: (i) a special pledge of tools, machinery and professional equipment; (ii) a pledge of fonds de commerce (enterprise); (iii) a possessory pledge; (iv) a finance lease; and (v) a reservation of ownership device. In some circumstances, equipment may be affixed to an immovable so that rights in the immovable extend to it. Articles 27–37 of the CRR regulate the legal effect of such affixation as well as the creation of security rights in movable assets already affixed to an immovable. Although the Law Related to [non-possessory] Pledges of Tools, Machinery and Professional Equipment47 was specifically adopted to facilitate the financing of equipment, it is essentially limited to acquisition financing since it does not allow a business to pledge equipment and machinery it already owns. Thus, this pledge can only be used by those providing acquisition financing to cover the purchase price.48 The wide spectrum of security devices available for encumbering equipment contrasts with the dearth of non-possessory security devices for inventory. At present, inventory may be encumbered only under a possessory pledge or a retention of title device. The lack of a non-possessory pledge over inventory that could be taken by financial institutions clearly has a stifling effect on asset-based lending.

41 Loi no 2000-35 du 21 mars 2000, relative à la dématérialisation des titres [Law No 2000-35 of 21 March 2000, Related to the Dematerialisation of Securities]. 42 Code de commerce maritime [Code of Maritime Commerce] 2010, Arts 101–06. Art 101 gives a definition of the phrase ‘maritime claims’ by including a list of 16 types of claims. 43 Stigall (n 41) 2. 44 ibid 8. 45 Code des sociétés commerciales [Commercial Companies Code (CCC)] 2016, Arts 314–86. 46 CCCP (n 40) Arts 404–09. 47 Law Related to Pledges of Tools, Machinery and Professional Equipment 2001 (n 1). 48 EBRD (n 1) 48.

290 Tunisia

2. Registration Systems The lack of a nationwide electronic registry for notices of all consensual encumbrances over movable assets is a major factor discouraging financial institutions from extending loans secured with movable assets. The main registration system in Tunisia is the Commerce Registry which registers enterprise pledges. It is not a notice-based system. The Commerce Registry has physical branches across the country, and a person or entity must be registered as a trader or business in the branch where its business is located. Branches of the Commerce Registry have been established in every locality that is under the jurisdiction of the first-instance court. An enterprise pledge (fonds de commerce) may be created only by a trader or business, and such a pledge must be registered in the local branch where the trader or business is registered. Local registrations are stored in a national database that is electronically searchable.49 However, the only evidence of a record found by a search which can be relied upon in court is a document certified by the Commerce Registry. In order to obtain this, an interested party must determine in which branch the physical record is stored and request an official certificate of the registry record from that branch.

a. Registrations The courts of first instance maintain a registry for pledges and finance leases over tools and machinery. Unlike the Commerce Registry, these systems are paper based and are not connected electronically so as to enable nationwide searches. Pledges over securities require registration in the records of the issuing company, and pledges and finance leases of vehicles must be registered with the motor vehicle authority. None of these systems are notice based.

3. Popular Security Devices A security right over fonds de commerce, or an ‘enterprise pledge’, is the preferred method for Tunisian creditors to secure obligations. However, a World Bank study found that fonds de commerce is often only secondary collateral securing a loan, and that the creditor usually relies primarily on a security over immovables.50 The enterprise pledge to encumber fonds de commerce is governed by Articles 189–268 of the CC. It covers some elements of an enterprise, such as goodwill, logo, intellectual property rights, lease of premises, machinery and equipment. However, it does not extend to inventory or other intangible assets, such as receivables.51 A fonds de commerce is a movable asset by nature and so must include movable assets and may not be composed of immovable assets only. This concept of fonds de

49 Access the Central Trade Register at: ‘Accéder a l’information légale sur les entreprises’ [‘Access to Legal Information on Companies’] (Registre National des Entreprises, 2009–19) www.registre-commerce.tn/search/ ExtraitRegistre.do?action=getPage; see also République Tunisienne Ministère de l’Industrie et des PME, ‘A propos de l’INNORPI’ [Republic of Tunisia Ministry of Industry and SMEs, ‘About INNORPI’] (Institut National de la Normalisation et de la Propriété (INNORPI)) www.innorpi.tn/fr/propos-de-linnorpi. 50 de la Campa (n 3) 7. 51 EBRD (n 1) 48. Although ‘enterprise’ is defined in Art 189 to include goods and merchandise, ‘enterprise pledge’ does not extend to goods and merchandise under Art 237.

Secured Transactions Framework – An Overview and Common Transactions 291 commerce differs from that found in some European countries that recognise an expanded concept of fonds de commerce, one that encompasses a certain percentage of inventory.52 The nature of the enterprise pledge and the elements that it should or should not include was once the subject of considerable scholarly debate in Tunisia – in particular, the question of whether an enterprise constitutes movable or immovable property; this was because it includes assets that fall under both categories. The Supreme Court in Tunisia settled the debate by ruling that an enterprise is a movable asset.53 The financing of receivables by outright transfer became popular in Tunisia in the late 1990s, especially after the establishment of Tunisia Leasing (which merged with Tunisia Factoring in 2018 to form Tunisie Leasing & Factoring (TLF)), which remains the market leader today.54 Many SMEs prefer to transfer their receivables to bespoke receivables financing companies, because they not only provide quicker financing than banks but also do not take a security right in the other assets of the transferor. In practice, it is customary for such companies to request a personal guarantee from the transferor. For transfers of receivables without any recourse, which accounts for about 20 per cent of its portfolio, TLF requires the financed business to take out credit insurance against the possibility that the receivables turn out to be uncollectible. In 2016, the overall default rate on receivables acquired by Tunisia Factoring was around 4 per cent.55 To minimise the risk that the receivables might previously have been transferred, TLF and other receivables financiers take possession of purchase orders, delivery confirmations and invoices. Since transfers of receivables do not require any form of public notice, receivables financing companies customarily consult one another. Given their small number in the Tunisian market, this is usually seamless.56 Taking possession of invoices and other documentation evidencing the debt, together with regular consultations of other receivables financing companies, minimises the risk of ‘secret liens’ that may have been created over the same receivable. Although notification to the debtor of the receivable is not required to render the transfer effective as between the transferor and transferee, it is a common practice for receivables financing companies to request the transferor of receivables to notify the debtor of the transfer. Tunisia’s leasing industry originated in the 1980s with the establishment of the first leasing company – Tunisie Leasing, which remains the market leader (as TLF).57 A World Bank study found that Tunisia and Morocco have the most developed leasing markets in the region.58 Lease financing approval is generally faster than approval of a bank loan.59 The Tunisian market is highly competitive, and most of the leasing companies are associated

52 eg, Belgium. Prior to the 2013 reform, Belgium allowed 50 per cent of inventory to be encumbered under a pledge of fonds de commerce. A Morrell and F Helsen, ‘The Interrelation of Transparency and Availability of Collateral: German and Belgian Laws of Non-Possessory Security Interests’ (9 March 2016) 3 European Review of Private Law 393, 400. 53 Cass civ (20215) 28 June 1989. Contrast, as well, the common law floating charge, see ch 1, E1b. 54 ‘A Propos’ [‘About’] (Tunisie Leasing & Factoring, 2018) www.tlf.com.tn/site/fr/tunisie-leasing.36.html. 55 Interview with Mr Raouf El Ferktaji, Adjunct General Director, Tunisie Factoring (Tunis, Tunisia, 29 September 2016). 56 ibid. 57 I Chérif, ‘Study on the Leasing Sector in Tunisia’ (MAC SA intermédiaire en bourse [broker], 2009) 4. 58 de la Campa (n 3) 7. 59 Interview with Mr Mohamed Fadhel Hamdi, Director, Attijeri Leasing Litigation Department (Tunis, Tunisia, 8 September 2016).

292 Tunisia with banks.60 A typical leasing contract provides the lessee with an option to buy the asset for the nominal price of one Tunisian dinar.61 The leasing companies do not offer operating leases. At times, leasing companies request a sizeable prepayment to cover the risk of depreciation, a payment that may represent as much as 50 per cent of the future lease payments. Loans provided to microbusinesses secured with rights in their business assets, such as an oven or a refrigerator, are also popular in Tunisia.62 In the absence of an efficient security right mechanism, lenders purchase these assets and then sell them back to their customers, reserving ownership until full payment. As an assurance of repayment of the loan, these lenders typically request a guarantee from a person with a salaried job. These transactions carry low risk because of the existence of an efficient secondary market for such assets.

D. Creation and Third-Party Effectiveness This section examines the requirements for the creation of security rights, primarily pledges. Creation and third-party effectiveness is covered in a single section rather than two sections because the Tunisian framework does not recognise the creation of a security right of pledge merely by an agreement between the grantor and the secured creditor, as is common in modern secured transactions legislation, including under Article 6(1) of the Model Law. The different types of pledges require, for their creation, two steps in addition to executing a security agreement. The first additional step involves some form of publicity. Thus, for example, a pledge over shares must be registered in the records of the issuing company and a pledge of receivables requires notification of the receivables debtor. The second additional step comprises one or more formality requirements, such as legalisation of the parties’ signatures on the agreement or alternatively the recording of the agreement with the tax authority (Recette des Finances) as prescribed in Article 450(2) of the COC. Legalisation of signatures is different from notarisation and requires the parties to sign a deed in front of an officer of an administrative municipality. In addition to signing the deed, the parties must add their signatures to the dated entry made in the register maintained by the officer. The act of legalisation of the signatures or recording of the pledge agreement with the tax authority renders those pledges for which no public registry has been established, such as a pledge of receivables, effective against third parties despite the fact that the publicity step has not yet been effected. In contrast, for those pledges that require registration, including the enterprise pledge, third-party effectiveness is achieved only on registration of the pledge in the applicable registry, which in the case of an enterprise pledge would be the Commerce Registry.63 Article 2 of the CRR differentiates between movable and immovable property, each being governed by a discrete set of rules. The CRR rules for immovable property are more developed and detailed because of the economic importance of land in Tunisia and probably also because land was the primary type of collateral at the time the CRR was adopted.

60 Stölting

(n 7) 12. This is similar to an English hire-purchase agreement, see ch 2, B2b. 62 Interview with Tayssir (Tunis, Tunisia, 20 September 2016). 63 See D3 below for discussion of how a pledge loses all effectiveness if it is not registered within a specified time. 61 ibid.

Creation and Third-Party Effectiveness 293 Movable property is defined in Article 13 as assets that are movable, ‘either by nature or by determination of law’,64 and the position is further clarified by Article 15 of the CRR which specifies that ‘movable property’ includes ‘all receivables, real rights and actions related to movable property’.65 Unharvested crops connected to the land by roots and fruits are immovable assets,66 and they become movable only after harvest. Livestock, equipment and items physically connected to the immovable that are necessary for the use of the land are considered immovable by destination, even though the manner of attachment to the land is not permanent.67 Accordingly, a security right in such assets cannot be created under the pledges that have been designed to apply to movable assets only, which are discussed in the following subsections. The removal of livestock, equipment and other items connected to the immovable does not extinguish any encumbrances created while these assets were connected, and the encumbrance follows the asset.

1. Possessory Pledge The CRR regulates the possessory pledge, which requires the pledgee to take actual or constructive possession of the encumbered asset. The pledge is created at the moment of delivery, and this is also the moment that the pledge becomes effective against third parties. Thus, for example, a pledge of merchandise stored in a warehouse is created by delivery of the warehouse receipt to the pledgee,68 and the pledge becomes effective against third parties from the moment of delivery.

2. Pledges over Intangible Assets The CRR also recognises pledges over intangibles, including receivables, which are, by their nature, non-possessory. Under Article 218, the creation of a pledge over receivables (or a bank account) requires the receivables debtor (or the bank) to be notified and to consent to the pledge. There are no registration requirements either for the creation or third-party effectiveness of a pledge over receivables or a pledge over a bank account. Shares (parts sociales) and other securities issued by private companies in a certificated form may also be pledged, but in such cases the issuing company must record the pledge in its own register for the pledge to be created.69 Salaries of individuals can also be pledged, although the

64 Specifically, CRR (n 37) Art 13 provides: ‘Les biens sont meubles soit par leur nature soit par détermination de la loi’ [‘The property is movable either by nature or by determination of law’]. The reference to ‘by determination of law’ authorises the lawmaker to designate, through legislation, an asset as movable. CRR (n 37) Art 14 expands on this definition, stating that ‘Sont meubles par nature, les corps qui peuvent se transporter d’un lieu à un autre, soit par eux-mêmes, soit par l’effet d’une force étrangère’ [‘Movable assets by nature are assets that can be moved from one place to another, either by themselves or by the effect of an external force’]. 65 Because Tunisian legislators are highly influenced by the French civil law system, there is a tendency to define every single concept in a code, thus limiting the court’s power to interpret the law. 66 CRR (n 37) Art 7. 67 ibid Art 9. 68 ibid Art 217. 69 ibid Art 221.

294 Tunisia c reation of such pledges is subject to a rather cumbersome process of court supervision.70 The grantor must request the court to issue a special declaration authorising the secured creditor to garnish her salary in the event of non-payment of the secured obligation. All such declarations are entered in a register maintained by the clerk of the court.71 Every court has its own register, and they are not centralised in a single electronic system. In practice, some banks obtain a power of attorney from grantors in order to request the declarations on their behalf. After the issuance of a declaration, the court clerk must notify the grantor’s employer within 48 hours and, upon recording the declaration in a register, the clerk must provide a confirmation of the declaration to the grantor. Recording the declaration in a register and notification of the employer are both required to make the pledge effective against third parties.

3. Enterprise and Equipment Pledges The pledge of a fonds de commerce must be evidenced by a notarised deed or a privately signed document.72 The creation of the pledge requires the pledge agreement to be registered in a registry in the place ‘where the fonds exist’, which is the Commerce Registry of the locality where the company is registered.73 To effect the registration, the registrant must submit two certified and true copies of the agreement to the Commerce Registry. Upon completion of the registration, the Registrar provides to the registrant a signed receipt evidencing the registration.74 The pledge is created and becomes effective against third parties on the date of registration as confirmed by the receipt. Multiple registrations in more than one branch of the Commerce registration may be required if the assets that form the enterprise are situated in different locations.75 Registrations must be completed within one month after execution of the pledge agreement.76 The Supreme Court of Tunisia, in its decision no 11446 of 1985, confirmed that if a secured creditor does not complete the registration in a timely manner, the pledge will have no effect even against the pledgor.77 The Law Related to Pledges of Tools, Machinery and Professional Equipment of 2001 stipulates that a pledge can be established without delivering possession of the asset to the secured creditor.78 The pledge must be evidenced by a contract that contains the mandatory information prescribed in Article 214 of the CRR,79 which includes the amount of the debt, its due date, and a description of the pledged assets by their nature, weight or other features in a manner that exactly identifies them. In addition, the pledge contract must be

70 CCCP (n 40) art 360. 71 ibid Art 361, para 1 – but it is not an electronic register. 72 CC (n 38) Art 238, para 1. 73 CC (n 38) Art 238, para 2. 74 Loi no 2010-15 du 14 avril 2010 modifiant et complétant certaines dispositions de la loi no 95-44 du 2 mai 1995 relative au registre du commerce [Law No 2010-15 of 14 April 2010 Amending and Supplementing Certain Provisions of Law No 95-44 of 2 May 1995 relating to the Commercial Register] Art 44 (nouveau). 75 CC (n 38) Art 238, para 3. 76 ibid Art 239, para 1. 77 Cass civ (11446) 3 June 1985. 78 Law Related to Pledges of Tools, Machinery and Professional Equipment 2001 (n 1) Art 1. 79 ibid Art 2.

Priorities 295 submitted to a specific registry maintained by the first-instance court that has jurisdiction over the area in whose Commerce Registry the debtor is registered. The Registrar will then register only certain information, including the names of the parties and a description of encumbered assets from the contract, and retain the contract in its files, which may be made available to searchers upon request. Registration must be concluded within one month after the pledge contract is negotiated; otherwise, the pledge will be void.80 If the debtor is not registered in any Commerce Registry, the registration of the pledge contract must be made with the court where the pledged asset is located or in the domicile of the pledgor.81

4. Finance Lease Under Article 1(1) of the Law Related to Finance Leasing, a lessee acquires possession and the use of an asset, but, as a general rule, the ownership passes only upon payment of the final instalment and an additional nominal sum.82 The Law allows the parties to agree otherwise so that the lessee may acquire equity in the asset as it makes payments.83 The lease must be evidenced by a written deed.84 Under Article 7, if the lease relates to equipment it must be registered in the Commerce Registry situated in the location of the lessee. As noted above, if the lessee is a company registered in a specific branch of the Commerce Registry, the lease must be registered in the same branch. If the lessee is not registered as a company, the registration of the lease must be effected at the court of the area where the leased asset is located. The lease agreement itself is registered, but the Registrar does not verify the accuracy or legality of the information in the lease agreement.85 Finance leases of vehicles are registered with the motor vehicles registry maintained by the Technical Agency of Land Transport (Agence Technique du Transport Terrestre).86 In all these cases, registration is effective against third parties. In the absence of a registration, the lessor’s ownership rights remain fully effective against the lessee, but not against third parties, including creditors and buyers who acquired their rights without knowledge of the lessor’s rights. Accordingly, failure to register a lease has a different effect on third parties to that of failure to register a pledge.

E. Priorities In light of the multiplicity of laws under which security rights encumbering certain types of movable assets such as equipment may be created, establishing priorities is challenging in Tunisia. The risks are further exacerbated by the absence of specific rules that would 80 ibid Art 4. 81 ibid Art 6. 82 Law Related to Finance Leasing 1994 (n 44) Art 1, para 1. 83 ibid Art 1, para 3. 84 ibid Art 1, para 2. 85 Interview with Mr Mohamed Fadhel Hamdi, Director, Attijeri Leasing Litigation Department (Tunis, Tunisia, 8 September 2016). 86 République Tunisienne Ministère du Transport, ‘Agence Technique du Transport Terrestre’ [Republic of Tunisia Ministry of Transport, ‘Technical Agency of Land Transport’] (2010) www.attt.com.tn/index.php?code_menu=7.

296 Tunisia determine whether the third-party effectiveness and priority of a pledge is affected by the pledgor’s relocation (eg from Tunis) and registration of a new company in a different branch (eg in Sfax) of the Commerce Registry. Upon relocation, the pledgor has a duty to notify the Commerce Registry branch where it has been registered so that the branch can forward its file to the branch of the pledgor’s new location. This process is cumbersome because the files, including any pledge agreements, are kept in a physical format. Finally, although the individual statutes provide rules allocating priorities between successive pledges created under that statute, there is no explicit rule determining priority between two pledges over the same asset created under two different statutes. Although one can assume that a court would apply the ‘first to become effective against third parties rule’, the absence of a statutory rule still creates a legal risk.

1. A Legitimate Cause of Preference The CRR provides the general priority rule that creditors share pro rata unless they establish a ‘legitimate cause of preference’.87 According to Article 193, the legitimate causes of preference are privileges, pledges and rights of retention. An example of a privilege is a worker’s claim for unpaid wages,88 and an example of a right of retention would be the right of a warehouse owner to retain possession of goods until its storage fees have been paid. Pledges have been discussed in detail above. The legitimate causes of preference may be created contractually or conferred by law. Article 194 makes it clear that legitimate causes of preference have priority over claims of unsecured creditors. Article 195 determines the respective priority among holders of the three types of preferences, with holders of privileges being ranked first (above pledges), irrespective of the time when the claim to the privilege arose. As a consequence, wages owed to workers will have priority over the claims of pledgees, whether or not the claims arose before or after the establishment of the pledge,89 although CRR Article 199 and CC Article 570, with respect to bankruptcy, stipulate that workers’ wages have priority only in respect of the last six months before bankruptcy or judicial settlement of the debts of the debtor.90 The manner

87 CRR (n 37) Art 192. ‘Les biens du débiteur sont le gage commun de ses créanciers et le prix s’en distribue entre eux par contribution, à moins qu’il n’y ait entre les créanciers des causes légitimes de préférence’ [The property of the debtor is subject to the rights of his creditors, and its value is to be distributed among them equally, unless a creditor holds a legitimate cause of preference]. 88 CRR (n 37) Art 194: ‘Le privilège est un droit réel de préférence que la loi accorde sur les biens du débiteur à raison de la cause de la créance’ [‘The privilege is a real right of preference that the law grants on the debtor’s property by reason of the cause of the claim’]. An exhaustive list of privileges is enumerated in CRR (n 37) Arts 199 and 200, and includes funeral expenses, taxes and workers’ wages. Privileges may be general or special. Eight privileges are recognised by the legislation. 89 Code du travail [Labour Code] (1996) Art 134 provides : ‘La rémunération des travailleurs de toutes catégories est déterminée, soit par accord direct entre les parties, soit par voie de convention collective, dans le respect du salaire minimum garanti fixé par décret. La rémunération des travailleurs relevant de secteurs non régis par des conventions collectives peut être-fixée par décret’ [‘The remuneration of workers of all categories is determined either by direct agreement between the parties or by a collective agreement in respect of the minimum wage fixed by decree’]. 90 CC (n 38) Art 199, para 2. ibid Art 570 provides that worker’s wages of the last six months prior to the bankruptcy or judgment of the judicial settlement of the debtor cannot be seized and have priority. ‘Judicial settlement’ is a phase in which a company in economic difficulties asks the judge to give it supplemental time to pay the debts

Priorities 297 in which this allocation of priority between secured creditors and preferential claimants holding privileges has been presented in Articles 199 and 200 of the CRR is consistent with the recommendation underlying Article 36 of the Model Law that requires enacting States to clearly set out the types and amounts of preferential claims that have priority over security rights.91

2. Priority of Competing Possessory Pledges According to the general rule of the CRR, priority between pledgees who have taken a possessory pledge is based on the date of the agreement that provided for the creation of the pledge.92 Given that there cannot be two pledgees who are both in actual possession of goods, this rule is primarily (if not solely) applicable in a scenario where a third party, such as a collateral manager, holds the pledgor’s assets for the benefit of the first pledgee but subsequently also undertakes to hold the same asset for the benefit of a second pledgee. Unlike the Model Law where the date and time of taking possession of the encumbered asset determines priority,93 under the CRR the date and time when the pledge agreements were executed determines their respective priority.94 Possessory pledges are rarely used in the Tunisian market and are commonly restricted to transactions involving only one pledgee, and therefore the application of these priority rules is theoretical.

3. Time of Registration The preceding rules (for possessory pledges) do not apply to pledges for which registration is required. Regarding non-possessory pledges, the Tunisian Supreme Court, in its decision no 11446 of 1985,95 confirmed that the order of their registration determines their priority, as stated in Article 269 of the CRR. The ‘time of registration’ rule applies not only to pledges governed by the CRR, such as the enterprise pledge,96 but also to pledges created under special laws, such as pledges over tools and equipment.97 The time of registration is

without going through bankruptcy procedures. In this context, the wage of a worker is divided into two parts. A part that cannot be seized (necessary for the worker’s basic needs), and a part that can be seized. CC (n 38) Art 571 provides that as for the rest of the wages (the part that can be seized), the workers can rely on CRR (n 37) Art 199 in order to claim payment of the remaining wages; however, ibid Art 199 does not specify the meaning of the ‘last six months’. As a result, the provisions of CC (n 38) Art 570 should be applied, so ‘last six months’ means the six months prior to the judgment of the judicial settlement or bankruptcy. ibid Arts 570 and 571 were recently adopted and incorporated in the CC by ‘ ‫ ددع نوناق‬36 ‫ ةنسل‬2016 ‫ يف خرؤم‬29 2016 (1 ( [Law No 36 of 29 April 2016, Relating to Collective Procedures], and they are available in Arabic at: www.legislation. tn/sites/default/files/fraction-journal-officiel/2016/2016F/038/Tf2016361.pdf. ‘Collective procedures’ include rescuing corporations in economic difficulties and bankruptcy rules. 91 LG, Recommendation 83. 92 CRR (n 37) Art 259. 93 ML, Art 29(b). 94 See CRR (n 37) Art 29(b). 95 See discussion of the Tunisian Supreme Court decision no 11446 of 1985, above at n 84 and text. 96 CC (n 38) Art 240; see discussion of the enterprise pledge, above C1–C3. 97 Law Related to Pledges of Tools, Machinery and Professional Equipment 2001 (n 1) Art 5, see C1 above.

298 Tunisia therefore both a priority point and the time when the pledge becomes effective against third parties;98 this is a position identical to that of the Model Law.99

4. Time of Agreement In the case of pledges over receivables, priority is determined by the date of creation, as evidenced by legalisation of the signatures on the pledge agreement or the recording of the agreement with the tax authority.100 This rule also applies to the priority of pledges over bank accounts.101 The application of this priority rule may be illustrated as follows. A bank enters into an agreement to create a pledge over the pledgor’s receivables on day 1. On day 3, the signatures of the parties are legalised. On day 5, the receivables debtor is notified of the pledge. The priority of this pledge dates from day 3. This is also the date on which the pledge becomes effective against third parties.102

5. Priority of Lessors Under Article 12 of the Leasing law, failure to register a lease agreement renders the lessor’s ownership ineffective against third parties, including other creditors of the lessee and buyers of the asset, unless the third party acquired the asset with actual knowledge of the lessor’s right. This rule applies only with respect to leases over those assets for which a registry has been established, including motor vehicles and equipment. If, however, the lessor does register the lease, third parties take subject to the lessor’s ownership rights. With regard to leases over assets for which no registration system has been established, such as household goods, third parties also take the asset subject to the ownership right of the lessor.

F. Enforcement of Security Rights Enforcement of security rights in an efficient and expeditious manner is a major challenge in Tunisia. Out-of-court enforcement is generally prohibited under Tunisian law, forcing parties to resort to the CCCP103 to seek judicial relief, including provisional seizure. The CRR stipulates that any clause empowering a secured creditor to dispose of a pledged asset without any judicial process is void.104 An exception to this general rule is found in Article 254 of the CRR, which provides that a pledge over a bank account may be enforced without a court order.105 However, enforcement of a pledge over a receivable requires a

98 See

discussion of when pledges become effective against third parties at D3 above. Arts 18 and 29. 100 CRR (n 37) Art 255; see discussion of pledges over receivables, above nn 55–56 and text (section C(iii)). 101 COC (n 36) Art 450, para 2, nos 1 and 4. 102 See discussion of date on which pledges become effective against third parties, D2 above. 103 CCCP (n 40). 104 CRR (n 37) Art 256. 105 EBRD (n 1) 48. cf South Africa B. 99 ML,

Enforcement of Security Rights 299 court order, and where there are multiple pledges over the same receivable, only the senior creditor has a right to collect the receivable.106 This approach to enforcement is inconsistent with the Model Law, which recognises extrajudicial enforcement107 and does not restrict enforcement rights only to senior secured creditors.108 The unavailability of extrajudicial enforcement rights with respect to pledged receivables is especially troublesome because the potential for physical violence and other ‘disturbance of the peace’ by the secured creditor and the grantor, present when repossessing tangible assets, does not exist when collecting receivables. In some cases, a third party may create a security right in its assets to secure an obligation of the debtor.109 Here, if the debtor defaults, the Supreme Court’s ruling no 975 of 2004 makes it clear that the secured creditor must attempt to collect the payment of the debt from the debtor, and only if such efforts do not fully satisfy the outstanding obligation can the secured creditor proceed to enforce its rights against the grantor.110 However, the Supreme Court has also interpreted the relevant COC provision111 to allow the third-party grantor to provide its assets under ‘a common pledge with solidarity’, which authorises the secured creditor to proceed against the grantor and the debtor concurrently. Article 72 of the Model Law is not similarly restricted and authorises the secured creditor to enforce any of its post-default rights, so that it may proceed first against an encumbered asset before attempting to enforce the debtor’s obligation even where the security right is granted by someone other than the debtor.

1. Enforcement of Possessory Pledges The CRR contains rules applying specifically to the enforcement of a possessory pledge.112 These rules are supplemented by the provisions of the CCCP that prescribe the requirements for judicial enforcement, including provisional seizure. Under the CRR, after a failure to perform a secured obligation, the secured creditor must send a notice via the bailiff to the grantor (and to the debtor if different from the grantor). If the secured creditor wishes to enforce by selling a pledged asset, it must request the court to authorise the sale of the pledged asset within eight days of receipt of the notice by the grantor (and debtor). The grantor (or the debtor) may object to enforcement by filing a petition with the court, but only within the eight-day time period. If the court rejects the objection, or if no objection is lodged within the time period, the secured creditor may proceed with the

106 CRR (n 37) Art 255. 107 ML, Art 73(1). 108 If a junior secured creditor enforces, a higher-ranked creditor has the right to take over enforcement in certain circumstances: ML, Art 76. Anyone taking an encumbered asset under an extrajudicial disposition, lease or licence from a junior creditor takes it subject to the rights of the senior creditor: ML, Art 81. 109 This contract, named cautionnement in French, is regulated in COC (n 36) Arts 1478–1531. This is a special security contract that used to be very common. Nowadays, it is barely used in practice because of a lack of trust between parties. It remains common between family members for small amounts of money. 110 This position was also adopted by the Supreme Court in a former decision (Cass civ (7047) 23 December 1982). 111 Art 1478, see n 109 above 112 CRR (n 37) Arts 247 et seq.

300 Tunisia disposal. The actual disposal of the pledged asset must be conducted pursuant to the rules of the CCCP.113 The secured creditor must notify the grantor 48 hours before the disposal. The notification must include the date, the hour and the place of the disposal, as well as the proposed price of the collateral. Failure to notify the grantor properly will invalidate the sale.114 This consequence contrasts with that prescribed in Article 81(5) of the Model Law, where only non-compliance with the enforcement requirements that materially prejudice the rights of the grantor or another person, and of which the buyer has knowledge, would affect the sale. Under the CCCP the grantor has the right to redeem the pledged asset up until the moment of disposal.115

2. Provisional Judicial Remedies The CCCP authorises the court to order provisional seizure of a pledged asset after default of the grantor. The meaning of what amounted to a default that was sufficient to allow the secured creditor to seek provisional seizure of an asset was analysed by the Supreme Court in its decision (Civil Ruling) no 40490 of 1996, in which, relying on Article 322 of the CCCP, it held that provisional seizure is available if the secured creditor proves that repayment of the debt is at risk, even though a monetary default has not yet occurred.116 The order will be granted at the request of the secured creditor who must present sufficient evidence to support the request. The judge will issue an order expeditiously without a hearing, on an ex parte basis.117 An application for a seizure must be brought before the court of the jurisdiction in which the debtor resides.118 The issuance of a provisional order may not affect the use of the asset, as the law permits the grantor to retain possession of the collateral unless (i) otherwise ordered by the judge, (ii) a receiver is appointed or (iii) until the provisional order becomes an executive seizure.119 According to the CCCP, the judge shall issue or deny the request for an order within 24 hours.120 However, in reality, these timelines vary according to the case-load of the judge. For instance, in the courts of the capital city, Tunis, the issuance of an order may take one to three weeks. Once granted, the bailiff has 10 days to execute the order. The bailiff must communicate the order to the grantor, after which the grantor will be able to retain the pledged asset if it (i) fully satisfies the secured obligation, (ii) enters into a new agreement with the secured creditor or (iii) files a claim with a court challenging the entitlement of the secured creditor to enforce its rights. If the grantor does not take any of these steps within 20 days from the day it was notified of the judge’s order, the provisional seizure will

113 ibid Art 247. 114 ibid Art 250, para 2. 115 ibid Art 252. 116 According to CCCP (n 40) Art 322, para 2, ‘Order for the seizure may be authorised for security of any claim appearing legitimate and whose recovery is at risk.’ 117 An order may be set aside, but a request to set it aside does not suspend its execution (CCCP (n 40) Arts 219–23). 118 The parties can agree to choose another jurisdiction (generally the clause is imposed by the creditor). 119 CCCP (n 40) Art 324. 120 ibid Art 216.

Enforcement of Security Rights 301 a utomatically be converted into executive seizure.121 Where the grantor refuses to surrender possession of the asset, the bailiff may seek assistance from the police.

3. Disposal of the Pledged Asset The CCCP prescribes the general rules applicable to the sale of pledged assets unless the relevant law provides for special procedures. For instance, securities law provides for enforcement against securities as discussed below. Pledged assets may only be disposed of by sale in an auction held no earlier than eight days after seizure of the asset.122 The secured creditor must ensure publication of notices announcing the auction in two newspapers.123 If the encumbered asset is estimated to have a high value, the bailiff may ask the court to appoint an expert in order to determine its approximate market price.124 The costs and expenses of the expert must be borne by the secured creditor.125 In cases that involve assets of significant value,126 the costs and remuneration of an expert can reach thousands of US dollars. An enterprise pledge may only be enforced by the sale of the entire fonds de commerce, rather than by its individual pieces; that is, by sale of the entire business.127 For the enforcement of pledges over securities, the Supreme Court in decision no 2603 of 2004 ruled that the secured creditor must follow the procedures prescribed in Article 407 of the CCCP.128 This has the effect that a secured creditor cannot dispose of pledged securities without a court order. The procedure for disposal of securities differs depending on whether the company that issued the securities is a public company (société faisant appel public à l’épargne) or a privately held company (société ne faisant pas appel public à l’épargne). Enforcement of a pledge of securities by a secured creditor can only be by way of sale, as there is no provision for the secured creditor to appropriate them.129 For securities issued by publicly traded companies, the court may authorise a bailiff to dispose of the securities in accordance with the rules of the securities exchange,130 including the General Regulations of the Stock Exchange Securities of Tunis.131 The sale must be effectuated following one of two methods authorised by the securities exchange, namely direct

121 CCCP (n 40) Art 390. 122 ibid Art 394. 123 ibid Art 396, para 2. The cost of complying with the publication requirement depends on the length of the announcement (ie the total number of words) and should not be too expensive. It should not exceed US$35–50. 124 ibid Art 394. 125 Generally, the creditor will bear the expenses, but if the appointment of an expert was requested by the debtor, the debtor will bear the expenses. ibid Art 394. 126 Assets with significant value such as jewellery. ibid Art 398. 127 EBRD (n 1) 48. 128 Cass civ (2603) 11 August 2004, available in Arabic from: République Tunisienne Ministère de la Justice, Formulaire de recherche [Republic of Tunisia Ministry of Justice, ‘Research Form’] (2019) jurisprudence.e-justice. tn/wwwisis/juris.10/form.htm. 129 cp UNIDROIT, ‘Convention on Substantive Rules for Intermediated Securities’ (25 September 2013) Art 33(1)(a)(ii). 130 CCCP (n 40) Art 406. 131 General Regulations of the Stock Exchange Securities of Tunis 1994 Arts 186–93, available in French at: Le Collège du Conseil du Marché Financier (CMF), ‘reglement general de la bourse des valeurs mobilieres de tunis’ [The College of the Financial Market Council, ‘General Regulations of the Stock Exchange Securities of Tunis’] (1994) www.cmf.org.tn/pdf/textes_ref/reglementations/Version_FR/bourse/rgb.pdf.

302 Tunisia selling or sale by public auction.132 The unavailability of extrajudicial disposal of securities and the right to accept securities in satisfaction of the secured obligation unnecessarily increases the cost of enforcement actions.

4. Lessors’ Remedies The Leasing law does not include any specific rules on enforcement of a lessor’s rights upon the default of a lessee. Although default is a breach of the contract that normally empowers the lessor to repossess the asset, in practice, lessees often refuse to surrender the asset. Bailiffs are also reluctant to provide their assistance solely based on the lease contract that confers such powers on the lessor, so the lessor is left with no choice but to seek a court order.133 Although a lessor may use expeditious judicial enforcement processes, the appeal process allows a lessee to delay the completion of enforcement significantly.134

G. Conclusion As mentioned above, Tunisia embarked on a reform of its secured transactions regime in the middle of 2016. An initial approach was to systematically modernise and rationalise the system of pledges, leaving finance leases and transfers of receivables aside. After some discussion, the Working Group was presented with a draft bill based on the Model Law. Although the draft bill followed the structure of the Model Law, it also included a few deviations from its model, such as specific rules on attachments to immovable property and serial number indexing and searching with respect to motor vehicles, which are rules commonly found in African secured transactions laws.135 One of the challenges was to translate the concepts of the Model Law properly into Arabic, the language in which all Tunisian legislation is drafted. At the time of drafting, an official Arabic translation of the Model Law was not available and, in any case, the Arabic translation of the draft Model Law available on the UNCITRAL website did not correspond to the Arabic terminology customarily used in Tunisian legislation. The discussion in this chapter has pointed out many of the shortcomings of the current Tunisian law in relation to secured transactions. Although it is possible for a lender to take security over many types of assets which a business, or an individual, would be likely to want to use as collateral, there are gaps. In particular, the inability of inventory to be given as collateral under a non-possessory pledge, even as part of a security right over the fonds de commerce, is problematic. The legal framework is fragmented so that different rules apply depending on the form of security device used. Thus, for example, equipment can be encumbered under a pledge, a retention of title device or a lease, and receivables can be the subject of an outright transfer or a non-possessory pledge, but part of a pledge over the fonds de 132 CCCP (n 40) Art 186. 133 Interview with Mr Mohamed Fadhel Hamdi, Director, Attijeri Leasing Litigation Department (Tunis, Tunisia, 8 September 2016). 134 ibid. 135 See Conclusion C1b and D5c.

Conclusion 303 commerce. Even the law relating to non-possessory pledges is not uniform and is governed by several different statutes, resulting in different rules for creation, third-party effectiveness and priority. The registries in Tunisia are also in dire need of modernisation, as they fail to provide the kind of public notice that prospective secured creditors need to assess lending risks. The ability of secured creditors to enforce their security is also hampered by the complex rules on enforcement and, in particular, the lack of effective out-of-court enforcement.

304

part iv Mixed Jurisdictions

306

14 Rwanda In 2013, Rwanda’s parliament enacted Law No 34/20131 on Security Interests in Movable Property (SIMP), which repealed Law No 11/2009 on Security Interests in Movable Property (‘the 2009 Law’) that also established a registry of security interests.2 Unlike the majority of other African economies that have reformed their decades-old regimes, Rwanda’s reform was aimed at addressing some deficiencies in the 2009 Law, which was itself an early attempt at reform.3 Implementation of the SIMP has led to a significant improvement in Rwanda’s ranking in the World Bank’s Doing Business Ease of Getting Credit rankings. This chapter discusses and explores the provisions of the SIMP and, where relevant, juxtaposes them against the provisions of the UNCITRAL Model Law. The SIMP is an example of a ‘simplified model’ developed by individual consultants, rather than a law based on an internationally recognised model. It is rife with conceptual and drafting errors that present many risks for prospective secured creditors. These misguided approaches and drafting errors call into question Rwanda’s high ranking in the World Bank’s Doing Business Report. Although the SIMP seemingly complies with the general benchmarks, such as whether the secured transactions law allows extrajudicial enforcement, the actual provisions render its practical utilisation uncertain and cumbersome.

A. Country and Economic Overview Rwanda is a land-locked economy in East-Central Africa.4 A member of the East A frican Community,5 it was colonised by the Belgian Crown, as a result of which its civil law framework is based on the Belgian civil law system,6 and attained independence in 1962.7 The economy declined in the 1990s as a result of civil war and genocide,8 but has since 1 The law came into force upon its publication in the Official Gazette no 21 of 27 May 2013. 2 Doing Business 2010: Reforming Through Difficult Times (Washington, DC, World Bank, 2009) 34. 3 Art 35 of the 2009 Law repealed Decree of 12 January 1920 on Security Interests in Business and Commercial Invoices and the Decree of 28 July 1934 on Bill of Exchange, Promissory Note and Protest. 4 ‘Rwanda Country Profile’ BBC (17 September 2018) www.bbc.com/news/world-africa-14093238. 5 East African Community, ‘Overview of EAC’, www.eac.int/overview-of-eac. The EAC aims at political, economic and social integration of the east Africa region. ibid. 6 E Musiime and F Kabasinga, ‘Update: Rwanda’s Legal System and Legal Materials’ Globalex (August 2013) www.nyulawglobal.org/globalex/Rwanda1.html. See also H Goitom, ‘Burundi; Rwanda: Possible Switch of Legal Tradition’ Library of Congress Global Legal Monitor (4 June 2009) www.loc.gov/law/foreign-news/article/ burundi-rwanda-possible-switch-of-legal-tradition/. 7 African Studies Center, ‘Rwanda – History’, www.africa.upenn.edu/NEH/rwhistory.htm. 8 CIA, ‘Economy: Rwanda’ (The World Factbook, 15 November 2016) www.cia.gov/library/publications/ the-world-factbook/geos/rw.html.

308 Rwanda recovered, averaging real GDP growth of over 7 per cent between 2001 and 2015.9 Mainstays of the Rwandan economy are agriculture, services and, to a lesser extent, industry.10 Agriculture remains the main employer, with industry and services a distant second.11 Tea, coffee, minerals and tourism are its chief exports.12 To reach its development goals, Rwanda relies heavily on foreign aid, which constitutes between 30–40 per cent of the government budget.13 The country, originally a civil law jurisdiction, is transitioning to a common law legal system,14 a move that would bring its legal framework more closely in line with its common law East African neighbours (Kenya, Tanzania and Uganda).15 It has been estimated that MSMEs in Rwanda account for approximately 97 per cent of businesses, contribute to 55 per cent of the total GDP, and employ around 41 per cent of the population.16 Rwanda has received foreign assistance in this transition process to a different legal system. For example, in 2011 Rwanda adopted Law No 45/2011 Governing Contracts.17 This Law was drafted with the assistance of US scholars.18 Further, in 2010, a team from the international law firm Allen & Overy commenced a three-year project to train Rwanda’s legal community in the common law system and general commercial law.19 In line with the common law practice of judicial precedents, the judiciary recently declared that decisions of the Rwandan Supreme Court are now binding on all lower courts.20 Rwanda’s efforts to improve access to credit have been rewarded with favourable rankings and considerable media coverage. Its secured transactions legal framework was awarded high scores in the World Bank’s 2016 Doing Business Report Getting Credit Indicator21 and is ranked in the 2019 rankings as the leading sub-Saharan African economy, tied with

9 World Bank, ‘Rwanda: Overview’ (8 April 2019) www.worldbank.org/en/country/rwanda/overview. 10 Agriculture employs over 80 per cent of the population. CIA (n 8). 11 ibid. 12 World Bank, ‘An Outlook of Rwanda’s Export Performance’, www.worldbank.org/content/dam/Worldbank/ document/Trade/Rwanda%20Infographic%20-%20An%20Outlook%20of%20Rwanda's%20Export%20Performance.pdf. 13 World Bank (n 9). 14 UK Department for International Trade, ‘Doing Business in Rwanda: Rwanda Trade and Export Guide’ (1 July 2015) www.gov.uk/government/publications/exporting-to-rwanda/exporting-to-rwanda. 15 WE Kosar, ‘Rwanda’s Transition from Civil to Common Law’ (2013)16(3) The Globetrotter 1–2. This is in line with the EAC’s efforts to create a common political federation with a common law legal system. As part of this process, Rwanda and Burundi are required to convert their legal frameworks from civil law to common law. Goitom (n 6); R Mugabe, ‘EAC Wants Rwanda, Burundi to Adopt Common Law System’ New Times (2 June 2009) www.loc.gov/law/foreign-news/article/burundi-rwanda-possible-switch-of-legal-tradition/. South Sudan, another common law country, joined the EAC in March 2016. J Jeffery, ‘South Sudan Joins East African Community Club’ CNBC Africa (4 March 2016) www.cnbcafrica.com/news/east-africa/2016/03/04/south-sudan-joins-eac/. 16 See A Singh, ‘MSME Finance in Rwanda – Status and Opportunities for Financial Institutions’ Microsave (March 2017) www.microsave.net/files/pdf/BN_170_MSME_Finance_in_Rwanda_Status_and_Opportunity_ for_Financial_Institutions.pdf. 17 Of 25 November 2011. The law was published in Official Gazette no 04bis of 23 January 2012. 18 F Crawford, ‘Law Professor Summers is Helping Draft Rwanda's New Code of Contract Law’ Cornell Chronicle (4 October 2006) www.news.cornell.edu/stories/2006/10/law-professor-helping-draft-rwandas-newcode-contract-law. 19 S Lind, ‘A&O Helps Rwanda to Transition from Civil to Common Law System’ Legal Week (26 May 2010) www.legalweek.com/sites/legalweek/2010/05/26/ao-helps-rwanda-to-transition-from-civil-to-common-law-syste m/?slreturn=20170325194309. 20 R Mbaraga, ‘Rwandan Courts Urged to Set Precedents’ The East African (13 December 2016) www.theeastafrican.co.ke/news/Rwandan-courts-urged-to-set-precedents-/2558-3485084-2su69oz/index.html. 21 K Agutamba, ‘Easy Access to Credit Earns Rwanda Top WB Ranking’ New Times (2 November 2015) www. newtimes.co.rw/section/article/2015-11-02/194030/.

Secured Transactions Legal Framework 309 Zambia, for access to credit.22 The Institute of Chartered Accountants in England and Wales has recognised Rwanda’s leadership in access to credit reforms that strengthen borrowers’ and lenders’ collateral laws, scoring Rwanda above 80 per cent on access to credit.23 It has been reported that the reforms have enabled financial soundness24 and made it easier to do business in Rwanda.25 Nevertheless, post-reform surveys indicate that although access to loans might have increased, the availability of credit has not, since the value of collateral required by banks exceeds the amount of credit made available by three- to five-fold.26 Furthermore, the processing of loans is time consuming, averaging two months.27 The same survey also found that banks do not innovate or tailor credit products to the needs of MSMEs. One recent contributor to the ease of doing business in Rwanda is the Law Governing Finance Lease Operations in Rwanda, which replaced Law No 06/2005 of 03/06/2005 – Establishing Regulations and Conditions Governing Lease Operations (‘the 2005 Leasing Law’). Leasing has been recognized as an innovation that holds the potential to enhance access to investment capital by SMEs, as it does not require collateral or a credit history.28 The process leading to the adoption of the new leasing law in Rwanda was fostered by the IFC. In 2009, the IFC developed a survey of the leasing market in Rwanda.29 The survey showed that the transportation and distribution sector constituted the bulk of leasing activity, accounting for 64.15 per cent of all leases in 2008.30 As a consequence, vehicles were the most heavily financed asset, followed at a distance by plants and machinery.31 The survey also revealed that 90 per cent of the finance leases were in the Kigali area (the capital of Rwanda), with values ranging from $5,000 to $150,000 and for an average period of three years.32 In 2015, Rwanda adopted the Law Governing Finance Lease Operations in Rwanda (‘the Finance Lease Law’). As will be shown below, the new leasing framework contains features far from the best international practice.

B. Secured Transactions Legal Framework Driven by a desire to improve its performance and implement good practices identified in the World Bank’s flagship Doing Business reports,33 in 2013, Rwanda enacted the SIMP, 22 See Doing Business 2019: Rwanda (Washington, DC, World Bank, 2019) and http://www.doingbusiness.org/ en/rankings. 23 K Brenda, ‘Rwanda Makes it Easy to Access Credit than any African Country’ KT Press (17 October 2016) www.ktpress.rw/2016/10/rwanda-makes-it-easy-to-access-credit-than-any-african-country/. 24 ibid. 25 Rwanda Eye, ‘Rwanda Leads Africa in Access to Credit’ African Markets (18 October 2016) www.africanmarkets.com/en/news/east-africa/rwanda/rwanda-leads-africa-in-access-to-credit-report. 26 See A Singh, ‘MSME Finance in Rwanda – Status and Opportunities for Financial Institutions’ Microsave, (March 2017) www.microsave.net/files/pdf/BN_170_MSME_Finance_in_Rwanda_Status_and_Opportunity_ for_Financial_Institutions.pdf. 27 ibid. 28 On the leasing market in Rwanda see IFC, ‘Survey of the Leasing Market in Rwanda’ (May 2009). 29 IFC, Rwanda Leasing Investment Guide (May 2009) 8. 30 ibid 10. 31 ibid 8. 32 ibid 9–10. 33 Since 2007, Rwanda has used the Doing Business Reports and its rankings as a benchmark for business environment and regulatory reforms, setting up a Doing Business Steering Committee for this purpose. Doing Business

310 Rwanda governing the creation, third-party effectiveness, priority and enforcement of security interests.34 As mentioned earlier, the improved framework for commercial lending has led to a significant boost in Rwanda’s ranking in the World Bank Doing Business Report Getting Credit Indicator. The country moved up to the number 2 spot in the 2016 Doing Business Report Getting Credit Indicator from being ranked 32 in the 2011 Report, and ranked joint third in the 2019 Report.35 This is despite the fact that the SIMP has not been modelled on any international standard, such as the Model Law, or other national law model, such as a PPSA or UCC Article 9.36 Rather, the SIMP is a typical example of a ‘simplified model’ that has been developed by a number of independent consultants familiar with the UCC Article 9/PPSA frameworks. As will be illustrated below, the SIMP, as well as many other statutes that implement the ‘simplified model’,37 suffers from a number of deficiencies such as gaps (eg the absence of any conflict of laws rules, uncertainty (eg the law provides for perfection by control but does not specify how control may be constituted and to which assets it relates), conceptual errors (eg a debtor is defined as a person that owes payment and includes a seller of receivables), drafting errors (eg references to words not defined in the Act, such as currency instead of money), and ambiguities (eg Article 6 states that: ‘The description of the collateral may indicate the existence of enforcement of one or several obligations by item or generally by kind …’). The desire to adopt a simplified law has not only removed the SIMP far from the best international practice, but created a secured credit framework filled with uncertainty and unpredictability that can be expected to have a negative impact on secured lending. The SIMP replaced the 2009 Law, which was repealed by SIMP Article 37.38 The repealed law, an earlier attempt at reform, had many shortcomings. For example, it failed to extend its scope of application to sales of accounts receivable, it required a searcher of the registry to demonstrate a purpose for the search that conforms with a number of statutorily identified purposes,39 it prescribed that the registry of security interests be maintained

2013: Smarter Regulations for Small And Medium-Size Enterprises (Washington, DC, World Bank, 2013) 38. In particular, the government has made improvement in Rwanda’s score on every area assessed by the Doing Business rankings a priority, and has engineered reforms to achieve this. J Gettleman, ‘The Conscience of a Strong Man’ New York Times Magazine (8 September 2013) MM38. See also D Kezio-Musoke, ‘Rwanda Makes More Reforms to Ease Doing Business’ Reuters Africa (19 May 2011) http://af.reuters.com/article/topNews/idAFJOE74I0DB20110519. 34 Doing Business 2014 (Washington, DC, World Bank, 2013) 92. 35 Doing Business 2019: Rwanda (Washington, DC, World Bank, 2019) 30. 36 For discussion of these models, see ch 1, B5. 37 Compare with UAE’s Federal Law No 20 of 2016 on Mortgaging Moveable Properties as Security for Debts; and Uganda’s Movable Property Security Interest Act 2019. 38 In addition, SIMP, Art 37 provides that ‘all prior legal provisions contrary to this Law are hereby repealed’. 39 SIMP, Art 30 requires that a searcher of the registry must have one of the following purposes: 1. establishing that there is any relation between the secured creditor or the debtor with the security interest; 2. establishing whether there is any reason which may lead to opposition to give the property in security interest; 3. establishing whether to provide credit to the person, company or any person who has interest in the guarantee; and establishing whether that security interest in the movable property availed by that person or company shall be valid; 4. establishing whether to invest in or with the person or company or related company registered in the register of security in movable property; 5. collecting information in case of an insolvency administrator and any other authorized person upon proving that the needed information from the register is related to the work he/she has been assigned.

Secured Transactions Legal Framework 311 in hard copy as well as electronically,40 it conferred on the debtor the right to register a security interest,41 and granted debtors the power to delay or block the enforcement of a security interest.42 Although the SIMP addresses these deficiencies, it fails the test of best international standards on many other counts. Since the SIMP replaced the 2009 Law under which a registry had already been established, the transition of ‘prior security rights’ was much easier.43 The SIMP came into force on the date of its publication in the Official Gazette. The SIMP was drafted in English, but it was published in the Gazette in Kinyarwanda, English, and French.44 Analysis of the 2005 Leasing Law45 revealed that it was inadequate for leasing development, did not correspond to global best standards and was inconsistent with other laws that affect leasing operations, such as tax laws.46 Among its shortcomings, the 2005 Leasing Law failed to distinguish between finance and operating leases, provided a limited list of assets that could be leased,47 and failed to regulate the rights and liabilities of the parties to a lease agreement properly.48 Armed with this analysis, the government of Rwanda included leasing law reform in its financial sector strategy for 2013–18 to enhance the ability of lenders to provide long-term credit for agricultural, industrial and transportation equipment.49 As a result, in 2015, Rwanda enacted Law No 41 bis/20 Governing Finance Lease Operations in Rwanda,50 (‘the Finance Lease Law’) replacing the 2005 Leasing Law. Chapter IV of the Finance Lease Law governs the registration of finance lease agreements. Article 14 specifically provides that: ‘Finance lease agreement must be registered in a registry designated for that purpose with the Registrar General.’ As will be seen below, the SIMP does apply to finance leases, and therefore this registration should be carried out in the SIMP registry. However, Articles 15–17 of the Finance Lease Law raise a number of important questions concerning proper coordination with the SIMP. Since 2009, there have been three different Companies Acts in Rwanda. The Companies Act 200951 was repealed and replaced by the Companies Act 2017,52 which was replaced in 2018 by another Companies Act.53 Although the 2009 Act included provisions on the registration of charges, these provisions do not appear in the two most recent Acts.

40 SIMP, Art 22 provides that: ‘The Register must be in electronic and in hardcopy [form]'. 41 SIMP, Art 24 provides that: ‘The registration shall be effected by the debtor on presentation of receipt of payment of due fees and shall mention the date and time of registration. The Registrar General shall issue a certificate of registration of security interest to the debtor and give a copy to the creditor.’ 42 SIMP, Art 17 provides that both the debtor and creditor must agree to any sale of collateral during enforcement: ‘Upon receipt of the request for sale of collateral by both parties, the Registrar General shall, without delay, issue a notice of sale to be effective in ten (10) working days’ (emphasis added). 43 See ch, 2J. 44 See SIMP, Art 36. 45 See A above. 46 IFC, A Survey of the Leasing Market in Rwanda (May 2009) 28. 47 ibid, citing Art 2 of Law No 06/2005 of 03/06/2005 – Establishing Regulations and Conditions Governing Lease Operations. 48 ibid 28–29. 49 Ministry of Finance and Economic Planning, Final Report: Rwanda Financial Sector Strategy 2013–2018 (2013) vii. 50 Published in Official Gazette no 07 of 16 February 2015. 51 Law no 07/2009 of 27/04/2009. 52 Law no 27/2017 of 31/05/2017, s 312. 53 Law no 17/2018 of 13/04/2018.

312 Rwanda This is consistent with the SIMP, since all security interests created by companies are to be registered in the SIMP registry. The registry of security interests in Rwanda is maintained by the Office of the Registrar General (ORG) at the Rwanda Development Board (RDB). Registrations may be submitted online if the user has obtained an electronic signature issued by the ORG.54 The registration procedure requires the debtor to pay registration and notarisation fees at the Bank of Kigali (or its branches). A copy of the payment slip must be scanned and attached to the registration along with a scan of the notarised abstract of a signed security agreement, when the debtor is an individual. In the case of a company debtor, an additional document is required, namely a notarised board resolution authorising the company to secure a loan.55 Until recently, registrations were permitted to be completed ‘in paper’, which was a cumbersome process where the applicant would submit a certified copy of proof of ownership of the collateral and a notarised copy of the security agreement to the registration officer at the ORG. A certificate of registration of security interests in movable property is issued at the end of the registration process.

C. Law on Security Interests in Movable Property The SIMP is divided into seven chapters that contain a total of 38 articles. Article 1 briefly sets out the purpose of the law, which is to govern security interests in movable property, and Article 2 includes definitions, which will be referred to below when they pertain to a particular question examined.

1. Scope Articles 3 and 4 establish the scope of the SIMP. The former reflects the functional approach, as it extends the application of the SIMP to any rights in movable assets, created by an agreement, that secure payment or other performance of an obligation. The SIMP applies to such rights irrespective of the form of the transaction, the type of the movable asset, the status of the parties and the nature of the secured obligation. However, Article 3 does not set out the full panoply of transactions to which the SIMP applies, such as sales of receivables, and must therefore be read with Article 4. The chapeau of Article 4 confusingly states that ‘movables subject to security interest shall mainly be comprised of the following’, and then provides under paragraph 4 that the SIMP also applies to ‘transfers of title to tangible assets for security purposes or assignments of receivables for security purposes, the various forms of retention of title sales and financial leases’. Obviously, paragraph 4 does not identify any movables, but rather transactions to which the SIMP applies. Paragraph 6 of Article 4 identifies additional transactions subject to the SIMP, including sales of

54 See Office of the Registrar General, ‘Register of Security Interests in Movable Property’, www.org.rdb. rw/?page_id=240. 55 See Office of the Registrar General, ‘The Law on Security Interests in Movable Property’ (2018) www.rdb.rw/ wp-content/uploads/2018/02/Booklet-Movable-Property.pdf.

Law on Security Interests in Movable Property 313 accounts receivable or chattel paper, and the lease of goods for more than one year. However, these three types of transactions are subject only to the SIMP’s rules on priority and registration of a notice and not to its other rules. To complete the picture of the SIMP’s application, the reader must look to paragraph 5 of Article 4, which provides that liens are subject to the SIMP for the purposes of priority and registration. A lien is defined in Article 2(20) as ‘a right in property which is created by operation of law, by an order of a court or other legal authority’, but excludes the right of retention which is defined separately in Article 2(21). Curiously, Article 4 does not specify whether the SIMP applies to rights of retention, although Article 13 provides for their priority as against security interests. Such cumbersome presentation of the fundamental scope provisions makes it difficult for the user to determine whether the SIMP applies at all to any type of transaction or interest, and, if so, which of its provisions apply. The starting point is the SIMP’s general application to all transactions that secure the performance of an obligation with a movable asset, thus adopting the functional approach which also encompasses financial leases.56 However, its rules on registration and priority also apply to some transactions whose function is not to secure an obligation, namely sales of accounts receivable, sales of chattel paper, leases of goods for more than one year and non-consensual liens. Finally, the SIMP’s priority rules also apply to rights of retention. One final point concerning the scope relates to the definition of movable property in Article 2(15).57 In this definition, ‘tangible property’ includes trees to be harvested. Arguably, the drafter attempted to replicate a similar provision of UCC Article 9,58 which defines goods to include ‘standing timber that is to be cut and removed under a conveyance or contract of sale’. It is questionable whether the mere intention of the debtor to harvest the trees would alter their nature from immovable to movable property. Under US law, it is the contract to cut that reclassifies the timber from immovable property to goods.59 Without the requirement of a contract, the actual moment when the trees are to be harvested, and thus become movable property, remains uncertain under the SIMP. The Model Law does not provide any guidance on the creation of a security right in growing trees before they are actually harvested. The movable property definition also states that intangible property includes chattel paper, documents of title, negotiable instruments and money. However, all of these assets would usually be physical embodiments of some intangible rights unless they are issued in an electronic form. This seems to be a conceptual error, an argument supported by the definition of money that refers to currency. Currency in the form of coins and bills is clearly not intangible property.

2. Creation Chapter II of the SIMP is entitled ‘Principles Relating to Creation and Effectiveness of Security Interests’. Article 5 provides the requirements for the creation of a security i nterest. 56 Ch 2, B2. 57 In addition to the defined term, SIMP interchangeably also refers to personal property and movables. For instance, see the definition of collateral in Art 2(8) for the reference to ‘personal property’ and the title of Art 4 for the reference to ‘movables’. 58 UCC, §9-102(a)(44) 59 See Feliciana Bank & Trust v Manuel & Sessions, 943 So 2d 736, 739–40 (Court of Appeals of Mississippi 2006).

314 Rwanda They are that a written security agreement has been entered into between the parties, that value is given and that the debtor has rights in the collateral. The drafter of this article took inspiration from the UCC Article 9/PPSA concept of attachment, and relabelled it as creation.60 However, unlike under the UCC Article 9/PPSA, where on attachment a security interest becomes enforceable against the debtor and third parties,61 under the SIMP, on satisfaction of the three requirements, the security interest is created and effective only against the debtor.62 This approach differs from the Model Law, which provides that a security right is created in an asset on execution of a security agreement, provided that the grantor has rights in the asset or the power to encumber it.63 Article 7 adds to the conceptual confusion by stating: ‘A security interest shall be effective upon its registration by the Registrar General.’ Given the placement of this article in Chapter II rather than in Chapter III, which deals with third-party effectiveness, one may wonder whether registration is a condition of effectiveness between the debtor and secured creditor or effectiveness as against third parties. If the former, it would add an additional requirement to the other three set out in Article 5, and be inconsistent with the Model Law as well as the UCC Article 9/PPSA model.64 If the latter, it would be a redundant statement of the rule covered in Article 8. At the same time, it would directly contradict Article 8, which recognises possession and control as alternative methods of third-party effectiveness. Following the conceptual structure of the Model Law would have avoided such ambiguity. Unfortunately, this attempt to transform the UCC Article 9/PPSA notion of attachment into creation is a standard feature of ‘simplified models’. The last article of Chapter II deals with the description standards for secured obligations and collateral in security agreements. Article 6 contains two provisions, one applicable to secured obligations and the other to collateral. The former is quite straightforward, stating that: ‘A security interest may secure one or multiple obligations, which may be described specifically or generally, and which may be present, pre-existing or future obligations.’65 However, the second provision is difficult to understand: ‘The description of the collateral may indicate the existence of one or several obligations by item or generally by kind, and its obligations may be existing ones, those which have existed or those whose existence is to come.’ Perhaps the intention of the drafter was to state that a security agreement may describe the collateral by item or generally by kind, and may also cover future property. In contrast, the Model Law contains clear rules on the description of both secured obligations and collateral, as well as on the creation of a security right in future property.66 Unlike the Model Law, the SIMP does not include a provision that would override the effect of anti-assignment clauses.67 In the absence of a specific provision, the prospective

60 See UCC, §9-203. For discussion of the distinction between the two concepts, see ch 2, C. 61 However, many third parties will then take free of, or achieve priority over, an attached security interest unless it is perfected. 62 UCC, §9-203(b) provides that on attachment ‘a security interest is enforceable against the debtor and third parties with respect to the collateral’. 63 Model Law, Art 6(1). See ch 2, C. 64 This is the position taken in the booklet ‘The Law on Security Interests in Movable Property’ (n 52), although the discussion in that booklet is not clear and further confuses the position. 65 Compare with ML, Art 9(1)(3). 66 See ML, Art 6(2). 67 See, eg, ML, Art 13 and UCC, §9-406.

Law on Security Interests in Movable Property 315 secured creditor would need to examine other relevant law, in this case, the Civil Code, to determine whether an anti-assignment clause in the underlying contract can affect a security interest in the receivable. The absence of provisions that override the effect of antiassignment clauses is symptomatic of the ‘simplified models’, but means that an important benefit is omitted, which could have a very adverse effect on access to credit.68

3. Third-Party Effectiveness Chapter III contains two articles that govern third-party effectiveness of security interests. Under Article 8, a security interest is effective against third parties when it has been created in accordance with Article 5 and one of the measures to achieve third-party effectiveness has been taken. The reference to Article 5 supports the argument, set out above, that Article 7 has no place in the conceptual structure of the SIMP and that a security interest is created and becomes effective against the debtor whether or not a notice thereof has been registered under Article 7. Article 8 sets out, in two paragraphs, three methods of third-party effectiveness. Under paragraph 1, the security interest (not a notice)69 may be registered in the Office of the Registrar General, and under paragraph 2, the secured creditor may take possession or control of the security.70 Third-party effectiveness by registration is covered in Chapter VI of the SIMP, supplemented by the relevant definitions of ‘notice’ and ‘registered security interest’ in Article 2. However, the SIMP defines neither possession nor control. To determine what constitutes possession for the purposes of achieving third-party effectiveness of a security interest, one would need to examine law other than the SIMP. In contrast, the Model Law clearly defines possession, thus reducing the transactional cost for the parties by eliminating the need for a legal analysis of the meaning of possession under the general law.71 The more significant issue relates to control. The SIMP neither defines control nor specifies what assets may be subject to a security interest that is made effective against third parties by control. Can a secured creditor who requests that the receivables debtor remit payments to a designated bank account claim that it is in control of such receivables, and thus make its security interest effective against third parties? The SIMP gives no answer to this question, leaving the door open for secured creditors to argue that they are in control of the collateral if they have failed to register a notice or the registered notice was ineffective. In contrast, the Model Law clearly defines control and identifies the assets to which this method is applicable, namely deposit accounts and non-intermediated securities.72 Only the courts will determine what constitutes control for the purposes of the SIMP. The second of the two third-party effectiveness articles deals with security interests in proceeds. Unlike the Model Law, the SIMP addresses the continuation of the security

68 See ch 1, E4a. 69 See ch 2, E1 for discussion of registration of notices. 70 The reference to ‘security’ rather than ‘collateral’ or ‘movable property’ – both defined in Art 2 – is a drafting error. 71 ML, Art 2(z) defines possession as ‘the actual possession of a tangible asset by a person or its representative, or by an independent person that acknowledges holding it for that person’. 72 ML, Art 2(g)(i)(ii).

316 Rwanda interest only in the context of third-party effectiveness.73 The first provision of Article 9 extends the third-party effectiveness of the security interest to proceeds automatically, where the security interest in the collateral was effective against third parties. However, under the second provision of this article, the effectiveness against third parties of a security interest in proceeds is merely temporary, lapsing 15 days after the debtor receives the proceeds, if the proceeds are not identifiable cash proceeds or described in the registered notice.74 Proceeds are defined in Article 2(4), but the SIMP does not provide any guidance as to which proceeds are cash proceeds and which are not. This is yet another attempt at simplification that renders the operation of an important rule uncertain. The drafter might again have been inspired by the UCC Article 9/PPSA notions, but did not incorporate the definition of cash proceeds contained in those frameworks. For example, UCC Article 9 explicitly defines cash proceeds to mean money, cheques and deposit accounts.75 Article 19(1) of the Model Law clearly identifies the types of assets (ie cash proceeds) in which a security right remains automatically effective against third parties, including money, receivables, negotiable instruments and rights to payment credited to a bank account. For non-cash proceeds, Article 9 of the SIMP provides a rule that allows the secured creditor, within a grace period of 15 days, to take action to ensure the continuation of third-party effectiveness in such proceeds. However, there are two problems with this provision. First, the provision states that: ‘If the proceeds are not cash …’. This should be read as: ‘If proceeds are not cash proceeds.’ Otherwise, the effect of this rule would be limited only to cash, understood to mean money, rather than the broader ‘cash and its equivalents’. This will have a negative impact on the rights of secured creditors in proceeds of their collateral. The second issue relates to the only action that the secured creditor may take to ensure continued third-party effectiveness, which is amending the registration to add a description of those proceeds. Article 19(b) of the Model Law is not similarly limited, and allows the secured creditor to ensure continued third-party effectiveness of a security right in non-cash proceeds by any method, including possession and control. Under the SIMP, if, for instance, a negotiable instrument is taken on the sale of some inventory, the secured creditor’s possession of the instrument might not ensure continued third-party effectiveness if instruments are not included within ‘cash proceeds’. This is an anomalous approach because the SIMP recognises possession as a method of third-party effectiveness for original collateral. As a result, taking possession of the instrument in this scenario would make the security interest effective against third parties only from the time the secured creditor took possession of the instrument, but not continuously from the time when the security interest in the inventory was originally made effective against third parties. Once again, the attempt at simplicity causes practical difficulties and conceptual anomaly. Article 9 contains one more deviation from the best international standards, which could either be attributed to a conscious policy choice or to an erroneous desire to replicate the concept of attachment under the SIMP. The 15-day grace period to amend the registered notice starts running when the debtor receives the proceeds. In contrast, under

73 Compare with ML, Art 10(1) on the creation of a security right in proceeds, which provides that a security right in an encumbered asset extends to its identifiable proceeds. See ch 2, C3. 74 The provision ambiguously refers to ‘proceeds described by the collateral description in the register’. 75 UCC, §9-102(a)(9).

Law on Security Interests in Movable Property 317 Article 19 of the Model Law this period runs from the time the proceeds arose. Arguably, the grace period under the latter will expire sooner because the proceeds must first arise before the debtor may receive them. Under UCC Article 9,76 the 21-day grace period runs from the time the security interest attaches to the proceeds. In an attempt to simplify the UCC rule, the drafter misconstrued the meaning of attachment in this context as when the debtor receives the proceeds. However, attachment may occur earlier, as soon as the debtor acquires rights in the collateral, rather than their actual receipt.

4. Priority Chapter IV, entitled ‘Priority Between Security Interests’, includes nine articles setting out rules addressing conflicts between competing claimants with respect to the collateral. The title of the chapter is not entirely reflective of its contents as it also governs the rights of lienholders and persons with a right of retention.77 The fundamental priority rule, set out in Article 10, is that the ‘first to register a notice or achieve third-party effectiveness by some other means has priority’.78 The rule applies equally to security interests and liens. The second statement in Article 10 of the SIMP provides for continuous priority in situations where third-party effectiveness has been achieved by one method (such as possession), but later changed into another method (such as where the secured creditor registers a notice before relinquishing possession).79 However, the statement that ‘priority shall be measured from the date of registration’ may be problematic, because it seems to suggest that the actual time of registration is immaterial in establishing priorities. As a result, if two registrations were made against the same debtor and the same collateral on the same day, they would have equal priority. In contrast, under the Model Law, it is the order of registrations that determines the priority of competing security rights. The general priority rule of Article 10 is subject to the other rules set out in the SIMP, one of which appears to be included in Article 11 covering acquisition security interests, entitled the ‘Acquisition security interest exception to priority rule’. Despite the title and the expectation of a user to locate a priority rule in Article 11, it does not set out any priority rule whatsoever. Instead, it merely defines an acquisition security interest.80 The absence of a priority rule giving an acquisition security interest priority over interests previously registered or otherwise made effective against third parties, and, ideally, distinguishing between different types of movable property (eg equipment, inventory) and specifying different conditions to be satisfied by each type in order for an acquisition security interest to enjoy priority,81 eviscerates the entire purpose of that provision. The reasons for this

76 UCC,

§9-315(d). SIMP, Arts 10 and 13. 78 It corresponds to the fundamental priority rule of the ML stated in Art 29. 79 The essence of this rule corresponds to ML, Art 31. 80 SIMP, Art 11: 77 See

Acquisition security interest occurs in case of: 1° a security interest is taken in collateral by a seller to the extent that it secures the obligation to pay all or part of the collateral purchase price; 2° a security interest taken in collateral by a secured creditor who gives value for the purpose of enabling the debtor to acquire rights in the collateral, to the extent that the value is applied to acquire those rights.

81 See

ch 2, F2 and ML, Art 38.

318 Rwanda super-priority protection for acquisition creditors are set out in Chapter 2.82 The failure to include a priority rule applicable to acquisition security interests essentially requires such creditors to negotiate subordination agreements with general secured creditors who may have already taken a security interest in the same type of asset. This is particularly detrimental to sellers and financial lessors of equipment. In the absence of a secured transactions law that takes the functional approach, sellers/lessors would be able to rely on their ownership rights.83 The SIMP has significantly undermined the protections for sellers/lessors, who instead need to search the registry to determine whether there are any pre-existing registrations against the type of asset being sold or leased, then contact the person identified as the secured creditor in that registration and seek a subordination agreement. Naturally, the transactional costs will be passed on to the debtor. Article 12 provides for the priority of assignees, and contains ‘taking free’ rules.84 The sole reference to assignees in the title is inadequate because the article also includes the rights of buyers of goods, and a rule on the priority of a transferee of money. ‘Assignees’ thus appears to be used to cover any outright transferee, and only relates to the first provision. Under that provision, an ‘assignee’ takes collateral free of a security interest if it gives value and takes delivery of the collateral without knowledge of the security interest and before registration, which seems to imply that a transferee does take subject to an unregistered security interest if it has knowledge, does not give value or does not take delivery. The Model Law approach is that a security right that is not effective against third parties is, by definition, not effective against a transferee.85 A rule such as that in the first provision of Article 12 is therefore not necessary. The provision would be conceptually sound if Article 5 provided that on creation a security interest becomes effective against third parties.86 However, since the SIMP’s concept of creation is limited to the effectiveness of a security interest as between the secured creditor and the debtor, it makes no sense for a transferee (in some circumstances) to take subject to a security interest which is created but not registered, and, therefore, there is no need for a ‘taking free’ rule providing an exception to this conclusion. This conceptual disconnect is a reflection of a poor attempt to simplify and mesh the approaches of UCC Article 9 /PPSAs87 with the Model Law. Under the second provision of Article 12, a buyer or lessee of goods in the ordinary course of business takes free of a registered security interest even if it takes with knowledge of the interest. The standard for the protection of the buyer/lessee in the ordinary course of business, also differs from that of the Model Law, contained in Article 34(4)–(5). Under the SIMP the buyer/lessee takes free even if the buyer/lessee has knowledge that the security interest exists, while to take free under the Model Law, the buyer/lessee must take without knowledge that the sale/lease violates the rights of the secured creditor. This is a material difference; the Model Law is more protective of the secured creditor while the SIMP prioritises the free flow of goods in the ordinary course of business over protection of the secured 82 Ch 2, F2. 83 See ch 2, B2b. 84 See ch 2, F3. 85 Other PPSAs take a slightly different position; for example, under the Australian and New Zealand PPSAs a transferee not for value takes subject to a security interest, as does a transferee who was a party to the creation of the security interest. 86 See above C2. 87 Under these systems, an unperfected but attached security interest can bind a transferee in some circumstances.

Law on Security Interests in Movable Property 319 creditor from bad faith.88 Additionally, the Model Law clearly defines the point in time at which knowledge is material – the time of conclusion of the sale/lease agreement. In contrast, the SIMP is mute on this point. The fourth provision in Article 12 provides that a person who receives ‘cash for value’ takes free of a registered security interest. First, the defined term in Article 2(1) is money, but given the context of this provision ‘cash’ should be understood as ‘money’. Second, the rule requires the transferee to provide value. This standard is different from the standard incorporated in Article 48 of the Model Law that prescribes two conditions for the transferee of money to take free: (i) the transferee must take possession; and (ii) without knowledge that the transfer violates the rights of the secured creditor. Accordingly, while a donee of money can take free of a security right under the Model Law, as well as other modern principles,89 it will not be protected under the SIMP. Furthermore, the SIMP does not define value, thereby generating uncertainty for transferees of money, especially those who might have provided value in the past. The reason for this departure from the modern principles of secured transactions law is unknown. However, the more troublesome issue is the absence of a similar protection for transferees of funds from a bank account.90 These types of transfers are much more common than transfers of physical coins and notes. It is important that these transferees know, ex ante, the conditions for them to take free of a security interest in the bank account. Articles 13 and 14 govern the priority of persons with a right of retention and security interests in accessions. The Model Law does not contain a specific article dealing with rights of retention; instead, it addresses them under the general label of preferential claims in Article 36. Nor does it reflect recommendation 89 of the Legislative Guide that addresses the priority of security rights in attachments to movable assets (accessions).91 Article 15 deals with commingled goods, which are defined in Article 2(3).92 Although the article is placed in the chapter on priorities, it covers the creation, third-party effectiveness and priority of security interests in commingled goods. The priority rule raises some concerns as it states that: ‘The priority of the security interest in the product or mass shall be measured from the time of registration of the security interest in the collateral that became commingled.’ This rule applies well in a situation where two security interests exist over the same asset that subsequently becomes commingled into a mass or product. In other words, the commingling would not alter the respective priority of the two security interests. However, Article 15 contains no priority rule for the more common situation where different assets subject to different security interests are commingled (for example iron ore and chemicals). In this situation, the ‘first-to-perfect’ formula does not provide a fair and predictable result similar to that stated in Article 33(2) of the Model Law, whereby

88 The SIMP does not have a comparable provision to UCC 1-201(9), which stipulates that a buyer in the ordinary course of business is one who buys goods in good faith, without knowledge that the sale violates the rights of another person in the goods. 89 Under UCC, §9-332(a), a ‘transferee of money takes the money free of a security interest unless the transferee acts in collusion with the debtor in violating the rights of the secured party’. In contrast, the New Zealand PPSA provides that a donee of money takes free unless it has knowledge of the security interest. 90 See ch 2, F3 and ML, Art 47(6). 91 See conclusion C1b. 92 Commingled goods are defined as ‘goods that are physically united with other goods in a way that their identity is lost in a new product or mass’. Compare with the definition of ‘mass and product’ in ML, Art 2(s) and (cc).

320 Rwanda the secured creditors share in the mass or product according to the ratio that each secured obligation bears to the sum of all obligations secured by all security rights. Furthermore, Article 15 does not take into account the possibility that one of the security interests that continues in commingled goods may be of an acquisition nature. Article 42 of the Model Law provides that the acquisition security right has priority over a non-acquisition security right granted by the same grantor in the same mass or product. Article 16 addresses the priority of a security interest in fixtures.93 Article 16 first states that ‘a security interest may continue in goods that become fixtures’, which suggests that a security interest may not be created in an asset after it has become affixed to an immovable. In contrast, recommendation 21 of the Legislative Guide is clear in stating that ‘a security right may be created in a tangible asset that is an attachment at the time of creation of the security right or continues in a tangible asset that becomes an attachment subsequently’. Under Article 16 of the SIMP, a fixtures secured creditor will have priority over a competing claim to the immovable if the security interest has been registered in the Office of the Registrar General before the registration of the rights in the immovable property. This priority rule is contrary to recommendations 87 and 88 of the Legislative Guide, which require the security interest to be registered in the immovable property registry94 in order to have a similar priority. Under the Legislative Guide recommendations, if the secured creditor only registered a notice in the registry of security rights, even before a competing claimant made a registration in the immovable property registry, the secured creditor’s right would be subordinate to that of the competing claimant. Perhaps the provisions in the SIMP were the result of a conscious deviation from the best international standards in order to reflect the realities of the local environment. Article 17 includes the only rule in the entire SIMP that has some bearing on crossborder transactions. As noted above, and unlike Chapter VIII of the Model Law, the SIMP does not include a chapter on conflict of laws. However, the SIMP’s Article 17 attempts to replicate the approach underlying Article 23 of the Model Law, and applies to a situation where an asset subject to a security interest is relocated from a foreign country to Rwanda. The SIMP’s approach is limited to relocation of goods, whereas the Model Law’s approach is broader as it also applies to relocation of the grantor. Irrespective of the difference in approach, the significant errors in the SIMP’s Article 17 render it inoperable. The essence of such provisions is to provide a grace period to the secured creditor to satisfy the requirements of the law applicable to the security right after relocation of the goods or grantor. If the secured creditor satisfies the third-party effectiveness of the applicable law within that period of time, the third-party effectiveness would be continuous, that is, measured from when it was first established under the prior law. However, the SIMP simply provides that ‘security interest shall continue to be effective in Rwanda and the security interest may be registered in Rwanda’. Accordingly, one could argue that the ‘foreign security interest’ never loses its third-party effectiveness after relocation of the collateral to Rwanda, but the secured creditor could register a notice under the SIMP if it wished, on the grounds that, as a matter 93 See SIMP, Art 2(22), which defines fixtures as ‘goods that are annexed or are intended to become annexed to immovable property in a manner that causes a property right in the goods to arise in a person who holds an interest in the immovable property’. For further discussion of fixtures, see Conclusion C1b. 94 LG, recommendation 43 says that registration in the immovable property registry is an option for a security right in an attachment.

Law on Security Interests in Movable Property 321 of fact, it makes the security interest more public in Rwanda. Article 17 also provides that, unless provided for by other law, the effectiveness of a security interest is determined by Rwandan law if the collateral is located in Rwanda. The practical application of both parts of Article 17 remains uncertain in the absence of a more comprehensive conflict of laws rule that would determine the application of the SIMP in the first place. For instance, if under the general law, the location of the debtor determines whether the SIMP applies, relocation of the collateral to Rwanda may be immaterial. The last article in the chapter on priority covers security in crops. Neither the Legislative Guide nor the Model Law includes a comparable recommendation or rule. The first portion of the rule, which gives priority to a security interest registered in the registry over any right of the owner or mortgagee of the land to the extent the debtor is in possession of the land or has a right to the land, seems to have been inspired by UCC Article 9.95 Curiously, this provision refers to ‘registry’ rather than the term ‘Office of the Registrar General’ used in other articles, raising the question of what registry is being referred to. Given the inspiration for this rule and its context, the reference to ‘registry’ should be read to mean ‘Office of the Registrar General’. The second portion of this priority rule provides for ‘production money security interest’, giving priority to a secured creditor who advances credit to a debtor for the production or harvest of the crop. This rule was eliminated from the last revision of UCC Article 9, but remains a part of Canadian PPSAs.96

5. Enforcement of Security Interests Chapter V’s four articles set out a skeletal framework for extrajudicial enforcement of security interests. In a similar way to other chapters of the SIMP, the simplified approach does not provide solid legal basis for enforcement of security interests. Article 19 enables a senior secured creditor (a secured creditor ‘with priority’) to repossess the collateral upon default of the debtor. This approach is inconsistent with the Model Law, which allows any secured creditor, including a junior secured creditor, to enforce its rights against the collateral. By stripping it of the power to enforce, the SIMP’s approach diminishes the value of a junior security interest and increases the cost of junior secured lending. Furthermore, the meaning of ‘with priority’ is uncertain. Consider a situation with three secured creditors A, B and C whose priority is in that order. Could B enforce because it has priority as against C, or is the intention of Article 19 to empower only A, which has the highest priority, to enforce? The difficulties are further exacerbated by the second provision of Article 19 under which the secured creditor must request an authorisation from the Registrar General in the form of a certificate authorising repossession of the collateral. Enforcement is a matter in which the registrar should not be involved. It is difficult to discern the rationale for this requirement in Rwanda. Some other African laws also require active involvement of the registrar in the enforcement process.97 The procedures for the secured creditor to submit a

95 UCC,

§9-334(i). eg, Saskatchewan PPSA, s 34(11). 97 See conclusion D6b. 96 See,

322 Rwanda request and for the Registrar General to assess it are not set out in the SIMP.98 Under Article 20, the Registrar General is authorised to issue general instructions governing the possession, lease and sale of the collateral in public auction. It remains uncertain whether such general instructions have been issued. Nonetheless, the rationale for the role of the Registrar General in the enforcement process is difficult to discern.99 Article 21 provides for the right to redeem the collateral, which is extinguished on sale of the collateral or when the secured creditor is deemed to have taken the collateral in satisfaction of the obligation secured by it. The reference to the remedy of accepting collateral in satisfaction of the obligation is problematic because Chapter V neither provides for such remedy nor authorises use of remedies set out in the parties’ agreement, other than those in Chapter V. Furthermore, this is the only instance where the Act refers to this remedy, so that, unlike in the Model Law,100 the procedures for its application are not set out in the SIMP. The last article of this chapter provides for the distribution of proceeds from a sale of the collateral and for the rights of purchasers. Under paragraph 2, the proceeds must be distributed (after deduction of the costs of enforcement) first to the senior security interest, a natural consequence of the enforcement rule that allows only the senior secured creditor to enforce its rights. The requirements for a purchaser to take free of any security interests, namely that it must act in good faith, pay value and take possession of the collateral, differ from those of the Model Law.101 Chapter V has far too many gaps and internal inconsistencies to provide a solid framework for enforcing security interests. First, the chapter does not expressly allow the parties to use any remedies that they might have included in their security agreement. Second, in the context of the right of redemption, it refers to a point in time when the secured creditor is deemed to have taken the collateral in satisfaction of the obligation, but neither generally provides for the remedy of taking collateral for this purpose nor regulates the relevant procedures for its application. Third, the right to dispose of the collateral seems to be restricted to the sale (or lease) of the collateral, although Article 22 only provides for the disposition of the proceeds of, and the consequences of, sale and not of a lease. Disposing of the collateral by a (lease or) licence does not seem to be an option. Related to that, from the context of Article 20, one could argue that only public sales or leases of collateral are recognised under the SIMP. Finally, the role of the Registrar General in the enforcement process raises a number of concerns. The chapter fails to provide many provisions necessary to enforce security interests, including rules on notifying the debtor and competing claimants about a sale of the collateral. This inadequate regulation of enforcement rights contrasts

98 SIMP, Art 20 provides that the ‘security interest agreement shall contain a clause authorizing the creditor to seek repayment from collateral taking into consideration the nature of the security interest and the market price’. One could speculate whether the Registrar General, inter alia, must examine the security agreement to verify that this clause has been inserted. 99 Some guidance is given in the booklet ‘The Law on Security Interests in Movable Property’ (n 52), but the discussion in that booklet is not clear and further confuses the position 100 ML, Art 80. 101 Compare SIMP, Art 22, para 2 (where collateral is sold to a purchaser who acquires the interest for value and in good faith and who takes possession of it, the purchaser acquires the collateral free from the interests of the secured creditor, debtor and any subordinate secured creditor) with ML, Art 81(1)(3).

Law on Security Interests in Movable Property 323 with Chapter VII of the Model Law that distinctly spells out the extrajudicial remedial rights of a secured creditor and the manner in which these rights are to be exercised.102 The Finance Lease Law provides its own set of unique remedies. However, it does not specify whether these remedies supersede those of the SIMP when the security interest was taken in the nature of a finance lease, or whether the finance lessor has a choice of remedies. As with the SIMP, the Finance Lease Law requires the involvement of the Registrar General in the enforcement of the lease. Under Article 24, the lessor must send a notice of the intended repossession not only to the lessee, but also to the Registrar General. If the lessee objects to the repossession, the lessor must request the Registrar General to issue an authorisation to repossess the asset. Based on this authorisation, the lessor can then seek the assistance of a court bailiff. The Finance Lease Law goes further than the SIMP because authorisation by the Registrar General empowers the lessor to seek assistance from the bailiff, and therefore has significant legal effect. In other words, the authorisation has a quasi-judicial authority. The lessee can lodge an opposition to the Registrar General’s authorisation, but this, however, does not have suspensory effect. The lessee may file a petition with a court if it is not satisfied with the response of the Registrar General to the filed opposition.103

6. Register of Security Interests in Movable Property Chapter VI addresses various aspects of registration and is the most comprehensive chapter in the SIMP. Article 23 designates the Registrar General of Companies and Business Activities as the Registrar General of Security Interests under the SIMP. The registry itself is maintained in electronic form.104 Article 25 provides that ‘the secured creditor or his/her agent, or a lien holder shall register a security interest’. This language suggests that registration is a duty of these parties. This is inconsistent with the Model Law, which does not impose a duty on secured creditors to make their security rights effective against third parties; instead, it is a choice of the secured creditors to do this, which they can make based on the consequences of not doing this. It is also inconsistent with the SIMP itself, which in Article 8 also recognises possession and control as alternatives to registration. Article 25 goes on to provide the minimal mandatory information that must be included in a registration. Generally, the information required for a registration under the SIMP is consistent with the items of information required for the registration of a notice under the Model Law. However, as noted earlier,105 the registration must be accompanied by a notarised copy of the abstract of a security agreement as well as a payment receipt, which means that registration in advance of entering into the security agreement is impossible,106 and the system is one of document filing rather than notice filing.107 Furthermore, some of the details of 102 The remedies available to a secured creditor include: (1) possession (Art 77); (2) disposition of the collateral (Art 78); and (3) acquisition of the collateral (Art 80). See also LG, 290–98. 103 Finance Lease Law, Art 26. 104 SIMP, Art 24. For a description of the registration process, see B above. 105 See B above. 106 See ch 2, E1 where registration in advance is discussed. 107 ibid.

324 Rwanda Article 25 raise a number of questions. For instance, under paragraph 1, if the debtor is a natural person, the registration must include the person’s name, full identification, address and identification number. First, it is uncertain what name and identification number of the natural person must be entered, as natural persons, for instance, may have their names stated differently in various identification documents. They may also have been issued multiple identification numbers.108 Second, the reference to ‘full identification’ is unclear, and the article does not explain how this component relates to and differs from name and identification number. The 2009 Law, and its regulations,109 do not contain any such requirement. Notably, for debtors that are organisations, paragraph 2 of Article 25 does not require ‘full identification’. Unlike for natural persons, this item does provide more clarity by stating that an organisation must be identified by a name and the unique number assigned to it on its incorporation. Nonetheless, it remains unclear how other types of organisations (partnerships) that may be formed without a formal act of incorporation must be identified in a notice, which is a major gap. Finally, under Article 32, only the name(s) of the debtor is a search criterion, so the full identification and identification number may be relevant only when the search retrieves multiple results (eg when the name of a natural person debtor is common). As noted above, the SIMP treats finance leases as security interests. The relationship between security interests and finance leases has been muddled after the enactment of the Finance Lease Law. For instance, Article 14 of the Finance Lease Law calls for the registration of the actual agreement, a duty of the lessor which is reinforced in Article 16.110 This requirement raises a question whether the Finance Lease Law contemplates the establishment or designation of a different registry for the registration of finance lease agreements. Furthermore, Article 17 of the Finance Lease Law sets out two important consequences of the registration of a finance lease agreement. First, the lease is effective vis-a-vis the lessee and third parties, and, second, the lessor has the right to recover the leased object. Accordingly, the failure to register the finance lease agreement will render the rights of the lessor ineffective against the lessee, including the right to repossess the object upon default. Such consequences of (non)registration are contrary not only to the Model Law, but also to the SIMP.111 Although Article 25 of the SIMP is essential for the determination of the effectiveness of a registration, and thus the perfection of a security interest, it is plagued by ambiguities. For instance, paragraph 4 requires that the notice contain ‘the nature and description of the collateral’. The meaning of the word ‘nature’ remains unclear, leaving the door open to future challenges of collateral descriptions (eg ‘all inventory’) that would be sufficient under the Model Registry Provisions.112 Paragraph 6 also requires an indication of the maximum amount of the secured obligation, which is an optional feature of the Model Registry Provisions. Under the SIMP, the statement of the maximum amount does not serve the function envisaged in the Model Law, which is to facilitate subordinate lending.113 The sole purpose 108 See ch 2, Eii where the optimal grantor identifier is discussed. 109 Instructions of the Registrar General No 01/2009/Org of 24/06/2009 Relating to Modalities of Registration and Management of the Register of Security Interests in Movable Property. 110 Again, this indicates that there cannot be registration in advance and that the system is one of document filing. 111 See ch 2, C and D. 112 MRP, Art 11. 113 See ch 2, C1.

Law on Security Interests in Movable Property 325 of the requirement to enter maximum amounts is to collect statistical information.114 After its entry into the registry, it is removed from the other items of information and stored separately in a database that is not accessible to searchers. Paragraph 5 of Article 25 allows the registrant to select the period for which the notice will be effective. However, the flexibility to choose the appropriate period is limited by Article 27(1) under which the registration of a security interest remains effective until the expiration of the term specified in the contract. Accordingly, contrary to the Model Registry Provisions,115 the duration of the registered notice is tied to the duration of the security agreement. Article 27 is internally inconsistent because under paragraph 3 the registered notice is to remain effective until its expiration, even if the secured obligation has been extinguished earlier. In this case, the duration and effectiveness of a registered notice is separated from the existence of the underlying secured obligation. Unlike the Model Registry Provisions, Article 28 of the SIMP also specifically addresses the situation where a security interest is assigned, in which case it requires the name of the assignee to be ‘immediately recorded in the register’. This approach is contrary to the modern secured transactions laws (eg the PPSAs and UCC Article 9) where the registration of an amendment reflecting an assignment of a security interest is merely optional, and has no effect on the validity of the assignment.116 This approach reflects the notice-filing nature of the registry system, as a searcher who discovers a registration that names the assignor as the secured creditor would be alerted that an asset which the prospective debtor is proposing as security may already be encumbered. In contrast, the SIMP mandates immediate registration of an assignment. Since it does not stipulate the consequences of a failure to register an assignment or a delay in doing so, one could argue that the failure to register it immediately, or at all, will render the assignment ineffective against third parties or even between the assignor and assignee. It is illogical to provide a rule on such a specific point as the amendment of the registration to reflect an assignment of a security interest, but not to include rules that would determine the effectiveness of a registration, address the effect of errors in a registration or specify when a registration has been authorised.117 The absence of these rules does not provide the certainty and predictability that users of registry systems expect. The requirements to conduct a search of the registry present yet another deviation from the Model Registry Provisions as Article 32 requires that a person must introduce himself/ herself to the Office of Registrar General, and also contemplates that a search result will be given in the form of a printed copy. Accordingly, it appears that a searcher must visit a registry office in person to request a search. This is a highly unusual, ineffective and costly procedure, especially given the fact that the records of the registry are maintained electronically. It is possible that searches may be made electronically over the internet, but the website of the registry does not provide sufficient information.

114 Cross-reference to other chapters of the book. cf discussion in ch 2, C1 of inclusion of a maximum amount in the security agreement and the registered notice. 115 MRP, Art 14, which includes several options for enacting States but none tie the period of registration to the duration of the security agreement. 116 UCC, §9-514(b). See ch 2, E1: the Model Registry Provisions merely enable an amendment notice to be registered reflecting the name of the new secured creditor (see Art 18 for global amendments). 117 See ch 2, E.

326 Rwanda Moreover, Article 15 of the Finance Lease Law requires the searcher to ‘sign a d ocument stating his/her commitment to observe confidentiality of data’. This requirement again raises the question whether the SIMP registry is the one for the registration of finance lease agreements. Article 32 of the SIMP sets out the search criteria, which are: (i) the names of the debtor, (ii) the unique number assigned to the collateral and (iii) the registration of the security interest. The names (in plural) of the debtor search criterion presents several problems and makes the user wonder which name(s) to search against. Article 25, which sets out the required data to be entered in a notice, requires the name of the debtor (in the singular). For the second search criterion (the unique number assigned to the collateral) the SIMP does not provide any guidance whatsoever. One could speculate that the criterion might be the serial number assigned to some assets by their manufacturers (eg vehicle identification numbers). However, the SIMP neither requires the identification of any assets by a number nor provides the consequences for a failure to identify that number accurately. From the context, it appears that a collateral-number search is merely a service to the users, allowing searches for those registrations where the registrants have voluntarily entered some collateral identification number. Rwanda’s policymakers and others who might be involved in reforming the SIMP will find useful guidance in the UNCITRAL Guide on the Implementation of a Security Rights Registry (2013).118 Articles 33 and 34 provide for the power of the Registrar General to set the fees and issue regulations.119 This allocation of such powers to the Registrar General is directly contrary to the approach of the Model Registry Provisions,120 which emphasise that the agency possessing such powers should be distinct from the agency that operates the registry. This is because the interests of the agency that operates the registry will often be contrary to the interests and needs of users. For instance, the operator may want to ensure that the registration fees cover not only the operational costs, but also other non-registry-related expenses. In contrast, the users will expect efficient access to the registry services at a low cost. The Rwandan model presents a number of risks and uncertainty.

D. Final Remarks Rwanda has been heralded as the top law reformer in Africa, including in the area of secured transactions.121 It has, indeed, enacted a new secured transactions law and established a collateral registry that, on their surface, reflect the best international standards. However, examination of the actual provisions and contrasting them with the Model Law reveals a deeply flawed secured transactions system that is rife with legal risks, unpredictability

118 ‘UNCITRAL Guide on the Implementation of a Security Rights Registry’ (Vienna, United Nations, 2014) (Registry Guide) 51–53. 119 SIMP, Art 33 states that ‘Fees for registration and searches shall be determined by instructions of the Registrar General’; whereas Art 34 states that the ‘Registrar General shall determine the modalities governing registration and the keeping of the register of security interests’. 120 MRP, Arts 27 and 33. 121 See IFC, ‘Rwanda Top Business Reformer’, www.ifc.org/wps/wcm/connect/news_ext_content/ifc_external_ corporate_site/news+and+events/news/rwanda_doing_business_report.

Final Remarks 327 and ambiguities. The enactment of the Finance Lease Law introduced further complexity and internal disharmony, distancing Rwanda’s secured lending framework from the best international standards. The SIMP exemplifies the many deficiencies of the ‘simplified models’. This chapter demonstrates that Rwanda would have been better off following the Legislative Guide or other international models. Given the many deficiencies of the SIMP, Rwanda should repeal it and replace it with a statute based on the Model Law. The public announcements touting Rwanda as the top African reformer and its high ranking in the World Bank’s Doing Business Reports, however, may mean that there is little incentive for the policymakers to reconsider the SIMP’s approach. Rwanda also presents a risk to future reforms in Africa, where countries may be led to believe that the SIMP is a model to be followed.122

122 See Sierra Leone C and C1, which discusses the modelling of the Sierra Leone reformed law on the Ghanaian law.

15 South Africa A. Introduction South Africa is an upper-middle-income economy of 55.4 million people.1 According to the 2018 South Africa Economic Update of the World Bank, the poverty rate increased from 16.5 per cent in 2010 to 18.6 per cent in 2017.2 High unemployment, slow economic growth since the 2008 financial crisis and weak enforcement of economic policies are the primary reasons for the rising poverty rate.3 Growth has contracted considerably in 2019: GDP fell 3.2 per cent in the first quarter of 2019.4 Services and industry make up 97 per cent of South Africa’s production (services account for 67.5 per cent, industry for 29.7 per cent and agriculture for only 2.8 per cent).5 These industries also provide virtually all employment.6 Mining is the single largest industry in South Africa, which is the world’s largest producer of platinum, gold and chromium.7 Besides mining, South Africa’s most important industries are automobile manufacturing, metalworking, machinery, textiles, iron and steel, chemicals, fertiliser, foodstuffs and commercial ship repair.8 Mining is by far the largest contributor to South Africa’s exports.9 Gold alone accounts for 20 per cent (US$20.7 billion) of all exports and diamonds contribute another 10 per cent (US$10.9 billion).10 The South African Reserve Bank oversees the banking system, which comprises local and foreign banks (see Table 15.1).11

1 CIA, ‘People and Society: South Africa’ (The World Factbook, 5 September 2018) https://www.cia.gov/library/ publications/the-world-factbook/geos/sf.html. 2 South Africa Economic Update: Jobs and Inequality (Washington, DC, World Bank, April 2018). 3 L Chutel, ‘Post-Apartheid South Africa Is Failing the Very People it Liberated’ Quartz Africa (24 August 2017) www.qz.com/africa/1061461/post-apartheid-south-africa-is-failing-the-very-people-it-liberated/. 4 ‘South Africa GDP Growth Rate’ Trading Economics (2018) www.tradingeconomics.com/south-africa/ gdp-growth. 5 CIA (n 1). 6 ibid. 7 ibid. 8 ibid. 9 Observatory of Economic Complexity, ‘South Africa’ (2018) https://atlas.media.mit.edu/en/profile/country/ zaf/. 10 ibid. 11 South African Reserve Bank Prudential Authority, ‘South African Registered Banks and Representative Offices’ (2019) www.resbank.co.za/PrudentialAuthority/Deposit-takers/Banks/Pages/South-African-RegisteredBanks-and-Representative-Offices.aspx.

Introduction 329 Table 15.1 Banks operating in South Africa Type of bank Locally controlled banks Foreign-controlled banks

Number 11 7

Foreign bank representatives

29

Branches of foreign banks

15

Mutual banks

4

The benchmark repo rate of the South African Reserve bank is currently 6.75 per cent.12 Inflation rose to 5.1 per cent in July 2018, a steep increase from the five-year low of 3.8 per cent recorded in March of 2018, before dropping to 4.0 per cent in February 2019.13 SMEs make up 91 per cent of businesses in South Africa, provide employment for 60 per cent of the labour force and account for 34 per cent of the country’s economic output.14 South Africa also has one of the world’s highest SME failure rates – about 75 per cent of SMEs fail within the first two years.15 Securing financing is one of the most significant hurdles for South African SMEs as banks reject a high proportion of finance applications.16 In 2018, slightly more than half of all credit applications were rejected.17 In 2015, loans to SMEs accounted for 26 per cent of total business lending.18 The most common problems for SMEs seeking financing include lack of acceptable collateral, lack of financial records and business plans, and a high SME failure rate.19 These constraints are compounded by weak judicial and legal frameworks supporting the most common forms of SME lending. A survey of financial institutions conducted by the Banking Association of South Africa (BASA) in 2009 showed that one of the primary factors in banks’ unwillingness to finance SMEs was the failure of SMEs to demonstrate that they met the bank’s evaluation criteria.20 In light of this, BASA recommended implementing programmes to help SMEs evaluate their financial status and conducting risk assessments in the prefinancing stages to ultimately ease the process of applying for and accessing credit.

12 ‘South Africa Interest Rate’ Trading Economics (March 2019) www.tradingeconomics.com/south-africa/ interest-rate. 13 ‘South Africa Inflation Rate’ Trading Economics (March 2019) www.tradingeconomics.com/south-africa/ inflation-cpi. 14 Banking Association of South Africa, ‘Small & Medium Enterprise’, www.banking.org.za/what-we-do/sme. 15 AT Mutezo, ‘Small and Medium Enterprise Financing and Credit Rationing: The Role of Banks in South Africa’ (University of South Africa, June 2015) http://uir.unisa.ac.za/bitstream/handle/10500/23479/Thesis_Mutezo_AT. pdf?sequence=1&isAllowed=y. 16 A study in 2009 found that two-thirds of applications were rejected. See M Herrington et al, ‘Global Entrepreneurship Monitor – South African Executive Report’ SSRN (24 November 2009) https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=1509253. 17 National Credit Regulator, ‘Consumer Credit Market Report: Third Quarter/September 2018’ (2018) 4, www. ncr.org.za/documents/CCMR/CCMR%202018Q3.pdf (‘The rejection rate increased from 50.06% for the quarter ended June 2018 to 53.65% for the quarter ended September 2018’). 18 S Timm, ‘Just 26% of Business Loans by SA Banks go to SMEs Reveals OECD Survey’ VentureBurn (25 July 2017) www.ventureburn.com/2017/07/just-26-business-loans-sa-banks-go-smes-reveals-oecd-survey/. 19 Mutezo (n 15). 20 Banking Association of South Africa, ‘Financial Institutions’ Hurdles to SME Financing’, www.banking.org.za/ what-we-do/sme/financial-institutions-hurdles-to-sme-financing.

330 South Africa It also suggested that banks need to develop more appropriate ways of evaluating risk, particularly in relation to collateral, and also to develop more suitable products for SMEs.21

B. Background of the Legal Framework South Africa’s legal system is rooted in Roman-Dutch law introduced by Dutch settlers. This law was adapted by the South African courts and legislature, who often drew on English law, especially during the nineteenth century and after 1909 when the Union of South Africa was created.22 This regime is subject to South Africa’s constitution, which was born of the struggle for independence and the end of apartheid.23 The 1996 Constitution goes to great lengths to spell out the rights it guarantees, including many that the apartheid regime denied. Foremost among those that relate to some aspects of secured transactions law are the right to a free and fair hearing before a judge and the prohibition of the taking of property without equitable compensation. To appreciate the weight and political significance of these constitutional guarantees, one need only look to the apartheid regime’s policies of forcible dispossession and evictions of black people from land designated for ‘whites only’.24 The constitutional protections present potential barriers to secured lending and contingent economic empowerment. The difficulty of enforcing security rights, judicially or otherwise, presents a serious impediment and disincentive to secured lending. Striking a responsible balance that upholds constitutional protections while serving the realities of modern commerce is essential to ensuring access to commercial credit. Neighbouring Zimbabwe has a colonial history similar to that of South Africa and many of its laws are rooted in South African law.25 Facing similar obstacles to secured lending, Zimbabwe reformed its secured transactions laws, but took a relatively conservative approach to enforcement of security rights.26 Some of the approaches pursued in Zimbabwe could, therefore, inform the direction of a future reform in South Africa. Extrajudicial enforcement of security rights in South Africa has been mired in controversy.27 The inherent conflict between the prohibition of self-help and the principle of freedom of contract is embodied in the history of Roman-Dutch law concerning the enforcement of mortgages and pledges under pactum commissorium clauses and, more controversially, parate executie clauses.28 In Bock v Duburoro Investments (Pty) Ltd, the Supreme Court of Appeal, in 2003, considered this controversy, and said that ‘our common law has always recognised that self-help is unlawful’. The court did, however, then draw a

21 ibid. 22 OD Schreiner, The Contribution of English Law to South African Law; and the Rule of Law in South Africa (London, Stevens & Sons Limited, 1967) 5, 10–11, 27, 33, 41, 43–44. 23 The current Constitution replaced the 1994 Interim Constitution in 1996 following a two-year drafting and approval process. 24 Schreiner (n 22) 92, 94; and see generally Group Areas Act No 41 of 1950. 25 Schreiner (n 22) 7. 26 See Zimbabwe C. 27 World Bank Group, ‘Secured Transactions Collateral Registries: Diagnostic for the Republic of South Africa’ (2015) 39 (on file with the authors). 28 LE Krause, ‘The History of Parate Executie’ (1924) 41 South African Law Journal 20.

Background of the Legal Framework 331 distinction between where the secured creditor is in possession of the collateral and where it is not.29 A parate executie clause is valid in the former situation. A pactum commissorium is a clause that allows a mortgagee to retain the mortgaged property upon the mortgagor’s default. It was first made illegal by Emperor Constantine in 326 CE. In Roman-Dutch law, as incorporated into South African law, a pactum commissorium agreement was also void,30 and this conclusion was reaffirmed by the Supreme Court of Appeal in Bock v Duburoro.31 The Supreme Court of Appeal explained, in contrast to a pactum commissorium clause, that a clause is valid which authorises the creditor to keep pledged property upon the grantor’s default, on payment of a price,32 provided that the price is fair, and determined after the grantor’s default and not when the agreement was entered into.33 The parties can agree that such a determination shall be, for example, by publicly listed market price, or expert appraisal.34 To the extent that the determined price exceeds the secured obligation, the balance belongs to the pledgor.35 Such agreements, known as quasi-conditional clauses,36 can be included in the original agreement or can be added at any time thereafter.37 The grantor has recourse to the courts in a dispute over whether the value determined is fair.38 Like these quasi-conditional clauses, parate executie clauses authorise a pledgee of tangible assets to sell the pledged assets extrajudicially.39 The Supreme Court of Appeal made it clear that such clauses are valid, but only in the case of a pledge of tangible assets, when the pledgee will already be in possession of the assets: there is no right to seize assets extrajudicially which are in the possession of the grantor.40 In exercising this right to sell, the pledgee must not prejudice the position of the pledgor: if it does so, the pledgor can seek the protection of the court.41 Thus, the pledgee is under a duty to ‘take all reasonable steps to obtain a fair price’:42 this duty can be conceptualised as arising from a relationship of agency, so that the pledgee is selling as agent for the pledgor. This latter formulation underpinned the reasoning of Sansole J in the Zimbabwe High Court in the case of Sakala v Wamambo,43 where he took the view that enforcing a parate executie clause ‘is tantamount to taking the law into one’s own hands’. Consequently such a provision in an agreement should be construed strictly’.44 As mentioned above, a parate executie provision is only valid 29 Bock v Duburoro Investments (Pty) Ltd 2004 (2) SA 242 (SCA) [14]. 30 ibid 20–21; R Brits, Real Security Law (Cape Town, Juta, 2016) 164–66; Krause (n 28) 20. 31 Bock v Duburoro (n 29) [8], citing Graf v Buechel 2003 (4) SA 378 (SCA) [9]–[11]. 32 Bock v Duburoro (n 29) [9], citing Graf v Buechel 2003 (4) SA 378 (SCA) [27] et seq. 33 Graf v Buechel 2003 (4) SA 378 (SCA) [29]; Brits (n 30) 166. 34 Brits (n 30) 166. 35 Graf v Buechel 2003 (4) SA 378 (SCA) [29]. 36 Brits (n 30) 166. 37 Krause (n 28) 20. 38 Brits (n 30) 168. 39 Bock and Others v Duburoro Investments (Pty) Ltd 2004 (2) SA 242 (SCA) [6], [14]; Krause (n 28) 21; Brits (n 30) 170. 40 Bock and Others v Duburoro Investments (Pty) Ltd 2004 (2) SA 242 (SCA) [13]–[15]; Nino Bonino v De Lange 1906 TS 120. Note that, in relation to consumer secured borrowing, this position is governed by the National Credit Act 2005, which appears to permit paratie executie clauses in pawn transactions, but not in other secured transactions. See R Brits, ‘Pledge of Movables Under the National Credit Act: Secured Loans, Pawn Transactions and Summary Execution Clauses’ (2013) 25 South African Mercantile Law Journal 555. 41 Osry v Hirsch, Loubser & Co 1922 CPD 531, 547. 42 See the Zimbabwean case, Change v Standard Finance Limited 1990 (2) ZLR 412 (SC). 43 Sakala v Wamambo 1991 (4) SA 144 (ZH). 44 ibid 149.

332 South Africa if the pledgee already has possession. South African law does not permit a creditor to use a self-help remedy to gain possession against the wishes of the grantor45 and contractual provisions which permit this will not be valid.46 When a grantor voluntarily complies with a request from a creditor to surrender collateral, the creditor must sell the collateral at a fair price. However, when a grantor is unwilling to surrender tangible collateral, a court order is required if the grantor opposes repossession.47 In the case of intangible property, when the grantor has already transferred legal control to the creditor by means of cession, the security agreement may authorise, upon the grantor’s default, that the creditor may retain intangible assets as security (eg freezing withdrawals from a bank account up to the amount claimed to be due from the grantor),48 or, for example in the case of receivables, that the creditor may directly enforce a right, without application to the court.49

C. Security Rights in Movable Assets In the World Bank’s Doing Business Report 2019, South Africa ranked 73rd out of 190 economies for ease of getting credit, with a score unchanged from 2018. South Africa performs well in the report’s credit information metrics, ranking slightly above the average score for OECD high-income economies and well above the average for sub-Saharan economies.50 This result is largely due to South Africa’s publicly accessible credit bureau that collects data for loans regardless of loan amount and covers more than 67 per cent of individual borrowers.51 However, South Africa ranks slightly below the sub-Saharan average in the report’s Strength of Legal Rights Index, which measures the degree to which collateral and bankruptcy laws protect the rights of borrowers and lenders.52 Under South African law, there are four devices which may be used to secure an obligation with a right in a movable asset under which the grantor remains the owner of the collateral: (i) pledge, (ii) cession (assignment), (iii) general notarial bond and (iv) special notarial bond.53 The oldest form of security right in South Africa is the pledge, which is based on the Roman law pignus and which requires the secured creditor to have possession of the

45 Nino Bonino v De Lange 1906 TS 120; First National Bank of South Africa Ltd v Land and Agricultural Bank and Others 2000 (6) BCLR 586 (O) 590. 46 Nino Bonino v De Lange 1906 TS 120. 47 This is also the case where a creditor wishes to perfect a general notarial bond; see C4d below. 48 Arctocel (Pty) Ltd v Firstrand Bank Ltd (32633/2015) [2015] ZAGPJHC 244 (21 October 2015). 49 World Bank (n 27) 41; Brits (n 30) 331, 174 fn 361. See discussion in C3e below. 50 South Africa scored 7 out of a possible 8 on the report’s Depth of Credit Information Index, compared to 3.3 for sub-Saharan economies and 6.7 for high-income OECD economies, with credit bureau coverage of 67.3 per cent of adults, compared to 8.9 per cent for sub-Saharan countries and 65.3 per cent for high-income OECD economies. See World Bank, ‘Ease of Doing Business in South Africa’, www.doingbusiness.org/en/data/exploreeconomies/ south-africa#DB_gc. 51 ibid. 52 ibid. South Africa scored 5 out of a possible 12 on the report’s Legal Rights Index, compared to 5.2 for subSaharan economies and 6.1 for high-income OECD economies. See ch 1, Table 1.1. 53 U Naumann, ‘Lending and Taking Security in South Africa: Overview’ Thomson-Reuters (2017) https://content. next.westlaw.com/Document/Id4aecad21cb511e38578f7ccc38dcbee/View/FullText.html?contextData=(sc.Defaul t)&transitionType=Default&firstPage=true&bhcp=1.

Security Rights in Movable Assets 333 pledged assets.54 Despite this requirement, which might be thought to limit the use of the pledge to tangible assets, it is possible to take a pledge over intangibles, by means of cession.55 This type of device can also be called a ‘security cession’.56 Cessions are effective when executed, and without any form of a public notice. There are also two non-possessory security devices, a general notarial bond and a special notarial bond,57 both of which are required to be registered in a local deeds registry.58 A special notarial bond, which is a security right created by statute,59 can only be granted over identified tangible assets,60 and is deemed to be a pledge despite its non-possessory nature.61 A general notarial bond, which is a common law security right, can be granted over assets of the grantor described generally, and can include future assets.62 It does not confer on the bondholder a real security right; instead the bondholder merely has a preference over unsecured creditors in insolvency.63 The grantor can dispose of its assets freely until its insolvency,64 or until the secured creditor takes possession of assets and therefore becomes a pledgee.65 The general notarial bond has some similarities to the floating charge in English law,66 although there are some differences.67 A single instrument may create a general as well as special notarial bond as long as they are clearly identified as such and distinguishable from one another.68 A number of specialised registries have been established for the recording of property rights to some assets, including aircraft, intellectual property and ships. The relevant laws provide for the registration of mortgages in the specialised registries covering those assets.69 Registered intellectual property rights, such as trade marks, patents and designs, may be given in security by way of hypothecation registered with the Companies and Intellectual Property Commission,70 but also by cession.71 For some intellectual property rights, such as unregistered copyright, cession is the only available security mechanism.72

54 See C2 below. 55 National Bank of SA Ltd v Cohen’s Trustee 1911 AD 235. 56 See S Scott, ‘One Hundred Years of Security Cession’ (2013) 25 South African Mercantile Law Journal 513 and C3 below. 57 Deeds Registries Act 47 of 1937, s 102. 58 ibid ss 61 and 62. 59 Security By Means of Movable Property Act 57 of 1993, s 1. 60 ibid s 1(1); Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA). 61 Security by Means of Movable Property Act 57 of 1993, s 1. This deeming provision is conceptually and practically problematic. R Brits, ‘Two Decades of Special Notarial Bonds in Terms of the Security by Means of Movable Property Act’ (2015) 27 South African Mercantile Law Journal 246, 252–61 and see C5 below. 62 See C4 below. 63 Insolvency Act 24 of 1936, s 102. 64 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [3]. 65 Firstrand Bank Ltd v Land And Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA) [4], the secured creditor can only take possession if the bond includes a clause permitting it to do so. 66 See ch 1, E1b. 67 N Locke, ‘Security Granted by a Company over its Movable Property: The Floating Charge and the General Notarial Bond Compared’ (2008) 41 Comparative and International Law Journal of Southern Africa 136, 137–38. 68 Reeskens v Registrar of Deeds 1964 (4) SA 369 (N) 372. 69 Brits (n 30) 266. 70 See Trade Marks Act 194 of 1993, s 41 (which requires the lodging of a deed of security); Patents Act 57 of 1978, s 60; Designs Act 195 of 1993, s 30. In relation to the two latter provisions, no deed of security if required to be lodged. S. Karjiker, ‘Intellectual Property as Real Security’ (2018) South African Intellectual Property Law Journal 1, 8–10. 71 ibid 8. 72 ibid 15.

334 South Africa

1. Reservation of Title Devices In addition to the security devices discussed above, creditors regularly deploy devices involving the reservation or acquisition of ownership in order to secure payment (in the widest sense), such as when an asset is sold on instalments or where a supplier sells an asset to a lender that in turn sells it subject to reservation of ownership to the buyer.73 There was little recognition of such a device, or its use as security, until the term ‘title interest’ was included in section 134 of the Companies Act 71 of 2008. Chapter 6 of this Act regulates business rescue proceedings of corporations, and section 134 provides that assets subject to either ‘security or title interest’ may only be disposed of with the prior consent of the creditor unless the proceeds of the disposition are sufficient to fully discharge the obligation owed to the creditor. The phrase ‘title interest’ was considered in 2017 by the court in the Gauteng Local Division, Johannesburg, in the case of Energydrive Systems (Pty) Ltd v Tin Can Man (Pty) Ltd.74 The case concerned the sale of equipment leased by the company being rescued. The lessor claimed the equipment from the buyer, but the court (noting that the term ‘title interest’ is novel in South African law) held that the term ‘title interest’ included a lease (including a reservation of ownership clause). The sale by the business rescue practitioner was permitted by section 134, but only if the proceeds of sale were used to pay the lessor; until this happened the lessor retained its ownership rights. The arrangement was, therefore, treated as a reservation of title for the purposes of security. The Insolvency Act 24 of 1936, which applies in the insolvency of natural persons, also deals with the effect of a reservation of ownership provided for in an instalment agreement.75 The effect of insolvency is the loss of the seller’s ownership, which passes to the insolvent buyer’s estate. Under section 84(1), the seller acquires a hypothec securing the amount owed by the buyer. Reservation of ownership protects the creditor when the asset is not expected to be sold. Although this security device may be deployed to finance assets that constitute the buyer’s inventory, there is no rule in South African law that automatically transfers the right of the creditor in an item of sold inventory to the proceeds.76 The creditor should thus take a separate security over any assets resulting from the sale or lease of the inventory, including cash, but also used vehicles traded in by buyers. For the latter type of assets, the grantor77 may be required to sell the trade-in to the creditor and the creditor will then sell it back to the grantor under a reservation of ownership.78 South African courts have struck down these arrangements on the basis that they attempted to simulate a pledge, but the requirements to make the pledge effective against third parties had not been fulfilled.79 In contrast, an arrangement where a creditor buys the inventory from the manufacturer and then sells it to the grantor on a reservation of ownership has been recognised as valid.80 73 Energydrive Systems (Pty) Ltd v Tin Can Man (Pty) Ltd 2017 (3) SA 539 (GJ) [16]; CG van der Merwe and C Smith, ‘Financing the Purchase of Stock by the Transfer of Ownership as Security: A Simulated Transaction?’ (1999) 10 Stellenbosch Law Review 303, 320. 74 Energydrive Systems (Pty) Ltd v Tin Can Man (Pty) Ltd 2017 (3) SA 539 (GJ). 75 Insolvency Act No 24 of 1936. 76 Van der Merwe and Smith (n 73) 304. 77 The term ‘grantor’ is used in this section in the interests of consistency with other chapters even though the proprietary interest is created by reservation and not grant. 78 ibid 305. 79 Vasco Dry Cleaners v Twycross 1979 (1) Sa 603; Nedcor Bank Ltd v Absa Bank Ltd 1998 (2) SA 830 (W). 80 Roshcon (Pty) Ltd v Anchor Auto Body Builders Cc 2014 (4) SA 319 (SCA) [36].

Security Rights in Movable Assets 335 Another option is to monitor the grantor’s business activity closely to ensure that the sale proceeds are in fact applied to reduce the debt owed.81 This may be achieved by requesting the grantor to deposit the cash proceeds into a specific bank account controlled by the creditor. The absence of a security right over fluctuating assets,82 especially inventory, complicates this type of financing and unduly increases the cost. The incorporation of Article 10 of the Model Law that extends the security right into proceeds would obviate the need for the techniques discussed above, but this is balanced by Article 41 which does not continue the super-priority of acquisition security rights as against security rights in receivables that were earlier made effective against third parties.

2. Pledge a. Nature of the Pledge It is well established that possessory pledges of tangible assets are not well suited for financing the needs of businesses. Nevertheless, the general principles of pledge, which are well established in South African law, remain important in that security cession is seen as a pledge,83 and special notarial bonds are deemed to be pledges.84 In South African law, the pledgor must transfer possession or legal control of movable assets, tangible or intangible, to the secured creditor (pledgee).85 While it is sufficient for the pledgee to have constructive possession, where the actual possessor, who holds the assets for the pledgee, is a third party,86 a pledge is not created where the pledgor holds the assets for the pledgee.87 The policy basis for this rule is the lack of publicity that this type of constructive possession gives to the pledge.88 The rule has been applied in a number of cases where the parties have used a simulated sale and buyback, in which a creditor extends a loan to the grantor who retains possession of the asset purportedly sold to the creditor for the amount corresponding to the loan, to achieve the effect of a non-possessory pledge. The courts have generally held that such transactions are disguised pledges, and, because the pledgor did not transfer possession or control of the asset to the pledgee, are ineffective against third parties.89

b. Creation As between the parties, the pledge can be created by mere agreement, but this will not be binding on any third parties, including unsecured creditors of the grantor, and merely 81 Van der Merwe and Smith (n 73) 304. 82 See ch 1, E4b. 83 National Bank of SA Ltd v Cohen’s Trustee 1911 AD 235. 84 Security by Means of Movable Property Act 57 of 1993, s 1(1). 85 Brits (n 30) 107. 86 Vasco Dry Cleaners v Twycross 1979 (1) Sa 603. 612 (A); Airkel (Edms) Bpk H/A Merkel Motors v Bodenstein En’ N Ander 1980 (3) SA 917 (A). See C2b below. 87 Lighter v Edwards 1907 TS 442,445; Goldinger’s Trustee v Whitelaw & Son 1917 AD 66, 89; Vasco Dry Cleaners v Twycross 1979 (1) Sa 603. 612 (A). 88 Goldinger’s Trustee v Whitelaw & Son 1917 AD 66. 89 Hofmeyer v Gous (1893) 10 SC 115, 117; Goldinger’s Trustee v Whitelaw & Son 1917 AD 66, 79; Vasco Dry Cleaners v Twycross 1979 (1) SA 603 (A); Roshcon (Pty) Ltd v Anchor Auto Body Builders Cc 2014 (4) SA 319 (SCA).

336 South Africa gives the pledgee a right to take action to obtain possession.90 Delivery (actual or constructive)91 is required for the pledge to become effective against third parties.92 Constructive possession will suffice, as will rights short of physical possession which result in the pledgee controlling the assets.93 Thus, when the pledge is of business assets, the pledgee can employ an agent to control the business or security guards to guard the building where the assets are located.94 Constructive delivery can take a number of forms. One is symbolic delivery, for example, the transfer of an instrument that gives the pledgee exclusive control over the pledged asset, such as a silo (warehouse) receipt or bill of lading, but also the keys to the building in which the assets are located, car keys, and so on.95 Another form is attornment, which takes place when a third party (eg a storage facility), acting as an agent of the pledgor, has control of assets, and undertakes to hold the assets on behalf of the pledgee.96 Delivery takes place when the third party becomes an agent of the pledgee.97 In Sterling NO v Landau,98 the court held that when an agent maintains multiple identical assets on behalf of a pledgor who wishes to pledge some, but not all, of them, the pledged assets must be separated for the pledge to be valid.99 A pledge may secure present or future obligations.100 In addition to the pledgor’s primary obligation, the pledge covers the pledgee’s costs of caring for and maintaining the collateral, as well as any enforcement costs.101 The pledgee has a duty to exercise due diligence, to take proper care of the collateral and protect the interests of the pledgor.102 Pledged objects can be individual, composite or collective. Composite objects are comprised of various individual items that have been combined, such as components of an automobile.103 Pledges over collective objects, such as a herd of cows or a business’s inventory, remain valid even though the individual objects fluctuate over time.104 A pledge extends to the fruits of the pledged object, and the pledgee must account for the fruits and return them to the pledgor, along with the pledged asset, when the pledge is extinguished.105 Pledges over future assets are valid, but they only become a real right when the pledgor 90 Stratford’s Trustees v London and SA Bank (1874) 3 EDC 439, 449; G Wille, Mortgage and Pledge in South Africa (Johannesburg, Hortors, 1920) 96. 91 As mentioned above, constructive delivery where the pledgor holds for the pledgee will not be sufficient to create a pledge; see Lighter v Edwards 1907 TS 442, 445. 92 ibid 122–23; Edwards v Van Zyl 1951 (2) SA 93, 98–99 (C); Danzas Trek (Pty) Ltd v Du Bourg 1979 (4) SA 915, 921 (W); Naumann, (n 53) 4; World Bank (n 27) 23. 93 In Pieters & Co v Landau Bros & the Trustees of the Insolvent Estate of I & J Goldberg 1914 SR 30, the Court held that possession of pledged assets by the pledgee’s lessee, although not technically an agent, gave the pledgee sufficient control. In Phillips v Soqaka 1915 EDL 37, the pledgee lent pledged cattle to his servant for the servant to plough his own field, but the court held that the pledgee retained sufficient legal control. 94 Brits (n 30) 124. 95 Heydenrych v Saber (1900) 17 SC 73; S v Buitendag 1980 (2) SA 152, 154 (T); Brits (n 30) 124–25. 96 Brits (n 30) 124–25; Air-Kel (Edms) Bpk H/A Merkel Motors v Bodenstein 1980 (3) SA 917 (A). 97 Brits (n 30) 128–29. 98 Sterling NO v Landau 1921 WLD 117. 99 Brits (n 30) 129–30. 100 Netherlands Bank of South Africa v Stern No 1955 (1) SA 667 (W); Brits (n 30) 117. 101 Brits (n 30) 115. 102 Simon v Mitsui and Co Ltd 1997 (2) SA 475, 497 (W); Brits (n 30) 107, 113, 145–47. 103 See CG van der Merwe and MJ De Waal, The Law of Things and Servitudes (Durban, Butterworths, 1993) 21. 104 ibid. 105 Freeman Cohen’s Consolidated Ltd v General Mining and Finance Corporation Ltd 1907 TS 224, 226; Judes v SA Breweries Ltd 1922 WLD 1, 8; Brits (n 30) 112–14, 145, 149.

Security Rights in Movable Assets 337 acquires rights to the future asset and the asset is subjected to the pledgee’s control.106 That a security right can extend to future assets and cover future obligations are essential principles of the Model Law.107

c. Extinguishment and Enforcement Satisfaction of the secured obligation extinguishes the right of pledge and the pledgee has a duty to return the property to the pledgor: this comes from the accessory nature of the pledge.108 It also has a duty to account to the pledgor, for example, for expenses claimed in relation to care of the pledged assets or enforcement, and for the fruits.109 However, because a pledge is indivisible, when more than one object is pledged, or when multiple pledgors are involved, the pledgee is not required to return any of the collateral until the obligation is fully discharged.110 The pledge ceases to exist if the object is destroyed.111 It is also extinguished if the collateral becomes permanently attached to land in such a manner that it loses its identity as a movable asset.112 South African courts apply a three-pronged test to determine whether a movable asset becomes part of the land: first, the nature and purpose of the movable asset; secondly, the manner and degree of annexation; and, thirdly, the intention of the owner of the movable asset or the person who attached the movable asset to the land.113 A pledge is enforced by the sale of the pledged assets by the pledgee and the application of the proceeds to the satisfaction of the secured obligation. The pledgor is entitled to the surplus and remains responsible for any deficiency.114 As explained above,115 a pledge can only be enforced by the pledgee pursuant to a court order, unless the pledge agreement contains a paratie executie (summary execution) clause. The duties of the pledgee when selling the goods are described above.116 If the pledged asset is seized and sold by a judgment creditor, the pledgee will have priority with respect to the proceeds.117 Pledged assets are part of the insolvent estate and the trustee is entitled to take them over at a value to be agreed between him and the pledgee (or the whole of the pledgee’s claim if appropriate), or, if the trustee does not do this, the creditor can realise the property and pay the proceeds, net of its claim, to the trustee.118 The Insolvency Act recognises pledgees as secured creditors,119 with a preference in relation to proceeds of pledged property, and as unsecured creditors for any remaining balance.120 106 Bokomo v Standard Bank van SA Bpk (1996) (4) SA 450, 454; Brits (n 30) 112. 107 See ML, Arts 6(2) and 7. 108 Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [20]. 109 Simon v Mitsui and Co Ltd 1997 (2) SA 475, 500–02 (W). 110 Brits (n 30) 116. 111 Lesedi Secondary Agricultural Co-operative Ltd v Vaalharts Agricultural Co-operative and Another 1993 (1) SA 695 (NC) 699; Brits (n 30) 107, 113. 112 Brits (n 30) 114. 113 I Knobel, ‘Accession of Movables to Land, South African Law and Dutch Law’ (2012) 45 Comparative and International Law Journal of Southern Africa 77, 79. 114 Brits (n 30) 160. 115 See B above. 116 ibid. 117 Lesedi Secondary Agricultural Co-operative Ltd v Vaalharts Agricultural Co-operative 1993 (1) SA 695 (NC) 700. 118 Insolvency Act No 24 of 1936, s 83. 119 ibid s 2. 120 ibid s 83(12).

338 South Africa

3. Cession a. The Nature of Security Cession Under South African law, a cession is a transfer of an intangible, that is, a right or rights against an obligor. The nature of a cession by way of security has long been a matter for debate in South Africa. One view is that it takes place by way of pledge, so that the pledgor (the cedent/transferor) remains the owner of the claim, subject to the right of the pledgee (the cessionary/transferee) to realise the claim if the cedent defaults on the secured obligation.121 The cedent’s ownership interest in the ceded debt has been described as a ‘bare ownership’ and ‘reversionary interest’.122 On this view, if the secured obligation is paid or otherwise discharged, the claim automatically reverts to the full ownership of the cedent, so that there is no obligation on the cessionary to transfer the claim back to the cedent.123 The idea of ownership in a right being separate from the right to sue the obligor (which the cessionary has on the default of the cedent) troubled many South African scholars, who put forward the alternative theory that a security cession was an outright transfer of the claim to the cessionary, who, on discharge of the secured obligation, then had a personal obligation to retransfer the claim back to the cedent.124 The South African Court of Appeal in Grobler v Oosthuizen rejected the second theory, and espoused the pledge theory, despite its theoretical difficulties, stating that the doctrinal debate must now be seen as settled.125 There remains the theoretical possibility for parties to structure the transaction as an outright transfer to the cessionary plus a contractual obligation to retransfer the right to the cedent on discharge of the secured obligation, although this is contentious.126 Although the courts have, on occasion, decided that this was the true intention of the parties,127 that intention has to be very clearly expressed, and therefore the default position is that, if a transfer has a security purpose, it will be construed as a pledge.128 The controversy discussed above has also led to some uncertainty as to the relationship between the parties to the cession and the debtor.129 Although the cases do not speak with one voice, the current position appears to be as follows, given the decision in Grobler that the correct analysis is usually one of pledge. Despite the fact that the cedent remains the owner of the claim, it is the cessionary who has the right to sue the debtor if the ceded debt

121 National Bank of SA Ltd v Cohen’s Trustee 1911 AD 235, 245. 122 Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 447–48 (SCA). 123 Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [20], [23]; Aussenkehr Farms (Pty) Ltd v Trio Transport CC 2002 (4) SA 483 (SCA) [20]. 124 Lief, No v Dettmann 1964 (2) SA 252, 271 (A); Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [17]. 125 Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [17]; Scott (n 56). 126 Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [23]. 127 African Consolidated Agencies (Pty) Ltd v Siemens Nixdorf Information Systems (Pty) Ltd 1992 (2) SA 739 (C). 128 Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [24], where the court stated that this is to avoid unacceptable commercial uncertainty. 129 For recent discussion, see S Scott, ‘The Evergreen Topic of Locus Standi and Security Cessions: Thekweni Properties (Pty) Ltd v Picardi Hotels Ltd and Picardi Hotels Ltd v Thekwini Properties (Pty) Ltd’ (2009) 21 South African Mercantile Law Journal 405; R Brits, ‘Right to Receive versus Right to Appropriate Proceeds of Insurance Policy Ceded in Securitatem Debiti: Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd and Others’ (2016) 28 South African Mercantile Law Journal 346.

Security Rights in Movable Assets 339 is not paid when it is due.130 This is unremarkable if the cedent is in default in relation to the secured obligation,131 but is potentially problematic if the position is that the secured obligation is not yet due. Although the matter is not free from doubt, South African law appears to be that, in this situation, the cedent does not have the right to sue the debtor,132 although it can take action to protect its reversionary interest.133 The position of cessionary vis-a-vis the debtor,134 then, depends on the terms of the cession agreement, express or implied.135 In some cases, the parties will intend the cessionary to collect in the debts, and apply the proceeds to the secured obligation, accounting to the cedent for any balance,136 or, at least, to have the ability to sue the debtor if necessary, and hold the proceeds on trust for the cedent.137 Courts have permitted this type of arrangement despite its close parallels to constitutum possessorium where the pledgee permits the pledgor to maintain possession and use of a pledged asset.138 However, as mentioned below, a security cession can be valid without giving notice to the debtor, in which case the cessionary cannot collect the debts. This arrangement is frequently the case in receivables financing transactions, where the parties agree that the cedent will continue to collect the receivables owed to it, either on its own behalf139 or as agent or mandatary for the cessionary.140

b. Creation of Security Cession There is no required form to transfer a right through cession as long as the intent to transfer is clear, which may be embodied in an oral agreement.141 Generally, the execution of a

130 Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [28]; Bank of Lisbon and South Africa Ltd v The Master 1987 (1) SA 276, 294 (AD); Goudini Chrome (Pty) Ltd v MCC Contracts (Pty) Ltd 1993 (1) SA 77, 87 (AD); Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 447–48 (SCA). 131 Enforcement of the security cession is considered in C3e below. 132 Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 447–48 (SCA); Picardi Hotels Ltd v Thekwini Properties (Pty) Ltd 2009 (1) SA 493 (SCA) [13]; Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [19]; Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [28]. 133 Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 447 (SCA); Grobler v Oosthuizen 2009 (5) SA 500 (SCA) [19]; Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [48]. 134 Of course, the cessionary could also contractually require the cedent to buy back the debt in this type of situation: this would be usual in a recourse financing arrangement. 135 Volhand & Molenaar (Pty) Ltd v Ruskin NNO 1959 (2) SA 751, 753 (W); Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 447 (SCA); 136 Volhand & Molenaar (Pty) Ltd v Ruskin NNO 1959 (2) SA 751, 753 (W); Bank of Lisbon and South Africa Ltd v The Master 1987 (1) SA 276, 294 (A). 137 Development Bank of Southern Africa Ltd v Van Rensburg 2002 (5) SA 425, 448 (SCA); Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [38]. 138 Brits (n 30) 332–33 citing, inter alia, Aussenkehr Farms (Pty) Ltd v Trio Transport CC 2002 (4) SA 483 (SCA) 491. 139 This construction is not free from conceptual difficulties; see Aussenkehr Farms (Pty) Ltd v Trio Transport Cc 2002 (4) SA 483 (SCA) Marais JA [4]. 140 Goudini Chrome (Pty) Ltd v MCC Contracts (Pty) Ltd 1993 (1) SA 77, 87(AD); Aussenkehr Farms (Pty) Ltd v Trio Transport Cc 2002 (4) SA 483 (SCA); Picardi Hotels Ltd v Thekwini Properties (Pty) Ltd 2009 (1) SA 493 (SCA) [13]. 141 Botha v Fick 1995 (2) SA 750, 762 (A); Fick v Bierman (1883–84) 2 SC 26, 33–34.

340 South Africa cession agreement is sufficient to render it effective against third parties.142 As mentioned above, unlike some civil law countries, a cession can take place without any notice being given to the debtor.143 Statutory and regulatory requirements governing certain types of rights may impose specific formalities to make the cession effective against third parties. For example, a cession (as well as a pledge) of interests in uncertificated securities must comply with the formalities prescribed by the Financial Markets Act 19 of 2012.144 A credit of the securities into the securities account of the creditor under the Financial Markets Act provides a mechanism for publicity that also strips the pledgor/cedent of the right to dispose of the securities, and so priority between interests in uncertificated securities is determined by date of credit entry.145 In general, the cedent is not required to deliver to the cessionary any documentation that proves the existence of the transferred right, such as a lease agreement entitling the cedent to rental payments from the lessee.146 However, courts have held that when the transferred right is so embedded in the documentation that it cannot be exercised without it, such documentation must be delivered to the cessionary to give legal effect to the cession.147 Negotiable instruments such as promissory notes are examples of instruments over which the cessionary must gain control to have an effective security right and that must therefore be transferred to the cessionary for the courts to recognise the cession as completed.148 Because these documents must be delivered to the cessionary, there is no practical difference between a cession and a pledge in this context. The lack of any publicity requirement, including a requirement to give notice to the debtor, has contributed to the popularity of this form of security but also increased the risk of fraud because third parties may have to rely on the representations of the cedent that it has not transferred its rights already, or take some other protective action.149 In practice, the lack of publicity has not been problematic and cessions function relatively well.150

c. Agreements Prohibiting Cession There is considerable controversy in South Africa about agreements prohibiting cession of a claim (pacta de non cedendo).151 The starting point was that claims were freely transferable,

142 World Bank (n 27) 27; Naumann (n 53) 6. 143 G Gretton and P Nienaber, ‘Assignation/Cession’ in R Zimmermann, K Reid and D Visser (eds), Mixed Legal Systems in Comparative Perspective: Property and Obligations in Scotland and South Africa (Oxford, Oxford University Press, 2004) 794. 144 Financial Markets Act 19 of 2012, s 39(1)(a)(i)–(ii). 145 ibid s 40(1). 146 Labuschagne v Denny 1963 (3) SA 538 (A); Botha v Fick 1995 (2) SA 750 (A); Standard General Insurance Co Ltd v SA Brake CC 1995 (3) SA 806, 812; Brits (n 30) 310. 147 Botha v Fick 1995 (2) SA 750 (A); Brits (n 30) 309–10 citing extensive case-law history and describing it as ‘not always perfectly clear or consistent’. 148 Brits (n 30) 311. 149 Brits (n 30) 306–07; see Tunisia C3 for a description of the practice of creditors taking possession of invoices and other evidence of the existence of a receivable. 150 Brits (n 30) 307, cf Britz NO v Sniegocki 1989 (4) SA 372, 480 (D), where the practical problems of lack of publicity led to the court calling for a central registration system. 151 See, eg, S Scott ‘Once Again: Agreements Prohibiting Cession’ (2008) 19 Stellenbosch Law Review 483; D Hutchison, ‘Agreements in Restraint of Cession: Time For a New Approach’ (2016) 27 Stellenbosch Law Review 273.

Security Rights in Movable Assets 341 but also that freedom of contract meant that the parties could, by agreement, make the claim non-transferable.152 This position was qualified, however, in Paiges v Van Ryn Gold Mines Estates Ltd,153 where the court held that such an agreement was not enforceable if ‘the debtor has no interest in the matter’, but that ‘if the pacta de non cedendo serves a useful purpose for the debtor, it is valid and binding’.154 Later on, in Trust Bank of Africa Ltd v Standard Bank of South Africa Ltd,155 the court made a distinction between restrictions on transfer that were external to the contract whereby the claim was created, to which the Paiges reasoning applied, and restrictions contained in the actual contract giving rise to the debt, which were valid even if the debtor had no interest in restricting transfer.156 This distinction was explained and supported in Professor Scott’s influential book on cession,157 and was adopted by the court of the Cape Provincial Division in 2008158 and by the Supreme Court of Appeal of South Africa in 2013.159 The current position, then, is that if a contract giving rise to a debt itself restricts transfer, or a company’s formation document restricts the transfer or pledge of shares,160 that restriction will prevent transfer vis-a-vis all third parties. As a consequence, certain receivables may not be used to access finance. If the agreement not to transfer is in a different, or subsequent, agreement, it will only be valid if the debtor has an interest in restricting transfer, which leaves a great degree of ambiguity as to which restrictions will and will not be enforced. It seems that a debtor of a receivable might not have an interest in whether the cedent transfers the receivable outright or under a pledge when the cedent continues collecting the receivable. The situation may be different if the cessionary notified the debtor of a receivable to remit the payment to a different bank account. It would seem that a company which restricted the transfer of its shares would have a great deal of interest in any transfer as that might result in a third party becoming a shareholder, and therefore such a restriction would seem to be enforceable. There is considerable dissatisfaction in South Africa at the present position, and Professor Scott herself has suggested that there are good policy reasons for limiting the effect of restrictive clauses to the parties to the agreement themselves (so that breach of the clause would be a breach of contract), but that they should be ineffective in relation to third parties.161 This is largely the position under Article 13 of the Model Law162 in relation to transfers of trade receivables and its adoption in South Africa would provide the needed clarity and certainty.

152 Paiges v Van Ryn Gold Mines Estates Ltd 1920 AD 600, 615. 153 ibid 600. 154 ibid 615. 155 Trust Bank of Africa Ltd v Standard Bank of South Africa Ltd 1968 (3) SA 166 (A). 156 ibid 148. 157 S Scott, The Law of Cession, 2nd edn (Cape Town, Juta, 1991) 212–14. 158 Capespan (Pty) Ltd v Any Name 451 (Pty) Ltd 2008 (4) SA 510, 515 (C). 159 Born Free Investments 364 (Pty) Ltd v Firstrand Bank Ltd (973/12) [2013] ZASCA 166. 160 Smuts v Booyens, Markplaas (Edms) Bpk en v Booyens 2001 (4) SA 15, 24 (SCA). 161 S Scott ‘Once Again: Agreements Prohibiting Cession’ (n 148), where Professor Scott considers the various international instruments which include such a provision, including the UNCITRAL Assignment of Receivables Convention. 162 ML, Art 13(2) provides that the debtor of the receivable may not avoid the contract as a result of breach of the restrictive clause.

342 South Africa

d. Cession of Future Rights Future rights may also be made subject to a cession,163 but there is debate as to the precise analysis of this in South African law.164 One view is that there can be a present cession of future rights,165 whereas another view is that there is a cession in anticipation.166 Whichever view is taken, it is reasonably clear that when the cedent acquires a right in future receivables, the cession automatically occurs without requiring any further action.167 Thus, for example, cessions of future rights are effective even after the cedent has filed for a business rescue, on the grounds that a security cession of receivables (book debts) is completed when the cession agreement is entered into.168

e. Enforcement The most usual way for the security cessionary to enforce on the cedent’s default in paying the secured obligation is for the cessionary to collect the debt (by suit if necessary) and to apply the proceeds to the reduction or extinguishment of the secured obligation, accounting to the cedent for any balance.169 Although the cessionary’s right is inherent in the security cession, it is usually provided for in the cession agreement.170 If the cessionary wishes to enforce by selling the ceded assets, it will have to proceed in the same way as a pledgee, namely, it must apply to court to authorise the sale unless the cession agreement includes a paratie executie (summary execution) clause.171 It can also take over the property at a fair price, if a quasi-conditional clause is included in the agreement.172 Insolvency rules for security cessions are the same as for pledges of tangible assets.173 The trustee may enforce the rights transferred in security and then use the proceeds to discharge the secured claim.174 In contrast, in the case of an absolute cession, the transferred right is outside of the insolvent estate.175 Because courts have rarely recognised a cession of rights as an outright transfer (preferring to construe cessions of intangibles as pledges of claims), no contemporary Supreme Court of Appeal ruling addressed an issue related to an absolute cession in an insolvency case.176 163 First National Bank of SA Ltd v Lynn No 1996 (2) SA 339, 346 (A). The same is true of contingent rights, see Standard General Insurance Co Ltd v SA Brake CC 1995 (3) SA 806, 815 (A). 164 Smit v Carniasaad 1998 (4) SA 877, 883 (SCA); Muller NO v Trust Bank of Africa Ltd 1981 (2) SA 117, 125–28 (N). 165 First National Bank of SA Ltd v Lynn No 1996 (2) SA 339, 360 (A); see also Headleigh Private Hospital (Pty) Ltd T/A Rand Clinic v Soller & Manning Attorneys 2001 (4) SA 360, 368 (W). 166 Muller NO v Trust Bank of Africa Ltd 1981 (2) SA 117, 125–27; Brits (n 30) 335. 167 Brits (n 30) 335. 168 BP Southern Africa (Pty) Ltd v Intertrans Oil SA (Pty) Ltd 2017 (4) SA 592 (GJ). 169 Kuranda v Boustred 1933 WLD 49, 53; Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [39]. 170 Brits (n 30) 331. It is possible that the cession agreement will limit the cession itself to the amount of the secured obligation, in which case only that amount can be recovered; see Van Staden NO v Firstrand Ltd 2008 (3) SA 530 (T) [23]. 171 See B above. 172 ibid. 173 See D1d above. 174 National Bank of SA Ltd v Cohen’s Trustee 1911 AD 235, 248. This does not apply on the death of a cedent; Retmil Financial Services (Pty) Ltd v Sanlam Life Insurance Company Ltd [2013] 3 All SA 337 (WCC) [54]. 175 Brits (n 30) 343. 176 ibid.

Security Rights in Movable Assets 343

4. General Notarial Bond a. Scope of General Notarial Bonds General notarial bonds have their origin in the Roman law concept of the hypotheca.177 They must cover all (as opposed to some) of the grantor’s movable tangible and intangible assets, both at the time of registration and in the future.178 This conclusion, while seemingly well established in South African law, comes from one case, Cooper NO v Die Meester,179 in which the court held that a notarial bond over specific movable property that remained in the possession of the grantor of the bond was not valid to create a preference over other creditors, that is, such a bond did not fall within section 102 of the Insolvency Act 24 of 1936. The actual problem in Cooper was statutorily remedied by the enactment of the Security by Means of Movable Property Act 57 of 1993 (SMPA).180 However, the decision in Cooper still appears to apply to a purported general notarial bond, which does not fall within that Act; for example, a bond which describes the collateral as all movable property with the exception of certain assets, so that such a bond will not give priority to its holder on the insolvency of the grantor.181 It is also not possible to grant a general notarial bond which includes immovable property.182 The restrictions mentioned here mean that the general notarial bond is neither able to cover all the assets of a business (including immovable property), nor can it cover part of those assets. However, it may cover part of those assets when the remainder is subject to a special notarial bond resulting in all the assets being encumbered under a bond. The court in Reeskens v Registrar of Deeds183 examined a bond which first described the collateral specifically, but then included a clause attempting to capture any property of the grantor that is not covered by the security of a special bond. The general bond thus purported to cover only unencumbered assets, but, in combination with the special bond, the entirety of the grantor’s assets were encumbered. The requirement that the general notarial bond must cover all assets of the grantor may thus be satisfied by extending to all assets that are not subject to any other form of security. Despite many similarities in operation, it will be seen from the discussion in this section that there are a number of differences between the general notarial bond and the common law floating charge.184

177 For a discussion of the hypothec in its earliest form, see A Watson (ed), The Digest of Justinian, vol 2 (Philadelphia, University of Pennsylvania Press, 1985) 20.1.34 (discussing the hypothec of a shop, which covers the shopkeeper’s stock in trade (inventory)). P Sacks, ‘Notarial Bonds in South African Law’ (1982) 99 South African Law Journal 605, 616–17 points out that the general notarial bond is wider than the hypothec of a shop, since it covers all the shopkeeper’s assets, including intangibles, and not just the inventory. 178 N Locke, ‘Security Granted by a Company Over Its Movable Property: The Floating Charge and the General Notarial Bond Compared’ (2008) 41 Comparative and International Law Journal of Southern Africa 136. 179 Cooper NO v Die Meester 1992 (3) SA 60 (A). 180 See C5. 181 Locke (n 176) 143–44. 182 A bond purporting to bind immovable property cannot be registered (Deeds Registries Act 47 of 1937, s 53(1)) and a registered bond over immovable property does not have priority on insolvency (Insolvency Act No 24 of 1936, s 86). See Firstrand Bank Ltd v Land and Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA) [13]. 183 Reeskens v Registrar of Deeds 1964 (4) SA 369 (N). 184 See ch 1, E1b for a discussion of the floating charge, and Locke (n 176) for a discussion of the general notarial bond in comparison to the floating charge.

344 South Africa

b. Creation and Registration A general notarial bond has some similarities to a security right under the Model Law, since it can cover present and future assets, generically described, and can secure future185 as well as present obligations. However, execution of a notarial bond requires more formalities than those required to create a security right under Article 6 of the Model Law. A notarial bond must identify the place where, and the date when, the bond was executed, the place where the notary that assisted with the execution of the bond is located, and the place of residence of the grantor and any place of business.186 Notarial bonds are frequently granted by businesses.187 The creditor must register the bond within three months of execution.188 Bonds not registered within the prescribed period are invalid and have no force or effect.189 However, unlike registration of a notice under the Model Law, registration of a general notarial bond does not give the creditor a real right which is effective against third parties. As will be explained below, the registration protects the bondholder in insolvency, but only as against unsecured creditors. Bonds must be registered in the ‘deeds registry for the area in which [the grantor] resides and in every deeds registry in the area in which [the grantor] carries on business’.190 This approach may require multiple registrations as well as close monitoring of the grantor’s activity to ensure that timely registrations are made in the areas where the grantor may be carrying on some business activity. These costs and risks do not exist under the Model Law where a single registration would suffice to make the security right effective with respect to any intangible assets of the grantor located in the relevant State as well as any tangible assets located in the same State. Bonds executed by limited-liability companies and close corporations are only required to be registered in the deeds registry for the area where the registered office of the company is located at the time of registration.191 A single nationwide registry would eliminate a number of risks faced and costs incurred presently by creditors.

c. The Nature of a General Notarial Bond The grantor of a general notarial bond retains possession and ownership of the property covered by the bond. As mentioned above,192 the general notarial bond does not create a security right in the assets it covers.193 This means that the grantor can deal with or dispose of the assets subject to the bond, and the bondholder cannot follow the assets into the hands of an acquirer, or prevent a judicial attachment.194 This rather broad statement about the 185 Deeds Registries Act 47 of 1937, s 50. 186 ibid s 61(3). 187 Locke (n 176) 139. 188 Deeds Registries Act 47 of 1937, s 61(1). 189 Brits (n 30) 195. 190 Deeds Registries Act (n 182) s 62(1). 191 ibid s 62(4). 192 See C4b above. 193 This has various legal consequences, including a shorter limitation period than a bond creating real security, Land and Agricultural Development Bank of South Africa v Phato Farms (Pty) Ltd 2015 (3) SA 100 (GP). 194 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [3]; Firstrand Bank Ltd v Land and Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA) [4].

Security Rights in Movable Assets 345 position of acquirers has been qualified by academics, who argue that the usual principles of property law should apply, so that a person acquiring the asset without giving any value or with actual knowledge of any restriction to dispose inserted in the bond acquires subject to the bond.195 This has the practical effect that acquirers take free of the bond when an asset is either disposed of in the ordinary course of business or they give value and take without knowledge in transactions outside the ordinary course of business.196 The fact that a general notarial bond does not create a security right also means that the grantor can grant a security right, such as a special notarial bond, over its assets.197 Again, however, the second bondholder would have priority, unless it had actual notice of it.198 Constructive notice is not enough, and, in any event, registration of the bond in the Deeds Registry does not give constructive notice.199 If the second holder has a real right, such as under a special notarial bond, it will obtain priority; if it takes another general notarial bond, it will only have priority if it takes possession of the assets first.200 The grantee of a general notarial bond obtains priority over the grantor’s unsecured creditors on insolvency. Without further steps being taken,201 a real right is not created because general bonds do not encumber specific assets described individually and therefore do not meet the requirements of publicity established in South African law.202 Third parties must be able to establish that an asset is subject to a security right without needing to assess external factors. This requirement is met when an asset is delivered to the pledgee or specifically described in a registration. However, a general description of assets in a bond would require a searcher to conduct further due diligence, thus failing the requirement of specificity.203 This approach is inconsistent with notice registration, as advocated by the Model Law,204 under which the function of a registration is to put the searcher on notice as to the potential existence of a security right.

d. Post-Default Rights Many general notarial bond agreements include a conditional ‘perfection clause’ allowing the bondholder to take possession of the assets and therefore acquire a real right by way of pledge.205 A holder of a bond which does not include such a clause, however, does not have a right to take possession,206 and merely has its right of preference over unsecured 195 Sacks (n 175) 615–17; Locke (n 176) 148. 196 Locke (n 176) 146, 148. 197 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [3]; Firstrand Bank Ltd v Land and Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA) [4]. 198 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [5]. It is not clear whether this doctrine applies even if there is no clause in the bond restricting the granting of security (a ‘negative pledge clause’). There was such a clause in the Chesterfin case. 199 Frye’s (Pty) Ltd v Ries 1957 (3) 575 (A). 200 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA). See 4d below. 201 See 4d below. 202 Brits (n 61) 246. 203 Brits (n 30) 198. 204 See ML, ch IV. 205 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [3]; Firstrand Bank Ltd v Land and Agricultural Development Bank of South Africa 2015 (1) SA 38 (SCA) [4]. 206 Eerste Nasionale Bank van SA Bpk v Schulenburg 1992 (2) SA 827 (T); Boland Bank Bpk v Spies 1993 (1) SA 402 (T); Boland Bank Ltd v Vermeulen 1993 (2) SA 241 (E).

346 South Africa c reditors in insolvency. Where there is such a clause, its terms govern the holder’s right to take possession, including the events which trigger that right, which will include the grantor’s default in payment, but also, often, likely detriment to the assets.207 As mentioned above,208 South African law will not permit a secured creditor to take possession of collateral without a court order, and therefore, if the clause is triggered and the grantor will not give up possession voluntarily, the bondholder must apply to the court for an order enabling perfection. This order is frequently an interim one made ex parte, which can be challenged by the grantor at the making of the final order, although probably only on the grounds that the creditor had no right to take possession.209 There are conflicting authorities on the amount of discretion the court has in making the orders and their terms. On one view, the court is making an order for specific performance, and therefore has discretion, which can be used to ameliorate the harshness of an order enabling the bondholder to take over the business of the grantor.210 On another view, the court has little discretion whether or not to make the order, if the bondholder is entitled to possession, although there could be some leeway in relation to its precise terms.211 Under South African insolvency law, a disposition of a company’s property after the commencement of insolvency is void unless pursuant to an order of the court.212 A holder of a general notarial bond therefore needs to take possession (if given voluntarily) or obtain a court order for possession before insolvency commences.213 An interim order suffices for this purpose, even if the grantor files for insolvency before a final order is issued.214 Thus, when insolvency is imminent, the general notarial bondholder is incentivised to race-to-court to ‘perfect’ its security under the bond.215 South African courts have issued ‘perfection orders’ since at least 1924 and they are now settled practice.216 It is clear that, provided there is agreement or a court order is sought as discussed above, a perfection clause in a general notarial bond is not unconstitutional.217 However, because of the gravity of taking possession of another person’s property, courts interpret perfection clauses narrowly.218 Provisions of a general notarial bond are unenforceable if they deny a grantor prior notice or due legal process before attachment and execution.219

207 Barclays National Bank Ltd v Natal Fire Extinguishers Manufacturing Co (Pty) Ltd 1982 (4) SA 650, 659 (D); for discussion, see J Roos, ‘The Perfecting of Securities Held under a General Notarial Bond’ (1995) 112 South African Law Journal 169, 171–73. 208 See B above. 209 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [8], [10]. 210 Barclays National Bank Ltd v Natal Fire Extinguishers Manufacturing Co (Pty) Ltd 1982 (4) SA 650, 654–56 (D). 211 International Shipping Co (Pty) Ltd v Affinity (Pty) Ltd 1983 (1) SA 79, 84 (C); Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [10]–[11]. 212 Companies Act 61 of 1973, s 341(2). 213 International Shipping Co (Pty) Ltd v Affinity (Pty) Ltd 1983 (1) SA 79, 85 (C). 214 Contract Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [8], [10]. 215 World Bank (n 27) 24, 33; Naumann (n 53) 16; Brits (n 30) 210–12. 216 J Roos ‘The Perfecting of Securities Held Under a General Notarial Bond’ (1995) 112 South African Law Journal 169, 169. 217 Juglal NO v Shoprite Checkers (Pty) Ltd t/a OK Franchise Division 2004 (5) SA 248 (SCA); SA Bank of Athens Ltd v Van Zyl 2005 (5) SA 93 (SCA) [14]. The decision in Findevco (Pty) Ltd v Faceformat Sa (Pty) Ltd 2001 (1) SA 251 (E), which declared paratie executie clauses unconstitutional, was distinguished. 218 Barclays National Bank Ltd v Natal Fire Extinguishers Manufacturing Co (Pty) Ltd 1982 (4) SA 650, 660 (D); Brits (n 30) 216. 219 Brits (n 30) 216; World Bank (n 27) 41.

Security Rights in Movable Assets 347 As mentioned above, upon perfection, the bondholder enjoys all the rights of a pledgee with a real security right that is effective against third parties, including in the grantor’s insolvency. It is therefore important that the court order enables the creditor to take possession of the assets. To achieve this, the court may order, inter alia, physical transfer of the assets to the creditor or control of access to premises where the assets are located – by possession of keys or employment of security guards – or both.220 Depending on the terms of the clause, the court may also find it necessary to place the creditor, or its agent, in control of the grantor’s business.221 This has some similarities to the appointment of an administrative receiver under a floating charge.222 If the bond is not perfected before the commencement of insolvency, it will have preference over unsecured creditors,223 including ‘non-preferent claims’,224 but not over other preferential claims.225 It will be seen, then, that the protection given to a creditor with a registered but unperfected general notarial bond is a great deal less than that of a creditor who has taken an all asset security and registered a notice under the Model Law.

5. Special Notarial Bond The special notarial bond creates a security right deemed to be equivalent to a pledge that becomes effective against third parties upon registration.226 The legal effect of this security right was statutorily recognised by the SMPA. Prior to enactment of the SMPA, the position in the Province of Natal was different from the rest of the country, in that registered special notarial bonds were permitted in Natal under the Notarial Bonds (Natal) Act,227 which reflected the pre-1916 case-law whereby a registered special notarial bond created a real security right, fully enforceable in insolvency and effective against third parties, without requiring delivery of the movable asset to the bondholder.228 The courts in other provinces interpreted common law to mean that special notarial bonds conveyed no real security without delivery of the asset to the creditor, but that in insolvency, non-Natal special notarial bonds were included in the category of general notarial bonds.229 As mentioned above,230 the court, in Cooper NO v Die Meester,231 held that special notarial bonds were not included in section 102 of the Insolvency Act and therefore had no preference over unsecured creditors in insolvency. This created an urgent need to reform national notarial bond law and led to the relatively quick enactment of the SMPA in 1993.232

220 Contract

Forwarding (Pty) Ltd v Chesterfin (Pty) Ltd 2003 (2) SA 253 (SCA) [14]. National Bank Ltd v Natal Fire Extinguishers Manufacturing Co (Pty) Ltd 1982 (4) SA 650, 663 (D). 222 See ch 1, E1b. 223 See Insolvency Act No 24 of 1936, s 102. 224 Listed in Insolvency Act No 24 of 1936, s 103. 225 ibid ss 96–101. 226 Security By Means of Movable Property Act 57 of 1993, s 1. 227 Notarial Bonds (Natal) Act 18 of 1932. 228 Nedbank Ltd v Norton 1987 (3) SA 619, 624–25 (N). 229 ibid. 230 See C4a above. 231 Cooper NO v Die Meester 1992 (3) SA 60 (A). 232 Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA) [14]–[15]. 221 Barclays

348 South Africa Although it was the decision in Cooper that triggered the passage of the SMPA, the enefits of expanding the Natal special bond regime across the country had been advocated b for a number of years and had been examined by the South African Law Commission.233 While acknowledging that possession provides a public notice, thereby justifying the preference enjoyed by pledgees against other creditors, the Commission recognised that delivery is an impractical way to satisfy modern business needs for credit secured with movables.234 The SMPA, like the Natal regime, strikes a balance that offers publicity without possession and a real security right effective against third parties and, in insolvency, a preference over holders of general notarial bonds and unsecured creditors. Special notarial bonds have not been widely used in practice to secure obligations, suggesting that reforms to the SMPA regime are needed.235 The SMPA classifies the right conveyed by a special notarial bond as a pledge of movable property ‘deemed to have been pledged to the mortgagee as effectually as if it had expressly been pledged and delivered to the mortgagee’.236 The idea of the deemed pledge is potentially problematic, given that the creditor does not actually have possession, and it can be questioned whether another reform technique would have been preferable.237 For example, despite the fact that it is strictly unnecessary, parties tend to include a perfection clause in a special notarial bond, in order to be able to apply to court to gain actual control of the assets.238 It is clear from the SMPA that the object of the bond must be specifically described.239 This has a number of consequences. Special notarial bonds may not be used to encumber fluctuating assets, such as inventory or the entire business,240 and intangible assets such as receivables and intellectual property rights may not be the subject of this type of a bond. They may, however, encumber assets to be acquired by the grantor in the future as long as they can be specifically described at the time of registration.241 The court used the deemed pledge analysis to reach this conclusion in Bokomo v Standard Bank van SA Bpk,242 because a pledge of another’s property, without the owner’s consent, becomes valid when the pledgor subsequently acquires the pledged property.243 Courts have interpreted the requirement of the SPMA that the special notarial bond specify and describe the tangible movable property ‘in a manner which renders it readily recognisable’244 as requiring the collateral to be specifically identified, so that upon searching the registry, third parties must be able to clearly identify which assets of the grantor remain unencumbered.245 In Ikea v BOE Bank, the court said: ‘It would thus not be s ufficient to 233 For details, see Brits (n 61) 248–51. 234 South African Law Commission, The Giving of Security by Means of Movable Property: Report (1991) para 1.5. 235 Brits (n 61)6. 236 Security By Means of Movable Property Act 57 of 1993, s 1. 237 Brits (n 61) 253–61. 238 Farmsecure Grains (Edms) Bpk v Du Toit 2013 (1) SA 462 (FB). 239 Security By Means of Movable Property Act 57 of 1993, s 1; See also Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA). 240 S Scott, ‘A Comparison Between Belgian, Dutch and South African Law Dealing with Pledge and Execution Matters’ (2010) 43 Comparative and International Law Journal of Southern Africa 93, 112. 241 Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA) [9]–[11]. 242 Bokomo v Standard Bank van SA Bpk (1996) (4) SA 450. 243 Brits (n 30) 110. 244 Security by Means of Movable Property Act 57 of 1993 (n 223) s 1. 245 Ikea Trading und Design AG v BOE Bank Ltd 2005 (2) SA 7 (SCA) [10].

Security Rights in Movable Assets 349 describe the property by reference to quantity and kind.’246 In Scheltema Beleggings v Comm. Truck and Trailer Sales, the court explained: Where a generic item is sought to be pledged, it is the unique item that is the subject of the pledge and it is not enough to describe it only with reference to its generic characteristic. … A member of the public must be able to establish from the information lodged at the deeds office whether particular assets of a debtor have been pledged (whether or not he requires expert knowledge to do so).247

In Ikea, the court held that the registered bond did not meet the descriptive requirements of the SMPA because third parties could not correlate the property in question to the description provided in the bond without consulting external evidence.248 Therefore ‘the bond did not create a deemed pledge over the property … and Ikea was not a secured creditor’.249 Similarly, the court in Rosenbach & Co (Pty) Ltd v Dalmonte250 held that: [I]t is not a compliance with the statute to describe the assets to be hypothecated in wide general terms, as goods, wares, merchandise, stock-in-trade, fixtures, fittings, furniture and appliances. It is necessary to know what are the goods, merchandise and so on, the nature of them and the types or kind of each of them, and also the number of them.

It appears from these cases that the focus of the legislation is on assets that could be identified by serial numbers or other unique identification methods, and accordingly, not only fluctuating inventory may not be subject to a special notarial bond, but such bonds may not even encumber assets that replace the asset originally described in the bond.251 A special notarial bond that does not meet the requirements of the SMPA is not even treated as a general notarial bond.252 Therefore, the holder of such a bond neither acquires a security right under the SMPA nor the preference of a general notarial bond under the Insolvency Act. The strict description requirements coupled with the ambiguous interpretation of those requirements by the courts have rendered the use of special notarial bonds impractical.253 Some commentators suggested an amendment of the SMPA to provide for ‘an exhaustive list of concrete methods in terms of which movables may be identified’.254 Although this suggestion may make the use of special notarial bonds more attractive, it does not address the underlying issue with a principle of specificity that requires that a description must enable third parties to identify the collateral without resort to extrinsic evidence. As a matter of practice, the searcher should always conduct due diligence beyond the registry record, including to inquire as to the status of the secured obligation owed by the grantor. Information about the collateral should be made a part of that inquiry. Incremental changes in terms of providing clear description standards meeting the principle of specificity do not correspond to international best standards or ordinary due diligence.255

246 Ikea

Trading und Design AG v BOE Bank Ltd (2004) SCA [19]. Scheltema Beleggings v Comm Truck and Trailer Sales (35420/08) ZAGPPHC 28 (2010) [10]. 248 Ikea (n 246) para 25. 249 ibid. 250 Rosenbach & Co (Pty) Ltd v Dalmonte 1964 2 SA 195 (N) 204G–205A. 251 Brits (n 61) 266. 252 Cooper NO en Andere v Die Meester en’ n Ander, 1992 (3) SA 60 (A). 253 Brits (n 30) 245. 254 ibid 250. 255 See ch 2, A2bii. 247 See

350 South Africa Specific information in the registry record that clearly identifies an asset is never sufficient for the searcher to determine whether to extend a loan against or buy that asset. Registration requirements for special notarial bonds are the same as for general notarial bonds: the bond itself must be registered in accordance with the Deeds Registries Act.256 The date of registration of the bond, not the date of its execution, determines priority;257 thus, registering a special notarial bond establishes both the third-party effectiveness and priority, whereas the priority of a general notarial bond is set by the date of a court order empowering the bondholder to seize the collateral. Because the effectiveness against third parties of a special notarial bond depends on registration, the holder of a special notarial bond has the right to follow the property into the hands of third-party acquirers.258 In Bokomo v Standard Bank van SA Bpk, the court confirmed that innocent third-party acquirers are not protected.259 Although holders of special notarial bonds are not protected from attachment by other creditors, they have priority to the proceeds of any sale.260 Additionally, the SMPA provides that movable assets subject to a special notarial bond are not affected by a landlord’s tacit hypothec.261 A legacy of Cooper is that special notarial bonds that do not comply with the requirements of the SMPA are not classified in insolvency as general notarial bonds and have no preference against unsecured creditors.262 If the property described in a special notarial bond is in the possession of the grantor, a creditor must apply for a court order to recover it on default.263 In the case of special notarial bonds, courts have ruled that because bondholders achieve third-party effectiveness by registration rather than possession, the creditor may only apply for an order of attachment if the bond agreement includes a conditional provision for delivery of the property to the creditor and the conditions of that provision have been met.264

D. Final Remarks Unlike many of its neighbours, South Africa’s economy has been relatively stagnant over the past decade, which has had a negative impact on the business prospects of SMEs. The country’s outdated and fragmented legal framework for secured transactions has been an

256 Security by Means of Movable Property Act 57 of 1993, s 1(1). 257 N Locke, ‘Aspects of Traditional Securitisation in South African Law’ (University of South Africa, 2008) 222, http://uir.unisa.ac.za/bitstream/handle/10500/2676/thesis_locke_%20n.pdf?sequence=1&isAllowed=y. 258 Brits (n 30) 253. 259 ibid citing Bokomo v Standard Bank van SA Bpk (1996) (4) SA 450 (C) 454. 260 Brits (n 30) 255. 261 Security by Means of Movable Property Act 57 of 1993, s 2; see also Brits (n 30) 435 explaining: ‘[O]riginating in Roman law, the landlord’s tacit hypothec is a security interest for the payment of outstanding rent, in movables brought onto the land. The hypothec is tacit because it arises implicitly from the contractual relationship between landlord and tenant rather than as a consequence of any express agreement, and cannot be contractually excluded. It becomes effective when the lessee defaults on its duty to pay rent and lapses when outstanding rent is paid in full.’ The Insolvency Act, s 2 recognises the tacit hypothec (landlord’s legal hypothec) as a preferred security right. 262 Brits (n 30) 239. 263 Bock v Duburoro (n 29) 7; see B above. 264 Senwes Ltd v Muller 2002 (4) SA 134 (T); Farmsecure Grains (Edms) Bpk v Du Toit 2013 (1) SA 462 (FB); Brits (n 61) 269–71.

Final Remarks 351 impediment to business growth. Although a variety of legal devices to secure obligations are available to creditors, the legal complexity and cost of structuring many types of secured transactions, but especially those secured with fluctuating assets such as inventory, make it challenging for South African microbusinesses and SMEs to access credit. The inefficiencies run through all the fundamental elements of a secured transaction, from creation where the execution of a notarial bond is attested by a notary public who must be identified in the bond itself and only limited exceptions to the effect of antiassignment clauses are recognised, to third-party effectiveness that is not dependent on registration of a notice, all the way to remedies where the law restricts the availability of extrajudicial enforcement. No security devices presently available are capable of satisfying the credit needs of South African businesses and providing the level of protection that creditors expect. Some South African commentators note that general notarial bonds do not pose significant practical problems, and therefore the current system is not problematic.265 However, the protections given to secured creditors are relatively weak as compared to the Model Law security right, including as against other secured creditors and transferees of the collateral. Although the general notarial bond has priority over unsecured creditors in insolvency, sections 96–101 of the Insolvency Act set out a list of claims that must be satisfied before any distribution is made on the general notarial bond. The lack of such protections is expected to be taken into account by creditors in a number of aspects. First, prudent creditors extend loans only to highly creditworthy borrowers with an established record, thus excluding SMEs in particular, as documented by the high number of rejected loan applications. Secondly, prudent creditors are expected to reduce the amount of credit available against the value of the collateral, which probably contributes to the unusually high rate of business failures. Finally, the cost of monitoring is expected to be much higher than for those secured creditors operating in a legal environment that protects their rights in insolvency of the grantor as well as against judgment creditors. There are a number of downsides for prospective borrowers, including the inability to encumber some, but not all, assets under a general notarial bond even when these assets would provide a sufficient level of security for the loan. Given the requirement for a specific description, special notarial bonds are unsuitable for any inventory financing.266 Structuring inventory financing transactions with reservations of ownership and fiduciary transfers of ownerships has its own peculiar challenges, but those devices might be adequate to at least support purchase money financing.267 Commentators have also suggested upgrading the existing infrastructure for the registration of motor vehicles to provide that registration confers ownership.268 As a natural consequence of that upgrade, the registry could then be used to register security rights as well (currently, motor vehicles are financed under retention of title). Again, this would be an incremental improvement, but a superior solution in the form of providing for security rights in motor vehicles registrable in a collateral registry should be considered, along the

265 Brits

(n 30) 271. der Merwe and Smith (n 73) 326. 267 ibid 325–26. 268 Brits (n 30) 272. 266 van

352 South Africa lines of the reforms in regional neighbours Malawi and Zambia.269 Others have suggested a reform of the insolvency law to provide for more protection and certainty of the various security rights holders under South African law.270 Finally, calls have been made for a comprehensive review of the law governing cessions in combination with security rights in tangible assets, and for the creation of a unified legislative scheme underpinned by an electronic registration system.271 The challenges examined in this chapter support this call for a comprehensive reform of the South African secured transactions law, which should be undertaken along the lines of the Model Law.

269 See Zambia D3 and M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in O Akseli and L Gullifer (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Hart Publishing, 2016) 198–99. 270 Locke (n 253) 154. 271 Brits (n 30) 349.

16 Zimbabwe A. Introduction Zimbabwe is a landlocked low-income economy in southern Africa with a population of 16.5 million people.1 It is a member of the Common Market for Eastern and Southern Africa (COMESA) and the Southern Africa Development Community (SADC). Although its economy is heavily dependent on the mining and agriculture sectors, services are the largest contributor to Zimbabwe’s GDP.2 Despite the developments outlined below, as of 2017 agriculture remains the main employer, engaging over 67 per cent of the labour force, followed by industry.3 The main agricultural products are tobacco, corn, cotton, wheat, coffee, sugarcane, peanuts, sheep, goats and pigs.4 Major exports include platinum, cotton, tobacco, gold, ferroalloys and textiles,5 while key imports consist of machinery and transport equipment, chemicals, fuels and food products.6 Zimbabwe’s primary trade partners are China, South Africa, Botswana, Zambia, India and the Democratic Republic of Congo. Once the breadbasket of Southern Africa, Zimbabwe has suffered a number of political, economic and climate misfortunes that have impeded its economic growth and agricultural productivity.7 Land-reform issues have plagued the nation for decades, beginning in the early 1900s.8 In the 10 years following independence from the United Kingdom in 1980, Zimbabwean agriculture provided 45 per cent of exports, 60 per cent of the raw materials used in industry and 70 per cent of employment.9 The Zimbabwean dollar declined in value, but the price of fertiliser, seed and transportation increased to the extent that, although large-scale commercial farming of crops such as tobacco grew, small-scale farmers suffered, particularly after losing subsidies the government could no longer afford to provide by the end of the twentieth century.10 Drought, failed government programmes, inflation and a lack of foreign exchange plagued the economy, and during this period demands for land reform

1 ‘Zimbabwe Country Profile’ BBC News (14 January 2019) www.bbc.com/news/world-africa-14113249. 2 CIA, ‘Africa: Zimbabwe’ (The World Factbook, 22 February 2019) www.cia.gov/library/publications/the-worldfactbook/geos/zi.html. 3 ibid. 4 ibid. 5 ibid. 6 ibid. 7 T Mushakavanhu, ‘Mugabe Is Asking Back the White Farmers He Chased Away’ Quartz Africa (20 July 2015) qz.com/458137/mugabe-is-asking-back-the-white-farmers-he-chased-away/; BBC (n 1). 8 BL Peters, ‘Why Zimbabwe Has Failed to Sate the Yearning for Land and to Fix Rural Hunger’ CFUZim. org (13 December 2016) www.cfuzim.org/index.php/newspaper-articles-2/the-courts/90-newspaper-articles/ land-issues?start=310. 9 ibid. 10 ibid.

354 Zimbabwe grew.11 Subsequently, Parliament revised the Constitution to legalise the e xpropriation of farms without compensation.12 As a result, some of Zimbabwe’s premier agricultural farms were reallocated to individuals, many of whom lacked the skills, capital and capacity to cultivate and maintain those farms.13 By 2004, only a quarter of arable land was cultivated, and Zimbabwe turned from a net exporter of food to a net importer.14 Previously, agriculture had provided 40 per cent of Zimbabwe’s export earnings and employed two million people.15 As of 2011, redistributed large farms remained unproductive and large commercial production never recovered.16 Between 2000 and 2010, agricultural productivity declined sharply, falling 60 per cent, and the economy shrank to half its size in 1980.17 Agricultural exports fell by over 50 per cent and agricultural revenue reduced by US$12 billion.18 In addition, by 2017, the government defaulted on its sovereign debt, leading to a freeze on foreign assistance from multilateral institutions, including the World Bank and the African Development Bank.19 In 2008, hyperinflation hit 500 billion per cent, the highest rate in the world for a country that was not at war.20 The local currency became worthless and the government abandoned it, primarily for the US dollar and South African rand.21 From 2009 to 2014, Zimbabwe experienced a period of relative economic stability and growth.22 However, by 2014, deflation set in, accompanied by a severe regional drought that led to poor harvests and the death of livestock and, thus, a decrease in exports.23 Deflation also led to a drop in Zimbabwe’s industrial capacity, from 57 per cent to 34 per cent.24 To address the shortages, the government banned imports of certain goods and issued bond 11 ibid. 12 ibid. 13 A Meldrum, ‘Mugabe Land Seizures Force Hundreds of Farm Owners to Flee’ The Guardian (2 June 2000); K Sieff, ‘Zimbabwe Seized White Farmers’ Land. Now Some are Being Invited Back’ Washington Post (14 S eptember 2015); Mushakavanhu (n 7). 14 ibid. 15 ibid. 16 ‘Zimbabwe: Current Issues and U.S. Policy’ EveryCRSReport.com (15 September 2016) www.everycrsreport. com/reports/R44633.html#ifn65. 17 D Smith, ‘Mugabe and Allies Own 40% of Land Seized From White Farmers: Inquiry’ The Guardian (30 November 2010); Mushakavanhu (n 7). 18 Mushakavanhu (n 7); Sieff (n 8). 19 Zimbabwe owed the World Bank and African Development Bank over US$1.7 billion. CIA (n 2); G Phiri, ‘Zim’s Public Debt Rises to $11.6bn’ dailynewsLive (26 June 2017) www.dailynews.co.zw/articles/2017/06/26/ zim-s-public-debt-rises-to-11-6bn. 20 M Dzirutwe, ‘Refile: Zimbabwe Plans New Currency as Dollar Shortage Bites: Finance Minister’ Reuters: Commodities News (12 January 2019) https://www.reuters.com/article/us-zimbabwe-economy-currency/ zimbabwe-plans-new-currency-as-dollar-shortage-bites-finance-minister-idUSKCN1P6080. 21 S Berger, ‘Zimbabwe Inflation Hits 231 Million Per Cent’ The Telegraph (9 October 2008) https://www. telegraph.co.uk/news/worldnews/africaandindianocean/zimbabwe/3167379/Zimbabwe-inflation-hits-231million-per-cent.html. The abandonment of the local currency took place pursuant to a policy known as ‘dollarization’. See, eg, J Noko, ‘Dollarization: The Case of Zimbabwe’ (2011) 31(2) Cato Journal 339, 340 fn 1. The government gave consent to the use of foreign currencies in transactions and the full dollarization of Zimbabwe in February 2009, ibid 348; M Ellyne, ‘Lessons From Zimbabwe’s Hyperinflation and Dollarization’ ResearchGate (2015) www.researchgate.net/publication/281150323. 22 World Bank, ‘The World Bank in Zimbabwe: Overview: Context’ (31 October 2018) www.worldbank.org/en/ country/zimbabwe/overview. 23 ‘Zimbabwe Declares “State of Disaster” Due to Drought’ The Guardian (5 February 2016). 24 D Pilling and A England, ‘Zimbabwe Swaps Hyperinflation for Deflation with Use of US Dollar’ Financial Times (25 January 2016).

Introduction 355 notes as a form of proxy currency, valued at 1:1 with the US dollar.25 The government also encouraged the use of mobile and electronic payments in order to ease hardships caused by severe cash shortages.26 In the rural areas, Zimbabweans resorted to trade by barter.27 The US dollar was, thanks to a thriving black market, worth up to four times more than bond notes, resulting in inflation and a shortage of food and fuel.28 Despite these issues, the Governor of the Reserve Bank of Zimbabwe (RBZ) announced in May 2018 that additional bond notes would be printed and circulated (increasing their issue by US$300 million, to US$500 million), while government increased borrowing via Treasury bills.29 The economy grew by 3.5 per cent in 2018.30 This led to improved predictions in real GDP growth, and the economy is now expected to grow by 4.2 per cent in 2019 and by 4.4–4.9 per cent in 2020, thanks in large part to the agricultural and mining sectors, which have received increased public and private financial support.31 At the end of 2016, commercial banks held $1.35 billion in Treasury bills issued by a government unable to pay its debts.32 Since mid-2013, credit to the government increased from 13 per cent of domestic credit to 42 per cent. Financing provided to smallholder farmers was limited because, until 2018, commercial banks were hesitant to provide loans to the agricultural sector, as the banking industry considered the 99-year land leases the farmers held to be ‘unbankable’ as collateral.33 This was in spite of the fact that section 69(a) of the Deed Registries Act permits a mortgage (via mortgage bond) of a registered long-term lease. Because ownership is vested in the State under the Constitution, banks argued that the leases only gave the farmers the right to farm the land, rather than full ownership, so the leases were not transferable in the event of default.34 As a result, Zimbabwe’s farmers lacked the capital they needed to operate and grow.35 Many indigenous farmers who acquired land in the 2000 land reforms were not able to access credit because of the lack of title deeds

25 B Latham and M Cohen, ‘“Death Spiral” Looms for Zimbabwe Economy as Cash Runs Out’ Bloomberg (6 February 2017) www.bloomberg.com/news/articles/2017-02-06/-death-spiral-looms-for-zimbabwe-economyas-cash-runs-out; L Chutel, ‘Zimbabweans Are Sleeping Overnight Outside Banks Again to Get Their Cash’ Quartz Africa (8 November 2016) qz.com/830776/zimbabwe-is-going-through-a-severe-cash-shortage-and-people-arequeuing-overnight-at-banks-and-atms/. 26 RJ Southall, ‘Bond Notes, Borrowing, and Heading for Bust: Zimbabwe’s Persistent Crisis’ Taylor & Francis Online (2017) www.tandfonline.com/doi/full/10.1080/00083968.2017.1411285. 27 ibid. 28 F Mutsaka, ‘In Zimbabwe’s Crisis, “We Can’t Talk of Christmas Anymore”’ AP News (10 December 2018) https://www.apnews.com/04bd573977754a17b60e91e881876a19. 29 ‘Zimbabwe: New Central Bank Measures’ ControlRisks.com (21 May 2018) www.controlrisks.com/ our-thinking/insights/zimbabwe-new-central-bank-measures. 30 African Development Bank Group, African Economic Outlook (2019) 185. 31 ibid; ‘Zimbabwe: Economic Forecasts: 2019–2021 Outlook’ TradingEconomics (2019) tradingeconomics.com/ zimbabwe/forecast. 32 T Hawkins, ‘Zimbabwe Is Staring at Currency Chaos, Again’ Financial Times (26 February 2017). 33 I Mafundikwa, ‘Zimbabwe’s Farmers Struggle to Feed the Nation’ United Nations: Africa Renewal Online (2014) www.un.org/africarenewal/magazine/special-edition-agriculture-2014/zimbabwe%E2%80%99s-farmersstruggle-feed-nation; ‘Zimbabwe: ‘99-Year Leases Work in Progress’ allAfrica: The Herald (20 January 2015) allafrica.com/stories/201501200689.html. 34 Mafundikwa (n 33); ‘Zim Land Reforms: 99-Year-Leases “Now Bankable, Transferable”, Says Official’ news24 (9 February 2018) www.news24.com/Africa/Zimbabwe/zim-land-reforms-99-year-leases-now-bankable-transferable-20180208. 35 Mafundikwa (n 33).

356 Zimbabwe and in any case, as at 2016, only about 200 leases had been issued.36 Prior to the 2000 land reforms, commercial farmers were borrowing approximately US$2 billion annually.37 The government and the RBZ attempted to address the access to credit challenges faced by farmers and other businesses through a comprehensive legal reform project that sought to revamp the secured transactions, companies and insolvency regimes. As part of that reform, in 2018, the RBZ announced that local banks agreed to accept 99-year land leases as collateral.38 The agreement came after the government, by amending the Constitution, revised the leases to add a provision that guarantees lenders compensation for losses if default resulted in the government’s taking of the farm.39 The Agribank of Zimbabwe agreed to accept the 99-year land leases as collateral shortly after the announcement from the RBZ, and set aside $105 million for that purpose during the 2018–19 growing season.40 It has also extended loans to tobacco, fruit and seed intermediaries as well as organisations that serve oilseed growers.41 One intermediary in turn on-lent to 18,000 tobacco farmers the funds necessary to purchase seed, fertiliser, vehicles and technical assistance.42 In July 2017, a new secured transactions law was adopted.

B. Pre-2017 Secured Transactions Framework The 2017 passage of the Movable Property Security Interests Act (MPSI) sought to replace Zimbabwe’s fragmented secured transactions legal framework, which featured a mix of common law and civil law secured transactions devices.43 Some statutes governing aspects of secured lending, such as the Hire-Purchase Act 1957 (HPA), were rooted in English law, whereas other security devices, such as notarial bonds registrable under the Deeds Registries Act 1959, were civil law legacies, particularly of the Roman–Dutch variety.44 The notarial bond, transplanted from South African law, is based on the Roman law hypotheca.45 This civil law influence is a result of the extension of the law of South Africa to Zimbabwe.46 Zimbabwe’s courts often resort to South African precedents, but at times, apply both English and Roman Dutch law simultaneously.47 36 A Mambondiyani, ‘Bank Loans Beyond Reach for Zimbabwe Farmers Without Land Titles’ Thomson Reuters Foundation (20 July 2016). 37 ibid. 38 ‘Zim Land Reforms’ (n 34); ‘99-Year Leases Now Bankable, Transferable’ The Herald (8 February 2018). 39 ibid. 40 ibid. 41 ibid. 42 ibid. 43 PR Wood, ‘Comparative Financial Law: A Classification of the Worlds’ Jurisdictions’ in R Cranston (ed), Making Commercial Law: Essays in Honour of Roy Goode (Oxford, Clarendon Press, 1997). 44 N Locke, ‘Security Granted by a Company over its Movable Property: The Floating Charge and the General Notarial Bond Compared’ (2008) Comparative and International Law Journal of South Africa 136, 138. 45 GL Gretton, ‘Reception without Integration: Floating Charges and Mixed Systems’ (2003) Tulane Law Review 307, 329. See South Africa C4a. 46 By a proclamation from the British High Commissioner in 1891 decreeing that the law applicable in the Cape of Good Hope shall be the law of Southern Rhodesia (Zimbabwe’s former name). 47 See B Dube, ‘Roman-Dutch and English Common Law: The Indispensable Law in Zimbabwe’ (2014) Afro Asian Journal of Social Sciences 1, 2.

Pre-2017 Secured Transactions Framework 357

1. Pre-MPSI Security Devices Zimbabwe’s pre-MPSI secured transactions regime featured two main security devices: the possessory pledge and the notarial bond.48 The common law possessory pledge could be used to secure obligations with both tangible and intangible assets. To be effective, the parties must agree (verbally or in writing) to create a pledge over the asset and transfer possession of it to the secured creditor for the duration of the pledge agreement.49 Intangible assets such as shares and receivables could be pledged through a letter of cession, which had the same legal effect as physical delivery of a tangible asset.50 In order to enforce a pledge, the secured creditor could not proceed extrajudicially but had to engage the court sheriff or liquidator.51 In Kufandirori v Chipuriro,52 a pledge of a motor vehicle was evidenced by an agreement that gave the pledgee the right to effect a change of ownership in the event of a default. The pledgor defaulted, and the pledgee transferred ownership of the vehicle to himself. The court, applying South African Roman–Dutch jurisprudence,53 decided that a pledge agreement could not confer on the pledgee a right to appropriate the collateral himself without giving credit for any surplus value, as such a clause constitutes a pactum commissorium.54 A notarial bond could be used to create a non-possessory security interest.55 A debtor could create either a general notarial bond or a special notarial bond, the distinction being that a general notarial bond, in a manner somewhat similar to a floating charge,56 encumbered all the movable assets of the debtor (tangible and intangible, present and future), whereas a special notarial bond, in a manner similar to a fixed charge, encumbered specific tangible assets and required the notification and consent of the secured creditor before any dealings with the asset.57 Unlike floating and fixed charges, which can only be granted by companies, notarial bonds could be granted by individuals.58 A notarial bond had to be prepared by a notary public and executed by the parties in their presence. It had to be registered in the Deeds Registry in accordance with the Deeds Registries Act.59

48 South Africa C. 49 V Muza, ‘Securing Loan Agreements’ Financial Gazette (25 March 2010). 50 ibid; see also PG Industries (Zimbabwe) Limited v Jones Holdings (Private) Limited (2015) HC 10678/13, 6; RH Christie, Business Law in Zimbabwe 2nd edn (Kenwyn, Juta, 1998). 51 N Scott, L Becker and K Neluheni (eds) Lex Africa Guide to Insolvency and Restructuring in Africa 2015 (2015) 120–21. 52 Kufandirori v Chipuriro and Others [2004] ZWHHC 12 (HH). 53 The court referred to B van Heerden, DP Visser and CG Van der Merwe, Wille’s Principles of South African Law, 8th edn (Kenwyn, Juta, 1991). See South Africa B. 54 The term pactum commissorium refers to an agreement between the parties that if the debtor does not, within a certain time, release the pledged asset by paying the entire debt, the full property in the asset will irrevocably pass to the creditor in payment of the debt. Kufandirori v Chipuriro (n 52). A pactum commissorium is prohibited in Roman–Dutch law and many other civil law jurisdictions. See South Africa B and N Manfred and J Voet, The Common Law of South Africa: A Treatise Based on Voet’s Commentaries on the Pandects vol 2 (London, Butterworth, 1904) 930. In Ncube v Nyathi (2015) ZWBHC 60 (HH), the High Court held that a clause in an agreement and the subsequent transfer of property to the respondent was illegal for being premised on a pactum commissorium. 55 Muza (n 49). The Deeds Registries Act 2001, s 2 defines the notarial bond as ‘a bond attested by a notary public hypothecating movable property generally or specially’. 56 See South Africa C and C4a. 57 Scott, Becker and Neluheni (n 51). 58 Locke (n 44) 139. 59 Muza (n 49).

358 Zimbabwe During the law reform process that led to the MPSI, some stakeholders argued that there was no need to introduce a new security device and a collateral registry. Instead, they suggested that the process of creating and registering notarial bonds should be enhanced. The problem with this argument is that the security conferred by notarial bonds would continue to be affected by the existence of other types of security rights over the same assets. Moreover, the notarial bond registration process was very cumbersome and costly: mere digitisation of the records and registration processes would not eliminate all unnecessary costs, as will be explained below. The Deeds Registry Act was amended in 2017 to provide for the digitisation of the Deeds Registry records.60 Furthermore, the Registrar of the Deeds Registry possessed powers inconsistent with the fundamental principles underlying noticefiling systems. Thus, the functions and operation of the Deeds Registry did not accord with the Model Registry Provisions. For instance, section 6 of the Deeds Registries Act empowered the Registrar to require proof of any fact in connection with any matter performed by the Registry and to rectify any error in the name of any person or in the description of assets subject to the notarial bond.61 Zimbabwean law also recognised devices based on retention of ownership (such as hire-purchase agreements, which were governed by the HPA) and common law. Trading companies commonly inserted reservation of ownership clauses in the terms and conditions of their contract of sale of goods. In addition to the security devices generally known to other jurisdictions, such as the possessory pledge, and notarial bonds unique to Zimbabwe and a few other jurisdictions, Zimbabwe enacted a series of special pieces of legislation forming various parastatals62 with the objective of channelling credit to specific sectors of the economy. For instance, the Co-operative Societies Act automatically granted a charge to cooperative societies over goods produced wholly or partly by their members from funds lent by the cooperative society.63 Another example is the Agricultural Finance Act (AFA), which created the Agricultural Bank of Zimbabwe (ABZ) to provide credit to farmers.64 The AFA created two types of devices with the function of securing advances. First, it permitted the ABZ to make advances in relation to which the borrower gave a ‘note of hand’ if the advance was for a period not exceeding five years.65 A note of hand is similar to a promissory note and the obligation it embodies can be secured by immovable property.66 This note of hand security enjoyed superior statutory remedies upon default. The ABZ was entitled, after a period of 10 days from the posting of a registered letter of demand addressed to the borrower, to take possession of any collateral and dispose of it.67 The ABZ’s security agreements commonly included a clause to that effect. In several decisions, the Supreme Court of Zimbabwe upheld the

60 See Deeds Registries Amendment Act (2017). 61 Compare the position under the modern principles and the Model Law, ch 2, E1. 62 A company or other organisation owned wholly or partly by the Government. 63 Co-operative Societies Act (2001, as amended) s 30. 64 Agricultural Finance Act (2001, as amended). 65 Agricultural Finance Act 2001, s 35(1). ibid s 35(2) stipulates procedures to be taken by the corporation where an advance for a period not exceeding five years on the security of a note of hand is to be secured against any land or registered real right. 66 Registration requirements are set out in Agricultural Finance Act 2001, s 35(2). 67 ibid s 38(2). This power overrides even a court’s power to attach or levy execution on the property. See ibid s 38 (4).

Pre-2017 Secured Transactions Framework 359 constitutionality of such AFA-authorised clauses.68 However, in Tembani v Z imbabwe69 the SADC Tribunal, a supranational court for the SADC Member States, disagreed with the decision of the Supreme Court of Zimbabwe, citing a decision of the South African Supreme Court70 as authority to declare the relevant section of the AFA unconstitutional and in breach of the principles of international law. However, the SADC Tribunal stated that seizure of immovable property is different from seizure of movable property subject to a pledge, thus limiting its ruling to the seizure of an immovable. The second device relates to ‘livestock and implements’71 either being acquired by the borrower or already purchased by him. Where the ABZ lent money for the purpose of such acquisition or to finance such assets already purchased, whether financing was provided directly to the supplier or by way of reimbursement to the borrower, the livestock or implements became the property of the ABZ and not of the borrower.72 The borrower was not allowed to dispose of the livestock or implements without the consent of the ABZ but was deemed to be the owner of the assets for certain purposes, such as tort or criminal liability.73 If the buyer defaulted or became insolvent, the ADZ had powers of seizure.74

2. Creation and Registration Requirements Security devices in Zimbabwe other than the possessory pledge had to be created in writing. For instance, under the HPA, it was mandatory for hire-purchase, credit-sale and financiallease agreements to be executed in writing and signed by or on behalf of the parties.75 A financial lease is defined under the HPA as a written agreement for the letting and hiring of capital goods to be used by the purchaser for the purposes of their trade, where the seller is a commercial bank, accepting house, discount house or a finance house (registered or required to be registered under the Banking Act).76 The civil law influence is evident in some creation requirements; for example, notarial bonds must be attested to by a notary public.77 The cost of creating pre-reform security devices was exacerbated by stamp duties prescribed by the Stamp Duties Act and the Finance Act.78 Stamp duties were payable to the Commissioner General of the Zimbabwe Revenue Authority (ZIMRA) within seven days of

68 ibid; see also Tembani v Republic of Zimbabwe (SADC (T) 07/2008) [2009] SADCT 3 (discussing, as cited in the text, John Nyamukusa v Agricultural Finance Corporation (SC 174/94) and Augustine Renusu Chizikikani v Agricultural Finance Corporation (SC 123/95)). 69 Tembani v Republic of Zimbabwe (n 68). 70 Barkhuizen v Napier (2007) 5 SA 323 (CC). 71 Pursuant to the Agricultural Finance Act 2001, s 2, ‘implements includes any fencing, piping, machinery, equipment, installation, engine or vehicle, whether or not it is affixed to or incorporated in any immovable property’. 72 Agricultural Finance Act 2001, s 41(1). 73 ibid ss 42 and 43. 74 ibid ss 48 and 49. 75 The agreements must also indicate the cash price of the goods (‘cash price’ being the price at which the goods would be purchased outright); otherwise, the goods will be deemed to have been sold to the purchaser outright. See also Agricultural Finance Act 2001, s 5(1). 76 ibid s 3(1). 77 Deeds Registries Act (2001, as amended) s 2, and the Companies Act (1952) s 106(2). 78 See the Stamp Duties Act (2009, as amended) and Finance Act (2009) ch II.

360 Zimbabwe the execution of an instrument.79 An instrument that had not been properly stamped was not registrable in the Deeds Registry.80 Registration requirements were generally onerous and based on document-registration systems. The Deeds Registry is still available for the registration of various types of deeds that convey an absolute interest in immovable property as well as bonds that evidence conveyances of rights in immovable property as security, while registration of notices relating to securities in movable assets under the MPSI will be effected in the new Collateral Registry that is yet to be established. Registration of a bond in the Deeds Registry entailed the submission of three original or certified copies of the bond, within three months from the date of its execution.81 Under the pre-MPSI regime, the following types of bonds could be registered: (i) mortgage and surety mortgage bonds over immovables; (ii) notes of hand secured by an immovable, relating to advances typically made by the ABZ; (iii) general and special notarial bonds over movables; and (iv) cession of book debts.82 Accordingly, any security right given over a specific asset or fluctuating assets, including an outright transfer of receivables, was subject to registration in the Deeds Registry. The last two instruments will no longer be registrable in the Deeds Registry after the operationalisation of the MPSI Collateral Registry. There are two Deeds Registries in Zimbabwe: one in Harare and another in Bulawayo. Each serves its respective area as prescribed in section 3 of the Deeds Registries Act. Any notarial bond created by a company must be registered in the company’s register of mortgages, notarial bonds and debentures within 14 days of its creation, rather than three months.83 Prior to the MPSI, notarial bonds were also registrable in the Deeds Registry,84 and could be searched at the public counter for a fee of US$1 or directly over the internet for US$5. Typically, fees for electronic services are lower than for in-person services in order to relieve the burden on the Registry Office. The rationale for a fee structure that charged more for online searches of the Deeds Registry was baffling. The MPSI repealed the requirement of the Companies Act that notarial bonds be registered in the Deeds Registry, stipulating that the provisions of the MPSI shall regulate the registration of notarial bonds.85 Rather oddly, however, the new Insolvency Act of 2018 still refers to ‘a general bond over movables, registered in the deeds registry’ in section 89(15) which provides for the priority of a general notarial bond in the insolvency of the grantor. However, pre-MPSI many security devices were effective against third parties without any form of registration. For instance, rights of owners under hire-purchase and instalmentsales agreements subject to the HPA were not subject to any registration requirements. 79 See Stamp Duties Act 2009, ss 5 and 12. 80 ibid s 30. 81 Deeds Registries Act 2001, s 55. 82 South Africa C3a. 83 Companies Act, s 107(1)(a). 84 ibid s 106(2)(4) and (5). 85 MPSI, s 12(8), amending Companies Act, s 106 so that it only applied to bonds relating to immovable property. See also MPSI, s 3(4), which says that from the date that the MPSI comes into force, no security interests may be registered by way of notarial bond in a deeds registry. The MPSI also amends Companies Act, s 107 by adding a subsection (9) saying that for the avoidance of doubt, any bond registered for the purposes of binding movable property was to be effected ‘in terms of the MPSI’. Although the drafting leaves something to be desired, this amendment probably achieves the intended effect, which is that notarial bonds are no longer registrable in the issuing company’s register. Companies Act 1952, s 12(8).

Pre-2017 Secured Transactions Framework 361 Regarding enforcement, most of the pre-MPSI security devices in Zimbabwe (eg pledges) required resort to the courts upon default of the debtor. Article 71(3) of Zimbabwe’s Constitution expressly prohibits depriving a person of their property, except as otherwise prescribed in a law of general application. Although special notarial bonds could be enforced without resort to a court, the holders of general notarial bonds were required to obtain an order from the High Court authorising enforcement.86 This prohibition was further reinforced in individual statutes such as the HPA, which expressly barred any provisions in a hire-purchase agreement that permitted the seller to enter upon any premises for the purpose of taking possession of goods.87 In addition, where the agreement was lawfully terminated after the seller had been paid 50 per cent of the purchase price, section 20 of the HPA provided that the goods must be sold by a person appointed by a magistrate. However, because, under section 4, the HPA only applied to hire-purchase transactions in which the purchase price did not exceed 3,000 Zimbabwean dollars, its protections became effectively meaningless.88 In UDC Ltd v Chawara Kapenta Fishing Co-Op Ltd,89 the purchaser in a hire-purchase agreement brought a claim against the seller who had, upon the purchaser’s default, repossessed and sold a truck. The High Court found that the disposal by the respondent (the seller) breached section 20 of the HPA. On appeal, the Supreme Court ruled that section 20 of the HPA did not apply. The decision was based on the limitation in section 4 of the HPA. As a result, the parties’ rights were governed exclusively by their agreement, which permitted the seller to take possession of the truck and dispose of it.90 This holding seems to be at odds with the constitutional restriction of extrajudicial enforcement to cases where it is specifically authorised by a law of general application, rather than provided for in the parties’ agreement. Some parts of contract law also influenced Zimbabwe’s secured transactions jurisprudence. For example, the Contractual Penalties Act (CPA) could be used to strike down enforcement provisions in agreements. In the Kufandirori case, the court applied section 4 of the CPA, which allows the court to grant such relief as the court considers fair and just to the parties.91 In that case, the court invalidated clause (d) of the pledge agreement that granted the pledgee ownership on default, on the grounds that the asset was worth more than the secured obligation, and therefore the clause created a penalty.92 Generally, the legal costs of enforcing debts are extremely high for ordinary creditors.93 These costs are exacerbated by the slowness of the court system, which is understaffed and overburdened with cases.94 The pre-MPSI legislation and jurisprudence exhibited distrust 86 See South Africa C4d where the position is the same. See N Smith, ‘The Power of Notarial Bonds’ Caveat Legal. com, www.caveatlegal.com/the-power-of-notarial-bonds/. 87 The Hire-Purchase Act (1982) s 8. 88 In 2015, due to hyperinflation, Zimbabwe phased out its local currency (the Zimbabwean dollar) in favour of the US dollar, at a rate of 35 quadrillion Zimbabwean dollars to 1 US dollar. P Bulawayo, ‘Zimbabwe Phases Out Local Currency at 35 Quadrillion to US$1’ Reuters (15 June 2015) www.rt.com/ business/267244-zimbabwe-currency-compensation-hyperinflation/. 89 UDC Ltd v Chawara Kapenta Fishing Co-Op Ltd (2002) ZWSC 29. 90 The Court included obiter dicta in its ruling, calling for a review of the 3000-dollar limit, contending that it was no longer relevant to the needs of present-day society, as nothing of value can be purchased for that amount. ibid. 91 Contractual Penalties Act (1992) s 4(2). 92 ibid. 93 Law Development Commission, ‘Discussion Paper on Debt Collection’ (Ministry of Justice, Legal & Parliamentary Affairs, 16 December 2016) 5. 94 ibid.

362 Zimbabwe in extrajudicial remedies which also influenced the policies underlying the MPSI enforcement framework.

C. The Movable Property Security Interests Act In 2017, Zimbabwe enacted the MPSI. Under a project funded by the World Bank Group, one of the coauthors of this book provided advice to the group of stakeholders that prepared the MPSI and has been assisting with its implementation. The Act is yet to be brought into force pending the establishment of the Collateral Registry. During the enactment process, the purpose of the Bill was misunderstood even by high-ranking government officials. Bloomberg reported statements such as: ‘Commercial banks in Zimbabwe will soon be compelled to accept livestock such as cattle, goats and sheep as collateral for cash loans to informal businesses under a new law presented to Parliament.’95 Laws of this nature do not compel lenders to take movable assets as collateral; rather, they create an enabling environment to allow lenders to engage in such lending. In 2015, Zimbabwe undertook reforms to overhaul its commercial law framework, drafting new secured transactions and insolvency bills. Since both reform projects were funded by the World Bank Group, the provisions of both bills were coordinated to operate in harmony with each other. However, although the MPSI was enacted in 2017, the insolvency bill became embroiled in prolonged discussions, particularly at the ministerial level, which led to a significant departure from the original intention of creating a stand-alone and comprehensive insolvency framework. Instead, some of the insolvency provisions have been inserted into the Companies Bill, which is also under revision. The Companies and Other Business Entities Bill was gazetted in October 2018, but has not been enacted as of April 2019.96 The reform efforts were reinforced by the enactment of a law to establish a commercial court as a specialised division of the High Court.97 The Judicial Laws Amendment 2016, deriving authority from Article 171(3) of the Constitution, provided the basis for the establishment of a commercial court with ‘a view to speeding up and otherwise facilitating the settlement of certain suits or actions, especially suits or actions of a commercial nature’. The MPSI is based on the Model Law, with the Canadian PPSAs of Ontario and Saskatchewan providing secondary sources of inspiration. Although largely tracking the Model Law, the MPSI contains a number of notable deviations, such as the inability to dispose of tangible encumbered assets extrajudicially. Some of these deviations were a result of a conscious decision by a large group of stakeholders (such as the designation of registration as the superior method of third-party effectiveness),98 whereas others were introduced by the Ministry of Justice during the drafting process (eg the restrictions on extrajudicial enforcement). The enforcement part of the initial draft Bill copied the provisions of the Model Law, and was supported by the group of stakeholders, especially from the lending community. However,

95 D

Kumbuka, ‘New Law Compels Zimbabwe Banks to Accept Cows as Collateral’ Bloomberg (11 April 2017). ‘Companies and Other Business Entities Bill 2018’ Veritas, www.veritaszim.net/node/2400. 97 See ‘Zimbabwe Moves to Set Up Commercial Court’ The Source (26 June 2017). 98 See C1, C3c and D below. 96 See

The Movable Property Security Interests Act 363 the ultimate version of the Bill submitted to the Parliament restricted extrajudicial enforcement to essentially security rights in intangible assets, also reflecting some concerns that the Parliament might reject the entire Bill given the historical sensitivity of this issue under the Zimbabwean law.99 The MPSI is unusually structured, with the Act consisting of 13 sections supplemented by two schedules. The First Schedule, entitled ‘Registration of notices with respect to movable security interests’, governs the operation of the Collateral Registry and the creation of security interests. It is unclear why the provisions on creation were inserted in a schedule that, for the most part, covers various aspects of registering notices. The Second Schedule, entitled ‘Legal regime pertaining to perfected movable property security interests’, regulates priorities, mutual rights and obligations of the parties, enforcement and applicable laws. The reform process was spearheaded by the RBZ, which was also designated as the host of the Collateral Registry.

1. The Main Body of the Act As noted above, the MPSI contains 13 sections which, inter alia, define its scope, set out the purpose of the Collateral Registry, and provide the transitional rules. Section 2 contains a number of definitions whose allocation between the MPSI and its Schedules does not seem to follow any particular logic. The MPSI uses a slightly different terminology to the Model Law. ‘Debtor’ is used instead of ‘grantor’, ‘security interest’ instead of ‘security right’, and ‘perfection’ instead of ‘third-party effectiveness’: this follows the nomenclature in the Canadian PPSAs.100 The MPSI also includes a definition of ‘financial lease’ from the Legislative Guide, a definition that was not carried over to the Model Law. Furthermore, section 2 defines movable property to include assets that may be or are affixed to immovable property. In this respect, the MPSI deviates from the Legislative Guide (the Model Law does not have any special rules on attachments to immovable property) in treating fixtures as movable property, rather than assigning them a special (hybrid) status, as is the case under the Legislative Guide. However, movable property does not include growing crops because Article 72(1) of the Constitution specifically defines land to include anything permanently attached to or growing on land. According to this definition, crops growing on the land are part of the immovable and thus outside the scope of the MPSI, until they are harvested. Section 3 generally sets out the scope of application of the MPSI, both in terms of scope and exclusions. The scope of the Act, as made clear by section 3(1) is limited to security interests in movable property. A functional approach is taken to what constitutes a security interest. This is made clear by the words at the end of section 3(2), which include in the scope of ‘security interest’ ‘any other transaction that secures payment or performance of an obligation’. In addition to this general statement, section 3(2) explicitly provides that rights created by means of a chattel mortgage, notarial bond, hypothec, note of hand, charge, secured debenture, pledge, trust indenture, trust receipt and financial lease constitute security interests. In including such a list, the MPSI again draws inspiration from the

99 See

100 See

C3d below. ch 1, D.

364 Zimbabwe Canadian PPSAs and does not rely only on the generality of the definition of ‘security right’ which provides for the functional approach in the Model Law.101 Section 3(1)(b) of the Saskatchewan PPSA provides that ‘the Act applies without limiting the generality of clause (a), to a chattel mortgage, conditional sale, floating charge, pledge, trust indenture, trust receipt, or to an assignment, consignment, lease, trust or transfer of chattel paper that secures payment or performance of an obligation’. Subsection 3(4) states: ‘Subject to paragraph 7 of the Second Schedule, security interests may be perfected by any formal or informal mode recognised by the common law.’ This provision was included at a late stage in the development of the Bill and was informed by discussions with the Ministry of Justice. Its inclusion in the scope section is rather odd, but its main purpose is clear: to ensure that security interests could still be perfected by the common law mechanisms available pre-MPSI, such as possession. A stakeholder forum had expressed some concern with the methods of perfection other than registration that were included in the first draft of the MPSI based on the Model Law. In order to make the Collateral Registry a more transparent and comprehensive source of information with respect to security interests, registration was made the only means of perfection in the subsequent draft of the Bill that had initially recognised the Model Law alternatives to registration. The late inclusion of section 3(4) does not undermine this general purpose since the MPSI gives priority to security interests perfected by registration, even when registration is effected after a competing security interest has already been perfected by a common law method. There is thus a strong incentive to register. Section 3(5) excludes a number of security interests, provision for which is found elsewhere, in legislation or otherwise. Section 3(5)(a) is particularly interesting and expressly excludes book-entry securities held through the Central Securities Depository (intermediated securities) from the scope of the MPSI. The Chengetedzai Depository Company (‘Chengetedzai’) runs Zimbabwe’s sole central securities depository (CSD), which went live in September 2014.102 Chengetedzai’s CSD Operational Rules govern pledges and hypothecation of securities.103 Under section 37(3) of those Rules, the CSD shall effect the pledge ‘by electronically locking or freezing the securities held in the account of the pledgor by transferring the securities in a subaccount marked “PLEDGED” and shall issue an advise of the same to the pledgor and pledgee’. This process must be completed within five days after receipt of a request from the pledgor.104 On the registration of a pledge, the securities will be ‘locked away’ so that the pledgor will have no access to them, even to create a junior pledge. The pledgor remains entitled to collect any distributions, such as dividends, and retains the power to exercise any shareholder rights.105 Sections 4–7 address the fundamentals of registration, including the establishment of the Collateral Registry and its purpose and functions, while deferring to the Schedules for detailed provisions. Section 9 ambiguously provides: ‘Every registered notice … shall

101 ML, Art 3(kk). This definition is replicated in MPSI, s 2. 102 See ‘Chengetedzai Depository Company Limited, Zimbabwe (CSD)’ CMA (2015) www.cma.se/customers/ exchanges/chengetedzai-depository-company.html. 103 Central Securities Depository Operational Rules (2014). 104 ibid s 37(3). 105 ibid.

The Movable Property Security Interests Act 365 be deemed to be the definitive record of any right or obligation recorded therein.’106 This ambiguity was highlighted by the CEO of the Bankers’ Association of Zimbabwe, Mr Sij T Biyam, during a workshop on the MPSI held in Harare on 21 August 2017. He correctly pointed out that a registered notice does not evidence any rights or obligations, as that evidence is usually contained in the security agreement. The ambiguity might have been introduced by the drafters who visualised the function of the Collateral Registry as registering underlying security agreements, in a manner similar to the Deeds Registry. Under section 10, the Collateral Registry shall not be liable for any actions taken in good faith, but an affected person may recover damages for any negligent actions. This liability standard does not reflect any of the three options in Article 32 of the Model Registry Provisions107 and leaves the Collateral Registry open to much more open-ended liability than any of the three Model Law options. MPSI section 8 provides important bases for the enforcement of security interests, leaving the details for Part V of the Second Schedule. Subsection 1 states: ‘Every registered notice of a security interest is deemed to be a liquid document enforceable by way of provisional sentence proceedings.’ This provision, inserted by the Ministry of Justice, was intended to ensure expedited judicial enforcement of security interests. Expedited judicial enforcement is also encouraged by Chapter VII of the Model Law, but as an alternative to extrajudicial remedies. However, section 8(1) of the MPSI is effectively a replacement for the extrajudicial remedies. Accordingly, with some exceptions (eg the collection of receivables),108 the MPSI does not permit extrajudicial enforcement, although, if the agreement so provides, it does permit a secured creditor to take possession as an interim measure.109 The terms used in section 8 require some explanation. Under Zimbabwean law, a document is liquid ‘if it demonstrates, by its terms, an unconditional acknowledgement of indebtedness in a fixed or ascertainable amount that is due’.110 A liquid document gives rise to a rebuttable presumption that the defendant is indeed indebted to the plaintiff in the amount stated in the document.111 The holder of a liquid document may use ‘provisional sentence proceedings’ to seek the enforcement of a debt.112 A provisional sentence proceeding is an extraordinary procedure in terms of which a plaintiff in possession of a document showing an indebtedness for a liquidated sum of money, for example cheques, negotiable instruments and any other form of document, including an acknowledgement of debt, may obtain provisional judgment for the amount payable on the face of the document prior to the trial date.113

Thus, section 8(1) confers on a registered notice of a security interest the status of a liquid document; that is, it is an unconditional acknowledgement of debt for a fixed or ascertainable

106 MPSI, s 9(1). 107 MRP, Art 32. 108 See section C3 and C4 below. 109 S 8(2) discussed below. 110 Twee Jonge Gezellen (Pvt) Ltd & Anor v Land & Agricultural Development Bank of South Africa & Anor (2011) ZACC 2, cited in Engen Oil Zimbabwe (Pvt) Ltd v Dowood Services (Pvt) Ltd, (2015) HC 1663/15. Also defined as a document ‘which evidences by its own terms, without the need for extrinsic evidence, an unconditional acknowledgement of indebtedness in an ascertained sum of money the payment of which is due’. Interfin Banking Corp Ltd (in liq) v Veanarcy (Pvt) Ltd (2013) HH-388-13. 111 Engen Oil Zimbabwe (Pvt) Ltd v Dowood Services (Pvt) Ltd; Sibanda v Mshapaidze HC 2961/09. 112 Engen Oil Zimbabwe (Pvt) Ltd v Dowood Services (Pvt) Ltd (n 112) 2. 113 ‘Kanokanga & Partners: Attorneys, Legal Practitioners, Notaries & Conveyancers’ (brochure, 2015) s 6.1.

366 Zimbabwe amount, which is enforceable through an expedited judicial procedure commanding the debtor to pay the sum due as established in the provisional sentence. However, it is uncertain how a registered notice can constitute a liquid document since, unlike negotiable instruments and acknowledgements of debts, it does not state the amount owed. The reference to the registered notice should be read to encompass the obligation secured with the collateral to which the registered notice relates. In other words, the registered notice would constitute a liquid document only in conjunction with the related agreement. To the extent provided for in a security agreement, section 8(2) authorises the secured creditor to repossess an encumbered asset. However, the secured creditor must obtain a court order so as to complete the enforcement. Accordingly, as a precaution, the secured creditor may seize an asset and thus minimise the risk that it might be damaged, hidden or disposed of. Although this provides some protection to secured creditors, the MPSI does not provide any exceptions to the general rule of requiring a court order to effect the final disposal, such as in the case of perishable commodities for which swift disposal would be necessary to preserve their value. Section 12 sets out a long list of consequential amendments to the various laws that regulated some aspects of secured transactions in Zimbabwe prior to the MPSI’s entry into force. This approach corresponds to Article 101 of the Model Law. As well as the common law, the following statutes comprised the fragmented secured transactions regime of Zimbabwe: the Bills of Exchange Act, the HPA, the Agricultural Finance Corporation Act, the Grain Marketing Act, the Tobacco Industry and Marketing Act, the Deeds Registries Act, the Companies Act, the Co-operative Societies Act, the Infrastructural Development Bank of Zimbabwe Act, and the Banking Act. Although section 12 contains amendments to most of these statutes (and others), none of them has been repealed because the MPSI only affects some aspects of their application. The provisions in those statutes referring to various security devices such as pledges and liens have been amended to conform to the MPSI, altering their effect to references to a security interest subject to the MPSI. Section 12 added a new subsection 3(2) to the previous Insolvency Act stating that a ‘notarial bond’ was, from the time the MPSI came into force, to be construed as a registered security interest under the MPSI. This section is omitted in the new Insolvency Act 2018, although the definition of ‘security interest’ in that Act does include the phrase ‘any equivalent right of property under the MPSI’. The position of the general notarial bond is particularly unclear. The MPSI appears to treat it as a security interest, whereas the Insolvency Act does not seem to do so and provides (as under the previous Insolvency Act) for its position in the waterfall of payments from the ‘free assets of the insolvent’s estate’.114 Finally, section 13 contains the transitional provisions, although they are severely limited as compared to Chapter IX of the Model Law. Section 13 defines ‘prior security interest’ and requires registration of such interests in the Collateral Registry within nine months of the MPSI’s entry into force in order to preserve their perfection and priority. Subsection (2) provides that security interests perfected by any formal or informal mode recognised by common law need not be registered in the Collateral Registry, since the MPSI recognises those modes of perfection as alternatives to registration, albeit inferior to registration in

114 Insolvency Act (2018) s 89(15). The practical effect, however, is similar to the provisions about preferential creditors in the insolvency legislation in some common law jurisdictions.

The Movable Property Security Interests Act 367 terms of their priority effect. Accordingly, a secured creditor in possession of the collateral under a common law pledge need not take any action for its pledge ‘to transition’ into the MPSI as a security interest.115

2. The First Schedule This Schedule is divided into three Parts: (I) Preliminary; (II) Creation of Security Agreement; and (III) Registration Requirements. The Preliminary Part is limited to a single section that sets out a number of definitions of terms, namely ‘designated field’, ‘inventory’, ‘intellectual property’, ‘negotiable document’, ‘negotiable instrument’, ‘perfected security interest’ and ‘registrant’. This list is a mixture of definitions relating to general provisions of the MPSI (eg inventory) and those specific to registration (eg designated field). One can only assume that the list is supposed to relate to terms only found in the First Schedule, since the definitions are said to apply ‘In this Schedule’. However, this is not necessarily the case as some of the terms (eg ‘inventory’) are found in the Second Schedule as well. Like Chapter II of the Model Law, MPSI Part II deals with the creation of a security interest. However, some terminology used in this Part should have been clearer. For instance, the title of this Part should have referred to ‘creation of a security interest’ rather than ‘creation of security agreement’. For a security interest to be created, the security agreement must be executed, signed by the parties and contain the prescribed information. The title of paragraph 2 ‘Creation and Registration of Security Agreement’ is not only ambiguous, but also incorrect, as the MPSI registration provisions do not contemplate or require that a security agreement be submitted to the Collateral Registry. As mentioned above in connection with the statement in section 9 of the MPSI that a registered notice is definitive proof of rights and obligations,116 the drafter might have been swayed by the current requirements for registration in the Deeds Registry. Paragraphs 2(1)–(4) are modelled on Article 6 of the Model Law, with some redrafting. For example, the minimum requirements set out in paragraph 2(4) for a security agreement to create a security interest are nearly identical to Article 6 of the Model Law. One difference is that the requirement that the security agreement should describe the ‘secured obligation’ does not apply to an outright transfer of a receivable:117 there is no secured obligation in such a transaction and so this qualification is entirely correct. Paragraph 2(6) is specific to Zimbabwe. It provides that a ‘security agreement … is enforceable, despite not complying with formalities imposed by other laws’, thereby ensuring that the pre-MPSI requirements on the constitution of various security devices, especially attestation by notaries and the payment of stamp duties, no longer apply to security agreements under the MPSI. This was a conscious decision by the group of stakeholders, who argued that the formal requirements unduly increase the cost of credit and delay the execution of transactions. Paragraph 2(3) (which deals with secured obligations and encumbered assets), paragraph 2(4) (which covers the description of collateral), paragraph 2(5) (which addresses the creation of security

115 See

MPSI, s 13. discussion of MPSI, s 9 in section C1 above. 117 MPSI, First Schedule para 2(4)(c). 116 See

368 Zimbabwe interests in tangible assets covered by negotiable documents and assets used in connection with intellectual property rights) and paragraph 2(6) (which addresses the creation of security interests in proceeds) all incorporate the relevant provisions of the Creation Chapter of the Model Law.118 The only difference is in paragraph 4(3), which requires serial-numbered collateral to be described by a serial number, as well as the make and the name of the manufacturer, unless it is held as inventory. The MPSI, as will be explained below, contemplates serial-number indexing of registrations, unlike the Model Law. However, ‘serial number descriptions’ are not necessary in security agreements, so that collateral that is a motor vehicle may be identified simply as ‘motor vehicle’. Furthermore, ‘serial-numbered collateral’ is not defined in the MPSI, a critical matter left for the Regulations. Part III provides a comprehensive set of rules related to registrations, searches, and other functions of the Collateral Registry. Paragraphs 7–25 are based on the Model Registry Provisions,119 with some redrafting to improve coordination with other legislation. Paragraph 23, which addresses the transfer of a security interest that is effective whether or not an amendment notice to reflect the full or partial transfer is registered, reflects a PPSA approach.120 On the transfer of a secured obligation, whether in full or partially, that carries with it the security interest, the registered notice may be amended to add the transferee. However, failure to register the transfer exposes the transferee to the risk that the transferor will amend or cancel the registration; the transferor retains such powers until an amendment identifying the transferee is registered. Two options are provided in Article 28(1) of the Model Registry Provisions, allowing enacting States to choose between a strict search logic that retrieves only exact matches of the criterion entered by a searcher, and a flexible search logic, which also retrieves close matches. Zimbabwe chose the former.121 As noted earlier, the MPSI also provides for serial number indexing,122 which is the capability to retrieve registrations according to a serial number, and paragraph 22(6) addresses the effect of an error in the registered serial number.123 It provides that any error in the serial number registered renders the registration ineffective but only as against a buyer or lessee of the relevant asset. The legal effect of errors in a serial number registered under the MPSI follows the approach of the Canadian PPSAs.124 Furthermore, serial-numbered collateral do not have to be identified by a serial number when held as inventory. Any omission or error in the serial number, to the extent the registered notice can be retrieved by a search according to the debtor identifier, will have no impact on the effectiveness of the registered notice as against competing claimants other than buyers and lessees. 118 ML, Arts 7, 8, 9, 16, 17, 19 (the last of these is actually in Chapter III of the ML, which deals with effectiveness of a security right against third parties). 119 See ch 2, E. 120 Saskatchewan PPSA, s 45. 121 MPSI, First Schedule para 21(1)(a). 122 The serial number must be included in the notice, since the description requirements from para 4 also apply to a notice. MPSI, First Schedule para 14(1). 123 On organisation of the registry record by serial numbers see the UNCITRAL Registry Guide paras 131–35. See also M Dubovec and H Sigman, ‘Some Thoughts (and Facts) about the Process of Secured Transactions Reform, with Special Emphasis on Registration, the Key to Achievement of Reform’s Goals’ in SV Bazinas and O Akseli (eds), International and Comparative Secured Transactions Law: Essays in Honour of Roderick A MacDonald (London, Bloomsbury, 2017) 157, 177–79. 124 Saskatchewan PPSA, s 30(6).

The Movable Property Security Interests Act 369

3. The Second Schedule The Second Schedule is divided into six Parts addressing respectively: (I) some definitions; (II) special provisions as to creation and transfer of security interests; (III) priorities; (IV) rights and obligations of parties and third-party obligors; (V) enforcement of security interests; and (VI) applicable law. Part I contains a number of definitions of terms such as ‘acquisition security interest’, ‘competing claimant’, ‘consumer goods’ and ‘obligor of the receivable’. All of these definitions are drawn from the Model Law. Paragraph 1 of this Schedule also defines an attachment to immovable property as ‘a tangible asset that, despite being physically affixed to immovable property, is treated as movable property by virtue of not having lost its separate identity, including an installation defined in section 19(2) of the Hire Purchase Act’. Assets that fall under this definition are treated as movable assets within the scope of the MPSI. Consequently, a right in the immovable to which the tangible asset is attached will not extend to the tangible asset.

a. Part II – Special Creation Provisions Part II incorporates various provisions of the Model Law. For instance, paragraph 2, incorporating Article 10 of the Model Law, provides that a security interest extends to identifiable proceeds and sets out the rule applicable to commingled money or funds. Paragraph 4 replicates Articles 13 and 15, which override the effect of anti-assignment clauses. As a result, clauses restricting or prohibiting the transfer of a receivable are ineffective as against the secured creditor.125 Oddly, paragraph 6, which addresses post-registration transfer of collateral, was placed in the Second Schedule with the other special creation provisions, unlike the Model Law, which deals with this matter in Article 26 of the Model Registry Provisions. From the three options set out in Article 26, Zimbabwe chose option B and selected a period of 10 days for the secured creditor to register an amendment adding the buyer as the debtor. Accordingly, to preserve its priority against competing claimants, the secured creditor must register an amendment adding the buyer of the collateral to the registered notice within 10 days from the time the secured creditor acquired knowledge of the transfer and the identifier of the buyer.

b. Part III – Priorities As in Part II, the provisions of this Part incorporate many Articles of the Model Law. For instance, paragraph 9 disregards knowledge as a factor that may determine priorities under the MPSI, replicating Article 45 of the Model Law, and paragraph 10 extends the priority of a security interest to future advances and future assets, emulating Article 44 of the Model Law. The main difference is in the fundamental priority rule of ‘first to register or first to perfect’ under Article 29 of the Model Law, which could not be reflected in the MPSI since, although section 3(4) of the MPSI preserves some methods of perfection other than registration (such as taking possession), they do not constitute a priority point

125 See

ch 1, E4a.

370 Zimbabwe as against registration in the Collateral Registry.126 Accordingly, the fundamental priority rule of the MPSI is ‘first to register’. Paragraph 7(3) provides a special priority rule to determine the priority between a security interest perfected by registration and a competing security interest perfected by any formal or informal mode recognised by common law. Regardless of the time of perfection, registration in the Collateral Registry is the superior mode of perfection, as it confers priority even over a security interest perfected under a common law mode earlier in time. A security interest perfected only under common law will remain effective against the insolvency trustee and other non-consensual creditors and will have priority over any other security interests subsequently perfected under the common law. The recognition of common law modes of perfection causes some challenges in the application of the priority rules. For instance, paragraph 20(1) provides a priority rule that allows a consensual transferee of a negotiable instrument to take free of a security interest perfected by registration. The MPSI lacks a corresponding priority rule for a situation where a security interest in the negotiable instrument is perfected under the common law. The remaining paragraphs of Part III incorporate the provisions of the Model Law on the priority of acquisition security rights,127 as rights of buyers, lessees and licensees of the collateral,128 and rights of non-consensual creditors.129 The term ‘non-consensual creditors’ is not defined, but means creditors who have a claim to assets of the grantor which does not result from an agreement, such as judgment creditors and tax authorities. Such creditors can register a notice in the Collateral Registry and their priority is then determined by the rules discussed above, qualified by paragraph 14(2).130

c. Part IV – Rights and Obligations of Parties and Third-Party Obligors Paragraphs 21–29 of this Part incorporate the provisions of Chapter VI of the Model Law. They deal with the duty of the person in possession of the collateral to exercise reasonable care,131 the duty to return the collateral to the debtor upon termination of a security interest,132 various rights, obligations and defences of obligors on receivables,133 as well as the rights and duties of financial institutions that maintain deposit accounts subject to security interests.134 Paragraph 27, which provides for the right to recover damages for the failure of any person to discharge its duties or perform its rights under any provision of the MPSI in a reasonable manner, does not have a counterpart in the Model Law. Furthermore, subparagraph 27(3) authorises the Regulations to prescribe statutory damages for the registration of a notice without authorisation of the debtor.

126 See

C1 above. Arts 38, 39, 41 and 42. 128 ibid Art 34. 129 Based on ML, Art 37, which refers to ‘judgment creditors’. 130 MPSI, Second Schedule para 14. 131 ML, Art 53. 132 ibid Art 54. 133 ibid Arts 61–65 (with some modifications). 134 ibid Art 69. 127 ML,

The Movable Property Security Interests Act 371

d. Part V – Enforcement of Security Interests This Part supplements section 8 of the MPSI examined above.135 It includes six paragraphs, five of which are drawn from the Model Law. Given the limitations on extrajudicial enforcement with respect to tangible assets, many provisions of the Model Law (eg the right of the secured creditor to obtain possession of the collateral under Article 77) covering out-ofcourt enforcement were not included. Even though paragraph 30(1)(b) allows the parties to provide for any other remedy in their security agreement, a stipulation (by the parties) that the secured creditor could repossess and dispose of the collateral extrajudicially would not be enforceable. Paragraph 31 addresses a situation where the secured creditor seeks to enforce its rights against an attachment to immovable property, reflecting recommendation 164 of the Legislative Guide, which was not incorporated into the Model Law. It allows the secured creditor to enforce its rights either under the MPSI or the applicable immovable property law when the obligation is secured by both movable and immovable property. This approach is also consistent with the PPSAs.136 However, it must be read in conjunction with section 8, which requires the secured creditor to obtain a court order before effecting the final disposal of the encumbered asset. There is no such requirement in the PPSAs. Paragraph 30137 provides a menu of enforcement options that include: (i) the remedies under this Part; (ii) any right provided for in the security agreement; and (iii) any right provided under other law, except to the extent it is inconsistent with the provisions of the MPSI. Subparagraph 30(3) renders any pre-default waiver of the debtor or borrower’s rights invalid. Paragraphs 32 and 33 cover the right of redemption and the right to take over the enforcement process, respectively. Both sections implement Articles 75 and 76 of the Model Law, which allow affected parties to terminate the enforcement process by satisfying the secured obligation in full, and grant the senior secured creditor the power to take over an enforcement process initiated by a junior secured or non-consensual creditor. However, subparagraphs 33(3)–(5), concerning enforcement by non-consensual creditors, do not have an equivalent in the Model Law (which does not deal specifically with non-consensual creditors). These subsections require a non-consensual creditor to notify a senior secured creditor who has registered a notice against the collateral prior to commencing any enforcement action. The secured creditor has a period of five working days after receipt of the notification to decide whether to take over enforcement. These provisions were inserted based on suggestions made during a stakeholder forum where a number of representatives of financial institutions expressed concern that other creditors, especially judgment creditors, could liquidate the collateral for an unreasonably low price and then refuse to turn over the proceeds to a higher-ranking secured creditor. Paragraph 34 provides for the acquisition of collateral in partial or full satisfaction of the secured obligation, implementing Article 80 of the Model Law. Accordingly, the secured creditor may propose to the affected parties, such as the debtor and other secured creditors, to take ownership of the collateral in exchange for fully cancelling or partially reducing the secured obligation. Arguably, this paragraph does not override the effect of section 8, which requires the secured creditor to obtain a court order before disposing of the encumbered

135 See

section C1 above. Saskatchewan PPSA, s 55(4)(b). 137 Based on ML, Art 72. 136 cf

372 Zimbabwe asset. On its terms, section 8 would seem to be limited to repossession of the encumbered asset in preparation for final disposal, requiring a court order whether or not the grantor consents to the repossession and final disposal. Given the policy concerns expressed with respect to extrajudicial enforcement against tangible assets, it would be odd to read paragraph 34 independently of section 8, so as to authorise the secured creditor to appropriate an encumbered tangible asset without a court order. Thus, although paragraph 34 would seem to authorise the secured creditor to proceed extrajudicially against an intangible asset, a combined reading of section 8 and paragraph 34, informed by the underlying spirit of the MPSI, would suggest that the secured creditor must obtain a court order before appropriating a tangible asset. Paragraphs 35 and 36 provide for the collection of receivables (as well as other ‘payment rights’ such as bank accounts, negotiable instruments and securities) by secured creditors, including outright transferees, following Articles 82 and 83 of the Model Law. There is no requirement in the MPSI for the secured creditor to obtain a court order before exercising its rights under paragraphs 35 and 36.138

e. Part VI – Applicable Law This Part contains eight paragraphs that implement the Articles of the Model Law with respect to the law applicable to security interests in tangible and intangible assets as well as the proceeds of such assets,139 the law applicable to enforcement of security interests,140 the law applicable to relationships of secured creditors and third parties141 as well as the rules to determine the location of the debtor142 and relevant time for determining that location.143 Paragraph 37 follows the Model Law144 by providing that parties may choose the applicable law to govern their mutual rights and obligations, but in relation to all matters that affect third parties, such as perfection of security interests, the MPSI provides mandatory rules. Thus, paragraph 38 provides that the law applicable to the creation, perfection and priority of a security interest in a tangible asset is the law of the country where the asset is located.145 Under paragraph 39, the law applicable to the creation, perfection and priority of a security interest in an intangible asset is the law of the country in which the debtor is located.146 Paragraphs 42 and 43 then provide the necessary rules to establish the debtor’s location at the relevant time.

D. Final Remarks Zimbabwe has joined the growing list of African economies that have reformed their secured transactions laws. Not only has the fragmented framework been brought under a

138 MPSI,

Second Schedule paras 35–36. Arts 85, 86 and 89. 140 ibid Art 88. 141 ibid Art 96. 142 ibid Art 90. 143 ibid Art 91. 144 ibid Art 84. 145 See ML, Art 85. 146 See ibid Art 86. 139 ML,

Final Remarks 373 single umbrella law, but the MPSI is based on the Model Law. Many provisions of the Model Law were incorporated as adopted and others with minor and substantively insignificant drafting deviations. Zimbabwean stakeholders also considered and selected from the various options provided by the Model Law, such as on the post-registration transfer of the collateral.147 There are certain areas where the MPSI notably departs from the Model Law, such as by designating registration as the superior method of perfection (supplemented by formal and informal common law modes of perfection) and the unavailability of extrajudicial enforcement with respect to tangible assets. The former was driven by a proposal from the group of stakeholders representing the lending community who viewed the Model Law’s alternative perfection methods, such as control agreements, as undermining the transparency and informational value of the Collateral Registry. Zimbabwe opted for a simplified perfection regime that in turn led to the simplification of the rules governing priorities. Although the stakeholders overwhelmingly supported the availability of extra-judicial remedies, due to concerns about their constitutionality and political considerations for the passage of the Bill by Parliament, extrajudicial remedies were limited to only intangible assets. The secured creditor may take possession of an encumbered asset, but its final disposition is suspended until it obtains a court order. The willingness to extend credit at a reasonable cost would thus depend on the efficiency of the Zimbabwean court system to issue orders expeditiously. This is a concern particularly with respect to the types of assets that many prospective Zimbabwean borrowers own, such as crops and livestock. Overall, the MPSI significantly modernises Zimbabwe’s secured transactions regime, which relied heavily on the security devices of Roman–Dutch law, as applied primarily in South Africa. Against this backdrop, the enactment of the MPSI marks the first successful adoption of the Model Law in a legal system of this nature. Zimbabwe’s MPSI has positioned itself as a model for South Africa, reversing the role of the two jurisdictions, as Zimbabwe’s laws have historically been influenced by the laws and judicial decisions of South Africa. Yet, the economic impact of the reforms remains uncertain due to the lack of bank capital that could be deployed for secured loans. The effect of the MPSI is also partially clouded by the ongoing reforms of the companies and insolvency laws that, in their most recent drafts, have departed from the original drafts, which were based on the best international practice and fully coordinated with the MPSI. Finally, almost two years have passed since the enactment of the MPSI without any progress with respect to establishing the collateral registry or capacity building, which raises additional questions.

147 cf

ibid, MRP, Art 26 and MPSI, Second Schedule para 6.

374

part v Final Chapters

376

17 A Modern International Approach to Equipment Financing in Africa: The Cape Town Convention and its Protocols ANNA VENEZIANO AND WILLIAM BRYDIE-WATSON*

A. Introduction The Cape Town Convention on International Interests in Mobile Equipment (‘the Cape Town Convention’ or ‘the Convention’), developed by UNIDROIT,1 is an innovative commercial law treaty system offering an internationally uniform legal framework to facilitate leasing and financing for the acquisition or use of specific types of high-value equipment.2 It does so by enabling parties to a financing or leasing agreement to create an international interest over the covered equipment, enforceable in all Contracting States through its registration in a dedicated international electronic registry, thereby reducing credit risk and enhancing predictability in these transactions. So far, three Protocols to the framework Cape Town Convention have been adopted, dealing with, respectively, airframes, aircraft engines and helicopters (the Aircraft Protocol),3 railway rolling stock (the Luxembourg Rail Protocol)4, and space assets, namely * Anna Veneziano is Deputy Secretary-General, UNIDROIT and Professor of Comparative Law, University of Teramo, Italy. William Brydie-Watson is Senior Legal Officer, UNIDROIT. 1 Founded in 1926, UNIDROIT is an intergovernmental organisation with 63 Member States from all continents, and is active in the harmonisation of private and commercial law. For more information see UNIDROIT, ‘History and Overview’ (22 April 2019) www.unidroit.org. 2 ‘Convention on International Interests in Mobile Equipment’ (jointly adopted at a diplomatic Conference in Cape Town in 2001 by UNIDROIT and the ICAO) (CTC). For a comprehensive overview of the Convention system, see R Goode, ‘From Acorn to Oak Tree: the Development of the Cape Town Convention and Protocols’ (2012) 17(4) Uniform Law Review 599. 3 ‘Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Aircraft Equipment’ (jointly adopted at a diplomatic Conference in Cape Town in 2001 by UNIDROIT and ICAO) (‘the Aircraft Protocol’). See R Goode, Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Aircraft Objects: Official Commentary, 4th edn (Rome, UNIDROIT, 2019) (‘the Aircraft Protocol Official Commentary’). 4 ‘Luxembourg Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Railway Rolling Stock’ (jointly adopted at a diplomatic Conference in Luxembourg in 2007 by UNIDROIT and OTIF) (‘the Luxembourg Rail Protocol’ or ‘Rail Protocol’). See R Goode, Convention on International Interests in Mobile Equipment and Luxembourg Protocol Thereto on Matters Specific to Railway Rolling Stock: Official Commentary, 2nd edn (Rome, UNIDROIT, 2014) (‘the Rail Protocol Official Commentary’).

378 A Modern International Approach to Equipment Financing in Africa satellites and their components (the Space Protocol)5. In addition, the fourth Protocol to the Convention on Matters Specific to Mining, Agricultural and Construction Equipment (‘the MAC Protocol’) has reached the final stage of development and will be adopted at a diplomatic conference in Pretoria (South Africa) in 2019.6 The two-tier architecture combines the framework Convention with each of the Protocols, depending on the category of equipment, and represents a unique feature of the Cape Town Convention model. One of its main advantages is that it introduces a basic uniform set of rules that the Protocols adapt to the specific requirements and needs of the category of equipment and related market.7 The Convention, however, will not apply in regard to any category of object until the entry into force of the Protocol relating to that category. The Aircraft Protocol entered into force in 2006, triggering the application of the Convention itself as regards aircraft objects. As of April 2019, these two instruments have reached the impressive number of 79 and 76 Contracting States respectively and one Regional Economic Integration Organisation (the European Union), with continuous growth in membership each year. The State Parties represent all continents, various legal traditions and different stages of economic development, including most emerging markets and major aviation economies.8 Equally impressive are the numbers regarding the practical impact of the treaties in aviation practice, with over one million registrations in the International Registry established under the Aircraft Protocol since it began operation in 2006, and covering aircraft objects worth an estimated US$500 billion.9 As we will see, the Cape Town Convention and its Aircraft Protocol have already been adopted by numerous States in Africa, most recently by Ghana. Although the Luxembourg Rail Protocol has not yet entered into force, it has already attracted the attention of a number of African governments and economies.10 The aim of this chapter is to present the Cape Town Convention model and look in particular at the application and significance of the Aircraft Protocol and the Luxembourg Rail Protocol in Africa. In addition, a special section is devoted to the MAC Protocol, which promises to become a powerful tool for the financing and leasing of equipment used in the mining, agriculture and construction sectors and for fuelling economic growth across Africa. 5 ‘Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Space Assets’ (adopted at a diplomatic Conference in Berlin in 2012 by UNIDROIT) (‘the Space Protocol’). See R Goode, Convention on International Interests in Mobile Equipment and Protocol Thereto on Matters Specific to Space Objects: Official Commentary (Rome, UNIDROIT, 2013). 6 See E below. For an overview, see CW Mooney, M Dubovec and W Brydie-Watson, ‘The Mining, Agricultural and Construction Equipment Protocol to the Cape Town Convention Project: The Current Status’ (2016) 21 Uniform Law Review 332. References to MAC Protocol provisions in this chapter are based on the draft MAC Protocol text prepared by the Committee of Governmental Experts in 2017 and approved by the UNIDROIT Governing Council in 2018. See below (n 60). 7 CTC, Art 6 prescribes that the rules of the applicable Protocol prevail in case of inconsistency; moreover, the two instruments have to be interpreted as a single one. 8 For the updated status of the Convention and the Aircraft Protocol see, respectively, ‘Convention on International Interests in Mobile Equipment (Cape Town, 2001) – Status’, www.unidroit.org/status-2001capetown; and ‘Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Aircraft Equipment – Status’, www.unidroit.org/status-2001capetown-aircraft. 9 For the latest published reports see International Registry of Mobile Assets, ‘Annual Statistical Report’, www. internationalregistry.aero/ir-web/annualStatisticalReport/findAll. See below (n 19). 10 Gabon is already a Contracting State to the Luxembourg Rail Protocol. The Protocol was also signed by Mozambique, while a number of other jurisdictions are considering to ratify it. In addition, two African countries (Burkina Faso and Zimbabwe) signed the Space Protocol.

The Cape Town Convention Model: The Basic Legal Framework 379

B. The Cape Town Convention Model: The Basic Legal Framework Although the Convention is a complex legal text, its basic architecture is simple. It provides for an ‘international interest’ which can be created over a type of high-value equipment covered by a Protocol. The notion of ‘international interest’ comprises not only a traditional right in rem (deriving from an agreement granting, or transferring to, the creditor a proprietary right to secure a loan), but also the right of a seller and the right of a lessor deriving from a conditional sale and a lease.11 It is important to note that the international interest is autonomous from domestic laws. This means on the one hand, that parties can create a valid and effective international interest even if the law of a Contracting State does not provide for a financing device of the type covered by the Convention or limits its validity and effectiveness. On the other hand, an international interest and a domestic one can coexist on the same equipment: for example, an aircraft can be encumbered by both an international interest created under the Convention and by an aircraft mortgage regulated by domestic law, and each of them will be valid and effective according to their respective legal regime.12 Furthermore, each of these two interests may be held by the same creditor. An international interest will be enforceable as against third parties (including the grantor’s13 insolvency administrator) if registered in an international registry, which also determines its priority status as against unregistered or later registered interests. A system of strong enforcement measures allows creditors to protect their interests in the case of default by the grantor. This uniform legal regime is rounded up by specific provisions on jurisdiction and conflict of laws. Since it is a treaty, it is necessary to determine the territorial scope of application; the connecting factor to a Contracting State is where the grantor is situated, whereas the location of the creditor, or of the equipment, is irrelevant.14 For the purposes of this chapter, we single out four elements of the Cape Town Convention model for closer consideration. Section B.1 discusses the application of a ‘functional approach’, section B.2 considers the international registry and its role, section B.3 examines the enforcement measures, and section B.4 looks at the system of declarations which is offered to Contracting States.

1. The ‘Functional Approach’ As already mentioned, the Convention expressly lists three types of international interests that are covered by the uniform law regime: (i) the one granted by the chargor under a 11 CTC, Art 2(2). 12 In terms of priorities, under the CTC registered international interests will prevail, as a general rule, over interests created under domestic law, whether registrable or not. See B2 below. 13 In the CTC, the term ‘debtor’ includes a chargor under a security agreement, a conditional buyer under a title reservation agreement and a lessee under a leasing agreement (Art 1(j)). The equivalent term used in this book is ‘grantor’, see ch 1, D. The terms ‘chargor’ and ‘chargee’ refer, respectively, to the grantor and the creditor under a security agreement only, in the few provisions where the CTC differentiates between security agreement and conditional sale or lease. 14 CTC, Art 3. The Aircraft Protocol adds another criterion for helicopters and airframes, ie the country of registration (Art IV(1)).

380 A Modern International Approach to Equipment Financing in Africa security agreement; (ii) the one vested in a person who is the conditional seller under a title reservation agreement; and (iii) the one vested in a person who is the lessor under a leasing agreement. The three categories are subject to the ‘same basic framework’:15 in particular, the same creation, registration and priority rules generally apply. Thus, the Convention basically embraces, in relation to these devices, the functional approach advocated by the Model Law. By way of exception, a few provisions, principally relating to enforcement, treat security agreements, on the one hand, differently from retention of title devices, on the other hand.16 The qualification of the agreement for the purpose of applying these rules depends on the applicable domestic law.17

2. The Role of the International Registries The functioning of this uniform law regime is based on the establishment of a wholly electronic, 24/7 accessible International Registry for each type of object covered by a Protocol. The main purpose of the registry is to render an international interest effective against competing claimants, including the grantor’s insolvency administrator, and establish its priority.18 The Cape Town Convention registry system has special characteristics. It is asset based and requires unique identification of the equipment. Furthermore, it is included in a complex regulatory architecture designed to ensure its efficient functioning at an international level.19 In more general terms, however, the registry is in line with the solutions adopted in modern secured transactions laws, mandating electronic registrations and searching and introducing a ‘notice filing’ system.20 As regards priorities, the Convention adopts a simple ‘first-to-file’ rule: an international interest registered in the International Registry prevails over later registered or unregistered international interests, notwithstanding any actual knowledge.21 It also prevails, as a general rule, in respect to interests created over the same asset under national law, whether they are registrable or not.22

3. Enforcement Measures Another central feature of the Cape Town Convention model is its enforcement rules. The creditor can avail itself of strong and effective enforcement rights, including those set out 15 See M Deschamps, ‘The Perfection and Priority Rules of the Cape Town Convention and the Aircraft Protocol: A Comparative Law Analysis’ (2013) 2(1) Cape Town Convention Journal 51, 53. 16 See CTC, Arts 8 and 9, which refer to a security agreement, and Art 10, on conditional sales and leases. 17 The applicable domestic law as determined by the rules of private international law of the forum State. For an illustration referring to a title reservation agreement, see Aircraft Protocol Official Commentary, para 4.60. 18 Registration is therefore not necessary to create a valid international interest inter partes, nor to grant the creditor the right to enforce its security upon grantor’s default. 19 On the international registry set up for the Aircraft Protocol and managed by Aviareto, see R Cowan and D Gallagher, ‘The International Registry For Aircraft Equipment – The First Seven Years, What We Have Learned’ (2014) 45 Uniform Commercial Code Law Journal 225. 20 See ch 2, A2b. 21 See CTC, Art 29. 22 There are a few well-defined exceptions, including those non-consensual interests that are the object of a declaration of a Contracting State under CTC, Art 39(1) and those which Contracting States declare to be registrable

The Cape Town Convention Model: The Basic Legal Framework 381 in the contract between the parties. Thus, for example, in the event of default, the holder of an international interest deriving from a security agreement may take possession or control of the asset, sell it or lease it, and collect or receive any income or profits arising from the management or use of the asset.23 It may also obtain the vesting of the object in satisfaction under the conditions set out in the Convention.24 All these remedies are similar to those available to a secured creditor under the Model Law. Whether a creditor needs to apply to a court, or can exercise self-help in enforcing its remedies under the Convention will depend on the choice made by each Contracting State under the mandatory Article 54(2) declaration in the Convention. Interestingly, the Convention also includes a provision granting all creditors, in the event of the grantor’s default, the right to obtain advance relief from a court. According to Article 13(1) of the Convention, Contracting States should ensure that, under specific circumstances, a creditor may obtain speedy relief from the competent national court, pending final determination of the creditor’s claim, in the form of one or more of the orders listed in the same provision, and as requested by the creditor. Moreover, all Protocols contain an additional provision that allows States, by declaration, to enhance the effectiveness of the remedy in question.25 This remedy is particularly important since it ameliorates the efficiency of judicial enforcement of creditor’s rights, guaranteeing effective creditor protection within court proceedings. The Protocols further add the right to procure the export and physical transfer of the encumbered equipment from the territory in which it is located.26 This provision, particularly relevant in cross-border situations, enables the creditor to invoke the cooperation of the administrative authorities in the Contracting State where the assets are located.27

4. Declarations under the Convention and Protocols A fourth, and innovative, feature of the Cape Town Convention model is the flexibility introduced by a system of declarations. Declarations allow Contracting States to make policy choices in relation to important aspects of the uniform legal regime. In particular, a limited number of the provisions of the main Convention may be either set aside, or their effects may be varied through a specific declaration. For example, Contracting States may make a declaration under the Convention listing those categories of non-consensual right or interest that enjoy priority under domestic law, within or outside insolvency, which will then have priority over a registered international interest.28 Contracting States may also, by

under CTC, Art 40 (to which the first-to-file rule will apply). See below (nn 28–29) and accompanying text; and more generally, Aircraft Protocol Official Commentary, paras 4.192 et seq. 23 CTC, Art 8. 24 CTC, Art 9. For more details, see Aircraft Protocol Official Commentary, paras 4.99 et seq. 25 Identical provisions in Aircraft Protocol, Art X; Luxembourg Rail Protocol, Art VIII; Space Protocol, Art XX; the same provision is reproduced also in the MAC Protocol, Art IX. 26 Aircraft Protocol, Art IX(1)(b); Luxembourg Rail Protocol, Art VII(1); the same provision is contained in the MAC Protocol, Art VIII(1). 27 eg for the Luxembourg Rail Protocol, see Rail Protocol Official Commentary, paras 3.20 et seq. 28 CTC, Art 39(1)(a). The majority of African Contracting States have opted for a specific declaration, listing a limited number of categories of non-consensual liens that enjoy priority under domestic law, the

382 A Modern International Approach to Equipment Financing in Africa declaration, list those categories of non-consensual right or interest which shall be registrable under the Convention to acquire priority.29 These declarations allow States to protect certain claims under domestic law, while at the same time permitting a degree of advance assessment of risk by creditors.30 Perhaps the most innovative declarations are those allowing Contracting States to opt into additional rules in the Protocols that strengthen the effectiveness of creditors’ rights, particularly in enforcement and insolvency. Although joining the Convention system is advantageous because of the introduction of an internationally uniform and predictable legal regime, it must be emphasised that different choices in respect to the available options have different economic effects. As will be seen below, the practical application in the aviation sector has clearly shown that crucial economic benefits are specifically and concretely linked to the implementation of certain key opt-in declarations relating to parties’ autonomy in relation to choice of law and to creditor’s enforcement rights within and outside of grantor insolvency (so called ‘qualifying’ declarations).31 The inclusion of these economically significant declarations in the Convention was the result of the synergy between governments and industry stakeholders during treaty negotiations and was aimed at enhancing the economic benefits linked to its adoption.32 This was formally recognised by the international financing community, particularly in the OECD Aircraft Sector Understanding (ASU) regarding the conditions under which export credit can be granted.33 Moreover, again in the aircraft sector, ratification with the requisite set of declarations enabled grantors located in those States to receive enhanced ratings by rating agencies, opening access to better conditions in capital market financing.34

most common of which are liens for workers’ unpaid wages and for repairers’ claims. For a complete list of declarations under CTC, Art 39(1) see, ‘Article 39(1) – Declarations Deposited under the Cape Town Convention on International Interests in Mobile Equipment’ (UNIDROIT, 1 April 2019) www.unidroit.org/ depositary-2001capetown?id=438. 29 CTC, Art 40. The most common non-consensual interests declared by African States to be registrable under the CTC are those protecting judgment creditors and tax claims. For a complete list of declarations under CTC, Art 40, see ‘Article 40 – Declarations Deposited under the Cape Town Convention on International Interests in Mobile Equipment’ (n 28). 30 Aircraft Protocol Official Commentary, para 4.279. It is important to underline that the Convention cannot, however, be used to expand the preferred rights existing in the jurisdiction of the State issuing the declaration. 31 A Saunders et al, ‘The Economic Implications of International Secured Transactions Law Reform: A Case Study’ (1999) 20 University of Pennsylvania Journal of International Law 309; V Linetsky, ‘Economic Benefits of the Cape Town Treaty’ Aviation Working Group (18 October 2009) www.awg.aero/assets/docs/economicbenefitsofCapeTown.pdf. 32 See J Wool, ‘The Case for a Commercial Orientation to the Proposed UNIDROIT Convention as Applied to Aircraft Equipment’ (1999) 4(2) Uniform Law Review 289; R Goode, ‘The Cape Town Convention on International Interests in Mobile Equipment: A Driving Force for International Asset-Based Financing’ (2002) 7(1) Uniform Law Review 3. 33 The OECD publishes an Arrangement on Officially Supported Export Credits negotiated to avoid unfair competition through subsidised export credit rates, which sets out the ‘most favourable terms and conditions on which officially supported export credits may be provided’. The Sector Understanding on Export Credits for Civil Aircraft (ASU) authorises participating agencies, under certain circumstances, to grant a reduction up to 10 per cent off the minimum premium rate on export credit (‘Cape Town Discount’) if the aircraft operator is based in a country that ratified the CTC making the qualifying declarations: ASU, Appendix II, ss 35–48 and Annex I containing the list of ‘qualifying declarations’ under the Convention and the Aircraft Protocol. See also C below. 34 Aircraft Protocol Official Commentary, para 3.1.

Facilitating Financing in the Aviation Sector: The Aircraft Protocol in Africa 383

C. Facilitating Financing in the Aviation Sector: The Aircraft Protocol in Africa The Aircraft Protocol adapts the uniform law regime to the specific requirements of the aviation sector. Being the first to be adopted as well as extremely successfully applied in practice, it has functioned as a model for subsequent Protocols. However, there are a few important elements that are unique to the Aircraft Protocol. For example, the identification criteria of the equipment in the agreement and for the purposes of registering the international interest are tailored to the specifics of aircraft objects and the practice of aviation finance.35 Under the Aircraft Protocol, unique identification is required both for registration purposes (as is the case for all Protocols) and for the constitution of the international interest. In particular, a description of an aircraft object that contains its manufacturer’s serial number, the name of the manufacturer and its model designation is both necessary and sufficient to identify the object in the agreement.36 Moreover, the Protocol provides for a special remedy whereby the creditor may procure the deregistration of the aircraft from a national registry (as a precondition to its export and physical transfer).37 De-registration is practically very important if the creditor wishes to exercise its enforcement remedies in an effective way, since there are no other international provisions relating to it. Another feature that is not present in all subsequent instruments38 is the extension to sales, whereby the rules on registration and priority apply to outright sales of aircraft.39 By and large, however, most provisions contained in the Aircraft Protocol were consistently reproduced in the subsequent instruments. This is particularly evident for the Articles that allow States to opt into rules strengthening parties’ autonomy and creditors’ rights. Under Article VIII, if a State so declares, parties may agree on the law which is to govern their contractual rights and obligations, wholly or in part. Article X allows Contracting States to opt into a more efficient regime regarding advance relief pending final determination. In relation to grantor’s insolvency, Article XI allows States to select one of two alternatives in determining the way creditors can exercise their remedies. The first of the two (Alternative A) offers the strongest protection to creditors.40 If a country does not make any declaration, national insolvency provisions will apply. Turning to Africa, the Convention and the Aircraft Protocol have enjoyed widespread political acceptance in the region. The Protocol in particular has been adopted so far in

35 See Aircraft Protocol, Arts VII and XX (1). 36 Aircraft Protocol Official Commentary, paras 5.36–5.37. 37 Aircraft Protocol, Art IX(1)(a). This provision is completed by Art XIII, which is one of the opt-in provisions States may decide to adopt through a declaration. Art XIII strengthens the remedy of de-registration because the registry authority is obliged to provide it and cannot impose additional procedural requirements. For an ample treatment of this remedy, see Aircraft Protocol Official Commentary paras 3.31 et seq and 5.45 et seq. 38 cf the Rail Protocol (D below) and the MAC Protocol (E1 below). 39 Aircraft Protocol, Arts III and V. This feature is shared by the Space Protocol. Both the Rail and the MAC Protocol provide instead for the possibility to register a ‘notice of sale’ in the international registry. Such notice has no legal effect under the uniform law but provides information and may produce effects according to domestic laws: see Luxembourg Rail Protocol Official Commentary, para 5.70; see also (n 8) and E below. 40 The Rail Protocol introduced a third option, Alternative C, which represents a middle solution between the other two. The MAC Protocol, as will be seen in E below, incorporated all three alternatives.

384 A Modern International Approach to Equipment Financing in Africa 24 African jurisdictions out of a total 76 Contracting States.41 There are some patterns in the way countries in Africa implemented the Protocol. A majority of Contracting States in Africa opted for compliance with the requirements of the ‘qualifying declarations’ referred to above. In particular, they elected Alternative A for the exercise of creditors’ remedies in insolvency, opted into the de-registration and export authorisation rules of Article XIII, and adopted the enhanced advance relief rules contained in Article X. Almost all African Contracting States also chose to permit the exercise of creditors’ remedies without leave of court under Article 54(2) of the Convention.42 This was clearly done to be able to access finance from export credit agencies and capital markets. As mentioned above, adoption of the qualifying declarations is one of the preconditions of eligibility for the ‘Cape Town Discount’ under the OECD Aircraft Sector Understanding in export-credit arrangements. In fact, some African jurisdictions figure in the OECD list of eligible countries.43 Although there are still issues of compliance with the terms of the Convention and the Protocol on the part of national authorities, the Aircraft Protocol has certainly enabled the African aviation industry to grow and will be instrumental in facilitating financing for the increasing demand for new aircraft and new services in this key economic sector.44

D. Facilitating Development of the Rail Sector: The Luxembourg Rail Protocol and Africa The Luxembourg Rail Protocol was adopted to complement the Convention in relation to railway rolling stock. It applies to all vehicles running on tracks or above, on or under a guideway (from high-speed trains to commuters and smaller carriages, to trams and metros)45 and, consistent with the main Cape Town Convention framework, creates an international system regulating the rights of financiers and lessors of this category of mobile equipment. As is the case for all Protocols, it is applied together with the main Convention.46 It also generally tracks the model of the Aircraft Protocol in relation to those provisions that add to, or derogate from, the uniform rules of the Convention.47 There are, however, some specific features of the Luxembourg Rail Protocol that are designed to take into account the different nature of the equipment as well as the different type of market. The Protocol offers an additional insolvency option (in Article IX)48 and, although it does not permit registration of outright sales, it allows the registration of a 41 Angola, Burkina Faso, Cabo Verde, Cameroon, Congo, Côte d'Ivoire, DR Congo, Egypt, Eswatini, Ethiopia, Gabon, Ghana, Kenya, Madagascar, Malawi, Mozambique, Namibia, Nigeria, Rwanda, Senegal, Sierra Leone, South Africa, Togo and Tanzania. Two more States (Seychelles and Zimbabwe) acceded to the main Convention without depositing the required declaration under CTC, Art 54(2), which is a condition for the entry into force of the instrument. 42 A minority of Contracting States went the other way: Côte D’Ivoire, Egypt and Namibia. Most recently, however, Côte d’Ivoire modified its declaration under Art 54(2) to allow out-of-court enforcement. 43 The current list encompasses Angola, Ethiopia, Kenya, Rwanda, Senegal and Nigeria. 44 The most recent official data on number of registrations and volume of collateral is available at the Aviareto – International Registry website. See ‘Facts & Figures’, www.aviareto.aero/information-centre/facts-and-figures/. 45 Rail Protocol, Art 1(2)(e). 46 See B2 above. 47 See C above. 48 This additional insolvency option is present also in the MAC Protocol, Art X. See E below.

The Future Mining, Agriculture and Construction Protocol 385 notice of sale, which has no legal effect under the Convention.49 One of the most relevant innovations concerns the identification of railway rolling stock for the purposes of registration in the dedicated International Registry. Whereas aircraft are invariably serialised and manufacturers are limited in number, the rail sector is much more fragmented and unique identification of each item of rolling stock is not always present or consistent. For this reason, on the one hand, the Rail Protocol allows a generic description of the railway rolling stock in the security agreement;50 on the other hand, it introduced a new global identification system for all railway rolling stock: a permanent and unique 20-digit number to be issued by the International Registry.51 Such a system will have the added value of streamlining the tracking and identification of railway rolling stock around the world.52 The Protocol is nearing entry into force, having being ratified by three States (including Gabon) and the European Union, and signed by six other States (including Mozambique),53 while the ratification process is advancing in a number of other countries. It is not surprising that there is a growing interest in the Rail Protocol among African economies. In fact, once operational, the Rail Protocol is expected to attract more private-sector lenders and lessors into the rail market, resulting in cheaper finance for operators and a wider choice in relation to types of financing. It would also allow States to concentrate public resources on the building of rail infrastructure, relying on private, or public–private, capital for procuring railway rolling stock. Moreover, it would foster sustainable economic and social development by enhancing urban and extra-urban transport of people and cross-border transport of freight by rail.54

E. The Future Mining, Agriculture and Construction Protocol In addition to the three existing Protocols to the Cape Town Convention regarding aircraft, railway rolling stock and space assets, UNIDROIT has almost completed work on a fourth Protocol, which will extend the application of the Convention to mining, agricultural and construction equipment. Having been considered and approved by various groups of

49 Rail Protocol, Art XVII. Although it is true that such notice, which is reproduced in the MAC Protocol, Art XVII, has no legal effect under the uniform law, it provides information and may produce effects according to domestic laws. See nn 38–39 above and E below. 50 Rail Protocol, Art V. 51 Rail Protocol, Art XIV. 52 B von Bodungen and H Rosen, ‘From the Luxembourg Rail Protocol to the Draft MAC Protocol’ (2018) 23(2) Uniform Law Review 190, 198. 53 For the updated status of the Protocol, see UNIDROIT, ‘Status – Luxembourg Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Railway Rolling Stock (Luxembourg, 2007)’, www.unidroit.org/status-2007luxembourg-rail. 54 See von Bodungen and Rosen (n 52), also for references to recent studies conducted on the prospective economic impact of the Luxembourg Rail Protocol. The potential benefits of the Protocol were recognised by the United Nations Economic Commission for Africa (UNECA) at its plenary meeting hosted by the Moroccan government in March 2019, which adopted a ministerial resolution mandating the Commission to raise the awareness of, and advocate the adoption of, the Luxembourg Rail Protocol across Africa. For the Report containing the recommendation of the meeting of the Committee of Experts, see ‘Report of the Meeting of the Committee of Experts’, E/ECA/CM/52/2/Rev.1 (UNECA, 23 March 2019) 23. See also V Songwe and H Rosen, ‘The Luxembourg Rail Protocol: Making the Railways Work for Africa’, Railways Africa (February 2019), www.railworkinggroup.org/ documentation/articlesreports.

386 A Modern International Approach to Equipment Financing in Africa experts and intergovernmental committees, the MAC Protocol55 will be adopted in November 2019 at a diplomatic Conference in South Africa.56 The future MAC Protocol will be the first Protocol to the Cape Town Convention expanding the system beyond the categories of equipment that were already envisioned when the Convention was adopted in 2001. Article 51 of the Convention expressly contemplates the extension of its applicability through future protocols to objects other than ‘airframes, aircraft engines and helicopters’, ‘railway rolling stock’ and ‘space assets’.57 However, the extension of the Convention is limited to other categories of high-value, mobile and uniquely identifiable equipment.58 Much of the preparatory work undertaken in relation to the MAC Protocol was aimed at ensuring it met the three criteria set out in Article 51. The remainder of this chapter will explore the operation of the MAC Protocol and discuss how it might impact on the mining, agriculture and construction sectors in Africa.

1. Basic Rules of the MAC Protocol Consistent with the Cape Town Convention’s two-tier architecture discussed earlier,59 Article 6 of the Convention provides that the future MAC Protocol and the Cape Town Convention are to be read and interpreted together as a single instrument, with the MAC Protocol prevailing over the Convention to the extent of any inconsistency. Article 6 allows the MAC Protocol provisions to tailor the application of the Convention’s legal framework to the MAC sectors. In light of the remarkable success of the Cape Town Convention and Aircraft Protocol, the MAC Protocol tries to adhere to the provisions of the existing Protocols wherever possible. This has resulted in a draft MAC Protocol60 that is largely consistent with the Aircraft Protocol, and thereby also the Luxembourg Rail Protocol. As in the existing Protocols, the MAC Protocol allows for the creation of international interests in MAC equipment, which can be done through security agreements, title reservation agreements or leasing agreements.61 In this regard, the MAC Protocol retains the functional approach in defining international interests, as established by the Convention. 55 The instrument’s formal title is the ‘Protocol to the Convention on International Interests in Mobile Equipment on Matters Specific to Mining, Agricultural and Construction Equipment’. 56 The MAC Protocol was developed under UNIDROIT’s standard working methodology which involves (i) preparatory research, (ii) preparation of a draft instrument by an expert Study Group, (iii) further negotiation by a Committee of Governmental Experts and (iv) adoption at a Diplomatic Conference. Further information and documentation regarding each stage of the process is available at UNIDROIT, ‘Study LXXII K – Development of a Fourth Protocol to the Cape Town Convention on Matters Specific to Mining, Agricultural and Construction Equipment’ (7 May 2019) www.unidroit.org/work-in-progress/mac-protocol. 57 CTC, Art 2(3) expressly lists ‘airframes, aircraft engines and helicopters’, ‘railway rolling stock’ and ‘space assets’ as categories over which an international interest can be constituted under the CTC. 58 CTC, Art 51(1). 59 See A above. 60 References to MAC Protocol provisions in this chapter are based on the draft MAC Protocol text prepared by the Committee of Governmental Experts in 2017 and approved by the UNIDROIT Governing Council in 2018 for transmission to a Diplomatic Conference for adoption. The draft instrument should be considered to be near its final form, but still subject to further negotiation at the Diplomatic Conference. Although most of the draft provisions are unlikely to be radically altered at the Diplomatic Conference, there will undoubtedly be some amendments, particularly in relation to issues that were not fully resolved by the Committee of Governmental Experts. 61 Art I of the MAC Protocol does not alter the definition of ‘international interest’ under the CTC or the types of agreements that constitute an international interest under CTC, Art 2.

The Future Mining, Agriculture and Construction Protocol 387 The basic formal requirements for the constitution of an international interest are the same as those in Article 7 of the Convention,62 and the connecting factor continues to be the grantor’s location in a Contracting State at the time of the constitution of the international interest.63 The MAC Protocol will establish a new International Registry for international interests and other registrable interests relating to MAC equipment. Registered international interests will have priority over unregistered interests, subsequently registered international interests and interests registered in domestic registries.64 Third-party effectiveness of an international interest through registration alone remains a core element of the MAC Protocol, which is consistent with the Cape Town Convention system as a whole. Like the Luxembourg Rail Protocol, the MAC Protocol does not apply to sales of MAC equipment, but does allow for the registration of notices of sale in the international registry. ‘Notices of sale’ have no legal effect under the Convention or Protocol, and are merely designed to give notice of a sale transaction.65 Registering a notice of sale might be useful to clarify the ownership of MAC equipment in circumstances where the owner is not in possession of the equipment, such as when large mining equipment is sold and the purchase price is paid prior to the delivery of the equipment to the buyer. The MAC Protocol retains the highly important default and insolvency remedies provided by the Convention and existing Protocols. In event of default by a grantor under a security agreement, the creditor may take possession or control of the asset, sell or lease it out, collect or receive any income or profits arising from the management or use of the asset, or procure its export and physical transfer.66 The inclusion of the Cape Town Convention’s strong approach to default and insolvency remedies in the MAC Protocol should not be undervalued. A recently released independent economic assessment of the MAC Protocol notes that the instrument will improve access to finance and reduce credit risk by removing some of the key uncertainties around asset recovery in the event of a grantor’s default or insolvency.67 The assessment specifically cites the stronger creditor remedial rights under insolvency Alternative A as an additional mechanism that would lower borrowing costs and reduce debt finance risks.68 The tangible economic benefits associated with stronger creditor rights is one of the core reasons the Convention has been effective in lowering borrowing costs in the aircraft sector.

2. Specialised Rules of the MAC Protocol Although the majority of the rules in the MAC Protocol are consistent with the existing Protocols, the very different nature of MAC equipment and the way it is financed, 62 An interest is constituted as an international interest where the agreement creating or providing for the interest: (a) is in writing; (b) relates to an object of which the chargor, conditional seller or lessor has power to dispose; (c) enables the object to be identified in conformity with the Protocol; and (d) in the case of a security agreement, enables the secured obligations to be determined, but without the need to state a sum or maximum sum secured. 63 CTC, Art 3(1). 64 CTC, Art 29. 65 Luxembourg Rail Protocol Official Commentary, para 5.70. 66 CTC, Art 8(1); MAC Protocol, Art VIII(1). 67 Warwick Economics and Associates, The MAC Protocol Economic Assessment (UNIDROIT, 2018) para 4. 68 ibid para 56.

388 A Modern International Approach to Equipment Financing in Africa as compared to the categories of equipment covered by the existing Protocols, necessitated the inclusion of specialised rules in the MAC Protocol. Three specialised rules of particular note are the mechanism used to determine the types of MAC equipment to which the Protocol applies, the treatment of equipment that has become associated with immovable property, and the application of the Protocol to MAC equipment held as inventory.

a. Application of the Protocol to MAC Equipment The Cape Town Convention system was designed to regulate interests in high-value, uniquely identifiable, mobile equipment.69 These terms were never specifically defined in the Convention or existing Protocols because aircraft, railway rolling stock and space assets were deemed to be high value and mobile by their very nature. This is a strong point of distinction as compared to MAC equipment, which encompasses a much larger and more diverse universe of equipment, including many types of lower-value equipment. It is undesirable for the MAC Protocol to apply to all types of MAC equipment, as its rules are designed for regulating commercial relationships between a creditor and a grantor, and not for the financing of low-value equipment which is not going to be used in a business activity. To limit the scope of the MAC Protocol and avoid its inappropriate application to all MAC equipment, the Protocol uses the Harmonised Commodity Description and Coding System (HS). Introduced in 1988, the HS is an international nomenclature for the classification of products used globally.70 It allows countries to classify traded goods on a common basis for customs purposes by dividing them into approximately 5,300 six-digit HS codes. The HS provided the MAC Protocol with a mechanism to identify types of MAC equipment that are predominantly high value and used commercially. It also assisted in satisfying the mobility criterion, as HS codes identify types of MAC assets that are traded internationally. Article II(1) of the MAC Protocol provides that the Convention applies to MAC equipment listed in Annexes 1, 2 and 3 of the Convention, which list 42 different 6-digit HS codes.71 Annex 1 lists codes relating to mining equipment, Annex 2 lists codes relating to agricultural equipment, and Annex 3 lists codes relating to construction equipment. The listing of MAC equipment separately in different Annexes also gives Contracting States the ability to opt out of the Protocol’s application to a particular category. For example, a State might decide that the MAC Protocol would improve the legal framework for the financing of equipment in the mining and construction fields, but was undesirable for agricultural equipment. In this circumstance, Article II(3) of the MAC Protocol would allow a State to declare it will only implement the Protocol in relation to Annex 1 and Annex 3, but not Annex 2.72 69 As noted earlier, CTC, Art 51(2) contemplates its extension to categories of ‘high value mobile equipment’. 70 The Harmonised System is governed by the World Customs Organisation and is the basis for the negotiation of bilateral, regional and international trade agreements. 71 It is expected that the number of codes listed in the Annexes may increase at the Diplomatic Conference based on proposals made by negotiating States in 2018. 72 It should be noted that where a type of MAC equipment is covered by an HS code which is listed in a MAC Protocol annex, that MAC equipment is within the scope of the MAC Protocol, regardless of its intended or actual use. This would mean that even where a country implements the Protocol in relation to mining equipment (Annex 1) but not construction equipment (Annex 3), a type of mining equipment covered by an HS code listed

The Future Mining, Agriculture and Construction Protocol 389 The use of the HS is an innovative mechanism to determine the scope of application of the Protocol in relation to MAC equipment. It will also allow the MAC Protocol to be adapted to apply to new types of MAC equipment that are invented in the future.73

b. Immovable-Associated Equipment Article VII of the MAC Protocol provides a set of rules to resolve potential conflicts between international interests in MAC equipment and domestic law interests in the equipment arising from the equipment’s association74 with immovable property. An example of this would be where an agricultural harvester covered by an HS code listed in Annex 2 of the MAC Protocol and subject to an international interest is used on a farm in a jurisdiction where the domestic law provides that a mortgage over the farm itself would extend to the harvester. Another example would be where a tunnel boring machine covered by an HS code listed in Annex 1, and subject to an international interest, is installed underground in a mine in a jurisdiction where the domestic law provides that the equipment would become a fixture and therefore subject to the real property mortgage. Given the challenges in harmonising interests in immovable property at an international level, the MAC Protocol allows Contracting States some flexibility in how to treat immovable associated MAC equipment, by providing three alternative sets of rules in Article VII. Alternative A generally gives priority to an international interest over an interest arising out of the equipment’s association with immovable property (in the first example above, this would mean that the international interest in the harvester would have priority over an interest in the harvester arising from the mortgage over the farm on which it is used). Alternative B resolves the conflict between the competing interests by determining whether the equipment loses its individual legal identity (in the second example above, an international interest in the tunnel boring machine would only be subordinated to the domestic law mortgage over the mine where the tunnel borer lost its individual legal identity under domestic law and had been used in the mine prior to the registration of the international interest). Alternative C simply defers to the domestic law in which the equipment is located to determine whether the international interest is subordinated to a domestic interest arising out of the MAC equipment’s association with immovable property. In implementing the MAC Protocol, every country is mandatorily obliged to make a declaration under Article VII stating that it will apply either Alternative A, B or C. This rule in the MAC Protocol provides a level of harmonisation in relation to immovable propertyrelated interests that was not achievable in other international instruments. For example, the Model Law does not provide for any specific rules on security rights in attachments to immovable property. The Guide to Enactment explains that this issue was not addressed specifically because the regulation of attachments to immovable property does not lend in Annex 1 would be within the scope of the MAC Protocol, even if it were subsequently used for construction purposes in that country. 73 MAC Protocol, Art XXXIII provides the rules in relation to future amendments to the Protocol, including amendments to the HS codes in the annexes. As one of the public international law-related provisions on the treaty, its final content will be decided at the Diplomatic Conference. 74 The word ‘association’ is used rather than ‘connection’, ‘fixture’ or ‘attachment’ to reflect that rules in certain jurisdictions allow for an immovable property interest to extend to a mobile asset even where there is no physical connection.

390 A Modern International Approach to Equipment Financing in Africa itself to harmonisation at the international level.75 In this regard, should African States adopt the MAC Protocol and uniformly apply either Alternative A or Alternative B under Article VII, the MAC Protocol would allow for a harmonised approach across Africa.

c. Treatment of Inventory The MAC Protocol contains an additional Article allowing Contracting States to make a declaration preserving existing inventory financing arrangements. Originally proposed by the private sector, Article XII of the MAC Protocol allows States to declare that an interest in inventory created or provided for by an agreement under which the dealer is the grantor cannot be an international interest. The effect of this rule is that lenders to dealers that hold large amounts of equipment as inventory for sale or lease that frequently turns over would not need to individually register and then deregister such interests in the international registry. Many African economies recently reformed their secured transactions law to provide for a security right that floats over rotating inventory that may be perfected by a single registration describing the collateral as ‘all assets’ or ‘all present and future inventory’.76 It would be expected that only States with well-functioning inventory practices would want to make a declaration under Article XII. Other States would be well advised not to make such a declaration, as a broader application of the MAC Protocol to inventory financing agreements could support the creation of an inventory financing industry in Contracting States where it does not currently exist. Furthermore, a State opting-out of the Protocol’s application to inventory financing would mean that lenders would not be able to benefit from the strong protections for their rights within and outside of insolvency.

3. The Relationship between the MAC Protocol and Domestic Law Implementation of the Cape Town Convention and the MAC Protocol will complement domestic secured transactions reforms. While the CTC regime is designed to operate independently from the domestic regime in almost all cases, it will generally be an improvement over domestic secured transaction regimes, which are usually broader in their application, for example, they usually apply to consumer transactions which require a different balancing of interests. The Cape Town Convention system is designed to be implemented in conformity with domestic reforms, such as those based on the Model Law. The Cape Town Convention and its Protocols apply to special categories of high value, mobile assets that are typically excluded from general domestic secured transaction regimes, a principle recognised by Article 1(3)(e) of the Model Law. However, while this has been the case for aircraft objects, all African secured transactions laws apply to MAC equipment, requiring greater care in the implementation of the MAC Protocol and education of the relevant stakeholders on the interaction of the international and domestic secured transactions regimes. The Cape Town Convention and its Protocols should be implemented in addition to domestic secured transactions reforms because the treaty system provides concrete

75 Guide 76 This

to Enactment to ML, paras 32–34. was already the case in the common law jurisdictions, see ch 1, E4b. See also Burundi C1, Zimbabwe C2.

The Future Mining, Agriculture and Construction Protocol 391 economic benefits for Contracting States. The 2018 MAC Protocol economic assessment shows that the instrument should enable more buyers to access credit or borrow on better terms against their existing equipment, either by lowering interest rates, extending loan durations or increasing loan-to-value ratios. The assessment predicts that the MAC Protocol will have an annual positive impact of US$23 billion on GDP in developing countries and of US$7 billion in developed countries.77 The estimates are focused on the potential economic benefits in the 63 UNIDROIT Member States. The global benefits will be greater if, as expected, countries that are not UNIDROIT Member States, many of which are in Africa, also ratify the Protocol (Egypt, Nigeria, South Africa, and Tunisia are the only UNIDROIT Member States in Africa). The economic impact of the MAC Protocol will also be greater if additional HS codes are added to the MAC Protocol annexes at the Diplomatic Conference, as is expected. The economic benefits provided by the Convention and Protocol are separate and additional to any economic benefits introduced by a domestic secured transactions law reform. African countries should be encouraged to implement improvements to their domestic laws and ratify the Cape Town Convention and its Protocols simultaneously. On a technical level, the MAC Protocol contemplates interoperability between domestic collateral registries and the MAC Protocol International Registry. Whereas the Cape Town Convention applies to special categories of assets that are typically excluded from domestic regimes, the MAC Protocol will apply to certain types of MAC equipment that will be ordinarily covered by general domestic regimes and collateral registries established thereunder. As discussed earlier, the Cape Town Convention’s priority rules establish the supremacy of international interests over interests registered in domestic registries. However, all Protocols to the Cape Town Convention contain a special provision to allow international registries created under specific Protocols to operate in conjunction with domestic registries. Article XVI(1) of the MAC Protocol allows Contracting States to designate ‘domestic entry points’ into the International Registry, which is designed to ‘link’ the two. This optional rule allows States to compel parties making a registration to go through a domestic authority, or ‘entry point’, in order to use the International Registry, when the grantor is situated in that State. Article XIX(1) of the Aircraft Protocol contains an identical rule allowing Contracting States to nominate a domestic entry point into the aircraft international registry. In practice, this usually requires a user of the aircraft International Registry to obtain an authorisation code from the relevant domestic authority before it can make a registration in the International Registry.78 Of the 76 parties to the Aircraft Protocol, 12 have made a declaration under Article XIX. Exactly how this will operate in practice for the MAC Protocol is yet to be determined, but the design should ensure that domestic creditors and grantors are aware of the existence of an international registry, and conduct proper due diligence. The existence of ‘domestic entry points’ into international registries allows the Cape Town Convention to supplement and build upon ongoing global efforts aimed at improving domestic asset-based financing regimes and increasing access to credit. To this end, the Convention perfectly complements the secured transaction reforms being implemented by multiple jurisdictions in partnership with various regional and international organisations. 77 Warwick Economics and Associates (n 67) paras 4–7. 78 Aircraft Protocol Official Commentary, para 5.89. See also the identical rule in Luxembourg Rail Protocol, Art XIII.

392 A Modern International Approach to Equipment Financing in Africa

4. The Importance of the MAC Protocol in Africa The future MAC Protocol is a vital tool that could contribute to sustainable development in Africa and support economic growth, lifting the living standards of the continent’s 1.25 billion people. There are three reasons why the MAC Protocol will be particularly beneficial to African economies: first, it will assist in increasing international private sector investment required for economic development; secondly, it targets sectors that are the engines of growth in Africa; and, thirdly, it will provide a uniform legal framework for investment in a region that has particularly disparate, and often outdated, rules.

a. The MAC Protocol as a Tool for Increasing International Private-Sector Investment in Africa The scale of investment required in Africa makes reliance on domestic finance impossible.79 Mobilising additional international private-sector investment in Africa will contribute to long-term sustained economic growth. As recognised in the Monterrey Consensus on Financing for Development, international private-sector investment is important because of its potential to transfer knowledge and technology, create jobs, boost overall productivity, enhance competitiveness and entrepreneurship, and ultimately eradicate poverty through economic growth and development.80 The Cape Town Convention and Aircraft Protocol has had a material impact in increasing international investment in the aviation sector. It has lowered the cost of credit in Contracting States, and it has made financing available in economies that could not previously attract investment. It is estimated that the Aircraft Protocol will generate savings of US$161 billion in aggregate financing costs between 2009 and 203081 and has reduced the cost of financing aircraft equipment by 30 per cent in certain sectors.82 It is anticipated that the MAC Protocol could have a similar impact in increasing the availability of finance in the MAC sectors in Africa. Although the MAC Protocol should improve lending conditions across Africa, it should be noted that it is not a panacea that will cure all underinvestment issues. There may be other factors that restrict access to credit. The MAC Protocol should be seen as an important element in improving investment in the relevant sectors in Africa, but not the only necessary action.83

b. The Mining, Agriculture and Construction Sectors in Africa Simply put, the MAC sectors across Africa are disproportionally important to the continent’s overall economic prosperity as compared to the rest of the world. As such, increasing the availability of credit for asset-based finance, including the acquisition and leasing of MAC equipment for commercial use, will have a much bigger impact in African economies than it would elsewhere. 79 Development Financing in Africa (Addis Ababa, UNECA, 2017). 80 ‘The Monterrey Consensus of the International Conference on Financing for Development’, A/CONF.198/3 (Monterrey, United Nations, 2002) para 20. 81 Linetsky (n 31). 82 HG Bollweg and S Schutleiss, ‘Das Berliner Weltraumprotokoll’ (2012) 61 ZLW 389, 390. 83 Warwick Economics and Associates (n 67) para 160.

The Future Mining, Agriculture and Construction Protocol 393 The mining industry is a driver of economic growth in Africa. Although accurate global data on the mining sector are sparse, the Mining Contribution Index (MCI) provides a global assessment of the significance of the mining sector’s contribution to national economies.84 The MCI 2018 indicates that of the 25 States most dependent on the mining sector, 16 are in Africa.85 Agriculture continues to be a major contributor to economic activity and employment across Africa. In sub-Saharan Africa, agriculture, forestry and fishing were responsible for 16.2 per cent of GDP in 2017, significantly higher than the global average of less than 4 per cent.86 Agriculture is also a major employer, accounting for 57.4 per cent of total employment across sub-Saharan Africa, and 17.8 per cent in North Africa.87 Given its importance to the African economy, the transformation of the agricultural sector is viewed as central to overall economic transformation in Africa.88 By increasing access to agricultural finance, the MAC Protocol would directly contribute to this transformation by boosting agricultural productivity, expanding cultivated area and raising yields, through mechanising and commercialising agriculture.89 Similarly to the mining and agriculture sectors, the construction sector in Africa is central to improving sustainable economic growth. The African Development Bank singled out a lack of investment in infrastructure as one of the key factors impeding industrialisation in Africa. The African Development Bank analysis suggests that the continent’s infrastructure needs amount to US$130–170 billion a year, and that a financing gap of US$68–108 billion currently exists.90 One estimate forecasts that Africa will experience the largest growth in the construction sector globally over the next five years, with an annual average growth of 6.4 per cent between 2018 and 2022.91 Although this level of projected growth is the highest globally, it will not be sufficient to address the large financing gap required to improve Africa’s core infrastructure. In increasing availability of financing for the acquisition of high-value construction equipment, the MAC Protocol should increase productivity and profitability for construction firms, which will directly support the building of new infrastructure across Africa.

c. Uniformity in the Legal Framework for the Financing of MAC Equipment across Africa Across the 54 countries in Africa, there is little commonality in the domestic laws regulating asset-based lending. As illustrated in this book, domestic secured transactions legal reforms accompanied by the establishment of collateral registries will assist in improving the secured 84 Role of Mining in National Economies: Mining Contribution Index, 4th edn (London, International Council on Mining and Metals, 2018). 85 Suriname, the Democratic Republic of Congo, Guinea, Burkina Faso, Mali, Sierra Leone, Liberia, Ghana, Namibia, Madagascar, Botswana, Senegal, Zimbabwe, Sudan, Mauritania and Zambia. 86 World Bank, ‘World Development Indicators: Data Catalog’, https://datacatalog.worldbank.org/dataset/ world-development-indicators. 87 ibid. Modelled estimates produced by the International Labour Organization. 88 African Transformation Report 2017: Agriculture Powering Africa’s Economic Transformation (Accra, African Center for Economic Transformation, 2017). 89 ibid. 90 African Economic Outlook 2018 (Abidjan, African Development Bank, 2018) 63. 91 ‘Global Construction Outlook to 2022: Q3 2018 Update’ GlobalData (September 2018) www.globaldata.com/ store/report/gdcn0010go--global-construction-outlook-to-2022-q3-2018-update/.

394 A Modern International Approach to Equipment Financing in Africa transactions regulatory framework in a number of economies in Africa. However, the MAC Protocol will provide a level of uniformity in the financing of MAC equipment that can only be provided by an international treaty. Having a clear, predictable, and uniform framework for the financing of MAC equipment across Africa has a number of benefits. If African countries uniformly adopt the MAC Protocol, financiers could allow the movement of MAC assets across borders in Africa to work on different farms, mines and construction sites with confidence that their international interests would be protected on the basis of a single registration in the MAC Protocol international registry. Uniform adoption of the MAC Protocol would also allow lenders to provide credit without the need for additional due diligence in relation to the local legal regime and its intricacies. The expansion of credit regionally could also provide increased export and trade opportunities for local manufacturers of MAC equipment. These benefits extend beyond uniformity across Africa, as the same rules for the financing of MAC equipment could apply across the world. The MAC Protocol envisages the same legal rules applying for the financing of a million-dollar crane used for construction projects in Brazil, Kenya, China and France.

F. Final Remarks The Cape Town Convention and its Protocols provide a widely adopted, harmonised and efficient international legal framework for the financing of certain categories of mobile, high-value equipment. By providing substantive rules governing the creation, registration and priority of international interests in equipment, the Convention allows grantors and creditors to conclude agreements with a high degree of clarity regarding their legal rights. The default and insolvency remedies of the Convention and its Protocols allow creditors to provide asset-based financing with confidence that their interest will be protected, even where the asset has moved across borders. During its 13 years of operation, the Cape Town Convention has already had a marked impact in improving asset based-financing in Africa. Almost half of all African countries have ratified the Cape Town Convention and its Aircraft Protocol, and a third of all Contracting States of the Aircraft Protocol are African. This strong representation of African States illustrates the value that the Convention and Aircraft Protocol has already had in increasing investment in the African aviation industry. The expansion of the Cape Town Convention through its Protocols provides African countries with further opportunities to improve their domestic financing arrangements. In particular, should the Luxembourg Rail Protocol and future MAC Protocol also become widely adopted across Africa, they would greatly increase investment in the African rail sector and in the MAC industries.

18 The Infrastructure Underpinning a Modern Secured Transactions Law GAVIN McCOSKER*

This chapter explores the key elements necessary to enable a secured transactions regime to achieve the important objective of improving the financing opportunities for MSMEs. It is written from the perspective of someone with many years of practical experience in the secured transactions field, both as Registrar of Personal Properties Securities in Australia and as a Senior Financial Sector Specialist and consultant for the World Bank Group focused on Africa. Rather than analysing the law, or the role played by the law in secured transactions reform, the analysis focuses on the other factors that have to be in place to enable a reformed law to achieve its purpose.

A. Introduction There are two primary, and equally important components of an effective secured transactions regime: the law and the registry. As critical as these ‘twin pillars’ are, a range of other essential elements need to be in place for a secured transactions regime to flourish. This chapter explores the components of a modern secured transactions regime and what is needed for them to be successful; in particular, an effective registry, and the the credit infrastructure environment necessary for that registry to achieve its purpose. That purpose is to increase the flow of productive credit, particularly for the benefit of SMEs. Although written from the perspective of secured transactions reform in Africa, the lessons from this chapter are equally applicable across the world, and not only in emerging markets. Although not an exhaustive account of all practices and approaches that can be taken to implement a successful modern secured transactions regime, it illustrates some of the complexity involved, as well as the fact that the necessary infrastructure goes well beyond establishing and overseeing the registry.1 In emerging markets, additional effort is required to build expertise and capacity among market participants. In particular, access to credit for many businesses, especially MSMEs, is limited or even denied because the financial infrastructure is inadequate or inadequately regulated.2 * Gavin McCosker is Adjunct Industry Fellow, Griffith Business School, Griffith University. 1 For an example of a country which established a registry first and then reformed its secured transactions law, see Nigeria D2. 2 See ch 1, B1.

396 The Infrastructure Underpinning a Modern Secured Transactions Law Of critical importance is the need to build institutional capacity for the registry, the registry users (eg creditors) and other key stakeholder groups. It is necessary for all of these actors to understand and integrate the revised approach into their business practices. Following the mantra ‘build it and they will come’ simply will not deliver the catalyst for change which is necessary to support sustainable economic growth. There is considerable research, including that undertaken after the global financial crisis (GFC), highlighting the important link between finance and the macro economy. For example, the Federal Reserve Bank of San Francisco points out that the ‘starkest illustration of this [macro-finance] linkage occurred in the fall of 2008, when the extraordinary financial market dislocations that followed the bankruptcy of Lehman Brothers coincided with a global macroeconomic free fall’.3 Guy Debelle, the Deputy Governor of the Reserve Bank of Australia, graphically described the point in 2018 as follows: Macroeconomics is like the model of the engine, finance is the oil that lubricates the engine. One can understand how the engine works without really needing to worry about the oil, as long as the oil is flowing. But at least a basic understanding of the plumbing is useful when the oil dries up.

He added that, although leverage ‘really matters’, ‘[a]t the same time, you need to keep the credit flowing to prevent the economy from seizing up’.4 In a market economy, the extension of credit, in its various forms, is a private act between willing participants. The maintenance of sustainable and stable credit markets is, however, of value to the broader society. This is because the benefits promote efficient use of capital and economic participation, which helps improve citizens’ overall standard of living. For this reason, it is generally accepted that the State should play a role in supporting the provision of credit by legally regulating the relationship between creditors and debtors, and by providing infrastructure tools to support the operation of the law. Although there is no clear definition of a ‘good’ upper limit of private credit as a percentage of GDP, the opportunity in many developing countries for increased access to capital for businesses to support growth is immense. For example, World Bank data from 2016 showed that high-income OECD countries had an average ratio of 145 per cent of private credit to GDP, but the sub-Saharan African average was 44 per cent and several countries fell well below this benchmark.5 Around the world, MSMEs play a crucial role in employment, job creation, investment, innovation and economic growth, accounting for ‘about 90% of businesses and more than 50% of employment worldwide’.6 Yet, in both developing and developed countries, access to finance remains a major constraint for MSMEs. Credit infrastructure,7 defined by the World Bank as ‘the legal, institutional and regulatory infrastructure that enables efficient 3 GD Rudebusch, ‘Macro-Finance Models of Interest Rates and the Economy’ (Federal Reserve Bank of San Francisco, January 2010) www.frbsf.org/economic-research/files/wp10-01bk.pdf. 4 G Debelle, ‘Lessons and Questions from the GFC – Address to the Australian Business Economists Annual Dinner’ (Reserve Bank of Australia, 6 December 2018) www.rba.gov.au/speeches/2018/pdf/sp-dg-2018-12-06.pdf. 5 World Bank, ‘Domestic Credit to Private Sector (%GDP)’, www.data.worldbank.org/indicator/FS.AST.PRVT. GD.ZS. 6 IBRD, ‘G20 Action Plan On SME Financing: Credit Infrastructure Country Self-Assessment Consolidated Report’ (2017) www.gpfi.org/sites/default/files/documents/GPFI%20-%20G20%20AP%20SME%20Financing%20 Self%20Assessment%20Consolidated%20Report.pdf. 7 See also in Conclusion, B1di where ‘credit infrastructure’ is described as built ‘primarily from three pillars: secured transactions, insolvency, and credit reporting frameworks’.

The Benefits of a Reformed Secured Transactions Regime 397 and effective access to finance, stability and socially responsible economic growth’,8 is key to removing this constraint. MSME access to finance can be improved by reforming three credit infrastructure priority areas.9 The first is the system for credit reporting, which gives lenders objective information, enabling them to expand their lending portfolios while managing risk. The second area is the secured transactions regime, which facilitates lending against movable collateral. The third area is the insolvency regime, which should, if operating effectively, save viable businesses while also promoting the smooth exit of businesses that are not economically efficient. Reforms in these three areas, but especially the secured transactions and insolvency frameworks, should be coordinated.10 This chapter focuses on the second of these areas, and, in particular, discusses some of the attributes necessary for registries to operate effectively as well as other aspects of the enabling environment required to encourage lenders to take movable assets (a core component of the borrowing base of SMEs) as collateral. It is necessary at this point to reflect briefly on the fundamental nature of credit, because this will inform the reasoning underpinning much of which follows in this chapter. There is a considerable amount of literature11 on the virtues of debt, wages and export-led growth, and this chapter will not analyse the arguments in any detail. The important conclusions of the debate are that a secured transactions regime supports access to finance, and that there is a well-established link between SME economic activity and a modern secured transactions regime.12 The ‘contest’ between SME-led, debt-enabled economic growth and stability is not expanded on here.

B. The Benefits of a Reformed Secured Transactions Regime This section considers the potential benefits of a secured transactions regime reformed along the lines of the scheme discussed in chapter 2. Such a regime helps to reduce the cost of, and increase access to, finance for SMEs in a number of ways. First, it broadens the asset base that can be used as collateral to secure lending.13 Typically, SMEs have limited or no immovable assets but have rights in a wide range of movable assets. Registries for security rights in movable assets help to address the critical asymmetry between these movable assets and the immovable assets preferred by traditional financial institutions, particularly banks. Secondly, it increases the ability of financial insitutions to assess and manage risk. Creating a central repository of much (or, in some jurisdictions,14 all)

8 World Bank, ‘Credit Infrastructure’, www.worldbank.org/en/topic/creditinfrastructure. 9 G20, ‘Action Plan on SME Financing: Joint Action Plan of G20 GPFI SME Finance Sub-Group and IIWG’ (2015) www.gpfi.org/publications/g20-action-plan-sme-financing-joint-action-plan-g20-gpfi-sme-finance-subgroup-and-iiwg. 10 See C1a below. 11 See, eg, EJ Medina-Smith, ‘Is the Export-Led Growth Hypothesis Valid for Developing Countries? A Case Study of Costa Rica’, Policy Issues in International Trade and Commodities Study Series No 7 (UNCTAD, 2001) www.unctad.org/en/docs/itcdtab8_en.pdf. 12 IBRD (n 6). 13 See Conclusion B1a. 14 This would be the case where registration was the only means of making a security interest effective against third parties, see C2a below and Ghana D3; Nigeria E2; Kenya C3; Sierra Leone C3.

398 The Infrastructure Underpinning a Modern Secured Transactions Law information relating to security rights in movable property (a registry) increases certainty and simplicity for creditors in making lending decisions by integrating all relevant information into a single source. Arguably, this reduces transaction costs, which benefits the borrower. The registry also helps protect buyers by providing a searchable source of security rights for buyers of second-hand goods, such as motor vehicles, and of intangible rights, such as intellectual property. The majority of the resulting additional credit generated in the economy is likely to be productive, that is, is used for capital investment that generates employment and income growth. This is the case, not only in relation to business financing, but also consumer credit. The OECD highlights the importance of consumer spending for economic activity, which typically makes up around 60 per cent of GDP,15 with varying percentages supported by credit. Because of the importance of this supply of credit, a modern secured transactions regime incorporates security rights over collateral for both business and consumer use. The inclusion of consumer credit in the regime is strongly related to the premise that the regime is particularly focused on supporting SME access to finance. For example, an SME may offer collateral which, on occasion, may be used by the business owner both for business and personally (eg a motor vehicle). Additionally, the SME is likely to sell goods to consumers on credit, and the receivables represented by these credit contracts may then also be used as collateral. Although not common, some registries capture information on whether the security right relates to an obligation with commercial or consumer property as collateral. For example, in Australia, as of March 2019, commercial property was the most common collateral type, accounting for 75 per cent of registrations.16

C. Components of a Modern Secured Transactions Regime As discussed in the previous sections, an effective secured transactions regime facilitates lending by allowing SMEs to use their movable assets as collateral. This flexibility must be built into the legal framework and be supported by a publicly accessible registry.

1. Legal Framework As discussed in chapter 2, the law must establish an efficient set of legal rules covering relevant aspects of the creation, effectiveness, publicity, priority and enforcement of security rights. It is also important that the law gives the registry legitimacy. Most public registries are essentially regulatory in nature, with the registry used either as a compliance tool or a means of publishing decisions in relation to individuals or entities – for example, business registries or occupational licencing registries. Supporting legislation for registries such as these ensures that the system operates in a fair and equitable manner, so that, for example, only those that meet certain criteria receive authority to perform their occupation.

15 OECD, ‘Household Spending’, https://data.oecd.org/hha/household-spending.htm. 16 For Australian PPSR statistics, see Australian Financial Security Authority, ‘PPSR Statistics’, www.afsa.gov.au/ statistics/ppsr-statistics; for an African example, see Liberia C5a.

Components of a Modern Secured Transactions Regime 399 The introduction of a modern security rights regime will involve a new legal regime brought about by codified legislation, which will entail the repeal of pre-existing legislation and which will replace previous case-law.17 However, although the legal regime is implemented by statute, the agreements underpinning credit are contracts between private actors, and are governed not only by the new legislation but also by other private law.18 In this sense, government maintains no direct interest in the relationships referred to in registered notices. Further, government does not play any role in creating those relationships nor, usually, is there any State requirement to complete the relevant registration process when creating the relationship.19 Rather, this type of registry exists as a form of economic infrastructure to be used by those entering into secured transactions for the benefit of commerce.20 However, in providing the legislative framework for the establishment and operation of the registry, the power of the State is usefully exercised in creating an environment that incentivises market participants to search and register notices with respect to their interests in the collateral. The next two subsections consider how the legal framework incentivises various stakeholders to use the registry, and what elements are required to maximise this use.

a. Incentivising Registration The first means of incentivising individuals and entities to publish information about their interests in collateral is to grant those who choose to publish information through a registration process certain rights. A key consequence of a modern secured transactions system is that secured creditors are given priority rights in the event of grantor default and insolvency.21 For this right to be enforceable, the system needs to be supported by legislation that provides secured creditors with a process for enforcing their priority rights.22 It also requires a level of harmonisation between secured transactions law and other legislation, particularly insolvency legislation, and its policy and practice.23 This harmonisation with the insolvency regime includes ensuring secured creditors gain the highest priority in terms of dealing with the collateral secured by their interest, or being first on the list of those paid (usually following costs of administering the insolvency),

17 See Conclusion C6. 18 See, eg, paras 19–20 of the draft UNCITRAL Practice Guide to the Model Law on Secured Transactions, which emphasise that many other types of law other than the secured transactions legislation will apply to an actual secured transaction, including contract law, and the ML, Art 3, which makes clear that, with very few limits, parties are free to include whatever provisions they wish in their security agreement. ‘Draft Practice Guide to the UNCITRAL Model Law on Secured Transactions’ (UNCITRAL A/CN.9/WG.VI/WP.79, 27 August 2018) www. undocs.org/en/A/CN.9/WG.VI/WP.79. 19 Note that in some African countries the criminal sanction for non-registration, which was part of the inherited English law legislation, has not been repealed, see Kenya B4. 20 The Finance Minister’s Cebu Action Plan includes the advancement of secured transaction (and insolvency) systems as key elements of ‘promoting an enabling financing environment’; see APEC, ‘Annex A – APEC Finance Ministers’ Process (FMP) Roadmap/Cebu Action Plan’ (2015) www.apec.org/Meeting-Papers/ Sectoral-Ministerial-Meetings/Finance/2015_finance/annexa. 21 See ch 2, D. 22 Although there is still some utility in creating a collateral registry in isolation of its use in attracting legal protections, it is likely such a system will only incentivise certain market sectors to participate. 23 For an example, see Conclusion C6bii.

400 The Infrastructure Underpinning a Modern Secured Transactions Law ahead of other creditor groups, including employees and tax authorities.24 The coordination between regimes must also relate to the enforcement rights of secured creditors, and should ensure predictable, transparent and affordable enforcement of rights for all secured creditors (defined widely under the functional approach)25 both outside and within insolvency. Coordination of the secured transactions and insolvency regimes can have wider benefits as well. The importance to secured creditors of effective registry and insolvency regimes is demonstrated by the following argument. Commerce is a system of commercial relationships that relies on contractual agreements between two parties. Since capital and credit, in their various forms, are the lifeblood of modern commerce, transparency, accountability and predictability are fundamental to sound credit relationships. The availability of credit at a reasonable price is predicated on the confidence of the creditor that the debt will be repaid. Pricing the credit risk, often represented by the interest rate offered, includes taking account of all information available to the creditor and its knowledge of competitive forces. The provision of information by the collateral registry and by credit information systems (credit bureaus or credit registries) helps achieve this end. A well-functioning insolvency regime is also critical, as competition in credit delivery is adversely impacted by lack of predictable legal mechanisms for debt enforcement, recovery and restructuring.

b. Incentivising Searching The second incentive is for prospective creditors and buyers of second-hand goods to search the registry, and is more difficult to achieve because it requires ongoing investment in awareness raising. As discussed in chapter 2, under a modern secured transactions law, a buyer of goods not in the ordinary course of business will take subject to a security right previously made effective against third parties.26 In this situation, the incentive for searching is buyer protection, to ensure that the buyer does not purchase an asset without knowing, or when they should have known, that it is already encumbered. Arguably, responsible lending could not occur without first searching the registry at least once during the lending process. Even where a financier will obtain ‘super-priority’ in relation to an asset, because it is financing the acquistition of that asset,27 searching the collateral registry is good practice as part of due diligence to determine the level of exposure and creditworthiness of the grantor. Lenders in advanced economies generally search the registry a minimum of twice, and often three times throughout the lending process.28 In Africa, following the introduction of registries (after the passage of a relevant law), experience indicates that uptake of searches is usually slow. Perhaps even more worrying in these jurisdictions is that the volume of searches is much lower than the level of registration activity.29 This indicates that even creditors extending new borrowing are not searching 24 Note, however, that the ML permits priority for these groups, up to a stated amount (Art 36) and see also ch 2, F4. 25 See ch 2, B2. 26 Ch 2, F3. 27 See ch 2, F2 for discussion of acquisition security rights. 28 Discussions between the author and lenders in such economies. 29 See, eg, Kenya C4 (in the first few months after the registry was launched, the ratio of registrations to searches was 100:1, and although it has reduced since is still very high) and Ghana D4b (in 2016 the ratio was 2.7:1 and in

Components of a Modern Secured Transactions Regime 401 the register as part of their due diligence. The IFC suggests a high ratio of seaches to new registrations, quoting examples of between three and six searches for every new registration as an indicator of a ‘prudent lending technique’.30 As a comparator to the African figures, the latest data from New Zealand show six searches for each registration31 and Australia around four.32 It is particularly important that search results are simple and easy to understand. A registry exists to deliver information to searchers. Therefore, simplification of searching should be the starting point for registry design. However, even the best design efforts are likely to cause some difficulties in practice. For example, a security right over inventory extends to the proceeds from the sale of that inventory, even though these proceeds are not identified in the registered notice.33 Although the majority of searches can be expected to relate to a direct encumbrance over the assets described in the notice, this example demonstrates that occasions exist where legal judgment will be required to determine the scope of the coverage of the registered security right. It is difficult to obtain data to assess whether the existence of a modern registry (where registration and searching are easy) actually leads to significantly more credit being advanced. The mere fact that there is a registration relating to a security right does not establish whether the borrower is a first-time borrower, or whether the loan would have been made anyway, irrespective of the introduction of the reformed registry. One possible approach, which is not infallible, is to include a specific field in the registered notice which records whether the grantor is a first-time borrower, thus generating statistical evidence.

c. The Role of Legal Infrastructure In addition to the incentives discussed in the last two sections, the registrar plays a critically important, and continuing, role in raising awareness about the registry, its purpose and incentives. Insolvency practitioners and the judiciary also play a significant part. Where a creditor loses priority due to a failure to enter a registration on the registry, it is important for the rule of law to be applied consistently. An example illustrating this proposition is the Australian case of Forge Group Power Pty Limited v General Electric International Inc.34 In that case, General Electric International Inc (GE) leased turbines to Forge Group on an operating lease which fell within the ambit of the Australian PPSA. GE failed to register the lease, and this was fatal, in that GE lost its title to, and interest in, the turbines, which (under the Australian PPSA) vested in Forge Group immediately before the appointment of voluntary administrators. As a consequence, GE lost a reported AU$50 million.

2018 2.2:1). The ratio of searches to registrations was better in Malawi (1.3:1 in 2016, see Malawi A) but still not optimal. 30 A Alvarez, S Downes and B Hennig, Making Security Interests Public: Registration Mechanisms in 35 Jurisdictions (Washington, DC, World Bank, 2012) www.doingbusiness.org/content/dam/doingBusiness/media/ Special-Reports/Making-Security-Interests-Public.pdf; see also Conclusion C4. 31 For information on the New Zealand PPSR, see New Zealand Companies Office, ‘Personal Property Securities Register’, www.ppsr.companiesoffice.govt.nz/. 32 For Australian PPSR statistics, see Australian Financial Security Authority (n 16). 33 See ch 2, D. 34 Official Committee of Unsecured Creditors of Motors Liquidation Co v JPMorgan Chase Bank [2016] NSWSC 52.

402 The Infrastructure Underpinning a Modern Secured Transactions Law In the US case of Official Committee of Unsecured Creditors of Motors Liquidation Co v JPMorgan Chase Bank,35 the lenders, JP Morgan, had entered into two separate secured transactions, the second of which was a loan of US$1.5 billion. The secured obligation in the first transaction was fully discharged, and a security right securing the obligation under the second transaction was mistakenly included in the termination notice relating to the security rights granted to secure the discharged obligation. US Court of Appeals for the Second Circuit held that the termination notice was effective according to its terms, which had the result that the effectiveness of the mistakenly referenced notice was terminated. This case demonstrates that failure to maintain effective control over the maintenance and procedures around the removal of registrations may be equally fatal to the commercial interests of an entity. The following quotation from a much earlier case36 is instructive: Although strict adherence to the [secured transactions law] requirements may at times lead to harsh results, efforts by courts to fashion equitable solutions for mitigation of hardships experienced by creditors in the literal application of statutory filing requirements may have the undesirable effect of reducing the degree of reliance the market place should be able to place on the [secured transactions law]. The inevitable harm doubtless would be more serious to commerce than the occasional harshness from strict obedience.

2. The Registry a. Variations in Approach to the Registration System in Africa Registration in the collateral registry is the primary mechanism for secured creditors to protect their rights, including ensuring priority over competing claimants if the grantor defaults or becomes insolvent. In many African States, the registry is a critical component, not only for parties to a security agreement, but also for non-consensual creditors, such as judgment creditors and the relevant revenue authority, whose rights must also be publicised by registration in the collateral registry.37 The purpose behind this approach is to make claims on the grantor’s assets transparent, thus increasing the information available to creditors on which they can make informed risk management decisions. Despite the critical role played by registration as a form of third-party effectiveness in economies with modern secured transactions regimes, in very many jurisdictions there are also other forms, such as possession and control.38 Nevertheless, some African jurisdictions have adopted a model where registration is the only form of third-party effectiveness.39 This has been the approach in Ghana,40 Nigeria,41 Kenya42 and Sierra Leone.43 In Zimbabwe,

35 In re Motors Liquidation Co. (2015) 777 F3d 100. 36 Security Nat Bank and Trust Co of Norman v Dentsply (1980) 617 P2d 1340, 1343. 37 See Conclusion D1b for more detailed analysis. Examples discussed in this book are Sierra Leone C4; Nigeria E4; Zambia D1; Kenya C4; Zimbabwe C3b; Rwanda C1: Liberia C4 and C5a; OHADA B7. 38 See ch 2, D. 39 See Conclusion D4. 40 Ghana D3. 41 Nigeria E2. 42 Kenya C3. 43 Sierra Leone C3.

Components of a Modern Secured Transactions Regime 403 although this approach was considered, it was ultimately rejected and common law methods of third-party effectiveness are still permitted, although registration carries with it advantages as to priority.44

b. Challenges in Establishing Registries in Africa Additional challenges arise with respect to the critical ‘twin pillars’ of the legal framework and registry where a legislative provision catering for the time needed to establish the electronic registry is not included in the statute enacted by the parliament. Zambia is one example where the parliament passed the legislation without any provision for time to establish the registry. The legislation became effective on proclamation, irrespective of the existence of a registry.45 If it had not been for the excellent work of the team from the IFC and the software vendor, supporting the local team from the host Patents and Companies Registration Authority (PACRA), significant disorder might have reigned for some time (at least in the commercial legal sense).46 However, there is still one ongoing issue in Zambia, which is the lack of clarity as to whether certain devices fall within the scope of the legislation. As a result, PACRA has had, of necessity, to give advice to potential registrants as to this issue, although this could have amounted to legal advice which the registry is prohibited from giving.47 In Kenya a registry was developed, initially without the vendor of the electronic registration system having a good understanding of some of the core requirements. To be effective, a registry should be easily accessible and searchable for those who wish to manage their risk by registering or searching the registry. Due to political forces, the Kenyan registry was initially launched with only the ability to make registrations and no search functionality was provided.48 The OHADA block of countries is also a challenging environment in which to establish effective registries. Since the significant achievement of the entry into force of the OHADA Uniform Act in 2011, the majority of jurisdictions have struggled to advance any form of electronic registry, thus substantially restricting the ability of the reform to achieve its development objectives.49 Although the work of the World Bank Group assisted greatly in overcoming challenges in each of these cases, these examples, drawn from the author’s experience, highlight the importance for jurisdictions to understand and consider the unique aspects of modern secured transactions reform. In particular, the establishment of a highly accessible registry is critically important. The legislature must be attuned to these factors, so as to be able to avoid some of the pitfalls arising from political forces. The starting point for guidance should be the Model Law, which provides an extensive view of relevant considerations relating to both the legal framework and the registry. However, it should be noted that some good-practice considerations, such as the registra-

44 Zimbabwe

C1. Zambia B and E. 46 See discussion in Zambia E. 47 See Zambia D2. 48 Kenya C. 49 OHADA I. 45 See

404 The Infrastructure Underpinning a Modern Secured Transactions Law tion of some non-consensual liens (eg tax liens) are not expressly provided for in the Model Law, although other competing claims, such as those of preferential creditors and judgment creditors, are included.50 Thus engaging with experts, such as those from the multilateral development institutions, or other entities with direct knowledge and experience, is also important, both for their knowledge of registry design and practice, and their understanding of nuances in the legal framework.

3. Key Elements that Support an Effective Registry The effectiveness of a registry ultimately depends on how easy and efficient it is for users to interact with it, and whether users have trust and confidence in the regime. These qualities are engendered by the elements discussed in the following subsections.

a. User-Centricity It must be easy both for those wanting to publish information on the registry to do so, and for those with an interest in the information to find it. The following recommendations are made based on the author’s experience of establishing and continually improving Australia’s Personal Property Securities Registry (PPSR). First, user-centred design must be harnessed to improve user experience. The government cannot be the arbiter of what customers need without having asked users what they want and having analysed and understood the responses. Engagement with users of the registry (including future users when designing and building a registry for the first time in any jurisdiction) will assist both in building knowledge by all parties as to the requirements, and in managing expectations of those who intend to use and rely on the registry. Secondly, it is important to work with key user groups to address usability issues. This assists with identifying and removing unnecessary barriers, and in streamlining processes to deliver user benefits, which encourages registry use. As highlighted in earlier chapters,51 motor vehicles are a commonly used form of collateral across Africa. Creditor risk management practices across the continent are resulting in consideration being given to linking the collateral registry with the local motor vehicle registry in many jurisdictions. In some cases (eg Mozambique)52 legal requirements dictate that only the owner of the vehicle has the right to offer the motor vehicle as collateral. This means that a ready and efficient link between the motor vehicle ownership registry and the collateral registry is of particular importance. Thirdly, it is critical to maintain ongoing stakeholder engagement by meeting regularly with a broad cross-section of stakeholders to test ideas, to drive an innovation agenda and to discuss issues that cut across industries. In Australia, this assisted the PPSR to understand how stakeholders interact with the registry, to build trusted relationships, and to address emerging risks before they became significant issues. Sometimes, decisions need to be made

50 ML,

Arts 36 and 37. See ch 2, F4. eg, Liberia A; Sierra Leone A. 52 Mozambique is not specifically covered in this book. 51 See,

Components of a Modern Secured Transactions Regime 405 in the interest of maintaining the integrity or functionality of the registry that do not meet the specific needs of certain industry groups. In such cases, in order to build a trusted relationship with key users, the registry’s operator should be open with industry about why decisions are being made, demonstrating fairness and upholding the purpose of the registry. Building a trusted relationship also assists in managing the response should unplanned events arise. For example, in Malawi the tobacco industry (who were a major supplier of credit to farmers and who took security over the tobacco crop) submitted a request to be able to link directly to the registry. Unfortunately, the registry in Malawi was developed without an application programming interface (API) channel53 which could support such access readily. Although such a request in an economy with the characteristics of Malawi might have been unexpected, this is an important reminder that any economy will include actors with varying levels of sophistication. The fact that the request from the tobacco industry came some time after the launch of the registry is an important result of ongoing stakeholder engagement. Additionally, this example highlights the need to take a flexible, modern approach to the design of the registry from the outset.

b. Operating Model Collateral registries will always remain the ultimate responsibility of a public body. However, certain activities and functions do not need to be directly controlled by the government, especially in a fully digitalised environment. Government can use a range of models to deliver and facilitate registry services. For example, it can own or lease all or some of the registry-related hardware and software. Although the government must retain ultimate responsibility for the registry and ownership of its data, all other aspects may be outsourced to public or private entities.54 A number of governments have outsourced the hosting of their collateral registries to the company that developed the collateral registry software, though this has not been the case in Africa.55 In choosing the right model, governments need to strike a balance between the desire to limit their role in keeping the technology up-to-date, and to harness private-sector innovation to develop a full suite of value-adding services, as well as the need for government to be able to intervene in a timely manner to protect users and access.

c. Administrative Processes Administrative processes must ensure the orderly and lawful creation of registrations and access to registry information. In the past, this would have been entirely achieved through 53 See below for an explanation of an API channel. 54 Secured Transaction Systems and Collateral Registries (Washington, DC, World Bank, January 2010) 73. Similar limits and allowances for outsourcing of IT and FinTech services apply to financial institutions. See, eg, Final Report on EBA Guidelines on Outsourcing Arrangements (London, European Banking Authority, February 2019) 6 (‘The responsibility of the institutions’ and payment institutions’ management body for the institution or payment institution and all its activities can never be outsourced. [] Functions that are considered critical [] may also be outsourced’). 55 New Brunswick, Newfoundland and Labrador, Nova Scotia, and Prince Edward Island formed the initial partnership with UNISYS in 1996. Northwest Territories and Nunavut signed on in 2001, and Yukon joined in 2016. ACOL, ‘Personal Property Registry System’, www.acol.ca/en/pprs/about/what-is-acol.

406 The Infrastructure Underpinning a Modern Secured Transactions Law rules and procedures underpinned by a control environment. Today, technology has the capacity to automate certain steps by using system-based business rules. One example is the introduction of functions such as automatic validation of company names and numbers at the point of completing a registration or search. Many registries aspire to this feature, though few have successfully implemented it.

d. Access and Availability In a sophisticated, highly digitalised financial sector that operates across time zones, participants must be able to register and search from anywhere at any time. Otherwise, there is a potential for collateral to be inadvertently dealt with by others. There are various critical enabling capabilities for access and availability. One is the use of mechanisms such as an open API for high-volume users to integrate their business systems with the registry. This improves efficiency for these types of users and facilitates greater reliance on the registry as a business tool by allowing industry users to build industry-specific solutions. One example is the integration of registration as a new step in the lending software of a bank as an alternative to having separate access to the registry for the credit officer. Another capability is the updating of the database in real time, rather than in batches, combined with minimal planned maintenance downtime and robust technology that prevents unplanned outages. The implementation of a disaster recovery site in Malawi is one excellent example of such an approach to the technology. However, the best technological implementation is only as effective as the manner in which it is used. Thus, appropriate, effective disaster recovery and business continuity planning that is regularly tested is also an essential element.

e. Data Integrity If the registry is to be a trusted means of assessing transactional risks, users must have confidence in the reliability of its information. It is therefore essential to preserve the integrity of data held on the registry. Failing to manage data integrity actively will, over time, lead users to lose faith in the results they receive, and will lead secured creditors to start questioning the effectiveness of the registry as a notice board. Data integrity should be considered from three perspectives: first, the need to prevent the corruption of data; secondly, the need to ensure that no data is lost; and thirdly, the fact that it is desirable to minimise the number of registrations that are no longer active or which contain incorrect or out of date information. i. Cyberattacks and Privacy Although digitalising public registries has delivered significant benefits by improving accessibility and usability of related services, it has come at a price, namely, the potential for cyberattacks or the inadvertent introduction of errors into the underlying database. Putting in place suitable measures to manage associated risks, while maintaining data confidentiality, integrity and accessibility, is extremely important. An appropriate security posture should be adopted by all registry operators. Importantly, a registry operator should consider all vectors of security, including physical and personnel access, along with the information technology practices. An overarching requirement should be an effective governance

Components of a Modern Secured Transactions Regime 407 framework to monitor and manage the ever moving threat environment to maintain relevant controls. Confidentiality at the individual transactional level is rarely a primary concern for operators of collateral registries, given that they are, by their nature, a form of notice board. That said, some elements of the operation of a registry create data, or links between data, that need to be protected. For example, in Australia, where there is a long-standing concern about privacy, the registrar is required to manage the risk that information about security rights of grantors who are natural persons could be accessed via the registry for an improper purpose. In practice, though, the system controls do not prevent a paying searcher from accessing this information. However, the search results provided, and the automatically logged IP addresses of the computer from which a search is conducted, are recorded and retained for use in any future enforcement actions. A key consideration when making decisions regarding these matters is to ensure that any controls implemented are proportionate to the level of risk. The failure to adhere to this principle will either mitigate the risk inadequately or will produce a situation where the control unnecessarily impacts usability and accessibility. ii. Incorrect Data Being Placed on the Registry Due to Human Error It is always possible that a registrant will include incorrect data in the registered notice. Mitigation strategies for this phenomenon include the incorporation of business rules56 designed to minimise incorrect information being entered, particularly where there is a numerical reference required in relation to collateral (eg the serial number of a motor vehicle). Another strategy is the integration of the registered information with other sources of truth to enable secured creditors to cross-check the details they want to enter against an authoritative source. For example, the Australian PPSR allows users to cross-check a company number with information held on the companies registry. Australia also provides information with respect to motor vehicle status and other details via a direct link to the National Exchange of Vehicle and Driver Identification System. A number of African economies are working towards adoption of similar principles. In particular they aim to connect motor vehicle and company registries directly with the collateral registry in order to assist the identification of assets as well as grantors and secured creditors. However, legacy technology or technological architecture (such as the adoption of an open API) or lack of digitisation across the spectrum of registries and databases has limited progress in achieving such integration to date. iii. The Unauthorised Placing of Incorrect or Inappropriate Data on the Registry Occasionally someone purporting to be a secured creditor or their agent makes an unauthorised registration. This category of registration includes those that are frivolous or vexatious. The primary mitigation strategy for this phenomenon is to require certain users to authenticate themselves as part of the transactional process, while ensuring that the level

56 ‘Business

rules’ in this context means the rules of the system.

408 The Infrastructure Underpinning a Modern Secured Transactions Law of verification is proportionate to the level of risk being managed. Where there is a low level of risk, authentication should be kept to an absolute minimum due to its impact on usability, whereas the presence of very high risks will result in users only being able to transact after completing some form of identity verification process. Given the broad coverage of transactions supported by a modern secured transactions regime, if access to the registry is subject to onerous procedures this is likely to add cost and time to the relevant transaction, which is contrary to the underpinning objectives of a notice filing approach. iv. Allowing Duplicate, Out-of-Date or Redundant Data to Remain on the Registry These types of data create ‘clutter’ that adversely affects the usefulness of search results. In practice, some registrations are not well managed or maintained by secured creditors. Occasions exist where the grantor has fulfilled the secured obligations but the secured creditor has not cancelled the registration. The grantor seeks to have the registration cancelled (through the relevant procedure provided under the law)57 but the secured creditor is not readily identifiable or contactable. This causes distress and potentially creates a loss for the grantor because a new creditor may not be willing to lend without the removal of the registration. Mitigation strategies for this problem include engaging with stakeholder representative groups to highlight the value of discharging and amending registrations in a timely manner. Another strategy is the linking of the collateral registry to pre-existing databases for authentication,58 which also assists in overcoming challenges in identifying the secured creditor. A third strategy is the implementation of transactional surveillance to identify potentially problematic registrations, which enables the registry to work with secured creditors to amend or discharge those registrations where necessary. v. Incorrect Information Entered Due to Misunderstanding Registry users sometimes enter incorrect information because they do not understand how to complete each field. This risk can be mitigated by engaging with user focus groups to identify the source of confusion and by developing simple, easy-to-read support materials in partnership with industry.

D. Credit Infrastructure Elements that Support Affordable Credit Although not the focus of this chapter, it is important to recognise that a secured transactions regime forms one part of a broader financial system, operating in a real economy. Thus, it is useful to highlight some key aspects in the enabling environment at this time. Prudential regulators across the globe constantly strive to strike a balance between the objectives of financial stability and economic growth. Within this general policy framework, international capital requirements have been drafted to try to ensure a safe and sound

57 See 58 See

ch 2, E1. C3Ei above.

Credit Infrastructure Elements that Support Affordable Credit 409 banking system without stifling banks’ support to the real economy, largely sustained by SMEs. However, international standard-setters have tended to gear capital requirements progressively to favour financial collateral (eg rated government and corporate bonds) and immovable assets (eg land) over physical collateral and receivables (movable assets). This trend has been reinforced by the global regulatory reforms59 since the 2008 GFC, when the shortage of highly liquid assets threatened the global financial system. As a result, despite the efforts to stimulate asset-based lending through legal reforms, the flow of affordable credit to SMEs remains limited. To help address this issue, various core elements need to be addressed. These are discussed in the following subsections.

1. Prudential Regulation Through the capital requirements enshrined in the Basel Capital Accords,60 the internationally harmonised regulatory framework strives towards ensuring the solvency of banks and the stability of the overall banking system. Within this context, through the considered balance of financial stability and growth policies, multilateral development programs could be used more effectively to encourage the extension of asset-backed loans if, when setting them up, consideration was paid to the credit ecosystem and to undertaking activities to further support and incentivise the take-up of movables-based lending. In particular, banks and other lenders need a set of incentives to accept movable assets as collateral. This means devising a sound regulatory strategy at the national level to ensure that the incentives brought about by secured transactions law reforms and capital requirements are aligned.61 Under the internal ratings-based approach of the Basel Accords, for a security right in a movable asset to be classified as eligible collateral, the financier must demonstrate that it has priority over competing claims, which is more difficult in an environment without an efficient registry system.62 The registry may also be an important source of information for regulatory agencies as activity may show signs of imprudent lending, evidenced by the disproportionately high number of registrations to searches. Supervisors should be encouraged to monitor prudence by requiring financial institutions to embed in their internal systems of controls evidence of performing an appropriate number of searches throughout the lending process, both in the collateral registry as well as the credit registry or bureau.

2. Lending Practices The lack of a good understanding of modern lending practices, without proactive loan portfolio management (particularly collateralised loans), is likely to contribute to heightened 59 Those already implemented and currently being discussed. 60 See, generally, Basel Committee on Banking Supervision, Minimum Capital Requirements for Market Risk (Basel, Bank for International Settlements, January 2019, revised February 2019). 61 Conclusion C5. 62 G Castellano and M Dubovec, ‘Global Regulatory Standards and Secured Transactions Law Reforms: At the Crossroad Between Access to Credit and Financial Stability’ (2018) 41 Fordham International Law Journal 531, 572.

410 The Infrastructure Underpinning a Modern Secured Transactions Law levels of non-performing loans, thus affecting the incentive to lend as well as the price of credit. Further, the absence of effective collateral management, including broad-based valuation expertise and lack of collateral monitoring expertise, impedes the lending process. As the lending environment matures, new asset classes emerge, creating new markets and industries. Initially, specialist collateral managers and valuation experts are employed to provide the necessary expertise in structuring transactions, as businesses seek to use the assets they have as collateral and financial institutions wish to enter the market to drive increased profits, while managing risk. Although these arrangements may increase transactional costs, they are a stepping stone to innovation that generates products and services not seen in the market previously. The use of collateral management arrangements may help banks obtain capital relief under the Basel Accords, but they are also an important risk mitigant that makes banks more comfortable lending in Africa.63

3. Secondary Markets for Collateral Competitive secondary markets, which provide confidence in the liquidation value of a certain asset after default, are key to diversifying the types of assets a lender will take as security for a loan. Moreover, efficient secondary markets for particular types of assets means that these can be added to the list of eligible collateral which can be used for calculating capital adequacy ratios. Ultimately, the more expansive this list, the more resources a financial institution has available for lending operations. The more assets on the ‘list’, the more likely there is to be a reduction in the risk attached to lending to SMEs. A mature and efficient secondary market, possibly supported by a FinTech delivery model, will also reduce transaction costs, and significantly improve the cost of enforcement for credit providers. It will also assist in dispute resolution and aid in the valuation of collateral at various stages of the lending cycle. The reduction in transaction costs should be reflected in risk-pricing models and reduce the price of credit available to SMEs.

4. Market Opportunity Assessment To create the business case for taking steps to improve the effectiveness of a secured transactions regime, a government, often with the support of multilateral development agencies, needs to assess the opportunities specific to its economy. The market assessment should identify strategic sectors ripe for targeted industry or product development, as well as all barriers to further advancing SME lending. Whether the opportunities are derived from domestic value chains and consumption or founded in export markets, the breadth of a modern secured transactions regime ensures their relevance. Additionally, consideration needs to be given by those implementing secured transactions reform to the broader ecosystem of financing options for SMEs. This requires goverments to focus on the development of capital markets and conducive environments for alternative lending. Many of these are supported by the introduction of a broad-based 63 R Spong, ‘Understanding Collateral Management’ Global Trade Review (17 January 2018) www.gtreview.com/ magazine/volume-16issue-2/understanding-collateral-management/.

Conclusion 411 secured transactions regime. The 2017 SME Finance Forum report points out that 40 per cent of SMEs globally are credit constrained.64 The most recent edition of the annual OECD report into SME financing65 indicates that, although debt financing remains the most prominent form of SME financing and has seen some growth following a depressed period after the global financial crisis, volumes are generally up for leasing, factoring, alternate finance, venture capital and private equity. Having identified market creation opportunities and the hurdles that must be overcome, a roadmap can be developed to help advance the flow of credit to SMEs. As a concept, movable asset-based lending will take time to develop in any new market. In a lower-income economy, the opportunities are likely to be found in supply chain financing (eg factoring) and leasing. Often, a pilot project can be delivered in close collaboration with local financial institutions and government counterparts to assess and track the impacts of relevant aspects of the credit cycle on the success of movable-based asset lending.

E. Conclusion The rationale for developing a modern and internationally harmonised legal framework for security rights in movable assets is to broaden access to finance at a lower cost, and to further international trade and economic growth. Such frameworks serve borrowers such as SMEs that, typically, fall outside banks’ business lines. This also allows banks to further develop their credit products.66 To ensure a secured transactions regime is able to achieve its full potential, governments have a role to play in making the enabling credit environment as supportive as possible. This includes putting in place an underpinning legal framework, with appropriate rules for secured creditors and supporting legislation that gives the registry legal effect and ensures the system operates in a fair and equitable manner. The registry itself must be operated effectively, ensuring it is easy to access and that users’ interactions with it are efficient. This will engender trust and confidence in the regime. Informed by the author’s experience, this chapter analyses some techniques which can be used to design, implement and run a registry which fulfils these criteria. In order to enhance credit markets, prudential regulation and lending practices will also need to be addressed. Equally, competitive secondary markets must be developed to support the effective pricing of credit. However, attention must also be paid to the broader ecosystem to support the financing needs of all SMEs, including through the various stages of growth. With these core elements in place, secured transactions systems will be more likely to operate effectively, enabling SMEs in particular to use their assets as security to generate productive capital. There are significant challenges, however, particularly in developing markets such as found across Africa. For example, there needs to be awareness of the revised legal framework 64 MSME Finance Gap:Assessment of the Shortfalls and Opportunities in Financing Micro, Small and Medium Enterprises in Emerging Markets (Washington, DC, World Bank, 2017). 65 Financing SMEs and Entrepreneurs: An OECD Scoreboard (Paris, OECD Publishing, 2019). 66 See ch 1, C4.

412 The Infrastructure Underpinning a Modern Secured Transactions Law and understanding of its application in practice in order for the market to take advantage of the opportunity presented by the modernisation of the secured transactions system. This awareness must extend to the judiciary and law enforcement bodies to ensure that enforcement, including extrajudicially, of security rights is effective. Further, the adoption of direct access notice-based registration practices is critical to the reform succeeding. Each of these considerations is essential to reducing transaction costs, which should result in a reduction of the cost of access to finance, a key element to addressing the well-documented financing gap, particularly for SMEs.

Conclusion A. Introduction As well as providing an overview of the state of secured transactions law in various countries in Africa, the discussion in the chapters in this book enables some general analysis to be undertaken concerning secured transactions law reform, both in Africa and elsewhere. This Conclusion provides a thematic analysis of the issues. It starts with an examination of the drivers towards reform. It then examines the reform process, and highlights a number of features which can prevent that process from operating smoothly and efficiently. The third part looks at some matters of substantive law, illustrating where deviations from the modern principles1 have been adopted because of local conditions or initial drafts that were not entirely based on international standards,2 and the challenges arising from this approach. The final section summarises the lessons learnt from the African experience so far, and considers the way forward for secured transactions law reform in the African continent.

B. Drivers for Reform The drivers towards reform vary considerably from country to country, but there are many common features. Some of these are not specifically about law: they stem from economic and other considerations, and these are discussed in the first part of this section. The second part discusses particular drawbacks in existing law and how these can be ameliorated by introducing a new regime based on the modern principles.

1. General Reasons a. The Desire for Economic Growth There is little doubt that one of the chief drivers towards reform is the desire to increase the availability and lower the cost of credit.3 These imperatives are, rightly, perceived as likely to lead to economic growth and increasing prosperity. They may also have distributive effects, since it is usually micro-businesses which have the least ability to access credit,4

1 As

discussed in ch 2. example, Ghana, ch 3 and Rwanda, ch 14. 3 The difficulties of accessing credit are noted in most chapters, see Zambia, ch 8, F. 4 See Nigeria, ch 7, A and B; Malawi, ch 6, A; Zambia, ch 8, B; Ghana, ch 3, B. 2 For

414 Conclusion and by improving this access the position of some of the poorest people in a country can be improved. This is also true for women.5 A related driver is the desire to widen the available collateral against which lenders will lend. In many countries with unreformed law, certain asset classes were just not used as collateral, partly (sometimes) because of the legal regime but also because of unfamiliarity on the part of both borrowers and lenders.6 In many cases, collateral was largely limited to land and motor vehicles, and receivables and inventory were hardly used.7 There is reasonable evidence from countries which introduced reforms long ago enough for significant statistics to be available that the introduction of secured transactions law reform has extended the types of assets now accepted as collateral.8 As mentioned below, most African collateral registries now collect statistical information to monitor access to credit.9 This information is helpful in assessing the impact of reforms in terms of the types of borrowers that now have access to secured credit, and also as to the types of collateral used.10 For example, in Malawi, where a reformed secured transactions law was brought into force in 2015, general economic statistics showed an increase in GDP in 2017 and a slowdown in the rate of inflation in 2018.11 Specific registry statistics showed 19,770 registrations in 2017 and 2018, and that about 15,000 of these related to lending to individuals or micro-businesses. It also showed, unfortunately, that the assets used as collateral did not appear to include inventory and receivables, as had been hoped.12 Similarly, in Nigeria, a report published by the central bank for 2018 indicates that in the total of 154,000 registrations, individuals were identified as borrowers in 146,000.13 These types of statistics can indicate lessons for the future. However, the fact that other reforms are also taking place in countries which have reformed their secured transactions law makes it difficult to isolate the effects of the secured transactions law reform. For example, access to credit in a country is expected to increase if a credit reporting system is implemented that allows creditors to determine the credit risk of prospective borrowers more easily. The establishment of a loan guarantee scheme, which protects secured lenders against losses, is also likely to lead to an increase in access to credit. Most countries have established schemes that guarantee the repayment of loans extended to qualified borrowers, typically SMEs.14

5 Nigeria, for example, collects statistics particularly about lending to women; see Nigeria, ch 7, F. See also Ghana (ch 3) where one of the largest users of the Collateral Registry are microfinance institutions financing women-owned start-up microbusinesses. 6 Liberia, ch 5, A. 7 Burundi, ch 11, A. 8 See, eg, Liberia, ch 5, A. 9 See C7 below. 10 See Nigeria, ch 7 F; Malawi, ch 6, A. However, statistics about the costs of credit in Ghana show that the reforms there have not really decreased the costs of credit, which may be due to the problematic nature of the reforms (see Ghana, ch 3, B). 11 Malawi, ch 6, A. 12 ibid. 13 Nigeria, ch 7, F. 14 Over 2,000 such schemes exist. See OECD, ‘Facilitating Access to Finance: Discussion Paper on Credit Guarantee Schemes’, 3, www.oecd.org/global-relations/45324327.pdf. The OECD highlighted that credit guarantee programs continue to be ‘the most widely used instrument that governments deploy to ease SME access to finance’. M Chatzouz et al, ‘Credit Guarantee Schemes for SME Lending in Western Europe’, Working Paper 2017/42 (European Investment Bank, June 2017) 10.

Drivers for Reform 415 Other events unconnected with law or infrastructure reforms also complicate the a ssessment of the impact of reforms, such as a macroeconomic crisis,15 an epidemic16 or interest rate caps.17

b. Appearance to Global Markets In many cases, secured transactions law reform in Africa has been partly motivated by a desire to improve how a particular economy is viewed by the world, particularly as a subject for international investment and as a trading partner. There are various signalling techniques which States can use. One is the Doing Business Report produced by the World Bank, which not only includes an account of the business infrastructure and commercial and regulatory law of each State, but also gives an evaluation of the ‘ease of doing business’ in various categories, the most relevant of which for present purposes is ‘Getting Credit’, discussed below. It also provides a ranking of countries in these categories, so that a desire to move up these rankings has become a significant driver towards legal reform.18 Although there is some merit in the Doing Business rankings, the fact that they are based on quite rigid criteria and only consider the law on the books rather than how it works in practice means that countries can rise up the list as a result of reforms, even though the actual operation of the law is still problematic.19 The Getting Credit indicator covers two aspects of the regulatory framework and infrastructure that affect access to credit. The first is the availability of credit information measured in the credit information index, and the second is the protection of creditor rights under secured transactions and insolvency laws measured by the strength of legal rights index. The first aspect, the Depth of Credit Information Index, includes the coverage, scope and quality of credit information available through credit registries and credit bureaus. The second, the Strength of Legal Rights Index, evaluates the degree to which movable property can be effectively used as collateral and the protection of creditor rights. The Doing Business reports have been subjected to considerable criticism for its methodology, including the selection of the criteria against which laws are benchmarked, and the way in which those criteria are applied in practice.20 One example of an unsatisfactory application of a criterion relates to the question investigating whether secured obligations may be described generally in security agreements. Liberia did not score a point on this question when it implemented Chapter 5 of the Commercial Code.21 However, it scored that point in the 2017 Report, which indicates that Liberia ‘adopted new laws on secured transactions that establish a modern, unified and

15 See Ethiopia, ch 12, A. 16 For example, the Ebola epidemic in East Africa, see Liberia, ch 5, A and Sierra Leone, ch 9, A. 17 See Kenya, ch 4, A. 18 Malawi, ch 6, A; Rwanda, ch 14, A; Burundi, ch 11, D; Ethiopia, ch 12, E4; Contrast Zambia (ch 8, B) where rising up the Doing Business rankings was not a driver for reform, since Zambia was already ranked at 8. 19 See, for examples, Rwanda, ch 14, A and B; Nigeria, ch 7, F. 20 See, eg, G McCormack, ‘Why “Doing Business” with the World Bank May Be Bad for You’ (2018) 19 European Business Organization Law Review 649, 650. 21 See Liberia, ch 5, C3; and Doing Business 2016 – Liberia (Washington, DC, World Bank, 2016) 57; and Doing Business 2015 – Liberia (Washington, DC, World Bank, 2015) 49.

416 Conclusion notice-based collateral registry’.22 It is unclear what laws were adopted in 2016 that would have had any effect on the assessment of this criterion. Further, Liberia does not score a point on the following question: Does the law allow parties to agree on out of court enforcement at the time a security interest is created? Does the law allow the secured creditor to sell the collateral through public auction or private tender, as well as, for the secured creditor to keep the asset in satisfaction of the debt?

Liberia clearly meets all three requirements of this criterion,23 yet it has not scored a point. For some African countries, especially Nigeria, the high rankings would not be justified upon more rigorous analysis of the law.24 The final example of the rankings not entirely reflecting the quality of secured transactions laws is the case of Zambia that in 2012 undertook to reform its framework based on the Canadian PPSAs. In the 2012 Doing Business Report’s Getting Credit indicator, Zambia ranked 8 whereas Canada ranked 24.25 The Strength of Legal Rights Index also shapes the particular policies and legislative solutions adopted by States, which are designed to score a point on the criteria measured in the Index. For instance, a country scores a point if its law gives priority to security rights over tax liens and wage claims, both outside and within insolvency.26 On this aspect, the Index goes beyond the Model Law, which recommends that enacting States identify the preferential claims in their secured transactions laws, but also specify the amount for each category of such claims up to which they will be given priority.27 Many African economies have chosen to prioritise security rights over tax liens, but in some cases this decision has not been driven by a desire to score a point in the Index, but rather a careful discussion of the underlying policies and engaging the tax authority.28 Another criterion measures whether any obligation may be secured, and whether they may be described in security agreements generically, including by a ‘maximum for which the assets are encumbered’. Applying the methodology requires the secured transactions law to contain a specific provision that permits such descriptions. Under the Model Law,29 enacting States have an option to incorporate the approach that caps the amount for which a security right may be enforced with an effect on the priority rules. African jurisdictions have incorporated this element of description for security agreements in their laws, but not because this approach facilitates subordinate lending or for other policy reasons,30 but to meet the requirements of the Index criterion measured. This is confirmed by the provisions in the registration rules that require registrants to enter a maximum amount in the notice, but such amounts are not displayed to third parties.31 In contrast, under the Model Law if the State provided for such descriptions in security agreements, the corresponding 22 See Doing Business 2017 – Liberia (Washington, DC, World Bank, 2017) 58. 23 See Liberia, ch 5, C7, which discusses the remedies available. 24 Nigeria, ch 7, F. 25 See World Bank, ‘Doing Business 2012 – Country Tables’, www.doingbusiness.org/content/dam/doingBusiness/media/Annual-Reports/English/DB12-Chapters/Country-Tables.pdf. 26 Question 9 in the Index states: ‘Are secured creditors paid first (ie before tax claims and employee claims) when a debtor defaults outside an insolvency procedure?’ 27 ML, Art 36. 28 Kenya, ch 4, C4. See also D1b below. 29 See ML, Art 6(3). 30 See discussion in ch 2, C1. 31 For example, Rwanda, ch 14, C6; Liberia, ch 5, C5a. See also D2a below.

Drivers for Reform 417 maximum amount is included in the notice retrieved in a search. Accordingly, any creditor wishing to extend a subordinate loan must inquire and negotiate with the senior secured creditor. Another signalling technique is for a State to ratify international conventions. This has been done by Liberia, which has ratified the UN Receivables Convention, although this was not done with the purpose of improving receivables financing in that country32 and is still not yet in force. More substantively, a number of African jurisdictions have ratified the Cape Town Convention and the Aircraft Protocol.33 These ratifications are not purely symbolic, and are likely to bring real economic benefits to those countries.34 Increasing access to secured credit through predictable and certain legal frameworks also facilitates a related objective to reinforce financial stability.35 Principle 5 of the G20/OECD High Level Principles on SME Financing expressly reiterates that regulation of SME financing instruments must be designed to ensure financial stability and investor protection.36

c. Infrastructure Improvements For many countries, a major driver towards reform is the desire to have a well-functioning system of registration for security rights. The importance of this is discussed in chapter 2,37 and there is no doubt that this is a critical component of a secured transactions law based on modern principles. It is not necessarily the case that a country will not have a registration system at all, but the existing system may be fragmented, inefficient or even lacking in basic functionality, such as the ability to be searched.38 A modern system includes registration of notices relating to all proprietary interests which have the function of securing obligations39 granted by all types of persons: individuals, whether as consumers or sole proprietors, as well as companies and other legal entities. Developments in technology have enabled the creation of fully electronic registries, which can fulfil the policy imperatives for a modern secured transactions law of being cheap and easy to use, and also efficient: information can be entered by a secured creditor and can be searchable without human intervention.40 An electronic registry can also meet the goal of being comprehensive, as pointed out in the last paragraph, in that it can include notices in relation to all types of proprietary interests (reflecting the functional approach) and all types of grantors. The law may also require information to be included on the registry other than notices relating to secured transactions, such as notices relating to other,

32 Liberia, ch 5, B. 33 See CTC, ch 17. The ratifying States are listed in ch 17, n 41. 34 See CTC, ch 17, B3. 35 See G Castellano and M Dubovec, ‘Global Regulatory Standards and Secured Transactions Law Reforms: At the Crossroad Between Access to Credit and Financial Stability’ (2018) 41 Fordham International Law Journal 531. 36 OECD, ‘Effective Approaches for Implementing the G20/OECD High Level Principles on SME Financing’ (July 2018) www.oecd.org/g20/Effective-Approaches-for-Implementing-HL-Principles-on-SME-Financing-OECD.pdf. 37 Ch 2, A2b. 38 For example, in Burundi, searching had to be by date of registration, which made entries impossible to find. Burundi B3. 39 This is part of the functional approach, see ch 2, B2. 40 Ch 2, A2b.

418 Conclusion similar, transactions such as operating leases41 and non-consensual interests.42 The registry does not examine the underlying nature of the transaction to which the notice relates, which makes it possible to also register notices relating to transactions that are not covered by the substantive secured transactions law at all, such as short-term leases or bailments. Such registrations would not have any effect under the secured transactions law. Despite the fact that the desire for a registry is a major driver for reform, it will not bring about the desired results without a legal regime that is consistent with modern principles. This consistency relates to many parts of the law, particularly its scope, the methods of third-party effectiveness and the priority rules. Trying to fit a modern registry system onto existing secured transactions law is unlikely to work.43 For example, the existing law may determine priority by date of creation or some other rule which does not depend on the time of registration, or may not permit extrajudicial enforcement. It may require the security agreement to include specific descriptions of collateral, which is inconsistent with more generic descriptions permitted in a registered notice, or may provide for document registration and verification of registration data by registrars, both of which are inconsistent with the operation of a modern electronic registry. It is therefore not surprising that the most successful registry systems are those operating with a completely overhauled secured transactions legal system, based on modern principles and using a coherent model.44 Moreover, limiting the law to only a part of the secured transactions system, such as credit extended by limited types of creditors,45 or to only certain types of transactions46 is likely to cause problems, as will the retention of previous registration systems under which security rights can still be registered, raising the possibility of dual registration requirements and priority conflicts.47 In 2019, countries completing their reforms inquire whether their collateral registries should be designed and implemented with distributed ledger technologies (blockchain). Presently, there are no collateral registries anywhere in the world, including the international registry operating under the Cape Town Convention,48 that use such technology for their core functions. This is not surprising as there are many disadvantages to using blockchain for collateral registries, including the slow process of storing the registration information in a decentralised database. African electronic collateral registries have been performing well from the technological perspective, providing instantaneous access to users, and without any incidents of data loss, thus protecting the integrity of information, which is one of the main benefits that blockchain enthusiasts highlight.

d. Legal Structure for Specialist Financial Activity A country may well want reformed law which particularly addresses a type of financing prevalent in that country, either because of local custom or because of a type of business prevalent in that country. These reforms are not a substitute for a comprehensive secured

41 See

D1b below. D1c below. 43 See, eg, the position in Nigeria before the latest reform, Nigeria, ch 7, D2. 44 See, eg, Zambia, ch 8. 45 Ghana, ch 3, D1; Sierra Leone, ch 9, C2. 46 See Sierra Leone, ch 9, C2. 47 Nigeria, ch 7, E1; Sierra Leone, ch 9, C6. 48 See further CTC, ch 17. 42 See

Drivers for Reform 419 transactions reform. Furthermore, their implementation should be coordinated with each other and with general secured transactions reform. This section details some examples of this discussed earlier in the book. i. Credit Infrastructure The World Bank Group promotes secured transactions reforms as part of a broader effort to foster the credit infrastructure of a country. These reforms are built primarily on three pillars: secured transactions, insolvency, and credit reporting frameworks.49 The pillars are designed according to a set of internationally recognised principles.50 The Principles for Effective Insolvency and Creditor/Debtor Systems, developed by the World Bank in collaboration with UNCITRAL, IAIR and INSOL, are an internationally recognised authority for norms concerning the protection of creditor rights in insolvency and debt resolution.51 The Principles address several aspects, including, first, the laws and institutions that recognise and enforce credit agreements (including security agreements); secondly, the legal framework for risk management and informal corporate workout systems; and, thirdly, formal commercial insolvency law frameworks. They also address the implementation of these systems through sound institutional and regulatory frameworks.52 The importance of coordinating insolvency and secured transactions reforms is highlighted throughout the book. The International Committee on Credit Reporting (ICCR), which is the internationally recognised standard setter in credit reporting, published the General Principles for Credit Reporting (2011).53 This is the global standard for norms concerning credit reporting systems. Although not examined in this book, secured transactions systems must be underpinned by effective credit reporting regimes that facilitate the process of establishing the loan applicant’s ability to repay the loan. It is well understood that collateral is only a secondary source of repayment, and lenders look primarily to their debtor’s ability to generate enough revenue. Accordingly, the effect of a modern secured transactions law will be muted if a country has not established an effective credit reporting system or does not adequately protect the rights of secured creditors in insolvency. ii. Warehouse Receipts and Commodity Trading Commodity production and trading is a vital part of the economy in many African countries. Although the production of some commodities is financed, and the loans repaid, from the proceeds of sales after a harvest, many commodities are stored in warehouses. Farmers and other depositors may then use warehouse receipts as collateral for loans. The systems supporting warehouse receipts have an impact beyond facilitating finance,54 49 IFC, ‘Secured Transactions Systems, Collateral Registries and Movable Asset-Based Financing Knowledge Guide’ (2019, forthcoming). 50 For secured transactions, see ch 2. 51 World Bank, ‘Principles for Effective Insolvency and Creditor/Debtor Rights System’ (2015) http://documents. worldbank.org/curated/en/557581467990960136/Principles-for-effective-insolvency-and-creditor-debtor-rightssystems. 52 ibid. 53 General Principles of Credit Reporting (Washington, DC, World Bank, 2011). 54 See further Warehouse Finance and Warehouse Receipt Systems: A Guide for Financial Institutions in Emerging Economies (Washington, IFC, 2013) vii.

420 Conclusion including the lowering of post-harvest losses, the reduction of seasonal price volatility and the improvement and encouragement of higher-quality production. Furthermore, they incentivise the development of many financial services such as insurance and the trading of rights to commodities (whether actual warehouse receipts or commodity contracts settled by delivery of warehouse receipts) on exchanges.55 For many years, commodities have been traded in some African countries using electronic warehouse receipts (EWRs) and many others are looking to implement such systems.56 EWRs are traded on local exchanges in addition to contracts for the delivery of commodities, namely, futures, options and forwards traded on traditional exchanges such as the London Metal Exchange.57 For example, a farmer deposits her crops in one of the warehouses accredited or owned by an Exchange. She obtains a confirmation of delivery, and then the Exchange’s system creates an EWR that is uploaded to its trading platform. Interested parties may either buy the EWR or take a security interest in it.58 An unreformed secured transactions law is unlikely to have specific rules to deal with this type of asset, and even a secured transactions law based on the Model Law will have no specific provisions relating to taking security over EWRs. Thus, the only method of third-party effectiveness will be registration, which is unnecessary as public notice may be also achieved by other means, such as a credit of an EWR to the creditor’s account. EWRs in many ways resemble exchange-traded securities for which alternative methods of third-party effectiveness have been developed.59 In these circumstances, a system of law covering the taking of security over EWRs is needed, whereby third-party effectiveness and priority are determined by control. It is possible for this set of rules to be part of a general reformed secured transactions law but it could also be a separate statute, such as in Malawi, where the Warehouse Receipts Act 2018 was developed to cover both taking security over EWRs and the trading of these assets.60 An approach where warehouse receipts are entirely excluded from the scope of the secured transactions law, as contemplated in Ethiopia,61 should not be pursued as its implementation would cause some of the challenges explained above associated with a fragmented secured transactions regime. iii. Securities Held through Intermediaries Increasingly, securities are not held directly but are held by investors with intermediaries, who themselves may hold with other intermediaries, forming a chain in which a Central Securities Depositary (CSD) or depositary holding for a CSD is at the top.62 The legal analysis of this form of holding is complex and varies between jurisdictions. The UNIDROIT Convention on Substantive Rules for Intermediated Securities, also known as the Geneva

55 A Guide to Warehouse Receipt Financing Reform: Legislative Reform (Washington, DC, World Bank, 2016). 56 Malawi, ch 6, C; Ethiopia, ch 12, D7. 57 M Dubovec and C Kambili, ‘Secured Transactions Law Reform in Malawi: The 2013 Personal Property Security Act’ in O Akseli and L Gullifer (eds), Secured Transactions Law Reform: Principles, Policies and Practice (Oxford, Hart Publishing, 2016) 198. 58 Ibid; Malawi, ch 6, C. 59 See ch 2, D. 60 Malawi, ch 6, B and C. 61 Ethiopia, ch 12, D7 and E1. 62 See L Gullifer and J Payne (eds), Intermediation and Beyond (Hart Publishing, 2019) ch 1.

Drivers for Reform 421 Securities Convention (GSC),63 is an attempt to provide a functional set of rules which could lead to substantive harmonisation. The Legislative Guide excluded all securities from its scope, an exclusion which the Model Law then narrowed down to intermediated securities. The Legislative Guide explained that transactions with securities raise a number of issues outside of a secured transactions law that merit special legislative treatment.64 The existence of the GSC was the reason for the omission of intermediated securities from the assets over which security rights could be taken under the Model Law.65 A set of model provisions supplementing the Model Law has been developed that may be used as a guide for reforming economies.66 This form of holding securities is significant in some African jurisdictions, but the treatment of security rights under reformed law has not been ideal. In Kenya, section 42 of the CDA provides for the creation of charges over intermediated securities, but the procedure is very cumbersome and differs from international best practice, as, for example, that included in the GSC.67 Other African jurisdictions also have separate laws dealing with taking security over intermediated securities (as part of a specific statute dealing with that asset class).68 In general, as part of the overall secured transactions reform process, reformers in Africa and elsewhere should take the opportunity to review the rules applicable to securing obligations with rights in intermediated securities, particularly the procedural requirements, and to take into account modern principles.69 iv. Leasing Leases facilitate the acquisition of various assets, but especially equipment. Many African jurisdictions that have reformed their secured transactions law have also undertaken modernisation of their leasing frameworks. Although secured transactions laws apply to finance leases, and some even apply to long-term operating leases, the specifics of leasing transactions are typically governed by special leasing laws, such as those based on the 2008 UNIDROIT Model Law on Leasing.70 Additional guidance for States and users may be found in the 2011 IFC Global Leasing Toolkit71 that focuses both on the establishment of a leasing entity, and the life cycle of a lease transaction. A particular challenge in reforming States has been the coordination of leasing and secured transactions reforms,72 as the former often attempt to disregard the functional approach and reinforce ownership rights of lessors which are effective without any form of public notice. For example, in Ethiopia, some types

63 See ‘UNIDROIT Convention on Substantive Rules for Intermediated Securities’ (Geneva, UNIDROIT, 2009). This Convention is not yet in force. 64 LG 40. 65 ML, Art 1(3)(c). A legislative guide on incorporating the GSC rules into a Model Law regime has been produced. See ‘UNIDROIT Legislative Guide on Intermediated Securities’ (Rome, UNIDROIT, 2017). 66 See ‘Update on the Model Provisions on Secured Transactions for Intermediated Securities’ NatLaw, www. natlaw.com/news_posts/update-model-provisions-secured-transactions-intermediated-securities/. 67 Kenya, ch 4, C1. 68 Zambia, ch 8, D1; Ethiopia, ch 12, D1; OHADA, ch 10, B6b and G. 69 As exemplified in the UNIDROIT Legislative Guide on Intermediated Securities (n 65). See also Ethiopia, ch 12, E1. 70 See ‘UNIDROIT Model Law on Leasing’ (Rome, UNIDROIT, 2008). 71 See IFC, ‘Global Leasing Toolkit’, www.ifc.org/wps/wcm/connect/industry_ext_content/ifc_external_ corporate_site/financial+institutions/resources/toolkits/global+leasing+toolkit. 72 See further C6biv below.

422 Conclusion of leases have been partially included within the proposed secured transactions bill (finance leases and hire-purchase agreements are included for registration and priority purposes), but a lessor still retains ownership under the regime. It is intended to follow broadly the non-unitary approach of the Legislative Guide in relation to these types of leases.73

2. The State of the Unreformed Law It is very rare that a law gets reformed just because it is unsatisfactory. Economic and other drivers also need to be present. However, there are several features of commercial law, including secured transactions law, which make transactions more difficult (and therefore more expensive) or even, sometimes, impossible. Although these features may not prevent the granting of credit, or even influence the cost of credit, to large prosperous businesses in developed economies, they can have serious ramifications in relation to the financing of SMEs and micro-businesses, and in developing economies. Some of these features are discussed in this section.

a. Complexity and Fragmentation In most jurisdictions, the law relating to secured transactions has grown up over a period of time out of the existing property law, often influenced heavily by the legal culture. It is also likely to reflect market practice as it varied from time to time, often in response to specific legal rules which affected particular parties adversely. As a result, most unreformed law is both fragmented and complex, the former leading to the latter, although extra complexity is often added by rules and exceptions that have grown up to balance the interests of different parties. This section considers two particular types of fragmentation: different treatment of various types of parties, and the proliferation of different devices, all of which have the function of creating proprietary interests which secure obligations. i. Parties At least some jurisdictions based on the common law are likely to have had a bifurcated system governing certain aspects of secured transactions law,74 particularly the registration system, so that different rules applied to the granting of security by companies and by unincorporated businesses and individuals. The origins of this system in the common law are discussed in chapter 1.75 This type of system is particularly problematic when a large number of businesses are unincorporated, as is the case in Africa, and, in particular, can give rise to overlapping interests if an unincorporated business grants a security right, registered under one regime, and then incorporates, or transfers the asset to a company who takes subject to it, but who has also granted a security right over its future assets.76 A bifurcated

73 Ethiopia,

ch 12, E1. Nigeria, ch 7, C1; Zambia, ch 8, C3; Ghana, ch 3, C2. 75 Ch 1, D1a 76 See discussion in Zambia, ch 8, B. 74 See

Drivers for Reform 423 reformed system was considered in Zambia, but the arguments in favour of a single system prevailed.77 A reformed system in accordance with modern principles will apply largely the same rules and, in particular, the same registration system, to individuals and companies alike.78 ii. Types of Devices A plethora of devices used to perform the function of security is very common in pre-reform African jurisdictions. Before reform, the common law jurisdictions tended to have a combination of ‘true’ security interests (the pledge, the lien, the mortgage and the charge) and other interests based on retention of title (conditional sale, hire purchase, finance leases), together with some devices covered by specialist legislation, such as agricultural charges.79 Civil law jurisdictions tended to have pledges (possessory and maybe non-possessory), security assignments of rights, retention of title devices, financial leases and some means of taking security over an entire enterprise. Mixed jurisdictions had notarial bonds, both over specific and fluctuating assets. In all jurisdictions, different devices had their own rules for creation, third-party effectiveness, priority and enforcement, and often their own registration infrastructure. This made financing complicated and potentially expensive for all but the most experienced lenders. It has also limited finance to customers of lenders that have had a longstanding relationship and built trust with each other. The benefit of the functional approach of a system based on modern principles80 is evident.

b. Lack of Functionality This section considers aspects of a jurisdiction’s secured transactions law which can actually make the system function less than optimally, and therefore can hamper financing transactions. i. Limits in what can be used as Collateral Certain legal rules can restrict the collateral that is available to secure credit. It is highly unusual for a jurisdiction actually to ban the use of certain assets as collateral, but restrictive requirements can have a similar practical effect. One particular example is the requirement that assets must be specifically described in the security agreement.81 This prevents the use of circulating assets as collateral, or assets which can only conveniently be described by class. Another example is that many laws are asset specific, such as for agricultural charges or hire purchase. Further, some laws allow only the owner of the asset to encumber it,

77 ibid. 78 Ch 2, B1. Note that the eighth question of the Doing Business Getting Credit strength of legal rights index reads: ‘Does a modern collateral registry exist in which registrations, amendments, cancellations and searches can be performed online by any interested third party?’ 79 See Zambia, ch 8, C2; Zimbabwe, ch 16, B1 80 See ch 2, B2. 81 Burundi, ch 11, B2. Jurisdictions where Bills of Sale legislation applied to security interests over goods granted by individuals (see ch 1, D1a) include: Nigeria, ch 7, C1; Zambia, ch 8, C3; Ghana, ch 3, C2.

424 Conclusion precluding lessees and licensees from using their rights (such as a lessee’s right to possession) as collateral.82 ii. No Extrajudicial Enforcement Whilst extrajudicial enforcement, at least to some extent, has been a part of the common law for many years, this is not generally available in unreformed civil law jurisdictions, though certain exceptions have been provided for, such as for possessory pledges.83 Given the cost and delays that can be part of court enforcement, the lack of the ability to enforce out of court, even to collect receivables, can be a severe functional disability reducing the availability of secured credit. iii. Excessive Formalities for Creation Rigid formalities for the creation of security rights may give some sort of protection to the borrower under some circumstances or reduce the risk of legal disputes, but they can also hamper the flow of secured credit and increase its cost. For example, the rigid requirements of the Bills of Sale Acts (or similar legislation)84 either acted as a disincentive to lenders, or encouraged them to use other devices.85 Many common law jurisdictions require the payment of stamp duties, though some have repealed such laws, such as Zambia,86 or made provisions in secured transactions law to exempt security agreements from stamp duties entirely87 or partially.88 In civil law jurisdictions, notarisation is another type of formality that can increase the cost of credit,89 as is legalisation of signatures.90 iv. Lack of Publicity The absence of an efficient registration system to give publicity to security rights is also a factor that disincentivises credit providers and, because it increases risk, is likely to increase the cost of credit. The registration systems in all countries pre-reform are less than optimal, but in some cases they were (or are) positively dysfunctional.91 In many countries, the only workable registration system was one for motor vehicles in which lenders routinely named themselves as owners or co-owners of vehicles that they financed.92 82 See further, ch 2, B2b. 83 South Africa, ch 15, A, B, D; Burundi, ch 11, B4; Tunisia, ch 13, F; OHADA, ch 10, D (even after reform) but cf Ethiopia, ch 12, C. 84 Nigeria, ch 7, C1. 85 Zambia, ch 8, C3. 86 ibid C. 87 See Kenya, ch 4, C2. 88 See also Dubovec and Kambili (n 57). 89 Ethiopia, ch 12, E1, E2. 90 Tunisia, ch 13, D. 91 For example, in Burundi the register for non-possessory pledges was paper-based and indexed manually in the order that they were received, so it was almost impossible for a third party to find a registration unless the searcher knew the exact date on which it was processed by the Registrar (Burundi, ch 11, B3). In Ghana, it remains unclear what the search criterion that determines the legal effect of a registration is (Ghana, ch 3, D4a). In Tunisia, registries are not centralised and one is established in every jurisdiction of the court of first instance (Tunisia, ch 13, C2). 92 See Kenya, ch 4, B; Sierra Leone, ch 9, A.

The Method of Reform 425

C. The Method of Reform 1. Use of Models Virtually no reform of commercial law is done without consulting external models, and frequently a number of models are looked at for comparative purposes. A variety of models have been used in Africa, with many countries using a mixture of models.93 One approach has been to use an international instrument, usually the Model Law or, before it was completed, the Legislative Guide, usually with certain modifications. Another approach has been to use a mixture of national models, such as the Canadian or New Zealand PPSAs, UCC Article 9 and local models, where available.94 A third approach, of which Rwanda is the only example in this book, is to use a simplified version of a modern secured transactions law.95 This section contains some reflections on the use of models in secured transactions law reform, using the experiences of the countries discussed in this book as illustrative examples.

a. Mixture of Models There are arguments for and against using a number of models when drafting a reformed law. In favour of this approach is the fact that it enables policy makers and drafters to see different ways of dealing with the same issues, and therefore enables a close match with the local economy and customs to be achieved. It also helps in deciding how to deal with gaps in the chosen model if one particular policy objective is not included: one example of this is that the Model Law does not include operating leases within its scope.96 However, against the use of multiple models, it can be argued that there is a danger they will not mesh well together. Each model is designed as a coherent whole and any attempt to use more than one conceptual structure for a reformed law could potentially lead to confusion. Rwanda is a case in point where the concept of creation that under the Model Law causes the security right to be effective between the grantor and the secured creditor has been confused with the UCC Article 9/PPSA concept of attachment where the security interest becomes enforceable against third parties.97 Thus, when using a mixture of models, care has to be taken to ensure that the provisions drawn from different sources work together in a coherent manner.

b. Deviations Not surprisingly, all reformed systems deviate somewhat from the model used. This is to be expected and, to some extent, encouraged, since legal transplants (which, to some extent, 93 Conclusion, ch 19. 94 For example, Zambia used the Malawian statute, inter alia (quite successfully, see Zambia, ch 8, F), and Sierra Leone used the Ghanaian B&L Act (much less successfully, see Sierra Leone, ch 9, C). 95 See C1d below. 96 Operating leases have been included, for example, in Zambia, which largely used the ML as a model, but used the PPSAs as a model for the provisions on operating leases. See Zambia, ch 8, D1. 97 Rwanda, ch 14, C2.

426 Conclusion the models are, despite the international character of the Model Law) must be adapted to local conditions for them to be successful.98 Moreover, it is vital that the local community ‘owns’ the reform, and one way of achieving this is for local stakeholders to have a say in the design of the system. Moreover, the models, especially the Model Law, provide a number of options for States, so their implementation is not uniform. However, it is important that the underlying conceptual structure is not undermined, as otherwise the law will not ‘work’. It is also important that deviations are carefully thought through, so that they operate well within the practical implementation of the reformed law. As mentioned above, it is in this regard that examining several models can help to achieve a successful result, although this cannot be at the expense of properly understanding the model used in all respects so that the deviation can be seamlessly incorporated. These points can be illustrated with several examples from the chapters in this book.99 The Legislative Guide included some recommendations dealing with security rights over assets which were or became attached to land or to movables,100 but these were not included in the Model Law. The general rules of the Model Law thus apply in these situations. A number of African jurisdictions have included specific provisions in their reformed statutes dealing with at least some aspects of attachments, in particular attachments to immovable property (sometimes referred to as fixtures). This issue involves difficult interrelations with the local law of immovable property, and is very susceptible to local conditions, including the efficacy of the local Land Registry. Most countries have not followed the Legislative Guide in totality, probably for this reason, with rather mixed results in terms of the robustness of the provisions.101 A particular example is Burundi, which had to accommodate the civil law concept of an ‘immovable by destination’; if the asset that is subject to the security right has become an immovable by destination,102 the security right cannot be enforced under the reformed law.103 Some other countries, inspired by the PPSAs, have introduced priority rules in respect of accessions to movable property.104 Ghana went further than just including security rights over attachments to land in its B&L Act, and included security rights over the land itself, which are now registrable in the collateral registry.105 The relationship between that registration and registration in the land registry, however, is not clear and the proposed amendments to the reforms include a more comprehensive extension of the regime to security rights over land, including priority rules regarding registration of other rights in the Land Registry.106 The 2019 B&L Act enacted in Sierra Leone provides for a more comprehensive and clearer regime that applies to security rights in both movable and immovable assets, but it is yet to be implemented. Another example of deviation from the Model Law approach, which reflects differences of approach taken by the national law models of UCC Article 9 and the PPSAs, is treatment 98 See ch 1, B6. 99 Other examples are covered in detail below when the substance of reform is discussed, eg D4a below (methods of third-party effectiveness). 100 LG, ch II paras 96–99 (recommendation 21), ch III paras 97–104 (recommendation 43), ch V paras 110–16 (recommendation 87–88 (immovable) and 89 (movable)). 101 Kenya, ch 4, C5; Zimbabwe, ch 16, C1, C3, C3d (although the deviations from the LG are not major); Zambia, ch 8, D6; Rwanda, ch 14, C4; Liberia, ch 5, C1. 102 Burundi, ch 11, D4. 103 Burundi, ch, 11, C4. 104 Sierra Leone, ch 9 C4; 105 Ghana, ch 3, D1. 106 ibid and E below.

The Method of Reform 427 of the taking free rules. The Model Law provides that the buyer of goods in the ordinary course of business takes free from all security rights, whether created by the seller or by anyone else (eg a previous owner of the goods from whom the seller acquired the goods subject to that security right).107 Other models, such as UCC Article 9, provide that the buyer only takes free of security rights created by the seller.108 A number of African jurisdictions follow this approach.109 Another issue on which there is deviation is the standard of knowledge required if the transferee is not to take money and funds in a bank account free of a security right. The criterion in the Model Law is that the transferee knows that the transfer violates the rights of the secured creditor under the security agreement.110 Some African jurisdictions follow this approach, although others require collusion (the standard in UCC Article 9 and some PPSAs),111 whereas others deviate in other ways, the reason for which is hard to understand.112 Finally, a number of African jurisdictions provide for specific quasi-judicial remedies involving the Collateral Registry, whereas under the Model Law the Registry plays no role whatsoever in enforcement. That approach has been chosen to facilitate the enforcement of security rights in an expeditious manner, such as in Nigeria.113 In contrast, sometimes it has been adopted to protect the interest of borrowers, as in Ghana.114 The involvement of the registry in each of these two jurisdictions produces a materially different result.

c. Structure One aspect where reforming jurisdictions can sometimes deviate from the relevant model is in the structure of the statute by which the reform is carried out. Although the various models do not all use the same structure, and certainly do not cover issues in exactly the same order, they do tend to cover the same core issues (creation, effectiveness against third parties, registration, priority, enforcement, conflicts of laws and transitional provisions), and to set them out in a conceptually coherent way. Some of the African statutes have adopted a very different approach. This may partly be because of local drafting culture,115 and partly because of the timing of the reform process.116 However, there are some deviations that have proved unfortunate and provide lessons for future reforming States. Zimbabwe, for example, included a great deal of important material in several schedules to the main Act, which led to difficulties in determining which definitions applied to which parts of the legislation.117 Ghana and Liberia included important rules about registration in regulations rather than the main Act.118 Also, some statutes include provisions that are inappropriate

107 ML, Art 34(4). 108 UCC, §9-320(a); and also Saskatchewan PPSA, s 30(2); Australian PPSA, s 46(1); New Zealand PPSA, s 53(1). 109 Sierra Leone, ch 9, C4; Zambia, ch 8, D6; Nigeria, ch 7, E4. 110 ML, Arts 47(6) and 48(1). 111 Zambia, ch 8, D6; Nigeria, ch 7, E4. 112 Sierra Leone, ch 9, C4; Rwanda, ch 14, C4 (where the law is unfortunately very unclear). 113 See Nigeria, ch 7, E6 and D6b below. 114 See the procedure described in Ghana, ch 3, D6 and D6a below. 115 See, for example, the engagement with local drafters in Zambia, although this did not result in deviations in structure from the Model Law. See Zambia, ch 8, B, F. 116 See C2 below. 117 Zimbabwe, ch 16, C, C2. 118 Ghana, ch 3, D4; Liberia, ch 5, C5 and E.

428 Conclusion for a secured transactions law, such as regulatory provisions119 or provisions dealing with unsecured lending or consumer protection.120

d. Simplified Approach Although the benefits of a reformed secured transactions law include reduction in fragmentation and complexity, the nature of the subject matter means that the law cannot be made too simple, otherwise important matters are just left out and the necessary balance between the parties121 is not achieved. The Model Law is relatively short (compared with other models such as UCC Article 9 and the PPSAs) and is designed to be conceptually coherent and therefore relatively straightforward to understand. Attempts to make the law even simpler can result in an unsatisfactory product. The simplification efforts have led to a number of consequences. First, a number of key elements are likely to be omitted, such as the provisions that override the effect of anti-assignment clauses or conflict of laws rules. Secondly, key concepts, such as creation and attachment, are often confused. Thirdly, approaches that have no application in the concerned jurisdiction, such as the rules on chattel paper, are introduced. One African jurisdiction which used a simplified model is Rwanda. As discussed in chapter 14, the collateral registry in Rwanda appears, on its surface, to comply with the best international standards, but the secured transactions system, taken as a whole, is rife with legal risks, unpredictability and ambiguities. Rwanda should repeal the existing statute and replace it with one based on the Model Law or other international models. Moreover, there is a danger that other African jurisdictions might follow the Rwandan model,122 which would be very unfortunate.

2. The Timing of the Reform Process The order in which the steps are taken in the reform process and the coordination between different parts of the process are critical if a reform is going to be successful. This section will, first, set out the optimal timeline for a reform process, and then discuss various examples from the African experience where deviation from this timeline has caused difficulties. A reform should be informed by a diagnostic that identifies the various obstacles to access to secured credit. A diagnostic is likely to reveal that the legal framework pertaining to secured transactions does not support certain credit products which are common in reformed jurisdictions, such as products relying on a pool of fluctuating collateral, or that the framework unduly increases their cost. The diagnostic should identify the deficiencies and consider general solutions to address them. However, this process should involve the relevant stakeholders, in order to get their buy-in to proceed with the reform. An appropriate model should then be chosen as the basis on which to draft new legislation.123

119 Ghana,

ch 3, D; Sierra Leone, ch 9, C1. ch 3, D. 121 See ch 2, A2b and 2c, E1, F2, F4 and H. 122 As, for example, Sierra Leone followed Ghana. 123 This has not been the case in Ghana, ch 3, Introduction and Sierra Leone, ch 9, C. 120 Ghana,

The Method of Reform 429 After multiple rounds of discussions of the various policies and approaches to different issues, the statutory language should be conformed to the local drafting conventions.124 In most cases, the legislative process is uncertain, and the enactment may take anywhere from a year to many years.125 Nonetheless, on enactment, the entry into force of the new statute should be delayed until the registry is fully operational. A premature entry into force could create significant legal risks, especially when the new statute repeals all or most of the preexisting legislation concerning secured transactions.126 Before the entry into force, affected stakeholders, including lenders, borrowers, legal professionals and regulators, should be appropriately trained. Statistics generated by African collateral registries reveal a number of deficiencies, including the minimal use of commercial assets, such as inventory and receivables, as collateral, which may be attributed to the lack of capacity of financial institutions. Some of these examples are highlighted below.127

a. Interrelationship between the Legislation and the Registry Given the central significance of the collateral registry in a secured transactions legal regime, it is critical that the registry is properly set up and established before the substantive legislation comes into force. This is not only important for the prospective operation of the system, but also to enable transitional provisions, which typically provide a grace period within which existing security rights must be reregistered to remain effective against third parties,128 to work properly. Furthermore, if the legislation comes into force too early, there is a danger that the development of the registry, and the relevant regulations,129 will be rushed and, therefore, more likely to be faulty. Zambia is a particularly egregious example of this phenomenon. Because of the local policy of expediting the entry into force of enacted legislation, the Zambian statute entered into force in April 2016, whereas the registry was not officially launched until January 2017, though it started operations in September 2016.130 The transitional grace period was six months, and registrations could not be submitted until two weeks before the end of that period, and even then, the registry was not fully functional and registrations had to be collected using Excel spreadsheets.131 Fortunately, through the diligence of PACRA, the transition took place without loss of security interests. The Zambian experience, however, is salutary. In Liberia, there was also consideration of a temporary registry to cover the period between the legislation coming into force and the collateral registry being operational. However, a temporary registry was not established, and lenders developed the practice of registering their security rights in the probate registry, even though this had no legal effect under the Commercial Code.132

124 Zambia,

ch 8, B and F. ch 3, Introduction and D. 126 Kenya, ch 4, C9; Zambia, ch 8, E. 127 See, especially, C3 and C4. 128 See ch 2, J. 129 See 2C below. 130 See Zambia, ch 8, B and E. 131 ibid E. 132 Liberia, ch 5, C5a. 125 Ghana,

430 Conclusion The position in Sierra Leone was also unfortunate. No date for the 2014 statute to come into force was specified, with the result that it came into force on the date on which it was enacted. The registry was not established by that date, nor was it operational within the six-month transitional grace period.133 However, the situation was less critical than in Zambia since the 2014 B&L Act did not repeal any pre-reform secured transactions law, which was the case in Zambia. In Kenya, the collateral registry was launched about two months after the legislation was enacted134 and two weeks after the legislation was actually brought into force.135 Not only did this lead to hasty drafting of the regulations,136 but it meant that there was no operating registry for the first two weeks of the regime. It also left little time for capacity building in relation to the use of the registry and for publicising it. This had the result that the registry has not, yet, been used satisfactorily, as shown by the statistical results indicating a highly disproportional ratio of registrations to searches and a significant majority of registrations being made against commercially insignificant assets, such as household goods, or even against assets not subject to the MPSR at all. However, the registry has since been upgraded and there is hope that the position improves.137 Another example is the OHADA Uniform Act, which came into force in May 2011 and was immediately directly applicable in all OHADA States. The Act requires registration in certain circumstances in a single registry (the Register of Commerce and Securities), but a registry had not been established in many OHADA States at the time the Act came into force, and, in some States, there is still no centralised registry.138 In addition to situations where secured transactions law has come into force before a collateral registry was established, there are many examples of a leasing law being enacted without any registry for leases having been developed.139 In some cases, a registry has not yet been developed at all, with consequential increased uncertainty and risk.140 Finally, in a number of African jurisdictions legislation has been enacted but its implementation is significantly delayed.141 For instance, Zimbabwe enacted legislation in 2017 but the Act is yet to be brought into force, since no collateral registry has yet been established.142

b. Coordination within Reform Process In some African countries, legislation relating to secured transactions is introduced at about the same time as other commercial law reforms. Although the broad purpose of all these reforms is usually similar, that is, to increase access to credit, to improve commercial transacting and to advance economic growth, these various reforms are sometimes not

133 Sierra

Leone, ch 9, C6. Kenya, ch 5, Introduction. 135 ibid C. 136 See 2C below. 137 See Kenya, ch 4, D. 138 OHADA, ch 10, D1 and I. 139 For example, Burundi, ch 11, C6. 140 For example, see Nigeria, ch 7, D1; Liberia, ch 5, C5a. 141 Burundi, ch 11, C. 142 Zimbabwe, ch 16, C. 134 See

The Method of Reform 431 coordinated. This can lead to uncertainty and inconsistences, which are likely to hamper the purposes for which they are implemented. One example of this phenomenon is Nigeria, where the ELA and the STMAA were being developed at the same time. There was, however, lack of coordination between the two initiatives, so that finance leases potentially fell within both regimes.143 Although the ELA has been enacted, no leasing registry has been established, unlike the collateral registry under the STMAA. The STMAA requires an automated interface between the collateral registry and any other public registry relating to transactions in movable assets, and once the ELA registry is established, there will need to be coordination to ensure that this takes place.144 Similar situations have occurred in Rwanda145 and Liberia.146 Some countries have also introduced new companies and/or insolvency legislation as part of a raft of reforms.147 As discussed in several chapters, coordination with, in particular, insolvency law is critical if a secured transactions regime is going to be successful. For example, the World Bank sponsored a reform project in Zimbabwe focused on both secured transactions and insolvency law reform, which involved good coordination between the two bills.148 However, after the secured transactions reform was complete, discussion at governmental level led to the insolvency reform being split between the new Insolvency Act and the companies bill, and resulted in departures in substance from best international practice, as well as some difficulties in coordination with the secured transactions legislation.149 Kenya also introduced new companies and insolvency legislation, based on the most recent English law.150 Since these statutes were very recent, there was little appetite to make substantive or other changes to align them to the new secured transactions regime and, despite some careful consideration, there is still some serious lack of coordination between the company and insolvency legislation and that covering secured transactions. This is discussed in detail in the Kenya chapter151 and below.152 However, a good example of effective coordination is found in Zambia, where secured transactions law reform was being developed at the same time as company and insolvency law reform, but by different agencies. When those considering secured transaction law were debating whether the reform should be limited to unincorporated grantors, the impact on the insolvency legislation was considered and was one of the factors that persuaded PACRA to pursue a reform which covered all types of grantors.153

c. Hasty Drafting Ideally, all parts of a new law, whether statutes or regulations, should be drafted carefully, with ample time for reflection and consultation. However, in some African countries, failure 143 Nigeria, ch 7, D1. 144 ibid. 145 See Rwanda, ch 14, A, B and D. 146 Liberia, ch 5, B and C5a. 147 See, eg, Liberia, ch 5, D. 148 Zimbabwe, ch 16, C. 149 Zimbabwe, ch 16, C, C1 and D. 150 Kenya, ch 4, Introduction and B4. 151 Kenya, ch 4, B4 and D. See also the lack of coordination between the MPSR and other existing legislation, such as the Hire Purchase Act; Kenya, ch 4, B2 and D. 152 See C6b i and ii below. 153 Zambia, ch 8, B.

432 Conclusion to observe the optimal timeline has led to hasty and ill-considered drafting, especially of regulations. In some States, the legislation introducing the regime was brought into force so quickly that the regulations had to be hurried through so that the registry could be usable. One prime example was Kenya,154 where the regulations were drafted very quickly, partly because of the timescale and possibly also in order for the reform to be included in the preparation of the 2018 World Bank Doing Business Report.155 The Zambian regulations were also prepared in a rush,156 and had to include a number of registration processes using paper, which would not have been needed had the electronic registry been ready in time.157

3. Engaging Local Stakeholders The importance of consultation has been highlighted by many commentators on law reform, especially legal transplants.158 Consultation with the local stakeholders is vital for at least two reasons. First, it enables the identification of local conditions, customs and idiosyncrasies to which the models used need to be adapted in order to succeed, and secondly, it encourages ownership of the reforms among the local population, which is essential if they are to be used widely and operated effectively. Capacity building is also vital for these purposes: that will be discussed in the next section. One example of effective consultation with stakeholders covering many aspects of the reformed law is the development of the Movable Property Security Rights Act in Kenya, where a Technical Working Group was established to develop the new statute, and, although advised by external parties, including one of the authors of this book, proceeded methodically in conjunction with stakeholders from the private and public sectors.159 Another example is Zambia, where local stakeholders were involved in the careful debate as to the scope of the reform, particularly whether it should apply just to unincorporated grantors or to all types of grantors.160 In Ghana, where there have been many problems with the B&L Act 2008, local stakeholders were extensively consulted over a number of years on the drafting of the 2017 amendments.161 Engagement with local stakeholders also uncovers challenges in the newly enacted statutes, as was the case in Sierra Leone with respect to the interaction of the B&L Act with the Companies Act.162 This section discusses two ways in which consultation with local stakeholders can lead to adaptation of models. It focuses largely on the law itself, while the implementation and operation of the reformed system will be discussed in the next section.

154 See

C2a above. ch 4, C and C4. 156 Zambia, ch 8, D3. 157 ibid. 158 See ch 1, B6. 159 See Kenya, ch 4, D. 160 See Zambia, ch 8, B and F. 161 Ghana, ch 3, D. 162 Sierra Leone, ch 9, C6. 155 Kenya,

The Method of Reform 433

a. Local Legislative and Drafting Practice Many African countries (in common with many countries around the world) have particular legislative procedures or drafting practices, which need to be followed if a statute introducing a reformed law is to be enacted smoothly and without undue alteration during the legislative process. It is therefore good practice to involve experts who can advise on these matters at an early stage. For example, in Zambia, the working group of local agencies and stakeholders was advised by an experienced and respected legislative drafter, who ensured that the draft legislation was submitted to the government in a format which complied with the standard drafting guidelines of the Ministry of Justice. This meant that the Bill was passed with very few amendments. The passage of the Bill was also speedy, compared with the two–three years it would have taken for the Ministry of Justice legislative drafters to revise it in compliance with the drafting guidelines.163 Some previous drafting conventions and practices, however, should be abandoned. For instance, in some African jurisdictions it was customary for regulations to include the forms that would be registered in the relevant registry. This was necessary because the forms were delivered in a physical form. This drafting practice serves no purpose when the registry is accessible by electronic means only. Several African jurisdictions included forms in their regulations, which also do not fully correspond to the electronic formats in the registry system.164

b. Local Legal Culture and Practices When introducing a reform based on an international model such as the Model Law, it is sometimes necessary to strike a balance between using the words and concepts contained in that text, and using language which is consistent with the prevalent legal culture, and which is familiar to the stakeholders. One example of this approach comes from Burundi, where, although it was decided to follow the non-unitary approach of the Legislative Guide,165 all security rights created by grant were merged into a single type of security right, which is consistent with the functional approach.166 However, it was also decided to use the traditional (civil law) word ‘pledge’ for this security right, so as not to confuse lawyers and other users.167 Perhaps less successfully, particularly given the overlap with the Companies Act, Ghana retained the common law term ‘charge’ as the generic term for a security right in the B&L Act. The term ‘charge’ is replaced by ‘security interest’ in the 2017 amendments to the B&L Act.168 It is also sometimes necessary to keep local practices, if they do not interfere with the modern principles of secured transactions to any great extent. For example, in Burundi there was a practice, before the reform, of attaching a distinctive sign to a pledged tangible asset indicating the existence of the pledge and where it was registered. This mechanism was

163 Zambia,

ch 8, B and F. Kenya, ch 4, C4; Zambia, ch 8, D3. 165 See ch 2, B2b. 166 ibid B2a. 167 Burundi, ch 11, C. 168 Ghana, ch 3, D1. 164 See

434 Conclusion retained in the reformed law in order to reinforce the confidence of creditors, even though, under the new legislation, the attaching of a sign to an asset does not have any legal significance. It was thought to have a prophylactic effect, since third parties who would not typically search the registry might well see the sign.169 The recently enacted Ethiopian law provides for the registry to have a function within the enforcement process, retaining the function provided by the current registries in an environment where creditors face difficulties enforcing their rights.170

c. Local Conditions As well as taking account of local law and practices, consultation may also clarify practical matters relating to the local economy, which need to be taken into account when drafting reforming legislation. For example, many African countries have a large agricultural sector, and this may be reflected in specific provisions relating to crops and other agricultural products. Zambia, for instance, has a specific provision concerning creation and perfection of security interests in farm products.171 Rwanda also has special priority rules in relation to crops.172 Although neither the Legislative Guide nor the Model Law include such rules, they are included in some national laws, such as the UCC Article 9173 and the Canadian PPSAs.

4. Capacity Building It is very important for the success of a secured transactions law reform that capacity is built for many different types of stakeholders. Unless those using the system understand it, and feel comfortable with the rules and their operation, they will not use it or they will not use it in the way that produces the benefits the reforms are intended to bring. Perhaps the most important category of stakeholder for these purposes are the lenders and their lawyers, because they will be using the system, particularly the registry, and it is confidence in the system that will result in an increase in access to credit. However, it is also important that borrowers understand the system, both so that they realise the consequences of entering into secured transactions under the new regime and also so that they do not have unrealistic expectations of what it can achieve. Others, such as judges and regulators, especially those that regulate and supervise financial institutions, also need training because if they do not understand the purpose and concepts well, their actions or decisions could also reduce the benefits. Moreover, for all these types of stakeholders, capacity building should not just take place at the time of the introduction of the new regime, though this is very important, but should continue, both to deal with newcomers to the market and to deal with new issues. Although the points made in the last paragraph are critically important, they are quite general. More concrete examples of the effect of lack of proper capacity building can be found in the chapters of this book.

169 Burundi,

ch 11, C2c. Ethiopia, ch 12, E5 and below C6b. 171 Zambia, ch 8, D5. 172 Rwanda, ch 14, C4. 173 Although removed in the latest revision. 170 See

The Method of Reform 435 Where lenders and other potential competing claimants do not understand fully the purposes of registration, they will not search properly, or, in some cases, at all. In such situations, the registry statistics will show a much higher proportion of registrations to searches, when it should be the other way round for several reasons. First, a registering secured creditor should search twice, once before it registers and once immediately afterwards to ensure that no other registration has been recorded in the interim. Secondly, one would expect that some searches would not result in a registration since some will be unsatisfactory and the lender will therefore not extend credit. Thirdly, parties other than secured creditors, such as some prospective buyers, should also search. In Ghana, the number of registrations has greatly exceeded the number of searches, indicating that a great deal more capacity building is required.174 The statistics in Kenya showed the same picture for the period after implementation of the registry, which was rushed and therefore did not involve any real training or public awareness activity.175 Another observable consequence of lack of capacity building is that users are likely to direct legal questions to the registry. Although this is understandable, it is very unfortunate, because the registrar typically will have no authority to give legal advice.176 The registry can publish general guidance as to the registration procedure, and this is quite common,177 but specific legal advice should be given by lawyers, who, unlike the registrar, bear the risk of error. Capacity building can take various forms, but the need for it to be substantial and intensive should not be underestimated. Donors and sponsors, such as the World Bank and the IFC, routinely organise training sessions. A workshop by an external expert for a few days, while helpful, will not be sufficient to enable bankers to fully internalise the rules and the lending practices necessary for the economic benefits of reform to be realised. Such workshops were delivered in Ghana five years ago, but the most recent statistics from the Collateral Registry illustrate that several deficiencies remain. Targeted workshops, based on proven need revealed by statistical information, can also be useful. An example is those run by the IFC in Malawi on asset-based lending, after statistics collected by the registry revealed that lenders were not taking inventory and receivables as collateral. The IFC also ran some workshops focused on women entrepreneurs, since the statistics revealed that this was a greatly underrepresented group of borrowers.178 Published guidance can assist, especially if it is rooted in the modern principles of secured transactions law and a deep understanding of how these work in practice. The UNCITRAL Practice Guide to the Model Law179 should, therefore, help, although face-to-face training will still be necessary. This Practice Guide includes sample documents designed to help

174 Ghana, ch 3, D4b. 175 Kenya, ch 4, C4. 176 See D5b below. 177 For Liberia, see Collateral Registry Liberia, ‘Help: General Questions’, www.registry.cbl.org.lr/Home/Faq; and for Zambia, see M Christopher and L Musonda, ‘The Zambian Movable Property (Security Interest) Act No 3 of 2016: A Step-by-Step Explanatory Guide to the Act’ (PACRA) www.pacra.org.zm/MPRS/docs/Guide%20to%20 the%20Movable%20Property%20(Security%20Interest)%20Act%20No.%203%20of%202016.pdf. 178 Malawi, ch 6, D. 179 The Practice Guide to the Model Law on Secured Transactions was adopted during the 52nd session of the UNCITRAL Commission, held in July 2019 in Vienna. Information and the latest draft is available at UNCITRAL, ‘Working Group VI: Security Interests’, https://uncitral.un.org/en/working_groups/6/security_interests.

436 Conclusion parties draft the necessary agreements and notifications, not sample registration notices, when engaging in secured transactions under the Model Law. Sample documents such as this are useful as a starting point, particularly for those engaged in low-value transactions or who are not repeat players. Zambia180 and Malawi181 have also included sample documents in their registry regulations.

5. Interaction with Regulatory Bodies In some African countries, the secured transactions law reform is instigated and heavily influenced by the central bank.182 This can have both beneficial and less beneficial effects. First, in some countries the central bank has developed and now runs the collateral registry.183 This is a step forward from the traditional model of a registry run by the country’s ministry of justice.184 In particular, central banks command respect in the economy, as they are perceived as more trustworthy institutions than other governmental agencies and this instils a higher degree of confidence in the system. Further, this structure allows the central banks to evaluate the impact of the reform on access to credit more efficiently by collecting statistical information,185 and to help in ensuring the overall sustainability of the registry. The registry’s infrastructure can also be supported by the central bank’s infrastructure, its IT department and so on. Secondly, the central bank can use its involvement in the reform to consider how prudential regulation supports secured lending and to foster the financial inclusion agenda generally.186 Some of these issues are explored in the UNCITRAL Practice Guide. Such involvement of central banks may help the development of a strategy that simultaneously promotes access to credit and financial stability through the implementation of regulatory and prudential policies enshrined in international standards and best practices.187 Deviation from those standards and policies could have far-reaching consequences, including a reduced availability of credit. Particular attention should be given to, first, prudential requirements for loan-loss provisioning, including loan classification for impaired credit facilities; secondly, guidelines and supervisory expectations for credit risk policies and risk management procedures; and, thirdly, capital regulation for banking institutions. The idea that the reform aids regulation can, though, lead to effects on the scope of the secured transactions law which are not entirely benign. For example, the scope of the law

180 Zambia, ch 8, D3 181 Dubovec and Kambili (n 57) C. 182 Ghana, ch 3, A, D3; Sierra Leone, ch 9, C1 and C3. 183 Ghana, ch 3, A; Sierra Leone, ch 9, C3 (the registry is administered by the Banking Supervision Department); Liberia, ch 5, C5a. This route is also being considered by Zimbabwe and Ethiopia. 184 Ghana, ch 3, A. 185 See C7 below. 186 See further G Castellano and M Dubovec, ‘Credit Creation: Reconciling Legal and Regulatory Incentives’ (2018) 81 Law and Contemporary Problems 63; G Castellano and M Dubovec, ‘Global Regulatory Standards and Secured Transactions Law Reforms: At the Crossroad Between Access to Credit and Financial Stability’ (2018) 41 Fordham International Law Journal 531; and G Castellano and M Dubovec, ‘Bridging the Gap: The Regulatory Dimension of Secured Transactions Law Reforms’ (2017) 22 Uniform Law Review 663. 187 Castellano and Dubovec, ‘Bridging the Gap’ (n 187) 685.

The Method of Reform 437 could be limited to regulated lenders, as is the case in Ghana188 and Sierra Leone.189 This is likely to result in a fragmented system, and in possible priority conflicts between lenders subject to two different regimes.190 Moreover, the link with regulation can also lead to regulatory matters being included in the secured transactions statute, which is inappropriate since these matters clearly belong in a different sort of legislation.191 Thirdly, in Ghana the central bank included in its secured transactions legislation regulatory measures designed to ensure prudent lending by regulated institutions, such as criminal penalties for failure to register security interests.192 The recent Bill, which includes amendments to that legislation, widens the scope of lenders included within the regime, but, because criminal penalties could not be imposed on non-regulated lenders, the ‘sanction’ for non-registration by such lenders is the inability to enforce extrajudicially. There seems to be no sound commercial or conceptual reasons for this approach, and it is an example of the fact that the involvement of central banks in reform activities and hosting collateral registries introduces a different perspective that working groups and funding agencies must consider.193

6. Interrelationship with Other Areas of Law a. Techniques: Express and Implied Repeal, Amendment Almost by definition, when any area of law is reformed, there will be previously existing rules which no longer apply, or apply in a more limited way than before. If, at least in a common law jurisdiction, those rules are established by case-law, no action needs to be taken for them to be supplanted by the new statutory law. However, care needs to be taken by the courts to work out which parts of the previous case-law still apply and which do not. This is not necessarily merely an administrative exercise and requires an understanding of the concepts and reasoning underlying the new law. In relation to secured transactions law, the best example of this process was the way the new law was dealt with by the courts in Canada, which worked out the relationship between the new law and the previous law over many years.194 A clear example of a common law rule inconsistent with a reformed secured transactions regime is the common law rule about when a transferee of money takes free of an earlier interest, which is that a bona fide purchaser of the legal interest in the asset without notice of the prior interest will take free of that interest.195 The rule in a reformed jurisdiction

188 Ghana, ch 3, D1. This will be changed under the amendments included in the draft Bill. 189 Sierra Leone, ch 9, C2. 190 ibid. 191 Ghana, ch 3, D; Sierra Leone, ch 9, C1. 192 Ghana, ch 3, D3 (B&L Act). 193 ibid. 194 See, eg, the Canadian cases on the application of the PPSA to floating-charge-style arrangements, Credit Suisse Canada v 1133 Yonge Street Holdings Ltd (1996) 28 OR (3d) 670 (ONSC); (1998) 41 OR (3d) 432; Royal Bank of Canada v Sparrow Electric Corp (1997) 1 SCR 411 and the discussion in A Duggan and D Brown, Australian Personal Property Securities Law, 2nd edn (Australia, LexisNexis Butterworths, 2016) 4.47–4.55; and on the treatment of priority competitions involving trusts, see R Cuming, C Walsh, and R Wood, Personal Property Security Law, 2nd edn (Toronto, Irwin Law, 2012) ch 9. 195 Pilcher v Rawlins (1872) LR 7 Ch App 259, 268–69; Miller v Race (1758) 1 Burr 452, 457–58.

438 Conclusion such as Zambia, which requires collusion for the ‘taking free’ rule not to apply,196 is clearly inconsistent with the previous law. An example of a rule that continues to apply, but in a limited way, is from Zimbabwe, where the MPSI recognised perfection by a common law method, but subordinated such method of perfection to registration, irrespective of the time of such perfection.197 Where the previous law is included in a statute, there are various techniques that can be used to deal with the fact that the new law will replace or modify the old law. The first, which is by far the most satisfactory, is for the new statute to include express provisions repealing or amending the previous statutes. Although it is time-consuming and difficult to work out which previous statutes require this treatment and what that treatment should be, the result promotes certainty and clarity and complements the transitional provisions in the new statute. A variation of this approach was incorporated in the 2011 OHADA Uniform Act that repealed the 1997 predecessor, but preserved its application to the security rights created thereunder.198 The second possibility is for the new statute not to repeal or amend the old law, but to include a general provision stating that the new statute takes precedence over the old law where there is inconsistency.199 Although this is easier to do, it does carry the danger that it will not be entirely clear exactly when that precedence applies. It may be appropriate sometimes when the desire is to not repeal the old statutory provisions, since they still apply to certain transactions, but it would still be better to amend them expressly. The third possibility is just to rely on the doctrine of implied repeal. This is likely to vary from jurisdiction to jurisdiction, but is basically that a later statute has precedence over an earlier one if they are inconsistent.200 The problem with the doctrine of implied repeal is the uncertainty of its operation. A variant of this is Burundi’s Law on Security Rights, which provides that it repeals all prior inconsistent laws;201 this is an express repeal of sorts, but it is so general that it suffers from the same level of uncertainty as implied repeal. The fourth possibility, which is the least satisfactory because it creates the most uncertainty, is where the law expressly preserves the application of the prior law, such as in Ghana where the B&L Act provided the authority to establish a collateral registry, but also expressly stated that the registration under this Act is in addition to any requirements applicable under the Companies Act.202 A different technique has been used for the Cape Town Convention and its Protocols, which do not repeal any domestic laws, but rather provide for the creation of an international interest that encumbers an object autonomously and separately from any other domestic law security right, if any.203 The interrelationship between the international

196 Zambia, ch 8, C6. 197 Zimbabwe, ch 16, C1. 198 OHADA, ch 10, H. 199 See, eg, Zambian Movable Property (Security Interest) Act, s 6. 200 See, in relation to English law, Dean of Ely v Bliss (1842) 5 Beav 574, 582; Thoburn v Sunderland Council (2003) QB 151 (37). 201 Burundi, ch 11, C6. 202 Ghana, ch 3, D3. 203 See CTC, ch 17, E6. As pointed out in that chapter, some of the high-value, mobile assets to which the Cape Town Convention and its Protocols apply are typically excluded from general domestic secured transactions regimes.

The Method of Reform 439 interest and a domestic security right is based on the priority whereby the former prevails over the latter.204 Of course, there are many areas of general law which are not affected directly by secured transactions law reform, but impact on its operation. One example is the general law of property that determines whether the grantor has sufficient rights to create a security right or whether a specific asset is movable or immovable property.205 Another example is from the law of agency, which will govern the relationship between an agent, a principal and a third party every time the agent does something on behalf of a principal (either a grantor or a secured creditor) that is required by the secured transactions regime.206

b. Interaction with Other Statutes and Common Law As discussed above, it is critical that any reform dovetails well into the existing legal system. When the techniques discussed in the previous section are not used to make the position clear, great uncertainty results. Moreover, it may be that the new law is not entirely inconsistent with the previous law, but that the interaction makes the law conceptually complex and operationally inefficient, which diminishes the advantages that the reform is meant to deliver. This section discusses some particularly problematic issues that have arisen because of the way in which reform has been implemented in common law jurisdictions. i. Companies Act Registration Regime As mentioned in chapter 1,207 under the common law there are separate registration regimes for individuals and companies. When a new secured transactions regime is introduced, one would expect the statute to repeal expressly the previous registration regime, although there might be a need to keep it or amend it for security rights over assets which are not included in the new regime, such as immovable property.208 Unfortunately, this was not done in some African countries, or was done in such a way that it is not clear whether one registration regime exists or two, and, if the latter, whether secured creditors have a choice or have to register in both registries. In Ghana, the new regime under the B&L Act was limited, at least as a matter of fact, to creditors who were regulated by the Bank of Ghana (largely banks).209 Thus, although the regime does, unusually, apply to security interests in land,210 there may still be charges granted by companies to non-bank lenders, which will need to be registered under the Companies Act regime. The policy choice made was, then, to retain both systems of

204 CTC, ch 17, E1. 205 Zimbabwe, ch 16, C1. 206 For example, registering a notice, repossessing collateral, signing the security agreement. Agency law is particularly important when the principal is a company, which can only act through its agents. 207 Ch 1, D1c. 208 See, eg, the Zimbabwean MPSI, s 12(8), which amends the Companies Act, s 106, relating to registration of special notarial bonds, to take out the words ‘or movable’, leaving the registration requirement to apply merely to such bonds over immovable property, Zimbabwe, ch 16, B2. 209 Ghana, ch 3, C1. 210 ibid.

440 Conclusion registration, and, moreover, to require registration under both systems.211 The potential difficulties this raises, both in relation to perfection, and to priorities, are discussed in the Ghana chapter.212 It also looks as though an opportunity may have been missed to improve the situation by amending the Companies Act and aligning it with the reform of the B&L Act.213 The new Borrowers and Lenders Bill merely contains a general provision stating that its provisions supersede ‘any provisions in existing enactments applicable to security interests in movable and immovable property’,214 while a more specific provision dealing with the question of dual registration would have been more conducive of certainty. A very similar situation pertained in Sierra Leone until the 2019 reform (which is not surprising as the Ghanaian system was used as a model), and the potential difficulties are discussed in that chapter.215 In Nigeria, section 2(3) of the reforming statute, the STMAA, provides that: ‘Nothing in this Act shall prevent the creation of security interest [sic] in the form of charges by companies registered under the Companies and Allied Matters Act.’ The possible interpretations of this section, and their ramifications, are discussed in the Nigerian chapter,216 including the possibility that dual registration is required or that registrants have a choice as to where to register. Neither solution is ideal; express repeal or amendment would have made the situation much clearer. Conversely, in Kenya, there was some express repeal and amendment, but this was not done with great clarity and some interpretation is required to reach the conclusion that there is no need for dual registration.217 Moreover, some of the provisions of the Companies Act that relate to the long-superseded position where the company was the party expected to register218 still remain, even in the amended Companies Act.219 ii. The Floating Charge and Insolvency As discussed in the Introduction to this book,220 if secured transactions law in a common law regime is reformed to follow the functional approach221 so that there is only one type of security right, the distinction between fixed and floating charges, which is an important feature of the common law, becomes legally irrelevant for most purposes because the functions of the floating charge are replicated in the reformed system. This has the consequence for a reforming jurisdiction that previous legislation referring to a ‘floating charge’ either has to be repealed222 (as in the case of a provision defining a floating charge or a requirement that a floating charge be registered in a register other than the new collateral registry) or amended to replace the ‘floating charge’ with another 211 ibid D3. 212 ibid. 213 ibid. 214 Draft Borrowers and Lenders Bill, s 77. 215 Sierra Leone, ch 9, C6. 216 Nigeria, ch 7, E1. 217 Kenya, ch 4, B4. 218 See ch 1, D1c. 219 Kenya, ch 4, B4. 220 Ch 1, D1b. 221 Ch 2, B2a. 222 Amendment rather than repeal might be necessary where floating charges could still be taken over assets outside the new regime, such as land.

The Method of Reform 441 concept, where the existence of a floating charge is a trigger for a particular consequence. The latter situation applies, in particular, to the ability to appoint a receiver and manager, or an administrator (if the insolvency law allows for this), and the means of determining over which charged assets preferential creditors (and sometimes other claims such as the expenses of insolvency) have priority.223 Various possible approaches for these amendments are discussed in chapter 1.224 In many of the African reforms, neither of these steps have been taken or they have only been taken to a limited extent. This is bound to lead to uncertainty, and is a missed opportunity. In Ghana, the parts of the Companies Act 1963 relating to floating charges have not been repealed or amended at all, either under the B&L Act, or under the proposed Bill.225 In fact, the definition of ‘security interest’ in clause 76 of the Bill includes ‘fixed and floating charge’. It would therefore seem that a security interest created under the current regime or the proposed regime that fell within the definition of ‘floating charge’ in section 87 of the Companies Act (as supplemented by case-law) would govern whether a receiver and manager could be appointed226 and also would determine over which assets preferential creditors had priority if a receiver is appointed or the company is wound up.227 A similar route has been taken by Sierra Leone.228 Unlike Ghana, Kenya did consider the question of what to do about the floating charge. There were some consequential amendments,229 but no amendments were made to deal with the two critical points mentioned above relating to the appointment of an administrator230 or priority of preferential creditors,231 and the possible models of New Zealand and Australia were rejected.232 In Zambia, which had very similar provisions to these countries in its Companies Act, there was no express repeal or amendment of that Act. Although some of the Companies Act may have been consequentially amended,233 it is not clear that this amendment will extend to these important floating-charge provisions.234 In Nigeria, the ambiguity of section 2(3) of the STMAA means that it is unclear whether the uncertainty discussed above in relation to other jurisdictions applies or whether there is just no provision at all for preferential claims.235 Again, these issues relating to the floating charge do not seem to have been specifically considered. iii. Protective Provisions for Certain Parties Some African countries have also failed to repeal types of previous legislation other than company legislation, on the basis that some of the provisions are important and need to 223 Ch 1, D1b. 224 Ch 1, D1b. 225 Ghana, ch 3, B6 and C5. 226 Companies Act 1963, s 88. 227 See Ghana, ch 3, C6 and D3. 228 Sierra Leone, ch 9, B1 and C6. 229 The Insolvency Act, s 534 was amended to make the MPSR determinative of the priority of floating charges if there is more than one. See Kenya, ch 4, B4. 230 Insolvency Act, s 534. 231 Oddly, tax claims remain preferential creditors, but are also registrable under the MPSR. Kenya B4. 232 Kenya, ch 4, C9. 233 Zambia, ch 8, C. 234 ibid C1. 235 Nigeria, ch 7, E1.

442 Conclusion be kept. A particular example is where there is a previous statute governing hire-purchase agreements, which is likely to include provisions dealing with contractual matters between the two parties and may well include some element of protection for weaker parties, including restrictions on enforcement. At best, then, the failure to repeal this statute is confusing, since it will then be necessary to work out what is affected by implied repeal, and, at worst, it can lead to neither statute being able to fulfil its purpose properly. Kenya is an example of this phenomenon. The drafters of the MPSR wanted to keep provisions from the HPA governing licensing of hire-purchase businesses, and also those providing for a protective regime on enforcement, but otherwise the intention was for hire-purchase agreements to be governed by the MPSR. The technique used, however, was unsatisfactory in at least two respects. First, the only part of the HPA that was expressly repealed236 was the part dealing with registration of hire-purchase agreements, and even those sections (sections 4 and 5) were, seemingly inexplicably, only partially repealed. Secondly, it would have been better to incorporate the protective provisions on enforcement expressly into the MPSR rather than just by reference, which is what is done in section 65(4) of the MPSR.237 Other countries have not expressly repealed any parts of their hire-purchase acts. In Zambia, the intention was to supersede the HPA, but this was done by a provision giving supremacy to the MPSIA rather than an express repeal.238 The HPAs were not expressly repealed or amended in Ghana239 or Nigeria.240 The position in both countries is thus uncertain, since it is governed by the doctrine of implied repeal. This doctrine depends on potentially difficult questions of whether the two statutes are inconsistent, and well-considered express repeal or amendment would have been a better approach. iv. Leasing Law Some African countries have undertaken broader reforms that led to the enactment of leasing laws. Secured transactions and leasing laws are often uncoordinated since they are drafted and discussed by separate teams and different stakeholders. One result of an uncoordinated approach is that there are provisions in the leasing laws that provide for the establishment of a lease registry. At the same time, the secured transactions law may treat some leases as security rights providing for the registration of notices with respect to those leases in order for the rights of lessors to become effective against third parties. The situations in Liberia and Nigeria are illustrative of this approach.241

7. Statistical Information for Assessment Even the basic registration of notices in a collateral registry generates useful information, such as the number of registrations, the type of collateral against which registrations

236 See

the schedule to the Kenyan MPSR. ch 4, B2, C7 and F. 238 See Zambia, ch 8, C4. 239 Ghana, ch 3, C7 and C8. 240 Nigeria, ch 7, C. 241 Liberia, ch 5, C5a; Nigeriam ch 7, D1. 237 Kenya,

The Substance of Reform 443 are made and the number of searches made. This information can be used to assess the consequences of introducing a reformed secured transactions system and to inform future capacity building.242 However, many States, especially where reforms have been funded by the World Bank Group, require additional information to be submitted to the registry when a notice is filed. This information is removed from the public record and is not searchable but is kept for statistical purposes. The additional information may relate to the grantor, such as the type of industry or industries in which the grantor is engaged, the size of the grantor (if it is an entity) and whether the business is majority-owned by women. It could also include whether the grantor is a new or existing client of the secured creditor. The information might also relate to the loan or credit facility itself; for example, the type of credit advanced (a line of credit, an instalment loan, or another type of credit), the duration of the credit facility, the actual or maximum amounts loaned and the interest rates on those facilities.243 Although legally irrelevant, in that it is not required for a security right to be effective against third parties and does not affect priority, this additional statistical information that can be generated by collateral registries can be valuable for policymakers and commentators.244 For example, it helps determine which sectors of the economy have or have not benefited from the new regime and whether the businesses who obtained secured loans are owned by women. These conclusions can assist the planning of future capacity building245 and decision making as to whether further steps are required to improve access to credit. However, the requirement to enter this information must be assessed in the light of the purposes for which collateral registries are designed and the burden it places on users to gather information from their loan applicants that they would not otherwise collect during the loan approval process. Moreover, many providers of secured credit are non-financial institutions, but are, for instance, suppliers and sellers, and these groups do not have access to such information or may not have the means to collect it. They may, thus, be discouraged from using the collateral registry or might simply enter information that is inaccurate. In addition, statistical information can delay advance registrations because the lender may need to undertake significant due diligence or make commercial decisions before it is able to register a notice. For example, a lender may want to register a notice in advance of entering into the transaction,246 but, at that moment, it does not know the relevant information about the grantor or has not yet decided finally on the maximum amount secured or the interest rate of the loan.

D. The Substance of Reform 1. The Scope of the Regime Some of the most important and difficult decisions that have to be made by a State when reforming its secured transactions law relate to the scope of the new regime, as

242 See

Ghana, ch 3, D4b; Malawi, ch 6, A, D. Liberia, ch 5, C5a; Kenya, ch 4, C4; Sierra Leone, ch 9, C3; Rwanda, ch 14, C6 (maximum amount). 244 See, eg, the detailed results set out in Nigeria, ch 7, F. 245 See C4 above; Malawi, ch 6, D. 246 For the advantages of this approach, see ch 2, E1. 243 See

444 Conclusion recommended by the modern principles.247 African countries, while seeking to follow these principles, have also had to take into consideration practical difficulties in implementation, local culture, sensibilities and conditions,248 and the advice and views of external agencies and sponsors. Some of the debates that have occurred about scope, and their outcomes, have already been discussed in this chapter.249 One particular example is the discussion that took place in Zambia as to whether the regime should include all grantors or just those who were unincorporated. After due consideration, the decision was made to adopt a comprehensive approach.250 On the other hand, some countries have limited the type of lenders to whom the regime applies, with potentially unfortunate results.251 In terms of the types of encumbered assets covered by the regime, one question has been whether intermediated securities should be included.252 Another question, usually answered in the negative, is whether land should be included, but Ghana has included land in its regime.253 This section discusses two important issues about scope which have arisen in one form or another in all African reforms, and the treatment of which has required considerable debate and explanation.

a. The Functional Approach One of the key foundations of a regime based on modern principles is the functional approach. As explained in chapter 2,254 this has two aspects. The first is that there is one concept and one label, which is used for all proprietary interests which have the purposes of security. The second is that devices with the function of security based on retention of title are included within the regime. The first of these is relatively straightforward to understand, and its benefits are usually apparent in a country where there are many types of interest which can be granted by a borrower in order to secure obligations, since a regime of this type gives rise to complexity and fragmentation.255 Thus, in both civil and common law jurisdictions, an attempt has been made to reduce the number of such interests, and in many cases they have been reduced to one, although the label attached to this interest varies.256 Many jurisdictions, to avoid confusion, include in the definition of that interest a list of former interests which

247 See ch 2, B. 248 See C3 above. 249 See C1b, C3 and C5 above. 250 Zambia, ch 8, B. 251 Ghana, ch 3, D1; Sierra Leone, ch 9, C2. See also Ethiopia’s proposed approach, Ethiopia, ch 12, E1. 252 Kenya, ch 4, C1; Zimbabwe, ch 16, C1; Zambia, ch 8, D1. 253 See discussion at C1b above. See also B1d iii on the treatment of security rights over intermediated securities. 254 Ch 2, B 2. 255 See B2a ii above. 256 Ghana, ch 3, D1 (B&L Act ‘charge’, 2017 Bill ‘security interest’); Kenya, ch 4, C1 (‘security right’); Sierra Leone, ch 9, C2 (‘security interest’); Zimbabwe, ch 16, C1 (‘security interest’); Zambia, ch 8, D1 (‘security interest’); Nigeria, ch 7, E1 (‘security interest’); Liberia, ch 5, C1 (‘security interest’); Rwanda, ch 14, C1 (‘security interest’); Burundi, ch 11, C (pledge); Ethiopia, ch 12, E1 (proposed reform) (‘security right’). OHADA is an exception, in that some types of security rights are still retained in the reforming legislation. See OHADA, ch 10, B. See also Dubovec and Kambili (n 57) 186.

The Substance of Reform 445 are now included in the one concept,257 even though this is not the approach of the Model Law.258 Although this reduction in the number of interests used for the purposes of security has many benefits, it does raise difficult questions as to the relationship between the new classification and existing legislation. Some of the difficulties arising in the common law context are discussed above.259 The second aspect of the functional approach can be much more difficult for reformers to understand,260 and to work out how to implement within their own legal culture. The difference between retention of ownership and the grant of a security interest is deeply rooted in both common and civil law jurisprudence and, although the benefits of including retention of title devices within the regime for registration and priority purposes can be fairly readily appreciated, the need to change the location of title may not be. However, all reformed common law and mixed jurisdictions discussed in this book adopted the ‘unitary approach’ of the Model Law and models such as UCC Article 9 and the PPSAs,261 although in jurisdictions which limit the types of secured creditors to financial institutions, this may have a limited effect.262 The main difficulty which arose in these jurisdictions, apart from drafting problems,263 was the potential overlap with existing or new leasing or hire-purchase legislation.264 The approach in the reforming civil law jurisdictions is different. In Burundi, it was felt that the unitary approach could have negative practical consequences and perhaps even endanger its passage through parliament, so it was decided to adopt the Legislative Guide’s non-unitary approach,265 although with some differences. For example, under the non-unitary approach, an unregistered finance lease is, like an unregistered security right, ineffective against all third parties so that the lessor loses its ownership and the grantor/ lessee becomes the owner. However, in Burundi, if a finance lease is not registered, the lessor’s ownership right remains effective against all third parties except secured creditors who have registered in the National Registry.266 The reforms in Ethiopia also follow the non-unitary approach for hire-purchase agreements.267 Under the OHADA Uniform Act, there is no overt adoption of the non-unitary approach, and retention of title devices are treated separately from other security devices. This treatment does not apply to finance leases, which are outside the scope of the Uniform Act.268 However, the practical effect of the Uniform Act’s approach to retention of title devices is similar to that of the non-unitary

257 See, eg, Ghana, ch 3, D1 (the list in the B&L Act is over-inclusive and confusing); Kenya, ch 4, C1; Sierra Leone, ch 9, C2; Zimbabwe, ch 16, C1; Zambia, ch 8, D1; Ethiopia, ch 12, E1. 258 See the definition of ‘security right’ in the ML, Art 2 (kk). This approach is followed by Nigeria, ch 7, E1; Malawi Personal Property Security Act (number 8 of 2013) s 2; Rwanda, ch 14, C1. 259 C6b i and ii. 260 Kenya, ch 4, C1; Dubovec and Kambili (n 57) 205–06. 261 Kenya, ch 4, C1; Zimbabwe, ch 16, C1; Zambia, ch 8, D1; Nigeria, ch 7, E1; Rwanda, ch 14, C1; Liberia, ch 5, C1. 262 Ghana, ch 3, D1; Sierra Leone, ch 9, C2. 263 See, eg, Nigeria, ch 7, E1 (it is not clear whether outright transfers of receivables are included); Rwanda C1 (the precise scope is obscured by the drafting of the statute). 264 See C2b and C6b iii above. See also Ghana, ch 3, D; Sierra Leone, ch 9, C2; Nigeria, ch 7, A and B. 265 See ch 2, B2b. 266 Burundi, ch 11, C2f. 267 Ethiopia, ch 12, E1. 268 OHADA, ch 10, B.

446 Conclusion approach. An unregistered device is ineffective against third parties, the priority position of a registered device is similar to that of an acquisition security right and, on enforcement, the retention of title creditor must account for any surplus value to the debtor.269

b. Treatment of Non-Consensual Interests Although the Model Law and other models are primarily concerned with consensual interests, they envisage provisions governing the priority of competing non-consensual interests, such as tax liens, non-consensual liens and the interests of judgment creditors.270 Many African jurisdictions have gone further and provide for the registration of some categories of non-consensual interests in certain circumstances. The position varies considerably from State to State. Ghana, for example, includes judgments but not liens arising by operation of law (although these are included in the 2017 proposed amendments to the B&L Act).271 Sierra Leone, Nigeria and Zambia also enable a judgment creditor to register as one form of completion of execution, giving the creditor priority against later-registered secured creditors.272 The Kenyan reforms include tax liens and judgment creditors within the registration and priority rules,273 and a similar system will apply in Zimbabwe.274 Rwanda and Liberia include ‘non-consensual liens’ within the system for registration and priority purposes.275 Under the OHADA Uniform Act, some non-consensual liens, such as tax liens, are only effective against third parties over a certain amount if they are registered.276 Many African common law systems also give priority to some creditors, such as tax authorities, over the interest of a floating chargee in charged assets. As discussed above, the interrelation between these provisions and a reformed system based on the functional approach is not clearly dealt with in some African jurisdictions.277

2. Content of Security Agreement This section discusses two provisions relating to the content of the security agreement where reformed law in Africa varies, and which are of considerable importance.

a. Secured Obligation The Model Law contains an option for States to include a provision requiring a m aximum amount for which the security right can be enforced in the security agreement (and in the registered notice). Chapter 2 discusses the pros and cons of the inclusion of such a 269 ibid B2. 270 See ch 2, F4. See, eg, Burundi, ch 11, C, C4; OHADA, ch 10, B7 (although some liens do appear also to be registrable, see D2). 271 Ghana, ch 3, D1 and D5. 272 Sierra Leone, ch 9, C4; Nigeria, ch 7, E4; Zambia, ch 8, D1. 273 Kenya, ch 4, C4. 274 Zimbabwe, ch 16, C3b. 275 Rwanda, ch 14, C1; Liberia, ch 5, C4, C5a. 276 OHADA, ch 10, B7. 277 See C6b ii above.

The Substance of Reform 447 provision.278 Some African jurisdictions have included this type of provision in their reformed legislation, although the actual drafting (and therefore the meaning) varies. A provision closely tracking the Model Law is included in the Zambian and Nigerian legislation.279 The reformed legislation in Burundi stipulates that the secured obligation must be defined specifically, including by amount. On first sight, this seems inconsistent with the Model Law concept of a maximum amount, but the amount of the obligation need not be determined, only determinable, and so, in effect, the requirement is the same as one to state the maximum amount.280 As mentioned above,281 some jurisdictions require registered information about the secured loan, which will, presumably, also be included in the security or loan agreement. For example, the Ghana and Sierra Leone B&L Acts require registration of the maximum amount secured by the charge.282 Kenya takes a similar approach, despite a decision not to exercise the option in the Model Law to include a requirement to state a maximum amount in the security agreement.283 In Rwanda and Liberia, the maximum amount of the secured obligation must be included in the registered notice, but this is merely for statistical purposes and is removed from the public record.284

b. Anti-Assignment Clauses The arguments in favour of including a provision overriding anti-assignment clauses are rehearsed in chapter 1.285 Despite these arguments, not all African countries that have reformed their secured transactions system include such a provision,286 but many do, even though some deviate from the precise rule in the Model Law.287 In unreformed systems, especially in South Africa,288 the lack of any override is a matter of dissatisfaction and calls for reform.

3. Special Rules for Consumers Given that a secured transactions system in accordance with the modern principles includes consumer grantors as well as business grantors, it is not surprising that States want to include some special rules in relation to consumers, either in the reformed legislation or elsewhere,

278 Ch 2, C1. 279 Zambia, ch 8, C2 and C4; Nigeria, ch 7, E2 and E3. 280 Burundi, ch 11, C1. Note that under the unreformed law in Tunisia, the exact amount of the secured obligation rather than a maximum amount must be stipulated in the security agreement, Tunisia, ch 13, A. 281 See C7 above. 282 Ghana, ch 3, D4; Sierra Leone, ch 9, C3. See also OHADA, ch 10, D2. 283 Kenya, ch 4, C2, C4. 284 Rwanda, ch 14, C6; Liberia, ch 5, C5a. 285 Introduction, ch1, D3. 286 Rwanda, ch 14, C2; Ghana (Borrowers and Lenders Act), ch 3, D2. 287 ML, Art 13; Zimbabwe, ch 16, C3a; Zambia does not limit its override to trade receivables, so any receivables are covered, Zambia, ch 8, D4; Nigeria, ch 7, E2 (although the drafting is somewhat defective); Ethiopia, ch 12, E2; Ghana includes this in the new Bill (Ghana, ch 3, D2). 288 South Africa, ch 15, C3d.

448 Conclusion and this is anticipated by the Model Law.289 It is not all surprising, then, that such special rules exist in the reforming African countries; some examples are given in this section. A particular example is the Zambian legislation, which contains a number of consumerspecific provisions. First, the security agreement must describe encumbered consumer goods specifically, and the secured creditor must obtain further written consent for the security right to extend to future consumer goods.290 Unlike under the Model Law, acquisition security rights in consumer goods must be registered,291 and on discharge of a secured obligation by a consumer, the secured creditor must register a discharge within 14 days, whereas a non-consumer grantor must request a registered discharge.292 The enforcement remedy of retaining the collateral in satisfaction of the secured obligation is not available in relation to consumer goods.293 African jurisdictions have not incorporated the approach for automatic third-party effectiveness of acquisition security rights in consumer goods. Liberia has some similar provisions. The security agreement cannot cover future consumer goods,294 and registration in advance of the execution of the security agreement is not permitted in relation to security over consumer goods.295 A registered financing statement in relation to consumer goods cannot last longer than five years.296 In Burundi, judicial enforcement is usually required if the grantor is a consumer.297 As discussed above, some protective provisions for vulnerable parties are found in existing hire-purchase legislation, and some States have not repealed or amended this legislation, so that it still applies post-reform.298 The protective legislation in Kenya, however, applies not only to consumers but also to unincorporated businesses in relation to small extensions of credit; such special treatment is not contemplated by the Model Law.299

4. Methods of Achieving Third Party Effectiveness Under the Model Law, and many other modern secured transactions regimes such as the PPSAs and UCC Article 9, there are a number of methods that can be used to make a security right effective against third parties. Some (registration and, to some extent, possession) are used as they give publicity of the potential existence of a security right to third parties.300 Others provide a less transparent notice, such as control,301 or, as in automatic

289 ML, Art 1(5); this provision is incorporated into the Zambian MPSIA, Zambia, ch 8, D1. The ML also includes special rules relating to acquisition security rights in consumer goods, see Arts 24, 34(9), 38(3) option A, 38(2) option B. 290 Zambia, ch 8, D4. 291 ibid D5. 292 ibid D2. 293 ibid D7. 294 Liberia, ch 5, C2. 295 ibid C5b. 296 ibid. 297 Burundi, ch 11, C5a. 298 See C6b iii above. 299 Kenya, ch 4, B2. 300 See ch 2, D. 301 ibid.

The Substance of Reform 449 or temporary perfection, are used because the costs of other means is inappropriate in particular circumstances.

a. Registration as Sole Method In the course of secured transactions law reform in Africa, a number of different approaches have been taken. In many reformed African countries, the only permissible method, under the reformed regime, of making a security right effective against third parties is by registration. In Ghana, this step was taken initially and preserved in the 2017 Bill because lenders generally prefer to rely on the public records found in the Collateral Registry, and did not see any utility in possession or control.302 In Kenya, the alternatives were considered impractical (possession), not sufficiently transparent (automatic), or unknown in the local environment (control).303 The same approach is taken in Nigeria304 and Sierra Leone.305 One effect of restricting the method of making a security right effective against third parties to registration is that the priority rules become a great deal simpler. The general rule found in the Model Law, UCC Article 9 and the PPSAs of ‘first to register or perfect’ is replaced by ‘first to register’, and specific rules relating to possession and control306 become unnecessary.307 Other countries have more of a hybrid view. In Zimbabwe, other methods were originally included in the Bill but were taken out after stakeholder feedback that the Model Law’s alternative perfection methods, such as control agreements, undermine the transparency and informational value of the Collateral Registry.308 However, a provision was included in the late stages of the Bill enabling third-party effectiveness to be achieved by ‘any formal or informal mode recognised by the common law’, which preserved possession as a method,309 although the priority rules give a registered security interest priority over one made effective against third parties by possession.310 In Ethiopia,311 the draft Proclamation limits the availability of possession as a mode of perfection of security rights only to negotiable documents, negotiable instruments and securities. It is not commercially reasonable to take possession of other assets, such as equipment and inventory, even though that is not an uncommon practice in Ethiopia today. Zambia adopts most of the Model Law methods, namely, registration, possession and control (although one type of control over a bank account, where the lender becomes the account holder, was rejected as unknown in the Zambian market).312 Rwanda also

302 Ghana,

ch 3, D3. E3. 304 ibid E2. 305 Sierra Leone, ch 9, C3. 306 Ch 2, D. 307 Kenya, ch 4, C3 (the conflict of laws rules were also simplified, C8); Zimbabwe, ch 16, D; Nigeria, ch 7, E4. 308 Zimbabwe, ch 16, D. 309 ibid C1. 310 ibid C3 ii. 311 Ethiopia, ch 12, E2. 312 Zambia, ch 8, D5. 303 ibid

450 Conclusion includes registration, possession and control as modes of making a security interest effective against third parties, but, unfortunately, does not define possession or control.313 Burundi has the same methods, but gives more guidance as to the meaning of possession and control.314 Liberia’s methods are registration, possession and automatic perfection, but not control.315 Unlike the Model Law, most African jurisdictions neither included automatic perfection of acquisition security rights in consumer goods nor temporary perfection, such as when the secured creditor releases a negotiable document to the grantor to retrieve the goods from a warehouse.316

5. The Registration Regime a. Manner of Registration (Notice/Document Filing) Generally speaking, in the reformed systems registration takes place by notice filing,317 although the same is not the case for unreformed systems.318 There are two aspects to this method in which there is some divergence. One is that registration can be effected in advance of entering into the security agreement, and the second, often connected, is that the security agreement itself has to be presented to the Registrar, so that registration does not take place without human intervention. Under the system operating pursuant to the B&L Act in Ghana, registration in advance is not possible: there is a 28-day window after creation in which to effect the registration.319 Although the Act itself contemplates document filing, the practice since inception of the registry has been to register a notice.320 Although registration in advance will be possible in Zimbabwe, and what will be registered is a notice, some provisions of the Act still seem to be influenced by the older Deeds Registry system.321 In Rwanda, a scanned notarised copy of the abstract of a signed security agreement, when the debtor is an individual, and, in the case of a company debtor, a notarised board resolution authorising the company to secure a loan, must be submitted;322 therefore, registration in advance is not possible.323 Under the OHADA system, a form and supporting documentation must be submitted which is then checked by the registry staff.324 All reformed jurisdictions, with the exception of those governed by the OHADA Uniform Act, support and have achieved electronic registration. A few did consider a paper

313 Rwanda, ch 14, C3. 314 Burundi, ch 11, C2, and C2b. 315 Liberia, ch 5, C3. 316 See ML, Art 26(3) and Liberia, ch 5, C3 (for temporary perfection). 317 Kenya, ch 4, C4; Sierra Leone, ch 9, C3; Zimbabwe, ch 16, C; Zambia, ch 8, D2; Nigeria, ch 7, D2; Liberia, ch 5, C5b; Burundi, ch 11, C2a. See ch 2, E1 for a discussion of notice filing. 318 Tunisia, ch 13, C2; Ethiopia, ch 12, C; South Africa, ch 15, C4 and C5. 319 Ghana, ch 3 D3. This system is continued under the draft Bill. 320 ibid D4. 321 Zimbabwe, ch 16, C1. 322 Rwanda, ch 14, B. 323 ibid C6. 324 OHADA, ch 10, D2.

The Substance of Reform 451 alternative, but the downsides of this are so great in terms of the efficiency of the system overall that it was discarded.

b. Administrative Role of the Registry Contemporary collateral registries are not only electronic, but automated in a sense that registrations submitted electronically by registrants are entered directly into the registry database. This method of direct entry without any verification of the data by the registry staff allows the registration to become publicly searchable almost instantaneously. The registry system is set up in a manner to ensure that notices with incomplete fields in which some information must be entered are not recorded in the registry. However, this is an automated process that alerts the registrant that some information must be entered, rather than one that entails verification of the accuracy or validity of that information.

c. Identifiers and Search Criteria As mentioned in chapter 2,325 the modern principles recommend a system of notice filing that is primarily based on registering and searching by a grantor identifier. It is up to a jurisdiction to decide what identifier or identifiers are required to identify the grantor. The Model Law uses the name of an individual or legal person as the identifier,326 and contemplates that enacting States will stipulate what official document is the designated source of that name. However, since in many African countries there is a system of unique ID number for individuals and an identifying number for legal persons such as companies, many of the reformed systems stipulate that this should be the grantor identifier for registration and searching.327 Where there is no system of unique ID numbers, the name remains the grantor identifier, although this can cause problems.328 Although the Model Law provides a choice between exact match searches (which only reveal precise matches to search criteria) and close match searches, nearly all African jurisdictions have opted for exact match systems.329 As well as requiring registration by grantor identifier, many African jurisdictions either require or permit inclusion in the registered notice of serial numbers of uniquely identifiable goods, and specify a serial number as a search criterion.330 From the registrant’s point of view, including serial numbers in the registration is only efficient, or even possible, when the tangible assets are equipment, which will be kept by the grantor for a reasonably long period of time. However, it does not work well where the tangible assets are used as inventory

325 Ch 2, E2. 326 MRP, Art 9. 327 Nigeria, ch 7, E3; Sierra Leone, ch 9, C3; Liberia and Zambia have dual systems: both name and ID number have to be included in the notice, see Liberia, ch 5, C5b; Zambia, ch 8, D2. See also Rwanda, ch 14, C6, which appears to have a similar system but the legislation is unclear. 328 Ghana, ch 3, D4a since many people have the same name: the position will be changed under the 2017 Bill. See also the proposed Ethiopian scheme, Ethiopia, ch 12, E3. 329 For example, see Kenya, ch 4, C4; Sierra Leone, ch 9, C3; Zimbabwe, ch 16, C2. 330 Kenya, ch 4, C4; Sierra Leone, ch 9, C3; Zimabwe, ch 16, C2; Nigeria, ch 7, E3; Zambia, ch 8, D2. Rwanda, despite having no requirement to register serial numbers, specifies a ‘unique number assigned to an asset’ as a search criterion, but gives no details as to what this is, see Rwanda, ch 14, C6. The draft bill in Tunisia, despite following the ML in many respects, does not provide for serial number registration, see Tunisia, ch 13, G.

452 Conclusion (eg a company selling motor vehicles) since the stock of assets keeps changing and registration would become very burdensome.331 Further, if inclusion of a serial number is compulsory, the issue arises as to which types of assets the rule applies, since not all tangible assets have serial numbers. Kenya dealt with this issue by only requiring a serial number to be included in relation to motor vehicles and, even then, only when ‘motor vehicle or trailer’ is selected from the drop-down menu of types of collateral, as opposed to the more generic types of ‘equipment’ or ‘all assets’ even if the collateral does in fact include a motor vehicle.332 Liberia also limits this approach to motor vehicles,333 although Nigeria also includes boats and planes.334 Although Zambia also limited serial number registration to motor vehicles, it does not include an exclusion for vehicles held as inventory, which could be cumbersome and increase costs.335

6. Enforcement One of the most critical parts of a secured transactions regime is the part dealing with enforcement, since it is the ability to enforce quickly, cheaply and successfully that incentivises financiers to extend credit. A regime based on the modern principles will include provisions permitting extrajudicial enforcement, including checks and balances in order to take the interests of the various parties into account, as well as the ability to enforce judicially. Most of the African jurisdictions which have reformed their law permit extrajudicial enforcement, although some limit the available methods of enforcement, and some include alternative mechanisms not included in the Model Law, which appear to be introduced because of local conditions and concerns.

a. Extrajudicial Enforcement Some African jurisdictions largely track the Model Law provisions on enforcement, including extrajudicial enforcement. An example is Kenya, which just made a few changes such as adding some more detail to some of the notification provisions.336 Nigeria also provides for a full set of extrajudicial enforcement remedies.337 Zambia permits extrajudicial enforcement, but requires registration of a notice of enforcement in the Collateral Registry first, to give publicity of the enforcement to all parties, including the debtor.338 There does not, however, appear to be a sanction for not registering a notice except a possible liability for damages. Liberia also has extensive provisions for extrajudicial enforcement.339 The legislation of some other reformed countries, however, includes considerable limits on the ability of a secured creditor to enforce extra-judicially. Ghana’s B&L Act requires a

331 For

this approach, see Kenya, ch 4, C4; Nigeria, ch 7, E3; Liberia, ch 5, C5b. ch 4, C4. Liberia has a similar system, see Liberia, ch 5, C5b. 333 Liberia, ch 5, C5b. 334 Nigeria, ch 7, E3. 335 Zambia, ch 8, D3. 336 Kenya, ch 4, C7. 337 Nigeria, ch 7, E5. 338 Zambia, ch 8, C7. 339 Liberia, ch 5, C7. 332 Kenya,

The Substance of Reform 453 notice of default to be sent to the borrower, who then has 30 days from receipt to cure the default (borrowers often elongate this period by managing to avoid ‘receipt’). Moreover, the secured creditor then has to register a notice of enforcement and, after 30 days, the registry has to issue a memorandum of non-objection before the creditor can enforce.340 The involvement of the Registrar in this process, and in the Rwandan process, is not in accordance with the modern principles: enforcement is not a matter in which the registrar should be involved. The position in Ghana is, however, improved in the draft Bill. Sierra Leone has a similar system to that of Ghana, except that there is no procedure requiring registration of intention to enforce and a memorandum of non-objection before enforcement.341 In civil law and mixed reformed jurisdictions, there is even less opportunity for extrajudicial enforcement. In Zimbabwe, although there can be extrajudicial enforcement against intangible assets, tangible assets cannot be sold extrajudicially. These limitations were introduced during the drafting process by the Ministry of Justice.342 However, possession of goods can be taken as an interim measure.343 In Rwanda, only a senior secured creditor can take possession extrajudicially, and it can only do so once it has applied for and obtained a certificate of authorisation from the Registrar. The provisions on out-of-court enforcement are very skeletal.344 There is no general right to enforce extrajudicially in Burundi, but it is possible in some limited circumstances.345 The proposed legislation in Ethiopia also permits limited extrajudicial enforcement, although other methods are also included so as to deliver the outcome most conducive to credit decisions.346

b. Alternative Forms of Enforcement One means of enforcement that is not included in the Model Law but is provided for in some common law jurisdictions is the ability to appoint a receiver, including a receiver and manager, over the charged assets. This was possible, at least with court involvement, in many pre-reform jurisdictions for secured creditors with floating charges.347 The ability of a floating charge holder, who had charges over the whole undertaking of the company, to appoint a receiver and manager enabled the business to be sold as a going concern, thus capturing extra value for all the creditors. It is therefore beneficial that the ability to appoint a receiver, or a receiver and manager, is preserved in the reformed law.348 One potential problem with an extrajudicial right to take possession is that, since it cannot be exercised if this will cause a breach of the peace, it is sometimes impossible to carry out, since the grantor may well object. Therefore, some jurisdictions bolster extrajudicial enforcement by permitting a secured creditor to obtain the assistance of the police

340 Ghana, ch 3, D6. 341 Sierra Leone, ch 9, C5. 342 Zimbabwe, ch 16, C, C5. 343 ibid C1. 344 Rwanda, ch 14, C5. 345 Burundi, ch 11, C5a. 346 Ethiopia, ch12, E6. 347 See ch 1, D1b. 348 Zambia, ch 8, D7; Ghana, ch 3, D4 (the appointment must be registered with the registry); Sierra Leone, ch 9, C5 (the appointment must be registered with the court); Kenya, ch 4, C7 (the parties can agree that the secured creditor can appoint a receiver); Liberia, ch 5, C7.

454 Conclusion in order to enforce peaceably. Thus, in Nigeria, a secured creditor who has registered a notice in relation to its security right, and who serves a copy of the security agreement and proof from the Registry that the registration was completed, can require the police to provide this assistance.349 Similarly, in Sierra Leone, a lender who is unable to enforce peaceably can obtain a warrant from the court entitling it to use the police or bailiffs to carry out repossession of the encumbered assets.350 In the Leasing Proclamation already in force in Ethiopia, the Ministry of Trade is under a duty to take measures to restore the leased goods to the lessor if the lessee will not return them and can order the police to facilitate this: this power proved successful in practice and the Collateral Registry has been given similar power in the recently enacted Secured Transactions Proclamation.351 Given the potential difficulties with extrajudicial enforcement, remedies of this sort could go a long way towards facilitating expeditious extrajudicial enforcement of security rights in a peaceable manner.

c. Expeditious Judicial Enforcement One of the main arguments against judicial enforcement is that it can be very slow. Not only does this mean that a secured creditor has to wait to recover the money that it is owed, but also that the value of the collateral may reduce, or, even, that the grantor may dispose of the collateral and disperse the proceeds in an irrecoverable way. Speedy judicial proceedings, both as an interim measure to preserve the collateral and its value, and as a final enforcement measure, go some way to ameliorating this problem. There are two caveats, however. The first is that sufficient protection must still be given to the grantor; this is a question of balancing the interests of all parties.352 The second is that speed does not necessarily solve the other problem with judicial enforcement, which is that it is costly. This is, of course, more of a problem in relation to low-value loans and collateral. Several reformed jurisdictions have provided for expeditious enforcement and dispute resolution proceedings within their secured transactions regime. Liberia has established a Commercial Court, which has ex parte procedures to enable a secured creditor to take possession of collateral quickly.353 In Sierra Leone, a Fast Track Commercial Court has been established to provide speedy relief, although only in disputes involving more than a certain amount of money.354 Burundi does not permit extrajudicial enforcement of leases, but has put in place strict timelines for judicial enforcement, including short time limits for appeal.355 Alternative dispute resolution can also play a part. The Nigerian STMAA provides for this type of dispute resolution, although it is unclear at present whether the Panel provided for in the legislation has actually been set up.356

349 Nigeria,

ch 7, E6 Leone, ch 9, C5. 351 Ethiopia, ch 12, E5. 352 Discussed further in ch 2, H. 353 Liberia, ch 5, C7. 354 Sierra Leone, ch 9, C5. 355 Burundi, ch 11, C5c. 356 Nigeria, ch 7, E6. 350 Sierra

Conclusion 455

E. Conclusion As demonstrated by the chapters in this book, many African countries have undertaken legal reforms with the primary objective of modernising the legal framework to facilitate access to credit. This modernisation process addresses the fundamental challenges which arose under the unreformed law. These include the limited scope of assets that could be used as collateral and the persons that could engage in secured transactions, excessive formalities for creation (eg stamp duties on the execution of security agreements or their notarisation in civil law countries), lack of publicity (eg security rights being effective against third parties without any form of a public notice), and inefficient remedies, often limited to judicial mechanisms. Those considering reform in Africa are encouraged by the emerging evidence that these reforms have had a positive economic impact globally, and by seeing the improvements their neighbours have achieved. The World Bank Doing Business report is less a driver of reform than an influencer of the approaches taken during the drafting of reformed law; for example, the approach to the priority of security rights over non-consensual liens. Probably the most compelling attraction for African countries is the electronic collateral registry, which replaces outdated and inefficient paper-based registries, and reduces the potential for corruption. At times, reformers pay more attention to how the registry will operate than to the legal rules for security rights. However, for a collateral registry to have an impact, it must be underpinned by a coherent and comprehensive legal framework based on modern principles. Early results show improvement, particularly for individual borrowers and micro-enterprises, but also some negatives, especially the minimal use of commercial collateral, such as inventory and receivables. There are five types of models chosen as the basis for a reform: (i) the Model Law; (ii) a domestic statute that reflects modern principles, such as a PPSA; (iii) a domestic statute of another African jurisdiction; (iv) a simplified model; and (v) a mix of two of the preceding types. The first two types have produced the most coherent and comprehensive frameworks. The third type has had mixed results: a positive outcome in Zambia (inspired by Malawi) and quite the opposite in Sierra Leone (inspired by Ghana). Combining the different models requires a great deal of care, as they often differ in approaches and drafting, though the fundamental underlying policies are similar. Liberia has encountered some challenges integrating UCC Article 9 and a PPSA. The simplified model has produced a legal framework in Rwanda that on its face appears to contain all the elements of a modern secured transactions system, but, on closer examination, is fraught with legal landmines. All African enactments deviate, to some degree, from the chosen models, often to accommodate local considerations, whether they be practices, other laws or drafting conventions. Examples of these deviations considered in this book range from the struggle to implement the unitary approach to all types of security rights, particularly with respect to finance leases, to the crafting of specific rules applicable to security rights in attachments to immovable property. Careful discussions with stakeholders have also led to simplified approaches, such as the recognition of registration as the sole method of third-party effectiveness and the consequential enhancement of the overall transparency of security rights, where control, automatic and temporary methods of third-party effectiveness were not viewed as providing the same degree of transparency. Such discussions also led to the development of

456 Conclusion alternative remedies where it was felt that neither extra-judicial nor judicial remedies would provide adequate protection for secured creditors. Involvement of local stakeholders during all stages of the reform process, from an assessment of the pre-reformed legal framework to capacity building, has proved to be essential. Local stakeholders also benefit from expert advice that explains the various differences among the models and the reasons for them. A critical aspect of any secured transactions reform is proper coordination with other legislation in two areas: the existing legislation, and other ongoing reforms. Both of these have presented a number of challenges in Africa. Some reforms already completed have not been coordinated properly, particularly with insolvency and company laws, but also with other specialist legislation, such as hire-purchase acts. The deficiencies in such coordination might not have caused any practical challenges yet, but, as these markets mature, legal problems are expected to emerge. Coordination with other reforms presents unique challenges caused primarily by donors supporting such reforms who proceed in a siloed fashion for example, providing experts well versed in the particular subject-matter who then limit the discussions to that subject-matter. Challenges also arise from the production of model texts by international organisations. These texts cannot take into account all the possible local idiosyncrasies affecting their integration into domestic frameworks. This process does not serve the countries well. The ratification and implementation of the Cape Town Convention, and its Protocols, raises less of a challenge as it operates autonomously from the domestic regime. The future MAC Protocol presents an opportunity, especially for those African jurisdictions with deficient frameworks as well as those struggling with the implementation of a collateral registry, to incorporate a modern regime for the financing and leasing of MAC equipment. It would be premature to measure the economic impact of secured transactions reforms in Africa. Most of the reforms examined in this book were completed only a few years ago, and many are in the process of being implemented. Using the statistical information generated by the collateral registries does not provide an accurate picture of the developments in the market, though it highlights some areas of progress as well as where improvements are needed. A thorough economic impact study of a reformed African jurisdiction has not yet been conducted. African countries contemplating reforms could, however, already learn from some lessons highlighted in this book, which are based on consideration of the reform processes elsewhere in Africa. Choosing the right model, engaging local stakeholders, operationalising the system only after the collateral registry is fully functional and stakeholders trained are key. Learning about the experiences of other jurisdictions helps those contemplating reforms avoid some of the challenges and pitfalls that plagued the implementation of reforms elsewhere. The 2012 pan-African event in Ghana became an incentive for many reforms in the years to come, but more events of this nature are needed, particularly to share experiences. As more collateral registries are implemented, registrars would also benefit from sharing their experiences, particularly about the operational aspects of collateral registries, different types of access, payment methods and features. Finally, this book hopes to contribute to legal education as scant attention has been paid to these reforms, particularly those in smaller African countries, despite their profound impact on the overall commercial law framework.

INDEX accession (attachment to movable property) Burundi 255, 258 Liberia 125–126 Rwanda 319 Sierra Leone 200 acquisition security rights 45, 55–56, 112, 254, 335, 370 administrator appointment of 26, 101–102, 401, 441 insolvency 79, 379, 380 advance registration 84, 230, 443 Africa Aircraft Protocol 383–384 economies 17–18 financing types leasing 20 microfinance 18–19 supply chain finance 20–21 warehouse receipts 19–20 legal systems 16–17 Luxembourg Rail Protocol 384–385 MAC Protocol 392–394 see also Mining, Agriculture and Construction (MAC) Protocol, Cape Town Convention secured transactions law reform in see reform of secured transactions law see also Burundi; Ethiopia; Ghana; Kenya; Liberia; Malawi; Nigeria; Rwanda; South Africa; Tunisia; Zambia; Zimbabwe after-acquired (future) assets 32–33 Aircraft Protocol, Africa 383–384 all-asset security 238, 347 anti-assignment clauses 28, 78, 155, 218, 351 breach 447 and receivables 31–32, 271, 315 applicable law 32 Ethiopia 282 Ghana 89 Kenya 103, 114 South Africa 320 Zambia 175 Zimbabwe 363, 369, 372 appointment of administrative receiver 347 of administrator 26, 101–102, 401, 441

of a receiver or manager see appointment of receiver/manager registration of 453 appointment of receiver/manager 441, 453 fixed and floating charges 24, 25, 26 in Ghana 71, 72, 82, 85 in Kenya 113 in Liberia 130 in Nigeria 146, 153, 154, 157, 160 in Sierra Leone 202, 208 in Tunisia 300 in Zambia 185 assets after-acquired 32–33 all-asset security 238, 347 circulating 24–26, 32, 116, 423 commingled see commingled assets description in security agreement 423 encumbered see encumbered assets fixed charges over 24–25 floating charges over see floating charges fungible 136, 232, 257 intangible see intangible assets movable property see movable property law pledged disposal of 301–302 intangible assets 220–225, 293–294 tangible assets 220 possession see possession/taking possession rights in 47 scope of secured transactions law 44–45 tangible see tangible assets assignment anti-assignment clauses 28, 78, 155, 218, 351, 447 priority issues 57, 70, 206, 224, 317, 446 receivables and anti-assignment clauses 31–32 Burundi 245, 248 Liberia 120, 123 Nigeria 141, 146 OHADA Uniform Act on Securities 214, 218, 221, 236, 238 Rwanda 312 Sierra Leone 197, 201

458 Index Tunisia 285 Zambia 175 attachment Burundi 250 concept 316, 317, 425 creation, relationship to 37, 46, 314, 428 general notarial bonds 346 to immovable property 426, 455 Kenya 111–112, 117 Liberia 124 Rwanda 314 Sierra Leone 205 Tunisia 302 Zambia 184 Zimbabwe 363, 369, 371 to land 293, 426 to movable property Burundi 255, 258 Liberia 125–126 Rwanda 319 Sierra Leone 200 of a security interest 46, 112, 180, 314 of a sign 250 special notarial bonds 350 Australia 3, 25–26, 116, 318, 401, 441 Personal Property Security Register (PPSR) 404, 407 authorisation for registration 453 in Burundi 249 in Ghana 73 in Kenya 109 in Liberia 128, 130 party autonomy 36, 382, 383 in Rwanda 321, 323, 325 unauthorised registration 52, 187, 407 aviation sector 382–384 bank accounts 48, 78, 89, 114, 181, 182, 249, 372 designated 315, 341 funds deposited 198, 221, 232 holding as security 234, 257, 335 instructions of a secured creditor 50 OHADA Uniform Act on Securities 220, 221 perfection of security interests 183, 186 pledges 29, 215, 221, 222, 225, 226, 247, 250, 256, 293, 298 secured transactions law 25, 27, 28, 59 security interests/rights 49, 89, 110, 162, 176, 181, 183, 316 tracing rules 48 transfers 201, 319, 427 bankruptcy 9 in Ethiopia 265, 266, 284 in Kenya 96, 116 in Sierra Leone 208

in South Africa 332 trustee in bankruptcy 96, 208 in Tunisia 296 in Zambia 174, 176 see also insolvency bills of lading 111, 136, 143–144 bills of sale 22–23, 27, 141–143, 146, 171, 424 absolute bills 22 bills by way of security 22 penalty for failure to comply 23 book debts see receivables Burundi 16, 239–261 assignment of receivables 245, 248 authorisation for registration 249 commingled assets 255 economic overview 239–241 encumbered assets 247–250, 252–257, 259, 260 expeditious judicial enforcement 454 grace period 244, 254 insolvency 255 Law on Security Rights and Leasing Law (2016) 245–260, 438 conflict of laws rules 258–260 creation 246–248 enforcement of leases 258 enforcement of pledges 256–257 enforcement of retention of ownership sales 257–258 operating leases 241 pre-reform legal framework 241–244 creation and effectiveness requirements 242–243 priority of security interests 243, 253–256 priority rules rights and obligations of parties 252–253 third-party effectiveness 248–252 transitional 258–260 substance of reform 445 third-party effectiveness 248–252 attachment of a sign 250 dispossession 250 finance leases 251–252 pledge, making effective against third parties 250–251 registration 248–250 retention of ownership sales 251 unsecured creditors 244 cancellations 108, 230 notices 52, 84, 107, 249 unauthorised 231 capacity building 434–436 Cape Town Convention and Protocols 33, 377–394 aviation sector, facilitating financing in 383–384 declarations under 381–382

Index 459 encumbered assets 379, 381 functional approach 379–380 insolvency 379, 380, 394 Mining, Agriculture and Construction (MAC) Protocol 387, 390 international registries 380, 418 Luxembourg Rail Protocol 384–385 Mining, Agriculture and Construction (MAC) Protocol 385–394 see also South Africa Central Securities Depository (CSD) 420 cession, security agreements prohibiting 340–341 creation 339–340 enforcement 342 nature of 338–339 South Africa 338–342 charges 23, 145–146 priority of security interests 80, 87, 101, 102, 146, 169, 170, 192, 441 see also fixed charges; floating charges; registration of security interests chattel mortgages 96, 104, 363, 364 chattel paper 428 Liberia 123, 125, 130 Rwanda 313 Sierra Leone 201 Zimbabwe 364 circulating assets 24–26, 32, 116, 423 civil law jurisdictions concepts 27–30 extrajudicial enforcement 29–30 fragmented and formalistic tradition 27, 29 priority disputes 29 security devices 423 see also Burundi; Ethiopia; OHADA Uniform Act on Securities; Tunisia collateral crops as 240 limits on what can be used as 423–424 over-collateralisation 38, 47, 126 proceeds 124, 285, 316, 371 repossession of 160, 186, 202, 321, 371 secondary markets for 410 seizure of 130 Collateral Registry, Ghana 65, 68, 427 performance 86 and Regulations 82–86 searchability and indexing 85–86 collection of encumbered assets 48 of receivables 114, 130, 186, 221, 234, 257, 299, 339, 365, 372 Commercial Code, Liberia see Liberian Commercial Code

commercial law, basic principles 35–36 commercial reasonableness 36, 39, 57, 89, 111 commingled assets Burundi 255 Ghana 87 Liberia 125 priority 319 Rwanda 319–320 Sierra Leone 198 Zambia 182–183 Zimbabwe 369 see also mass or product commodity exchanges, Malawi 137–138 common law jurisdictions concepts Bills of Sale regime 22–23 fixed and floating charges 23–26 registration regime 26–27 floating charges, dealing with 25 and reform methods 439 see also Ghana; Kenya; Liberia; Malawi; Nigeria; Sierra Leone; Zambia conflict of laws 60 Burundi 258–260 Ethiopia 282–283 Ghana 89–90 Nigeria 161–162 OHADA Uniform Act on Securities 236–237 transition provisions Burundi 258–260 Ghana 89–90 OHADA Uniform Act on Securities 236–237 constructive possession 30, 270, 280, 293, 335, 336 consumers, special rules for 447–448 control agreements 219, 373, 449 creation of security rights 37–38 attachment, relationship with 37, 46, 314, 428 Bills of Sale Acts regime 23 in Burundi 242–243, 246–248 cession 339–340 civil law 28 in Ethiopia 280 general notarial bonds 344 in Ghana 77–78 in Kenya 105–106 in Nigeria 155–156 pledges, South Africa 335–337 rights in the asset 47 in Rwanda 313–315 scope of the security right 47–48 security agreement 46–47 in Sierra Leone 198–200

460 Index in South Africa cession 338–339 general notarial bonds 344 pledges 335–337 trade receivables 48–49 in Tunisia 292–295 in Zimbabwe 359–362, 369 see also pledges credit affordable, infrastructure supporting 408–411, 419 common barriers to accessing secured credit 31–33 increasing access to 8 lending practices 409–410 prudential regulation 409 secondary markets for collateral 410 Credit Information Index 4–7, 9, 415 creditors of lessees 149, 273 receivables creditors 31, 32, 260 see also judgment creditors; preferential claims/ creditors; secured creditors; unsecured creditors crops as collateral 240 financing against 19 growing 19, 125, 182, 353, 363 immovable property 363 priority issues 279–280, 321 unharvested 293 crystallisation of floating charges 24, 146, 192 currency Burundi 240 Ethiopia 264 Ghana 67 Kenya 93, 97 Liberia 121 money as 310, 313 proxy 355 Rwanda 310, 313 Zimbabwe 354, 361 data on registry cyberattacks and privacy 406–407 duplicate, out-of-date or redundant, allowing to remain 408 human error 407 incorrect or inappropriate 407–408 misunderstanding, incorrect information entered due to 408 debtors 21, 311 consumer 127, 130 in Ethiopia 266, 269, 270 individual 134, 137

organisations 324 professional 215 protecting 113, 241 receivables see receivable debtor unincorporated 168 vulnerable 98–99, 113 in Zambia 179, 182, 183 default 8, 39, 56, 98, 153 abusive conduct by creditor 113 clauses 87 curing 173, 201–203, 208, 235 defining 87, 186 events of 87 grace period Burundi 244, 254 Ghana 87, 88 Kenya 112 Liberia 125 Nigeria 158, 159 OHADA Uniform Act on Securities 235 Rwanda 316–317, 320 secured transactions law 53, 56 Sierra Leone 202 transitional 429, 430 Zambia 183 notice of 87–88, 113, 114, 149, 160, 201, 208 OHADA Uniform Act on Securities 214, 233, 234 post-default remedies 98–99, 113, 130, 161, 179, 186 rules 56, 57–58 deficiency, and enforcement of security interests 59–60 Depth of Credit Information Index 9, 415 dispossession 220, 248, 250, 330 Doing Business Report (World Bank) 9, 109 Doing Business index/rankings 3, 9 Getting Credit Indicator 3–4, 9, 132, 133, 141, 167, 239, 261, 263, 308, 310, 415, 416 Legal Rights Index 9, 46, 168, 189, 239, 332, 415, 416, 423 reform drivers for 415, 416 method of 432 substance of 455 Ease of Doing Business rankings 4–7, 286 economies, Africa 17–18 Burundi 239–241 Ethiopia 262–264 Ghana 66–69 Kenya 92–95 Liberia 119–121 Nigeria 140–141 Sierra Leone 191–192 Tunisia 286–288

Index 461 electronic warehouse receipts (EWRs) 10, 134, 136, 137, 277, 420 encumbered assets acquisition in satisfaction 234, 237 collection of 48 consumer goods 448 description 78, 82, 448 enforcement 58, 59 grantor-based registration system 53 legal issues 25, 29, 31 possession 297, 373 proceeds 48, 254 repossession 269, 366, 372, 454 Rwanda 325 scope of security rights 47 security agreement 46, 47, 48 Sierra Leone 203 South Africa 343 tangible assets 250, 256, 372 third-party effectiveness 38, 49 Tunisia 285, 289, 293, 295, 297, 299, 301, 302 Zambia 175, 177, 183, 187, 444 Zimbabwe 357, 362, 367, 371–373 enforcement of security interests 39, 452–454 acquisition of encumbered asset in satisfaction 234, 237 alternative forms 453–454 appointment of receiver see appointment of receiver/manager in Burundi 243–244, 256–258 Cape Town Convention and Protocols 380–381 cession 342 collection 48n91, 114, 186, 221, 365, 372 by court 59 deficiency 59–60 in Ethiopia 282 extrajudicial 59, 424, 452–453 in Ghana 87–89 in Kenya 113–114 leases 258 in Liberia 130–131 maximum amount 46, 84, 105–106, 126, 156, 157, 176, 180, 247, 280n156, 446 in Nigeria 160–161 OHADA Uniform Act on Securities 233–235 pledges 256–257, 337 possessory and non-possessory 243–244 proceeds 58, 59, 60, 235 retention of ownership sales 257–258 in Rwanda 321–323 in South Africa 337, 342 surplus 59–60 taking possession see possession/taking possession

in Tunisia 298–302 disposal of pledged asset 301–302 lessors’ remedies 302 possessory pledge 299–300 provisional judicial remedies 300–301 in Zambia 185–187 in Zimbabwe 371–372 enterprise pledges 28, 29 Burundi 242, 243, 244 Tunisia 290–292, 297, 301 enterprises large 95, 264 lien of seller of an enterprise 221, 223, 224 microbusinesses/microenterprises 10, 18, 40, 95, 141, 169, 263, 455 small and medium enterprises (SMEs) 8, 40, 263 equipment financing, modern approach to see Cape Town Convention and Protocols immovable-associated 389–390 lessors 148, 149, 384 MAC Protocol 388–394 pledges 294–295 possession 42 priority of security interests 394 errors identifiers 52, 54, 108, 368 registration of security interests 407 in serial numbers 177, 199, 208 effect on lessee 53, 108, 128, 368 Ethiopia 262–284 advances on securities 271 alternative forms of enforcement 454 applicable law 282 assignment of rights 270–271 bankruptcy 265, 266, 284 currency 264 current secured transactions regime overview 266–269 proposed reform 277–283 security devices 269–277 economy 17, 262–264 intermediated securities 279 leases 272–273 legal system 264–266 mortgage of a business 273–275 negotiable documents 280, 449 operating leases in 267, 272, 279, 284 pledges 270, 271 warehouse receipts 275–277 priority of security interests 281–282 proposed reform 277–283 conflict of laws 282–283 creation and third-party effectiveness 280

462 Index enforcement 282 priority 281–282 registry 280–281 scope 277–280 transitional provisions 283 registration as sole method 449 registration fees 267 retention of title 271–272 substance of reform 445 warehouse receipts, pledges 275–277 expeditious judicial enforcement 454 extinguishment of pledges, South Africa 337 extrajudicial enforcement 59, 424, 452–453 civil law 29–30 repossession 160, 269 factoring, supply chain finance 20 fees, registration Ethiopia 267 Ghana 71, 72 Kenya 98 Liberia 128–129 Nigeria 143 Rwanda 326 Zambia 177 see also registration of security interests; registry finance leases 20, 73, 251–252 third-party effectiveness 251–252 financial considerations commodity trading 419–420 credit infrastructure supporting affordable credit 408–411, 419 lending practices 409–410 market opportunity assessment 410–411 prudential regulation 409 secondary markets for collateral 410 financing types, in Africa leasing 20 microfinance 18–19 supply chain finance 20–21 warehouse receipts 19–20 fixed charges 24–25 in Australia 116 distinguished from floating 117, 440 in Ghana 82 in Kenya 101 in Nigeria 145, 146, 153 in Sierra Leone 192 in Zambia 171 fixtures Liberia 123, 124, 131 OHADA Uniform Act on Securities 223 Rwanda 320 Sierra Leone 205

Zambia 184 Zimbabwe 363 see also immovable property, attachment to floating charges 23–26, 440–441 alternative forms of enforcement 453 in Australia 116 crystallisation 24, 146, 192 development 23–24 distinguished from fixed 117, 440 effects 25 features 24 in Ghana 72, 81–82, 84, 87, 441 in Kenya 101, 102, 105, 115 in New Zealand 115 in Nigeria 145–146, 153, 154 qualifying floating charge holder 26, 101, 102 repealed or amended in insolvency 440–441 in Sierra Leone 192, 193, 204, 207 in South Africa 333, 343 and tax authorities 446 uncrystallised 101, 117 in Zambia 168, 170, 171, 172, 185, 441 see also charges; fixed charges functional approach, secured transactions Cape Town Convention 379–380 fixed and floating charges 25 interests that can be included 43–44 Kenya 104–105 retention of title devices 41–43 single concept of security right 41 substance of reform 444–446 fungible assets/goods 136, 232, 257 future assets see after-acquired (future) assets G20 see Group of Twenty (G20) general notarial bonds 30, 361 creation 344 mixed jurisdictions 31 nature of 344–345 origin 343 post-default rights 345–347 registration 344 scope 343 South Africa 343–347, 348, 350, 353 Getting Credit Indicator (World Bank Doing Business Report) 3–4, 6, 9 and Burundi 239, 261 and Ethiopia 263 and Liberia 132 and Malawi 133 and Nigeria 141 reform 4–7t, 415, 416 and Rwanda 308, 310 and Zambia 167 see also Doing Business Report (World Bank)

Index 463 Ghana 65–90 applicable law 89 appointment of receiver/manager in 71, 72, 82, 85 authorisation for registration 73 Bank of Ghana (BoG) 65, 66 Borrowers and Lenders Act (B&L Act), 2008 65, 74–90, 450 Collateral Registry and Regulations 82–86 conflict of laws and transition 89–90 creation 77–78 enforcement 87–89 prior to 69–73 priority 87 scope 75–77 third-party effectiveness 78–82 capacity building 435 Collateral Registry 65, 68, 427 impact of 15–16 performance 86 and Regulations 82–86 searchability and indexing 85–86 users 19 commingled assets 87 currency 67 economy 17, 66–69 extrajudicial enforcement 453 floating charges 72, 81–82, 84, 87, 441 Industrial Sector Support Programme (ISSP) 67 micro, small and medium-sized enterprises (MSMEs) 68 non-consensual interests, treatment 446 non-performing loans (NPLs) 68, 69 pledge, customary 69 priority of security interests 87 Registrar-General’s Department 71, 72 registration fees 71, 72 registration rules 427 unsecured creditors 74, 75, 78 warehouse receipts 20 goods consumer 448 fungible 136, 232, 257 sale of 70–71, 99, 358 tangible 23 see also assets grace period 112, 125, 183, 202, 235 in Burundi 244, 254 in Ghana 87, 88 in Nigeria 158, 159 in Rwanda 316–317, 320 secured transactions law 53, 56 transitional 429, 430 grantors default 331, 332, 346, 380, 381, 387 grantor-based registration system 53

identifiers 52, 54, 108, 281, 451 representatives of 46, 49 Group of Twenty (G20) G20/OECD Effective Approaches for Implementing the G20/OECD High Level Principles on SME Financing 8, 417 SME Finance Action Plan (2015) 8 hire purchase Ghana 73 Kenya 97–99 Zambia 172–173 history of international instruments and models 11–13 hypothecation 131, 333, 334, 343, 349, 350, 357, 364 legal 350 Roman law 356 identifiers 207–208, 281, 368 errors 52, 54, 108, 177, 368 grantor 52, 54, 108, 281, 451 impact of a change of 53 legal persons 53, 281, 451 natural persons 53, 281, 451 searches/search criteria 53, 85, 108, 281, 451–452 immovable property, attachment to 426, 455 in Kenya 111–112, 117 in Liberia 124 in Rwanda 314 in Sierra Leone 205 in Tunisia 302 in Zambia 184 in Zimbabwe 363, 369, 371 see also fixtures; movable property infrastructure, modern secured transactions law 395–412 benefits of a reformed secured transactions regime 397–398 components of a modern regime 398–408 see also data on registry; registration of security interests; registry insolvency 6, 8–11, 16, 24, 31, 32, 116, 456 aviation sector 382–384 in Burundi 255 Cape Town Convention and Protocols 379, 380, 394 Mining, Agriculture and Construction (MAC) Protocol 387, 390 collective 255 in Ethiopia 265, 266, 274, 281 expenses 441 and floating charges 440–441 legal issues 31 legislation, common law jurisdictions 26 in Liberia 131 in Nigeria 146, 152

464 Index OHADA Uniform Act on Securities 213 rail sector 384 reforms 397, 399, 431, 440–441 representatives of grantor 46, 49 rights of secured creditor 419 in South Africa 333, 334, 342–348, 350–352 trustees 108, 251, 370 in Tunisia 287 in Zambia 115, 167, 168, 176 in Zimbabwe 356, 360, 362, 366, 370, 373 see also bankruptcy insolvency administrator 79, 379, 380 intangible assets, pledge of 220–225, 293–294 OHADA Uniform Act on Securities 220–225 Tunisia 293–294 intellectual property, pledge of 223 intermediated securities in Ethiopia 279 in Kenya 105, 117 OHADA Uniform Act on Securities 223, 235 in Sierra Leone 206, 208 in Zambia 175 International Finance Corporation (IFC) 16, 18 Global Leasing Toolkit 421 international instruments and models, history 11–13 inventories 388, 390, 401, 414 in Burundi 240, 243, 250, 251, 254, 257 defining 183 in Ethiopia 277, 278, 280, 284 financing 120, 179, 280, 351, 390 in Ghana 83, 85 in Kenya 108, 110, 112 legal issues 25, 28, 30 in Liberia 119, 120, 125, 128 MAC Protocol 390 in Malawi 134 in Nigeria 157 OHADA Uniform Act on Securities 220, 231 pledges 220, 243 and/or receivables 1, 2, 18, 32, 134, 224, 263, 414, 429, 435, 455 and reform 451, 452 in Rwanda 316, 324 secured transactions law 42, 47, 56 in Sierra Leone 197–198, 200, 208 in South Africa 334, 335, 336, 348, 349, 351 in Tunisia 285, 289, 290, 291, 302 in Zambia 183 in Zimbabwe 368 judgment creditors 57, 446 OHADA Uniform Act on Securities 235 secured transactions law 402, 404 Sierra Leone 206

South Africa 351 Zambia 175, 185 Zimbabwe 370, 371 judicial enforcement expeditious 454 extra-judicial 59 Tunisia 300–301 Kenya 91–118 applicable law 103, 114 appointment of receiver/manager in 113 authorisation for registration 109 bankruptcy 96, 116 capacity building 435 coordination within reform process 431 currency 93, 97 economy 17, 92–95 extrajudicial enforcement 452 floating charges 101, 102, 105, 115 identifiers and search criteria 452 intermediated securities 105, 117 interrelationship between the legislation and registry 430 Movable Property Security Rights Act 103–117, 432, 442 applicable law 114 creation 105–106 enforcement 113–114 functional approach 104–105 general provisions 114–117 priority 110–112 registration 107–110 rights and obligations of parties 112–113 scope 104–105 third-party effectiveness 107 third-party obligors 112–113 negotiable documents 107, 111 retention/transfer of ownership 99–100 registration as sole method 449 registration fees 98 security agreement 447 stakeholder engagement 432 transfers of receivables 100, 114 outright transfer 104, 105 unsecured creditors 101 warehouse receipts 20 Law on Security Rights and Leasing Law, Burundi (2016) 245–260, 438 conflict of laws rules 258–260 creation 246–248 enforcement of leases 258 of pledges 256–257 of retention of ownership sales 257–258

Index 465 priority rules 253–256 rights and obligations of parties 252–253 third-party effectiveness 248–252 attachment of a sign 250 dispossession 250 finance leases 251–252 pledge, making effective against third parties 250–251 registration 248–250 retention of ownership sales 251 transitional rules 258–260 leasing in Burundi 251–252 in Ethiopia 272–273 finance leases in Burundi 251–252 defining 20 in Ghana 73 in Tunisia 295 in Ghana 73 Global Leasing Toolkit (IFC) 421 lease agreements 148, 149, 380 in Nigeria 148–149 operating leases 401, 418, 421, 425 in Burundi 241 in Ethiopia 267, 272, 279, 284 long-term 22, 37, 123, 124, 131, 132, 189, 421 in Nigeria 160, 163 in Sierra Leone 197, 206 in Tunisia 292 in Zambia 174, 181, 189 priority of lessors 298 remedies of lessors 302 in Sierra Leone 194 in Tunisia 295, 298, 302 UNIDROIT Model Law on Leasing 421 see also lessees; lessors legal issues civil law concepts 27–30 common barriers to accessing secured credit after-acquired (future) assets, taking security over 32–33 anti-assignment clauses 31–32 common law concepts Bills of Sale regime 22–23 fixed and floating charges 23–26 registration regime 26–27 encumbered assets 25, 29, 31 mixed jurisdictions, concepts in 30–31 see also legal transplants Legal Rights Index, Doing Business Report (World Bank) see Doing Business Report (World Bank) legal systems 16–17, 264–266 legal transplant theory 13–15, 425–426

lending practices 409–410 lessees 104, 125, 208 as bailees 273 creation of security rights 273, 279 creditors 149, 273 default by 149, 258, 302, 350 effect of errors on serial numbers on 53, 108, 128, 368 failure to return or surrender asset 269, 302 and lease agreements 148, 149 operating 22, 181 ownership rights 20, 44, 73, 246, 248, 272, 273, 278 increments of ownership 279 proportional 272, 273 rights of 42, 47, 184, 185, 252, 253 option to purchase 20, 44, 73, 246, 272, 279, 292 ownership 20, 44, 73, 246, 248, 272, 273, 278 to possession 273, 295 use of asset 253, 258 see also leasing; lessors lessors duties 324 equipment 148, 149, 384 financial 148, 151, 181, 217, 279, 318, 323 immovable property 225, 233 and lease agreements 148, 149, 380 operating 152, 160, 185, 272, 273 ownership rights 104, 248, 252, 272, 273, 295, 298, 334, 421, 422 removal 278–279, 445 priority 149, 298 rights of 29, 73, 123, 126, 149, 245–246, 251–252, 324, 379 enforcement/remedies 252, 258, 269, 302 ownership 104, 248, 252, 272, 273, 279, 295, 298, 334, 421, 422 security 278 see also leasing; lessees Liberia 119–132 appointment of receiver/manager in 130 assignment of receivables 120, 123 authorisation for registration 128, 130 chattel paper 123, 125, 130 Commercial Code see Liberian Commercial Code commingled assets 125 consumer legislation 448 economic overview 119–121 extrajudicial enforcement 452 fixtures 123, 124, 131 identifiers and search criteria 452 insolvency 131 negotiable documents 124 Receivables Convention, ratifying 417 registration as sole method 450 registration fees 128–129

466 Index registration rules 427 security agreement 447 Liberian Commercial Code 121–122, 415–416 Chapter 5 122–131 creation and attachment 124 enforcement 130–131 perfection 124–125 post-implementation issues 129–130 priorities 125–126 registry development 126 fees, payment of 128–129 registration provisions 127–128 scope 122–124 licensees 56, 184, 208, 370, 424 liens Nigeria 144 OHADA Uniform Act on Securities 223–225 pledge of 223–224 liquidation 101, 112, 131, 207, 227, 410 liquidators 100, 103, 146, 193, 357 see also administrator; appointment of receiver/ manager; insolvency; receiver; winding up MAC Protocol see Mining, Agriculture and Construction (MAC) Protocol, Cape Town Convention Malawi 16, 33, 133–138 capacity building 436 commodity exchanges 137–138 negotiable documents 136 Warehouse Receipts Act (2018) 20, 134–137 manager, appointment of see appointment of receiver/manager market opportunity assessment 410–411 mass or product 48, 183, 319, 320 see also commingled assets maximum amount secured 109, 128, 199, 200, 229, 270, 324, 443 enforcement of security interests 46, 84, 105–106, 126, 156, 157, 176, 180, 247, 280, 446 notices 416, 417 security agreement 45, 47, 51, 109, 158, 325, 447 micro, small and medium-sized enterprises (MSMEs) access to credit, lack of 8 Ghana 68 microbusinesses/microenterprises 10, 18, 40, 95, 141, 169, 263, 455 microfinance 18–19 Mining, Agriculture and Construction (MAC) Protocol, Cape Town Convention 33–34, 385–394 application to MAC equipment 388–389 uniformity in legal framework 393–394

basic rules 386–387 and domestic law 390–391 immovable-associated equipment 389–390 importance in Africa 392–394 relevant sectors in Africa 392–393 specialised rules 387–390 treatment of inventory 390 mixed jurisdictions concepts in 30–31 see also Rwanda; South Africa; Zimbabwe Model Law (UNCITRAL Model Law on Secured Transactions) 15 anti-assignment clauses 32 deviation from approach 426–427 electronic warehouse receipts (EWRs) 420 Guide to Enactment 21–22 maximum amount concept 447 method of reform 427 simplified approach to reform 428 terminology 21 models deviations 425–427 simplified approach 428 structure 427–428 money commingled 369 as currency 310, 313 equivalents 25 mobile applications 93 purchase-money security interest (PMSIs) see purchase-money security interest (PMSIs) mortgages 24, 27, 131 business 28, 268, 270, 273–275 of capital assets 120 and charges 41, 145 chattel 96, 104, 363, 364 contractual 274 enterprise 28 equitable 145 goods 142 land 23, 119 motor vehicles, civil law 29 movable property attachment to Burundi 255, 258 Liberia 125–126 Rwanda 319 Sierra Leone 200 collateral, movable assets as 10 credit secured with movable assets, unwillingness to extend 10 MSMEs see micro, small and medium-sized enterprises (MSMEs)

Index 467 negotiable documents 13, 50, 368 and Ethiopia 280, 449 and Kenya 107, 111 and Liberia 124 and Malawi 136 and Sierra Leone 208 and Zambia 171, 184 negotiable instruments 13, 159, 164, 249, 316 examples 340 extrajudicial enforcement 59 free rules, taking 56 legal issues 25 priority 54, 87, 110, 114 versus registered notices 366 third-party effectiveness 49, 50 transferees 184, 370 New Zealand 25–26, 115, 425 Nigeria 139–164 alternative forms of enforcement 454 appointment of receiver/manager in 146, 153, 154, 157, 160 assignment of receivables 141, 146 bills of sale 142–143 Central Bank of Nigeria 149–151 charges 145–146 economy 17, 140–141 encumbered assets 153 Equipment Leasing Act (ELA) 139–140, 147–149, 163, 431 Equipment Leasing Association of Nigeria (ELAN) 139–140 expeditious judicial enforcement 454 floating charges 145–146, 153, 154 grace period 158, 159 leasing reforms 148–149 liens 144 non-consensual interests, treatment 446 operating leases in 160, 163 other security/quasi-security devices 146–147 pledges 143–144 pre-reform secured transactions framework 141–147 reforms 151–162 method 431 recent 147–151 registration fees 143 secured transactions law reform conflict of laws 161–162 creation 155–156 enforcement of security interests 160–161 miscellaneous matters 162 perfection 155–156 priority 158–159 registration 156–158

rights and obligations of parties 159–160 scope 151–155 security agreement 447 transfers of receivables 152, 160, 163 non-intermediated securities 13, 59 Ethiopia 315 Kenya 110–114 legal issues 22 Nigeria 162 OHADA Uniform Act on Securities 221 secured transactions law 49, 50, 60 Zambia 175 non-possessory pledges Burundi, non-possessory 242–244 enforcement rules 243–244 priority 243 third parties 242–243 see also pledges notarial bonds 423 effectiveness 30–31 execution 30 general see general notarial bonds mixed jurisdictions 30–31 special see special notarial bonds in Zimbabwe 356–361 see also South Africa; Zimbabwe notary public 30, 344, 351, 357, 359 notice filing 51–53 notice of default 87–88, 113, 114, 149, 160, 201, 208 notice of disposition 187 notice of registration 450–451 numerus clausus principle 27–28 obligations, secured 45–46 OHADA Uniform Act on Securities 213–238 assignment of receivables 214, 218, 221, 236, 238, 245 civil law 28, 29 conflict of laws 236–237 default 214, 233, 234 encumbered assets 216, 220, 221, 229, 231, 232, 234, 236 enforcement of security interests 233–235 fiduciary transfer of funds 219 fixtures 223 intermediated securities 223, 235 interrelationship between the legislation and registry 430 manner of registration 450 non-consensual interests, treatment 446 pledges 220–225 priority 231–232 receivable debtor 218, 219, 221, 226, 229, 232, 233, 236

468 Index registration of security rights in Register of Commerce and Securities 227–231 amendments, discharges and searches 230–231 general overview 227–228 initial requirements 228–230 reservation of ownership 217–218 retention, right of 216–217 rights and obligations of parties 232–233 scope 213–225 security transfer of receivables 218–219 substance of reform 445 third-party effectiveness 225–227 transfers of receivables 218–219, 231–233, 236 transition provisions 236–237 types of security rights 213–225 unsecured creditors 227, 235 operating leases 401, 418, 421, 425 in Burundi 241 in Ethiopia 267, 272, 279, 284 long-term 22, 37, 123, 124, 131, 132, 189, 421 in Nigeria 160, 163 in Sierra Leone 197, 206 in Tunisia 292 in Zambia 174, 181, 189 ordinary course of business assets, dealing with 39, 47, 56, 71, 101, 145–146, 254, 345 buyers, position of 25, 56, 87, 122, 125, 137, 159, 163, 201, 318, 400, 427 leases 125, 184, 318 Organisation for the Harmonisation of Business Law in Africa (OHADA) formulating agency, as 10 Uniform Act on Securities see OHADA Uniform Act on Securities outright transfer of a receivable 76, 95 in Ethiopia 276 in Kenya 104, 105 in Nigeria 151, 152, 160, 163 OHADA Uniform Act on Securities 236 secured transactions law 37, 41, 43, 44, 47 in Tunisia 291, 302 in Zambia 169, 181, 185 in Zimbabwe 360, 367, 372 see also receivables over-collateralisation 38, 47, 126 ownership Burundi 251, 257–258 Kenya 99–100 lessees, rights of 20, 44, 73, 246, 248, 272, 273, 278 increments of ownership 279 proportional ownership 272, 273

lessors, rights of 104, 248, 252, 272, 273, 295, 298, 334, 421, 422 removal 278–279, 445 reservation 217–218 retention Burundi 251 Kenya 99–100 security transfer of 99–100 transfer of 99–100, 216, 217, 225, 230, 248 parties, secured transactions law 40–41 party autonomy 36, 382, 383 protective provisions for certain parties 441–442 rights and obligations 57–58 Burundi 252–253 Kenya 112–113 Nigeria 159–160 OHADA Uniform Act on Securities 232–233 Zimbabwe 370 see also third-party effectiveness partnerships Incorporated Private Partnerships Act (1962), Ghana 71 pledge of rights 221–223 party autonomy 36, 382, 383 perfection of security interests bank accounts 183, 186 Liberia 124–125 Nigeria 155–156 versus third-party effectiveness 21 Zambia 181–182 see also third-party effectiveness pledges 220 bank accounts 221, 222, 225, 226, 247, 250, 256, 293, 298 in Burundi enforcement 256–257 making effective against third parties 250–251 non-possessory 242–244 competing possessory 297 creation 335–337 deemed to be 333, 335, 347–349 disposal of pledged asset 301–302 enforcement 256–257 enterprise 223–224, 294–295 equipment 294–295 in Ethiopia 270 of financial instruments 234 in Ghana 69 intangible assets 220–225, 293–294 intellectual property rights 223 lien of seller of an enterprise 223–224 nature of 335

Index 469 Nigeria 143–144 non-possessory in Burundi 242–244 enforcement rules 243–244 notarial bonds 31 priority 243 third parties 242–243 OHADA Uniform Act on Securities 220–224 partnership rights, securities and financial instruments 221–223 possessory see possessory pledges proceeds of pledged assets 249–250, 337 in South Africa creation 335–337 enforcement 337 extinguishment 337 nature of 335 third parties, making effective against 250–251 in Tunisia enterprise and equipment 294–295 intangible assets, over 293–294 possessory pledge 293 warehouse receipts 275–277 see also enterprise pledges possession/taking possession 55, 381, 387, 449, 453 of assets 56, 101, 130, 144, 250, 254, 297, 345, 347, 361, 373, 381, 387, 449, 453 encumbered 297, 373 pledged 144 tangible 49, 130, 250 in Burundi 250, 254, 256 of collateral 112, 156, 158, 183, 198, 280, 322, 346, 358, 367, 370, 371 constructive possession 30, 270, 280, 293, 335, 336 of documents 135, 136, 365 of equipment 42 in Ethiopia 280 of goods 44, 70, 184, 361 as interim measure 365 of invoices 291 in Kenya 101, 111 lessees, rights to 273, 295 in Liberia 130 in Malawi 136 of negotiable instruments 28 in Nigeria 144 before passing of title 71 retaining possession 28, 44, 233, 296, 300 right to take possession 346 in Rwanda 315, 319, 322 by secured creditors 31, 41, 49, 54, 453, 454 in Rwanda 315, 316, 323 in South Africa 332, 333, 346 in Zimbabwe 365, 366, 367, 371 see also secured creditors

secured transactions law 56 in South Africa 345, 346, 347, 348 transferring possession 30 in Tunisia 291, 297 of warrants 240 in Zambia 184 in Zimbabwe 361, 365, 369, 373, 381, 387 see also repossession possessory pledges Burundi 243–244 civil law 28 competing 297 enforcement 299–300 Tunisia 293, 297, 299–300 post-default remedies 98–99, 113, 130, 161, 179, 186 general notarial bonds 345–347 preferential claims/creditors 404, 441 in Ghana 72, 87 in Kenya 101, 115 and maximum amount concept 112 in Nigeria 154 priority of 25, 26, 57, 72, 101, 154, 207, 281, 441 in Sierra Leone 207 priority of security interests 39, 455 acquisition security rights 45, 55–56, 112, 254, 335, 370 basic rule 54–55 bills of lading, holders of 111 bona-fide purchasers 97 in Burundi 243, 253–256 charges 80, 87, 169, 170 fixed charges 192 floating charges 101, 102, 146, 192, 441 collateral 158 commingled assets 319 competing interests/assignments 57, 70, 206, 224, 317, 446 crops 279–280, 321 equipment 394 in Ethiopia 281–282 fixtures 320 floating charges 24 general notarial bonds 350, 360 in Ghana 87 intangible assets 114, 372 in Kenya 110–112 lessors 149, 298 in Liberia 125–126 licensees 370 liens 87, 225 in Malawi 135 mass or product 319 in Nigeria 158, 161 by non-companies 168

470 Index non-consensual interests, competing claimants with 57, 206, 224, 446 non-possessory pledges 243 OHADA Uniform Act on Securities 231–232 pledges 231, 243, 248, 251, 296 possessory pledges, competing 297 preferential claims 25, 26, 57, 72, 101, 154, 207, 281, 441 prior security rights 237 proceeds 158, 182, 200, 337 purchase-money security interest (PMSIs) 183, 184, 200 retention liens 87 in Rwanda 317–321 secured transactions law 39, 54–57 in Sierra Leone 200, 206 taking free rules 56 tangible assets 161, 372 time of agreement 298 time of registration 297–298 transferees 318 transition provisions 60–61 in Tunisia 295–298 competing possessory pledges 297 legitimate cause of preference 296–297 lessors 298 time of agreement 298 time of registration 297–298 warehouse receipts 135 in Zambia 168, 176, 182–185 in Zimbabwe 369–370 proceedings 206 civil 154 court 58, 59 dispute resolution 454 enforcement 58, 59, 244, 277 insolvency 57, 176, 255 judicial 39, 454 provisional sentence 365 rescue 334 proceeds arising of 317 collateral 124, 285, 316, 371 defining 47–48, 158, 198, 254, 316 disposition 60, 101, 158, 249, 322, 334 distribution 89, 114, 161, 235, 322, 454 encumbered assets 48, 254 enforcement 58, 59, 60, 235 identification 198, 249–250, 316, 369, 401 insurance 47, 158 money 48 non-cash 107, 125, 249, 316 of pledged assets 249–250, 337 priority of 158, 182, 200, 337 of sale 43, 158, 202, 253, 256, 334, 335, 350

security rights in 48, 78, 107, 125, 182, 255, 315, 316, 335, 368, 372 third-party effectiveness 316 on trust 147, 339 unperfected 125 property immovable see immovable property, attachment to movable see movable property prudential regulation 409 purchase-money security interest (PMSIs) 10, 25, 55, 87, 123, 321, 351 Nigeria 148, 151, 158, 160 rail sector, facilitating development in 384–385 railway rolling stock 377, 384–386, 388 reasonableness, commercial 36, 39, 57, 89, 111 receivable debtors 44, 260, 292, 315 OHADA Uniform Act on Securities 218, 219, 221, 226, 229, 232, 233, 236 secured credit, common barriers to accessing 31, 32 receivables account 123, 174, 263 assignment and anti-assignment clauses 31–32, 271, 315 Burundi 245, 248 Liberia 120, 123 Nigeria 141, 146 OHADA Uniform Act on Securities 214, 218, 221, 236, 238 Rwanda 312 Sierra Leone 197, 201 Tunisia 285 Zambia 175 and bank accounts 221 as circulating assets 32 collection of 114, 130, 186, 221, 234, 257, 299, 339, 365, 372 debtors of see receivable debtor financing 32, 99, 192, 229, 291, 339 Liberia 120, 122 Nigeria 146, 147 secured transactions law 41, 49 fixed charge over 219 future 218–219, 221, 270, 342 and/or inventories 1, 2, 18, 32, 134, 224, 263, 414, 429, 435, 455 multiple 270 outright transfer 76, 95 and Ethiopia 276 and Kenya 104, 105 and Nigeria 151, 152, 160, 163 OHADA Uniform Act on Securities 236 secured transactions law 37, 41, 43, 44, 47 and Tunisia 291, 302

Index 471 and Zambia 169, 181, 185 and Zimbabwe 360, 367, 372 see also receivables pledges over 247, 257, 357 OHADA Uniform Act on Securities 221, 231–233, 234, 236 Tunisia 293, 298 receivables creditors 31, 32, 260 sales 312 security rights in 25, 32, 58, 106, 113 Nigeria 150–151, 163 OHADA Uniform Act on Securities 224, 235 Receivables Convention (2001) 12, 341n161, 417 receiver administrative receiver 347 appointment of see appointment of receiver/ manager official receiver 96, 185 see also administrator reform of secured transactions law 8–15 amendments 437–439 background in Africa 15–21 benefits 397–398 capacity building 434–436 common law, interaction with 439–442 consumers, special rules for 447–448 content of security agreement 446–447 coordination within reform process 430–431 drivers for 413–422 economic growth 413–415 general reasons 413–422 global markets, appearance 415–417 infrastructure improvements 417 legal structure for specialist financial activity 418–422 state of unreformed law 422–424 formulating agencies 10–11 functional approach 444–446 hasty drafting 431–432 history of international instruments and models 11–13 intended beneficiaries (microbusinesses) 10 legal matters 22–33 leasing law 442 legal transplant theory 13–15 local legal culture and practices 433–434 legislative and drafting process, local 433 local conditions 434 method of reform 425–443 capacity building 434–436 interaction with regulatory bodies 436–437 interrelationship with other areas of law 437–443 local stakeholders, engaging 432–434 models, use of 425–428

statistical information for assessment 442–443 timing of process 428–432 models deviations 425–427 mixture of 425 simplified approach 428 structure 427–428 registration see registration of security interests regulatory bodies, interaction with 436–437 repeal, express and implied 437–439 scope of regime 443–446 stakeholder engagement 432–434 status 4–7t statutes (other), interaction with 439–442 substance of 443–454 techniques 437–439 terminology 21–22 third-party effectiveness see third-party effectiveness timing of reform process 428–432 coordination within reform process 430–431 hasty drafting 431–432 interrelationship between the legislation and registry 429–430 treatment of non-consensual interests 446 see also unreformed jurisdictions registration of security interests 50–54 access and availability 406 administrative processes 405–406 advance 84, 230, 443 authorisation for 453 Burundi 249 Ghana 73 Kenya 109 Liberia 128, 130 Rwanda 321, 323, 325 unauthorised registration 52, 187, 407 Burundi 248–250 challenges in establishing registries 403–404 data integrity 406–411 document filing 450–451 effective registry, key elements 404–408 Ethiopia 280–281 fees see fees, registration general notarial bonds 344 grantor-based system 53 incentivising registration 399–400 incorrect data placed on registry due to human error 407 Kenya 107–110 Liberia development of registry 126 fees, payment of 126–129 registration provisions 127–128

472 Index in movable property 323–326 Nigeria 149–151 notice filing 51–53 notice of registration 450–451 operating model 405 regime see registration regime registration regime 26–27 administrative role of the registry 451 Bills of Sale Acts regime 23 common law 26–27 Companies Act 439–440 identifiers and search criteria 451–452 manner of registration 450–451 role of international registries under Cape Town Convention 380 Rwanda 323–326 searches see searches, registration South Africa 344 third-party effectiveness, achieving 449–450 time of 297–298 Tunisia 290, 297–298 user-centricity 404–405 variations in approach to African system 402–403 Zimbabwe 359–362 registry administrative role 451 components of a modern regime 402–408 duplicate, out-of-date or redundant data to remain on, allowing 408 Ethiopia 280–281 fees, Liberia 128–129 Ghana’s Collateral Registry 15–16, 82–86 incorrect or inappropriate data placed on 407–408 interrelationship with legislation 429–430 Liberia 126 specialised 275, 333 Zambian Collateral Registry and Office 176–180 see also registration of security interests regulatory bodies, interaction with 436–437 repossession 73, 88, 98, 131, 149, 160–161, 173 authorisation 321, 323 of collateral 160, 186, 202, 321, 371 encumbered assets 269, 366, 372, 454 extrajudicial 160, 269 intended 113–114, 323 opposition to 332 peaceable 161, 208 right to repossess 202, 324 tangible assets 256, 299 see also possession reservation of title devices 41–43 Ethiopia 271–272 notarial bonds 31

OHADA Uniform Act on Securities 217–218 sales 251, 257–258 South Africa 334–335 retention of security 216–217 reverse factoring, supply chain finance 20 right of redemption 322, 371 rights and obligations of parties 57–58 in Burundi 252–253 in Kenya 112–113 in Nigeria 159–160 OHADA Uniform Act on Securities 232–233 in Zimbabwe 370 Roman-Dutch law jurisdictions 30, 31 Rwanda 307–327 assignment of receivables 312 authorisation for registration 321, 323, 325 chattel paper 313 commingled assets 319–320 country/economic overview 17, 307–309 currency 310, 313 encumbered assets 325 extrajudicial enforcement 453 fixtures 320 grace period 316–317, 320 movable property, law on security interests 312–326 creation 313–315 enforcement of security interests 321–323 priority 317–321 Register of security interests in movable property 323–326 scope 312–313 third-party effectiveness 315–317 registration fees 326 secured transactions legal framework 309–312 security agreement 447 simplified approach to reform 428 taking possession in 315, 316, 323 sale bills of sale 23, 27, 141–143, 146, 171 contract of 42, 246, 248, 272, 274, 313, 358 of goods 70–71, 99, 358 power of 69, 144 proceeds of 43, 158, 202, 253, 256, 334, 335, 350 retention of ownership 251 searches, registration 54 criteria 451–452 Ghana 85–86 incentivising searching 400–401 Register of Commerce and Securities (OHADA) 230–231 see also registration of security interests secured creditors 125, 234 bank accounts, instructions regarding 50

Index 473 first 45, 55, 128 junior 42, 43, 45, 47, 172, 273, 299, 321, 371 legal issues 25–28 rights, in insolvency law 419 secured transactions law 30–33, 38, 39, 41, 46, 48–52, 54, 55 senior 60, 180, 299, 321, 322, 371, 417, 453 subordinate 322n101 taking possession by 49, 453, 454 in Rwanda 315, 316, 323 in South Africa 332, 333, 346 in Zimbabwe 365, 366, 367, 371 see also possession; preferential claims; priority of security interests; taking possession; unsecured creditors secured transactions law 35–62 assets see assets components of a modern regime 398–408 credit infrastructure 408–411 legal framework 398–402 registry 402–408 conflict of laws 60 creation see creation of security rights defining 8 enforcement see enforcement of security interests functional approach 41–44 grace period 53, 56 infrastructure underpinning 34, 395–412 credit 408–411 legal 401–402 modern principles 36–40 obligations 45–46 parties 40–41 priority see priority reform see reform of secured transactions law registration see registration rights and obligations of parties 57–58 scope 36–37, 40–46 third-party effectiveness 38, 49–50 transfers of receivables, outright transfer 37, 41, 43, 44, 47 transitional provisions 60–61 see also specific countries securities advances on 271 intermediated Cape Town Convention and Protocols 420–421 Ethiopia 279 Kenya 105, 117 OHADA Uniform Act on Securities 223, 235 secured transactions law 45 Sierra Leone 206, 208 Zambia 175

non-intermediated 22 Ethiopia 315 Kenya 110–114 Nigeria 162 OHADA Uniform Act on Securities 221 secured transactions law 49, 50, 59, 60 Zambia 175 pledge of 221–223, 271 see also secured transactions law; security agreement; security devices; security interests/rights security agreement anti-assignment clauses 31–32, 447 cession, prohibiting 340–341 content 446–447 creation of security rights 46–47 description of assets in 423 maximum amount secured 45, 47, 51, 109, 158, 325, 447 secured obligation 446–447 time of 298 security devices civil law 28, 29 Ethiopia 269–277 Nigeria 146–147 non-possessory 28 quasi-security devices 146–147 quasi-security devices, Nigeria 146–147 Tunisia 290–292 types 423 unreformed law, complexity and fragmentation 423 Zimbabwe, pre-MPSI 357–359 security interests/rights attachment 46, 112, 180, 314 creation see creation of security rights distinctions becoming irrelevant 25, 117, 440 enforcement see enforcement of security interests in proceeds 48, 78, 107, 125, 182, 255, 315, 316, 335, 368, 372 purchase-money see purchase-money security interest (PMSIs) registration see registration of security interests scope 47–48 single concept 41 security transfer of ownership 216, 225, 230 serial numbers 51, 53, 127 errors in 108, 128, 177, 199, 208, 368 Sierra Leone 191–209 alternative forms of enforcement 454 appointment of receiver/manager in 202, 208 assignment of receivables 197, 201 bankruptcy 208

474 Index Borrowers and Lenders Act (B&L Act), 2014 195–208 approach and models 195–196 creation and perfection 198–200 enforcement 201–203 issues 196–198 priority rules 200–201 Regulations 205 transition 203–204 chattel paper 201 commingled assets 198 Companies Act 192–193 Corporate Affairs Commission 193 economic/financing overview 191–192 expeditious judicial enforcement 454 Fast Track Commercial Loan Court 202–203, 454 fixtures 205 floating charges 192, 193, 204, 207 future steps 208–209 interaction with regulatory bodies 437 intermediated securities 206, 208 interrelationship between the legislation and registry 430 judgment creditors 206 Leasing Guidelines 194 legal framework affecting secured transactions 192–195 negotiable documents 208 non-consensual interests, treatment 446 operating leases in 197, 206 security agreement 447 unsecured creditors 192 small and medium enterprises (SMEs) 8 South Africa 328–352 anti-assignment clauses 447 applicable law 320 background to legal framework 330–332 bankruptcy 332 cession, security agreements prohibiting 340–341 creation 339–340 enforcement 342 future rights 342 nature of 338–339 economy 17 encumbered assets 343 floating charges 333, 343 general notarial bonds 343–347, 348, 350, 353 judgment creditors 351 movable assets, security rights in 332–350 cession 338–342 general notarial bonds 343–347 pledges 335–337 reservation of title devices 334–335 special notarial bonds 347–350

pledges creation 335–337 enforcement 337 extinguishment 337 nature of 335 reservation of title devices 334–335 security rights in movable assets 332–350 taking possession in 332, 333, 346 unsecured creditors 333, 335, 337, 344, 345–346 see also Cape Town Convention and Protocols special notarial bonds mixed jurisdictions 30–31 South Africa 347–350 specialised registry 275, 333 stamp duty 30, 94, 98, 100, 106, 115 standard setters, international 10, 14 State-building, and legal transplants 14 statistical information 414, 442–443 Strength of Legal Rights Index 9, 415, 416 subordination agreement 172, 200, 204, 318 supply chain finance 20–21 surplus, and enforcement of security interests 59–60 tangible assets encumbered 250, 256, 372 mixed jurisdictions 30 notarial bonds 31 pledge of 220 possession 49 repossession 256, 299 terminology 21–22 third parties of non-possessory pledges 242–243 see also third-party effectiveness third-party effectiveness 38, 49–50 attachment of a sign 250 Borrowers and Lenders Act, Ghana (2008) 78–82 civil law 28 dispossession 250 Ethiopia 280 finance leases 251–252 Law on Security Rights and Leasing Law, Burundi (2016) 248–252 methods for achieving 448–450 Movable Property Security Rights Act, Kenya 107 OHADA Uniform Act on Securities 225–227 versus perfection 21 pledge, making effective against third parties 250–251 proceeds 316 registration 248–250 retention of ownership sales 251 Rwanda 315–317 Tunisia 292–295

Index 475 third-party obligors Kenya 112–113 Zimbabwe 370 timing of reform process 428–432 coordination within reform process 430–431 hasty drafting 431–432 interrelationship between the legislation and registry 429–430 trade receivables 32, 43, 48–49, 181, 218, 341 transfer of ownership 99–100, 217, 248 security 216, 225, 230 transition provisions 60–61 and conflict of laws Burundi 258–260 Ghana 89–90 OHADA Uniform Act on Securities 236–237 Ethiopia 283 ‘transplant effect’ 14 trustee in bankruptcy 96, 208 trustees 227, 237 insolvency 108, 251, 370 security rights 337, 342 Tunisia 16, 285–303 appointment of receiver/manager in 300 assignment of receivables 285 bankruptcy 296 creation 292–295 Ease of Doing Business (EODB) 286 economic overview 286–288 encumbered assets 285, 289, 293, 295, 297, 299, 301, 302 enforcement of security interests 298–302 disposal of pledged asset 301–302 lessors’ remedies 302 possessory pledge 299–300 provisional judicial remedies 300–301 finance leases 295 insolvency 287 operating leases in 292 pledges competing possessory 297 enterprise and equipment 294–295 intangible assets, over 293–294 possessory 293, 297 priority of security interests 295–298 competing possessory pledges 297 legitimate cause of preference 296–297 lessors 298 time of agreement 298 time of registration 297–298 secured transactions framework 288–292 multiplicity and overlap of laws 289 registration systems 290 security devices 290–292

third-party effectiveness 292–295 unsecured creditors 296 UNCITRAL (United Nations Commission on International Trade Law) Convention on the Assignment of Receivables in International Trade 32 formulating agency, as 10, 11 Legislative Guide on Secured Transactions 12, 13, 21, 421, 425, 433 Model Law see Model Law (UNCITRAL Model Law on Secured Transactions) Practice Guide to the Model Law 435 UNIDROIT (International Institute for the Unification of Private Law) 10 Convention on Substantive Rules for Intermediated Securities (Geneva Securities Convention) 420–421 Model Law on Leasing 421 United Nations Commission on International Trade Law (UNCITRAL) see UNCITRAL (United Nations Commission on International Trade Law) unreformed jurisdictions after-acquired (future) assets, taking security over 33 collateral, limits in what can be used for 423–424 complexity and fragmentation 10, 11, 422–423 excessive formalities for creation 424 extrajudicial enforcement, lack of 424 lack of functionality 423–424 parties 422–423 publicity, lack of 424 security devices 423 unsecured creditors 31, 49 Burundi 244 Ghana 74, 75, 78 Kenya 101 OHADA Uniform Act on Securities 227, 235 Sierra Leone 192 South Africa 333, 335, 337, 344, 345–346 Tunisia 296 see also secured creditors warehouse receipts electronic see electronic warehouse receipts (EWRs) Ethiopia 275–277 financing of 19–20 Malawi 134–137 pledges 275–277 winding up 154, 192 World Bank Doing Business index see Doing Business Report (World Bank)

476 Index Principles for Effective Insolvency and Creditor/ Debtor Systems 10, 419 and Zimbabwe 431 World Bank Group 14–15, 419, 443 Zambia 16, 165–190, 416 applicable law 175 appointment of receiver/manager in 185 assignment of receivables 175 bankruptcy 174, 176 capacity building 436 Collateral Registry and Office 176–178 commingled assets 182–183 consumer legislation 448 coordination within reform process 431 District Farmers Association (DFAs) 19 encumbered assets 444 extrajudicial enforcement 452 fixtures 184 floating charges 168, 170, 171, 172, 185, 441 intermediated securities 175 interrelationship between the legislation and registry 429 interrelationship with other areas of law 438 judgment creditors 175, 185 Movable Property (Security Interest) Act (MPSIA) background to 167–169 Collateral Registry and Office 176–178 creation 180–181 enforcement 185–187 framework prior to 169–173 general provisions and offenses 187–188 implementation challenges 188–189 perfection 181–182 preliminary provisions 174–176 priority 182–185 Registry Regulations 178–180 structure 173–188 negotiable documents 171, 184 non-consensual interests, treatment 446 operating leases in 174, 181, 189

parties, secured transactions law 422–423 priority of security interests 182–185 registration as sole method 449–450 registration fees 177 Registry Regulations 178–180 security agreement 447 stakeholder engagement 432 substance of reform 444 transfers of receivables 181, 185 Zimbabwe 16, 353–373 applicable law 363, 369, 372 chattel paper 364 commingled assets 369 currency 354, 361 encumbered assets 357, 362, 367, 371–373 fixtures 363 insolvency 360, 362, 366, 370, 373 interrelationship between the legislation and registry 430 interrelationship with other areas of law 438 judgment creditors 370, 371 manner of registration 450 Movable Property Security Interests Act (MPSI), 2017 creation and registration requirements prior to 359–362 First Schedule 367–368 framework prior to 356–362 main body 363–367 Second Schedule 369–372, 427 security devices prior to 357–359, 362–372 notarial bonds 356–361 priority of security interests 369–370 Second Schedule of MPSI (2017) 369–372, 427 Part II (Special Creation Provisions) 369 Part III (Priorities) 369–370 Part IV (Rights and Obligations of Third Parties/ Third-Party Obligors 370 Part V (Enforcement of Security Interests) 371–372 Part VI (Applicable Law) 372 taking possession in 365, 366, 367, 371