Economic Analysis of the DCFR : The Work of the Economic Impact Group Within CoPECL [1 ed.] 9783866538559

Citation preview

Economic Analysis of the DCFR

Economic Analysis of the DCFR The work of the Economic Impact Group within CoPECL

edited by

Pierre Larouche Filomena Chirico

ISBN (print) 978-3-86653-113-0 ISBN (e-book) 978-3-86653-855-9

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de. © 2010 by sellier. european law publishers GmbH, Munich. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior permission of the publisher. Design: Sandra Sellier, Munich. Production: Karina Hack, Munich. Typeface: Goudy Old Style and Goudy Sans from Linotype. Printing and binding: Friedrich Pustet KG, Regensburg. Printed on acidfree, non-ageing paper. Printed in Germany.

Foreword This book is the outcome of the work of the Economic Impact Group (EIG) created as part of the CoPECL (Common Principles of European Contract Law), a Network of Excellence funded the EU’s FP6.1 Pursuant to the CoPECL funding agreement, the EIG was tasked with “assess[ing] the economic impact of the results achieved by the joint research of the network”2 and providing “a broad assessment of the economic implications of the rules taking into consideration the needs of the economic operators in the internal market”.3 The Tilburg Law and Economics Center (TILEC) was entrusted with the management of that workpackage, under the leadership of its directors Eric van Damme and Pierre Larouche. When the work began in 2005, a number of points became immediately clear. First of all, the timeline for the work of the EIG would be heavily dependent on the progress made in the drafting groups of CoPECL, namely the Study Group and the Acquis Group. Indeed given the EIG’s mission, it is preferable to work on the basis of drafts from those groups as opposed to dealing with the Europeanization of contract law in abstracto. Secondly, the size and scope of the Draft Common Frame of Reference (DCFR) would make it impossible to carry out an economic analysis of all its provisions. It would be more effective to focus the work on a number of key elements of contract law covered in the DCFR (and a few key elements outside of contract law, when it became clear that the DCFR would extend beyond contract law). Thirdly, the workpackage would be best carried out – both as regards feasibility and quality – by forming a network of European law and economics experts administered and supported by TILEC. The EIG network was formed in 2006. It brought together well-established scholars and young talents. Elements of interest in the DCFR were identified, and a number of EIG members volunteered to contribute on one or more of these elements. The members of the EIG are (in alphabetical order, contributors being indicated with an asterisk): Prof. Roger van den Bergh, Erasmus University Rotterdam Dr. Rogier van Bijnen, De Brauw Blackstone Westbroeck Prof. Eric Brousseau, University of Paris X Dr. Filomena Chirico, European Commission and Tilburg University* Dr. Assunção Cristas, Universidade Nova de Lisboa* Dr. Katalin Cseres, University of Amsterdam* Prof. Eric van Damme, Tilburg University* Prof. Winand Emons, University of Bern Prof. Michael Faure, Erasmus University Rotterdam Prof. Nuno Garoupa, University of Illinois at Urbana-Champaign*

1

2 3

Joint Network on European Private Law (CoPECL: Common Principles of European Contract Law), 6th EU Framework Programme for Research and Technological Development (FP6), Priority 7 – FP6-2002-CITIZENS-3, Contract no. 513351. Ibid., Article 4.1, p. 21. Ibid., Article 6.2.7, p. 33.

VI

Foreword

Prof. Gerrit de Geest, Washington University in St. Louis* Prof. Fernando Gomez, Universitat Pompeu Fabra (Barcelona)* Dr. Robert Hardy, Stibbe* Dr. Geerte Hesen, Maastricht University* Prof. Martijn Hesselink, University of Amsterdam Dr. Tomas Kontautas, Vilnius University* Prof. Mitja Kovač, University of Ljubljana* Prof. Pierre Larouche, Tilburg University* Prof. Patrick Leyens, University of Hamburg* Hanneke Luth, LLM MSc, Erasmus University Rotterdam* Prof. Anthony Ogus, Erasmus University Rotterdam and University of Manchester* Prof. Hans-Bernd Schäfer, Bucerius Law School (Hamburg)* Prof. Urs Schweizer, Universität Bonn* Prof. Jan Smits, Tilburg University Dr. Ann-Sophie Vandenberghe, Erasmus University Rotterdam* The EIG organized its work as follows. Each contribution was meant to be presented to the group and discussed in plenary. It was also refereed by at least two anonymous members of the group (single-blind review). Contributors were also free to present their work in other fora, in order to gather more comments, which many did. The aim was to create a collegial atmosphere and to produce a set of high-quality contributions which already withstood peer review. The EIG held three plenary meetings; at each meeting, a number of contributions were discussed. The first meeting was held in Brussels at the Fondation Universitaire / Universitaire Stichting in April 2007. The second meeting was held in Barcelona in December 2007; the EIG is grateful to Professor Gomez for hosting this meeting at Universitat Pompeu Fabra. The third meeting was held in Venice at Venice International University in June 2008. A conference to be held in Brussels in November 2009 will present the outcome of the work of the EIG to a broader audience. In closing, in the name of the EIG, I wish to thank all of the individuals and institutions who made our work and this book possible. Funding for the operations and activities of the EIG was received from the European Union, via the 6th Framework Programme. Prof. Maurits Barendrecht of Tilburg University played a role in launching the idea of the EIG when the CoPECL Network of Excellence was formed. Prof. Hans Schulte-Nölke and Dr. Christoph Busch ensured the link with the rest of the CoPECL Network of Excellence. The day-today business of the EIG was coordinated by Filomena Chirico during her time in Tilburg, from 2005 to 2008, and the EIG is very grateful for all her dedication in organizing meetings and ensuring that the work of the EIG progressed along schedule. Angela Maria Noguera provided valuable assistance in the final phases of the project. The TILEC secretariat took care of administrative matters and reporting. sellier. european law publishers undertook the editing and production of the book. Our thanks to all. Pierre Larouche

Table of contents Foreword List of contributors

Introduction Filomena Chirico & Pierre Larouche

V IX

1

1. Contract Law – General Aspects The Function of European Contract Law: An Economic Analysis Filomena Chirico

9

The Economic Function of Good Faith in European Contract Law Filomena Chirico

31

Non-Discrimination in the Common Principles of European Contract Law Ann-Sophie Vandenberghe

45

2. Formation and Interpretation of Contracts The Formation of Contracts in the Draft Common Frame of Reference – A Law and Economics Perspective Gerrit De Geest & Mitja Kovač

67

Contract Interpretation – Interpretive Criteria Geerte Hesen & Robert Hardy

83

Judicial Control of Standard Terms and European Private Law Hans-Bernd Schäfer & Patrick Leyens

97

3. Performance of Contracts Specific Performance, Damages and Unforeseen Contingencies in the Draft Common Frame of Reference Gerrit De Geest

123

VIII

Table of contents

Measure of Damages: Expectation, Reliance and Opportunity Cost Anthony Ogus Remedies for Non-Performance:Chapter 3 of Book III of the DCFR from an Economist’s Perspective Urs Schweizer

133

147

4. Termination of Contracts Termination and Compensation in Long-Term Distribution Contracts: An Economic Perspective of EU Law Fernando Gomez

171

Variation and Unilateral Termination of an Agreement in the Draft Common Frame of Reference Mitja Kovač

203

5. Specific Areas of Contract Law Principles of European Insurance Contract Law: Law and Economic Insights Tomas Kontautas The DCFR and Consumer Protection: An Economic Assessment Hanneke A. Luth &Katalin Cseres

227

235

6. Non-contractual Liability The Boundary between Torts and Contracts: A Law and Economics Perspective Assunção Cristas & Nuno Garoupa ‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR Pierre Larouche

Conclusion Filomena Chirico, Eric van Damme &Pierre Larouche

277

295

319

List of contributors The following members of the EIG have authored or co-authored one or more of the contributions in this book. They are listed in alphabetical order. Filomena Chirico Filomena Chirico is an official at the Directorate General for Competition of the European Commission. She holds a law degree from LUISS University in Rome (1999), a European Masters in Law and Economics (2002) and a Ph.D. in Economics and Institutions from the University of Rome La Sapienza (2005). She was the first assistant in the European Law and Economic Analysis programme at the College of Europe. At the time she contributed to this project, she was Assistant Professor and senior member of the Tilburg Law and Economics Center (TILEC), Tilburg University, where she coordinated the Economic Impact Group (EIG). Assunção Cristas Assunção Cristas is Associate Professor of private law at the Faculty of Law of Universidade Nova de Lisboa. Between 2002 and 2005 she was General-Director at the Ministry of Justice of Portugal. She graduated from the University of Lisbon (1997) and she obtained her Ph.D. at the Universidade Nova de Lisboa (2005). She was a visiting researcher in the Max-Planck Institute for Private and Comparative Law (Hamburg), the Institute for Advanced Legal Studies (London), the University of Rome La Sapienza and Columbia University. Her teaching and research interests, generally on contract law, law of obligations and consumer law, include particularly the transmission of contractual claims and of property rights. In 2009 she was elected Member of the Portuguese Parliament. Katalin Cseres Katalin Cseres obtained her university degree in law (1999) at the Eötvös Loránd University of Budapest, Hungary, and her Ph.D. at the University of Utrecht (2004), with a thesis on ‘Competition Law and Consumer Protection’. She moved to the Law Faculty of the University of Amsterdam in 2003, where she now is associate professor of law. She is a research fellow at the interdisciplinary Amsterdam Center for Law & Economics (ACLE). Her main fields of interest are competition law, consumer protection, comparative law and economics and institutional economics. Eric van Damme Eric van Damme is Professor of Economics at the CentER for Economic Research, Tilburg University, and co-Director of TILEC (Tilburg Law and Economics Center). He studied mathematics at the Katholieke Universiteit Nijmegen, and received his Ph.D. from the Technische Universiteit Eindhoven. He was affiliated with the Technical University of Delft (The Netherlands), the Kellogg School of Management at Northwestern University (Evanston, USA) and the Universität Bonn. His research interests involve game theory, economic theory, competition policy and regulation, experimental economics, bounded ra-

X

List of contributors

tionality and bargaining. Eric van Damme is a fellow of the Econometric Society. He has been a member of the Council of the European Economic Association. From 1999 to 2008, he held the position of secretary and treasurer of the Game Theory Society. Nuno Garoupa Nuno Garoupa joined the University of Illinois at Urbana-Champaign in 2007 as Professor of Law after serving as Professor of Law and Economics at Universidade Nova de Lisboa, and as Associate Professor of Economics at Universitat Pompeu Fabra (Barcelona). He received his Ph.D. in Economics from the University of York (UK) and also holds an LL.M. from the University of London with specialization in criminal justice and criminology. His research interests lie in the economics of comparative law and legal institutions as well as empirical legal studies. He is a member of the editorial board of the International Review of Law and Economics, and the co-editor of the Review of Law and Economics. Gerrit de Geest Gerrit De Geest became Professor of Law and Co-Director of the Center for Law, Innovation & Economic Growth at Washington University in St. Louis in 2007, after having been Professor of Law and Economics at Utrecht University. He holds degrees in law (1983) and economics (1986), as well as a Ph.D. (1993) from Ghent University. He is a past president of the European Association of Law and Economics. He is series editor of the forthcoming 2nd edition of the Encyclopedia of Law and Economics and the co-editor of the New Horizons in Law and Economics book series (Edward Elgar). He specializes in the law and economics of contract and tort law and of comparative law. Fernando Gómez Fernando Gómez obtained his law degree from the Universidad Complutense de Madrid (1985) and his Ph.D. from the Real Colegio de España in Bologna, Italy (1988). He is Professor of Law at the Universitat Pompeu Fabra in Barcelona and visits regularly at European and American universities. A member of the managing board of the European Association of Law and Economics (2001-2007) and the Society for European Contract Law (2004), he serves on the editorial board of the International Review of Law and Economics, the Review of Law and Economics, the European Review of Contract Law, and is co-editor of InDret, the first e-legal journal in Spanish. His research interests are in the area of economic analysis of private law, rules and institutions. Robert Hardy Robert Hardy studied law at Maastricht University, the University of Cambridge, the Université de Paris I Panthéon-Sorbonne, and Harvard Law School, where he obtained his master’s degree in 2007. He defended his Ph.D. at Maastricht University in 2009. In 2005, he was a visiting scholar at the University of California Berkeley (Boalt Hall). He is now working in Stibbe’s Competition and Regulation Department.

List of contributors

XI

Geerte Hesen Geerte Hesen holds degrees in Economics (2004) and Law (2005) from Maastricht University, where she obtained her Ph.D. in 2009. She also studied at the Université de Paris I Panthéon-Sorbonne, and was a visiting scholar at the University of California Berkeley (Boalt Hall), Columbia University and Stanford University. She was President of the European Law Students’ Association. She is now an associate at De Brauw Blackstone Westbroek (Amsterdam) and Fellow at Ticom. Tomas Kontautas Tomas Kontautas is attached to the Faculty of Law of Vilnius University, and he is a partner at the Lithuanian law firm Sorainen. He holds a law degree and a Ph.D. from Vilnius University, as well as a European Masters in Law and Economics (Hamburg and Rotterdam). He was involved in drafting insurance laws and regulations in Lithuania and the EU. His practice and his research interests centre on insurance and banking law. Mitja Kovač Mitja Kovač graduated in law from the University of Ljubljana and went on to obtain an LL.M. and a Ph.D. in the area of comparative contract law and economics from Utrecht University. He is now assistant professor at the Faculty of Economics of the University of Ljubljana. His areas of research are comparative contract law and economics, consumer protection and contract theory. Pierre Larouche Pierre Larouche is Professor of Competition Law at Tilburg University and Co-Director of the Tilburg Law and Economics Center (TILEC). He is also a professor at the College of Europe (Bruges). He graduated from the Faculty of Law of McGill University (Montreal, 1990). He clerked at the Supreme Court of Canada. He obtained his masters degree from the Universität Bonn (1993) and his Ph.D. from Maastricht University (2000), following three years in practice with Stibbe. His teaching and research interests include competition law and economic regulation, telecommunications law, media law, basic EU law and the common European law of torts. He was one of the chief editors of the journal “Competition and Regulation in Network Industries”. Patrick Leyens Patrick C. Leyens is Junior Professor at the Institute of Law and Economics, University of Hamburg. He holds an LL.M. from the University of London (2000) and a Ph.D. from the University of Hamburg (2006). He was a senior research fellow at the Max-Planck Institute for Comparative and International Private Law, Hamburg. He is now the Director of the Erasmus Mundus European Doctorate in Law and Economics for the University of Hamburg. His main areas of research include German and European contract, corporate and capital market law with a focus on comparative law and law and economics.

XII

List of contributors

Hanneke Luth Hanneke Luth graduated from the Erasmus University Rotterdam with degrees in both Business Economics and Law. She also obtained the European Master in Law and Economics. Since 2005, she has been a member of the Rotterdam Institute for Law and Economics (RILE), where she is pursuing a Ph.D. in the European Doctorate in Law and Economics programme. Her work concerns the economic analysis of consumer protection regulation, and in particular standard terms in consumer contracts. Anthony Ogus Anthony Ogus is Emeritus Professor at the School of Law of the University of Manchester (where he was Dean 1990-92); he is also part-time Erasmus Professor of the Fundamentals of Private Law at the University of Rotterdam. He has held visiting positions at the Universities of Antwerp, California (Berkeley), Maastricht, Paris II and Toronto, and the Bucerius Law School at Hamburg. He has written books and articles on the law of damages, social security law, law and economics and regulation. He was joint founding editor of the International Review for Law and Economics. In 2002 he was nominated Commander of the British Empire for his services to the Social Security Advisory Committee. He is a Fellow of the British Academy. Hans-Bernd Schäfer Hans-Bernd Schäfer is Professor of Law and Economics at Bucerius Law School in Hamburg (Germany). He studied economics and business administration at the University of Cologne and holds a Ph.D. in economics from the University of Bochum. He previously taught at the University of Bochum and the University of Hamburg. He was Director of the Institute of Law and Economics, the Ph.D. programme in law and economics and the European Master Program in Law and Economis (EMLE) at the University of Hamburg. A former president of the European Association for Law and Economics, his research interests lie in the economics of the civil law and in law and development. Urs Schweizer Urs Schweizer is Professor of Economics at the Universität Bonn. His research interests cover different fields of microeconomic theory. In recent years, he mainly published on contract and tort law from a game theoretic perspective. He is on the board of CASTLE, the newly created centre for advanced studies in law and economics at the University of Bonn and he is the managing director of the Bonn Graduate School of Economics (BGSE) as well as the collaborative research centre on governance and the efficiency of economic systems (SFB TR 15), a joint endeavour of Universities in Berlin, Bonn, Mannheim and Munich. Ann-Sophie Vandenberghe Ann-Sophie Vandenberghe is Assistant Professor of Law and Economics at the Erasmus University Rotterdam. She holds degrees in law (Ghent University, 1996) and education (Ghent University, 1996), as well as a European Master in Law and Economics (1997) and a PhD (Utrecht University, 2004). She was a visiting scholar at Columbia Law School and a visiting professor at National Law School of India University. She specializes in comparative and behavioural law and economics of contracts, especially employment and consumer contracts.

Introduction Filomena Chirico* Pierre Larouche** The publication of the Draft Common Frame of Reference (DCFR) at the end of 20081 is the latest step in the ongoing creation (or for many, rebirth) of a European private law. Along the way, what started as an academic project for a core group of comparative law and private law scholars in the 1980s2 blossomed into a large academic enterprise. In addition to the Study Group and the Acquis Group, which prepared the DCFR, the broader academic community has joined the discussion. Policymakers have taken up the idea of European private law, with the European Parliament in particular adopting since 1989 a series of resolutions at various stages of the discussion.3 The European Commission followed suit with a series of documents in the early 2000s, leading to the Communication on European Contract Law and the revision of the acquis in 2004, on the basis of which the DCFR was commissioned.4 The Council has also been active in the policy discussions.5

*

** 1

2

3

4

5

Official at the Directorate General for Competition of the European Commission – At the time she contributed to this project, she was Assistant Professor and Senior Member, Tilburg Law and Economics Center (TILEC), Tilburg University, and coordinator of the CoPECL Economic Impact Group. The opinions expressed in this article represent only the author’s personal views and not necessarily those of the European Commission. Professor of Competition Law and Co-Director, TILEC, Tilburg University, [email protected]. Christian von Bar et al., eds., Principles, Definitions and Model Rules of European Private Law – Draft Common Frame of Reference (DCFR), Outline edition (Munich: Sellier, 2009). The Commission on European Contract Law, chaired by Ole Lando, which produced the Principles of European Contract Law (PECL). See Ole Lando and Hugh Beale, eds., Principles of European Contract Law – Parts I and II (The Hague: Kluwer, 1999) and Ole Lando and Hugh Beale, eds., Principles of European Contract Law – Part III (The Hague: Kluwer, 2003). See the latest Resolution of 3 September 2008 of the common frame of reference for European contract law, P6 TA(2008)0397, which refers to the previous ones. Communication on European Contract Law and the revision of the acquis: the way forward, COM(2004)651 final (11 October 2004). In the past two years, the Council (Justice and Home Affairs) has reached conclusions on the Common Frame of Reference on three occasions: see the conclusions of the meeting of 4-5 June 2009 (10551/09) for the most recent pronouncements.

Filomena Chirico & Pierre Larouche

2

Against that background, it will come as no surprise that law and economics scholars also joined the fray. Several prominent authors, such as Roger Van den Bergh,6 Ugo Mattei7 and Jan Smits8 , sought to analyse the creation of a European private law from a law and economics perspective. In addition to that, it has become common parlance amongst legal scholars, when discussing the Europeanisation of private law, to acknowledge the need to justify any top-down legislative intervention which would take place at European level: typically, this is done by invoking transaction costs that need to be reduced in order for the internal market to function. Some awareness of the significance of, and the need for, economic analysis is therefore already discernable.

1. The Economic Impact Group (EIG) Notwithstanding such contributions, more work is needed in order to provide a broad assessment of the economic implications of the rules, taking into consideration the needs of the economic operators in the internal market. Within the European Network of Excellence on Common Principles of Contract Law (CoPECL), this was indeed the mission given to the Economic Impact Group (EIG), which must assess the economic impact of the results achieved by the joint research of the network. The EIG aims at improving the existing analysis in two ways. First, its contributors draw upon the extensive discussion on the costs and benefits (and desirability) of harmonisation of contract law and bring it one step further: instead of dealing with contract law, or even private law, in general, the EIG looks – to the extent possible – at the case for or against harmonisation of specific elements of contract law, as they are defined further below. The EIG was not in a position to analyse the whole of the DCFR, given its limited scope and resources. It focussed on specific elements of the DCFR, as discussed further below. These elements were chosen because they are amongst the most interesting to study from an economic analysis perspective, thereby allowing the EIG to deliver the most added-value. They are usually also central elements of the DCFR and of any system of private law. Secondly, for each of these specific elements, the EIG also studied the content of the DCFR itself, leaving aside issues of harmonization. The question here is whether, from an economic perspective, the drafters of the DCFR settled on the best rules and principles, i. e. whether a “benevolent legislature” setting out contract law without any a priori would have chosen these rules and principles.

6

7

8

See Roger Van den Bergh “Subsidiarity as an Economic Demarcation Principle and the Emergence of European Private Law” (1998) 5 Maastricht J Eur Comp L 129; “Forced Harmonisation of Contract Law in Europe: Not to be Continued”, in Stefan Grundmann and Jules Stuyck (eds.), An Academic Green Paper on European Contract Law, (The Hague: Kluwer, 2002) 245; “Towards a European Private Law: To Harmonise or not to Harmonise, That is the Question” in Hans-Bernd Schäfer and Hans-Joachim Lwowski, Konsequenzen wirtschaftsrechtlicher Normen – Festschrift für Claus Ott (Wiesbaden: Gabler, 2002) 327. Ugo Mattei, The European Codification Process – Cut and Paste (The Hague: Kluwer Law International, 2003). Jan Smits, ed., The Need for a European Contract Law: Empirical and Legal Perspectives (Groningen: Europa Law Publishing, 2005).

Introduction

3

1.1. The appropriateness of harmonization As mentioned, academics – especially comparatists – have already contributed extensively to the debate on the need and the form of unification of contract law. Already in the Lando & Beale Principles of European Contract Law (PECL), uniform European rules were advocated in order to: a) facilitate cross-border trade, b) strengthen the common market, c) create an “infrastructure” for future European interventions in the contract law field, d) provide guidelines for national courts and legislators, and e) construct a bridge between common law and civil law.9 The current mainstream position of the proponents of unification of European private law (or at least contract law) is that it would be required primarily because differences across Member States constitute a non-tariff barrier to trade.10 Some scholars11 and the European Parliament12 have already criticised that line of reasoning for turning the discussion on the European private law into a purely technical debate, thereby ignoring the policy issues. In recent years, the Commission seems to have moved away from the idea of a binding instrument and is now focusing on other rationales besides the internal market, namely the improvement of the quality of EC law.13 Besides, there is also a fundamental empirical problem with the trade barrier argument: it is very difficult to assess the magnitude of the costs in question and whether they are really obstructing trade to a significant extent. Respondents to the Commission consultations were rather sceptical about the benefits of an exercise in Europeanizing private law – except for the academic community. There seems to be no conclusive evidence so far, save for some proposals on how to make estimates in certain limited areas where harmonisation has taken place and has had time to produce its effects.14

1.2. The appropriateness of the rules and principles chosen in the DCFR Even if for the sake of argument it is agreed that harmonization or unification of private law (or at least contract law) in Europe is needed, such a finding does not give much guidance for the subsequent step, namely ascertaining what that European private law should entail. It cannot be that the reduction of transaction costs and the other benefits invoked above are so massive that any law, as long as it is harmonized at European level, will do. Indeed, some

9 10

11 12 13 14

See the Principles of European Contract Law – Parts I and II, supra, note 2, at p. xxi. C. f. Ulrich Drobnig, “Europäisches Zivilgesetzbuch – Gründe und Grundgedanken”, in D. Martiny and N. Witzleb (eds.), Auf dem Wege zu einem Europäischen Zivilgesetzbuch (Heidelberg: Springer, 1999) 109 or Christian von Bar, “A Civil Code for Europe” [2001] Juridisk Tidskrift 3. See among others Ugo Mattei, supra, note 7. As recalled again in the Resolution of 3 September 2008, supra, note 3. See the 2004 Communication, supra, note 4, at pp 2-5. Indeed in the 2004 Communication, ibid. at 11, the Commission mentions that studies have shown that “there are no appreciable problems arising from differences in the interaction between contract law and tort law in the different Member States”.

Filomena Chirico & Pierre Larouche

4

scholars claim that the aim of the DCFR, beyond the reduction of transaction costs, is much more fundamental, i. e. “drafting a contract law of good quality”.15 Yet we submit that here as well economic analysis can inform and enrich legal debates, this time differently, by taking into consideration a more general notion of transaction costs, not restricted to cross-border ones. The economic analysis of contract law suggests that one of the goals of this branch of law is to facilitate allocation of resources through bargaining, by reducing the transaction costs that parties face. There might be other policy objectives as well, which coincide more or less with these goals, such as freedom, justice or solidarity. Accordingly, when it comes to deciding which rule to pick as the common one, economic analysis can focus on whether the law works efficiently. Is the law enabling these policy objectives to be fulfilled by creating an efficient market and guarding against market failures? Relevant questions can then be: Is the chosen rule minimizing transaction costs for parties to a contract? Is it allocating efficiently the information burden? Is it correcting moral hazard? With the help of the literature on the economic analysis of law, a standard can be established for assessing whether European private law is truly of ‘high quality’.

2. Methodological issues At the outset, a few methodological issues must be raised. First, the work of the EIG is mainly based on the existing literature on the economics of harmonisation and on the economic analysis of private law, whose insights are applied to the DCFR as an object of analysis. Secondly, the analysis tries to keep conceptually separate the questions concerning “what the law is”, “what is really going on” and “what the law should be”. With respect to the first question, much of the work has been done by the drafting groups who prepared the DCFR. Nevertheless, it remains necessary to translate the law into terms that are understandable for economists (or to put the law into the analytical grids of economics). For instance, when referring to the analysis of a certain “rule”, it may be that in fact more than one provision is covered at the same time because identifying “what the law is” implies the need to go beyond the content of one particular article. The second question (“what is going on”) refers to the “positive” economic analysis of legal rules. It requires investigating the incentive structure that such rules establish and their consequences on individuals’ behaviour. This kind of analysis aims at verifying the “fitness for purpose” of the rule, whether the rule achieves the stated objective. When it comes to assessing the “fitness OF purpose”, the third of our questions, i. e. the desirability of the rule itself, it is then necessary to choose a normative standard. Economic analysis typically uses the maximization of welfare as standard. In the context of the DCFR, it could be argued that the standard should be further refined as the maximisation of consumer welfare.

3. The elements of the DCFR retained for analysis At the outset, it was clear that time and budgetary constraints would not enable the EIG to analyse the DCFR in its totality. A number of elements were chosen because they are central

15

See Grundmann and Stuyck, supra, note 6.

Introduction

5

to the areas covered in the DCFR or because they have been dealt with at length in law and economics literature.

3.1. Contract Law – General Aspects The first set of contributions is concerned with general aspects of contract law. In her first contribution, Filomena Chirico expands on the issues raised in this introduction by looking at how the function of contract law is defined in the literature and in the DCFR. Good faith is one of central notions running through contract law – at least in continental systems – and it has been studied at length in both law and economics and comparative law. In her second contribution, Filomena Chirico examines the treatment of good faith in the DCFR in the light of this literature. The introduction of a general principle of non-discrimination in the DCFR (Article II.–2:101 and ff.) distinguishes the DCFR from previous exercises. It reflects the state of EC law, but at the same time it is an interesting intrusion of a public law element in a private law instrument. Ann-Sophie Vandenberghe puts this in perspective in her contribution.

3.2. Formation and Interpretation of Contracts Contract formation typically occupies a large place in contract law (at least in how contract law is conceived and taught). The DCFR follows this tradition, but also adds a substantial section on pre-contractual information duties at Article II.–3:101 and ff. Gerrit de Geest and Mitja Kovač investigate how the two relate to each other, from an economic perspective. Geerte Hesen and Robert Hardy, for their part, examine the law and economics of contract interpretation as set out at Article II.–8:101 and ff. DCFR, in the light of contract theory. Specific validity and interpretation rules apply to standard contract terms at Article II.–9:401 and ff. DCFR, in line with Directive 93/13 on unfair terms in consumer contracts.16 Standard terms touch upon central tenets of contract theory, and Hans-Bernd Schaefer and Patrick Leyens analyse the DCFR provisions from that perspective in their contribution.

3.3. Performance of Contracts Contract performance has gained in significance as an area of research in no small part due to law and economics literature, which brought legal thought forward with key notions such as efficient breach, as well as expectation and reliance as measures of damages. Accordingly, three separate contributions are dedicated to the DCFR provisions concerning contract performance. Gerrit de Geest looks into the general choice of remedies: specific performance or damages. Anthony Ogus, for his part, concentrates on the standard used for the measurement of damages: reliance, expectation or other (including the use of penalty clauses). Finally, Urs Schweizer revisits these provisions from a game theoretical perspective.

16

Directive 93/13 of 5 April 1993 on unfair terms in consumer contracts [1993] OJ L 95/29.

Filomena Chirico & Pierre Larouche

6

3.4. Termination of contracts The general rules on contract termination, at Article III.–1:108 and 1:109 DCFR, interact with the issue of contract performance. In his contribution, Mitja Kovač goes deeper into this issue with the help of the law and economics literature. In the specific case of the termination of long-term relational contracts (agency, franchising, distributorship), the rules of the DCFR (Article IV. E.–1:101 and ff.) must be analysed from a game-theoretical perspective, considering that the contract is an indefinitely repeated game. In his contribution, Fernando Gomez assesses whether the drafters of the DCFR correctly assumed that the rules of Directive 86/653 on commercial agents17 could be extended to other relational contracts.

3.5. Specific Areas of Contract Law Even if the Principles of European Insurance Contract Law were ultimately not incorporated in the DCFR, they touch upon a number of issues which have been discussed extensively in the law and economics literature, as Tomas Kontautas shows in his contribution. One of the most interesting issues arising out of the DCFR process is how, in the work of the Acquis Group in particular, a number of EC directives which were primarily concerned with consumer protection were generalized, and what is left of them in the DCFR from the point of view of consumers. Kati Cseres and Hanneke Luth tackle this issue in their contribution.

3.6. Non-Contractual Liability The last two contributions venture beyond contract law to touch upon the other parts of the DCFR, in particular Book VI on non-contractual liability. Nuno Garoupa and Assunçao Cristas use law and economics literature to revisit the classical set of issues arising on the boundary between torts and contracts. Finally, Pierre Larouche looks at a central element of Book VI, namely the notion of ‘legally relevant damage’ which is meant to act as an a priori limit to non-contractual liability. 17

Council Directive 86/653 of 18 December 1986 on self-employed commercial agents [1986] OJ L 382/17.

1. Contract Law – General Aspects

The Function of European Contract Law: An Economic Analysis* Filomena Chirico**

Abstract This paper aims at addressing a fundamental but somewhat neglected issue regarding the discussion of the function of European Contract Law. At a time when both academics and political actors show their interest in drafting a comprehensive body of rules meant to be available to market operators across Europe, a thorough analysis of the aims of such legal instrument is fundamental and cannot be sufficiently addressed by an all-encompassing list of values that European Contract Law is expected to pursue. The work tries to show how using an economic method can be of help at the core of European Contract Law, to identify its function(s) and to provide guidance as to the appropriateness of the rules set to become part of it. Such guidance includes both suggestions as to the appropriate regulatory level, going beyond the incomplete cross-border transaction costs argument, as well as indications as to what constitutes contract law of good quality.

Introduction Common Principles of European Contract Law1 are currently being drafted by groups of researchers and will be proposed to the European Commission in a coherent structure for use in the EU legislative process (leading, ideally, to the so-called Common Frame of Reference – CFR – for European Contract Law). An arguable novelty in this kind of academic endeavour is the involvement of a group of economic experts, entrusted with the task of ‘assessing the broad economic impact’ of the rules and principles constituting European Contract Law. To perform such exercise, a first important step is the identification of what economic analysis has to say about the function of European Contract Law. This initial effort is useful to provide both a basis for the economic assessment of the rules and a first economic check on *

**

1

This article first appeared in the European Review of Contract Law 2009. Reprinted with permission. Official at the Directorate General for Competition of the European Commission – At the time this article was written, she was Assistant Professor and Senior member of the Tilburg Law and Economics Center (TILEC), Tilburg University, coordinator of the CoPECL Economic Impact Group. The author wishes to thank Pierre Larouche, Fernando Gomez and Eric Brousseau for their feedback on earlier drafts, the participants to the First Roundtable of the CoPECL Economic Impact Group and the participants to the TILEC WIPs for their useful comments. The opinions expressed in this article represent only the author’s personal views and not necessarily those of the European Commission. Hence the acronym of the project: CoPECL.

Filomena Chirico

10

the function of European Contract Law, as stated in the Draft Common Frame of Reference (DCFR).2 Accordingly, the present work aims to provide a general framework of analysis and puts forward a method for the economic analysis of European Contract Law. It refers to the main economic principles that are relevant for carrying out an economic analysis of legal rules. The paper relies on existing literature in contract law and economics, new institutional economics and economics of federalism. Therefore, not many details will be given with respect to the proof and explanation of the claims made in the various sections, and the reader is referred to the underlying literature. After a brief introduction to the political and academic background in Section 1, the underlying economic focus permeating the discussion is highlighted (Section 2). Section 3 presents the DCFR position as to what the function of European Contract Law is, which it terms ‘Aims and underlying values’. Section 4 explains the economic method of analysis and Section 5 is devoted to showing a possible economic interpretation of the function of European Contract Law. Section 6 concludes.

1. The Background For many years, the creation of a European Civil Code has been a major discussion topic in the academic world, especially amongst comparative lawyers. Efforts to elaborate and write down principles common to the whole of ‘Europe’3 sprung from the seemingly simple consideration that the creation of a European Union with common rules, common citizenship and a common (albeit differentiated) identity would naturally require also a common civil code.4 The other side of the debate, however, has usually pointed to a different albeit seemingly just as simple reflection, namely that the alleged common identity is in fact not so uniform and that in reality ‘legal cultures’ are diverging so much that talking about a common civil code is just meaningless if not perverse.5 With respect to contract law in particular, this debate has also reached the European political level, notwithstanding the limited competences attributed to the EC. European legislative measures have taken place thus far only in specific fields and to solve specific problems perceived as single market issues.6 However, the desire for greater coherence and a more gen2

3

4 5

6

This work refers to the academic exercise referred to as Draft CFR (DCFR), published as an interim article-only outline in December 2007 by Sellier – European Law Publishers. The Principles of European Contract Law (PECL) by the Commission on European Contract Law, the Common Core project, Casebooks project and others. Echoing the nineteenth century drive of national codifications. See, among others, P. Legrand, ‘A diabolical idea’, in A. Hartkamp et al (eds), Towards a European Civil Code (3rd ed, Nijmegen: Ars Aequi Libri, 2004) 245. Ten directives have been enacted on contract law topics and nine more touching issues related to contract law. Inter alia: Council Directive 1993/13 of 5 April 1993 on unfair terms in consumer contracts, OJ EC 1993, L 95/29; Directive 1999/44 on sale of consumer goods, OJ EC 1999, L 171/12; Directive 1999/93 on electronic signatures, OJ EC L 13/12; Directive 2000/31 on electronic commerce, OJ EC 2000, L 178/1; Directive 2005/29 on unfair commercial practices, OJ EC 2005, L 149/22; Directive 97/7/ EEC on distance contracts, OJ EC L 144/19, Directive 90/314 on package travel, OJ EC L 158/59; Directive 85/577 on contracts negotiated away from business premises, OJ EC 1985, L 372/31;

The Function of European Contract Law: An Economic Analysis

11

eral ‘plan’ behind sectoral interventions spurred the inter-institutional ‘trialogue’. After a few resolutions of the European Parliament7 and some statements from the European Council,8 the European Commission enacted the 2001 Communication on European Contract Law,9 followed by the 2003 Action Plan on a more coherent European Contract Law10 and its subsequent Communications and Progress Report.11 Following the Action Plan, both the Parliament and the Council of Ministers issued a number of resolutions12 on the matter. The outcome of the discussion took the form of an intervention bringing together these two worlds, the academic one and the political one: an academic network has been entrusted with the delivery – after appropriate stakeholders’ consultation – of a Common Frame of Reference (CFR) for European contract law, to be used by European institutions for various purposes.13 The specific focus on contract law, rather than on the whole of private law, is probably the consequence of the perceived more ‘technical’ character of the contract branch of civil law, as compared not only to family or labour law, but also to neighbouring areas of civil law such as property law and extra-contractual liability/tort law.14 In such fields, policy considerations play a more obvious role and may render a Europe-wide agreement more difficult to

7

8

9

10

11

12

13

14

Directive 2000/35 on combating late payment in commercial transactions, OJ EC 2000, L 200/35; Directive 2002/65 on distance marketing of financial services, OJ EC 2002, L 271/16. The European Parliament called for action to bring in line the private laws of the Member States, for the first time as early as in 1989 with a Resolution (Resolution of 26 May 1989 on action to bring into line the private law of the Member States, OJEC 1989 C 158/400) making explicit reference to a European code of private law. Such call has been reiterated in later documents. The 1999 Tampere meeting of the European Council called for ‘greater convergence in civil law’. To be precise, it requested a study on the need of harmonisation in civil law; the conclusions of the Presidency can be read at http://www.europarl.europa.eu/summits/tam_en.htm. Communication from the Commission to the Council and the European Parliament on European contract law, OJEC 2001 C 255/1. Communication from the Commission to the European Parliament and the Council – A more coherent European contract law – An action plan, COM(2003) 68 final, OJEC 2003 C 63/1-44. Communication from the Commission to the European Parliament and the Council – European Contract Law and the revision of the acquis: the way forward, COM(2004) 651; Report from the Commission – First Annual Progress Report on European Contract Law and the Acquis Review, COM(2005) 456 final. While, not surprisingly, Parliament’s resolutions are usually calling for more integration and Europeanisation, the Council is more cautious and refers to the ‘usefulness’ of measures to facilitate cross-border contracts, although the emphasis was chiefly on the need for ‘internal’ consistency of European normative activity in the contract law field and its national transpositions. See for example the European Parliament resolution of 12 December 2007 on European contract law and the Council Resolution of 22 September 2003 on ‘A More Coherent European Contract Law’, OJEC 2003 C 246/1. See also the Conclusions of the Competitiveness Council, 28-29 November 2005, available at http://ec.europa.eu/consumers/cons_int/safe_shop/fair_bus_pract/cont_law/con clusions_competitiveness_council_en.pdf. See EU Commission’s Action Plan and follow-up, n 10 above. More recently the Commission seems oriented towards drafting a White Paper on a Common Frame of Reference (CFR) integrating some parts of the DCFR. Obligations deriving from non-contractual liability were eventually included within the scope of the DCFR, in Book VI.

12

Filomena Chirico

reach. In fact, although stressing the ‘technical’ character of contract law seems to enhance its neutrality and acceptability (as opposed to having to reach a political agreement),15 at the same time it can turn out to be a weakness in the methodology when it implies denial of the policy choices that, too, underlie this branch of the law.16 The purposes of this exercise, according to the initial statements, are:17 the improvement of the Community acquis in the contract law field, the promotion of European Standard Terms and Conditions, the starting point for a future optional Community instrument in the field. The CFR is seen as potentially useful also to national governments to improve their own contract law. Following further reflection and the input of stakeholder groups, the first goal (improving the acquis) has been given prominence and the European Parliament has invited the Commission not to enact any measure concerning contract law before the CFR is completed.18 Conversely, the second objective stated above (standard terms) seems to have been set aside, whilst the third and perhaps most controversial one (concerning the optional instrument) has never been dropped but is often said to need further reflection.19 Of these multiple purposes, the idea of a ‘toolbox’ or handbook for legislators seems recently to have become predominant.20 Several groups of researchers21 are involved in the drafting of European Contract Law, of which the main ones are the Study Group on a European Civil Code22 (‘Study Group’) and the Research Group on the Existing EC Private Law23 (‘Acquis Group’). The two groups employ two different methodologies to identify and formulate legal principles common to the whole of Europe. The former uses an ‘inclusive comparative law method’, based on national experiences but transcending them: the chosen rules are said to be those ‘best suited

15

16

17 18 19

20

21

22 23

A discussion on this approach to law, especially supranational European Law can be found in M.P. Maduro We, the Court, Hart Publishing, Oxford 1998, p. 10 and sources referred to therein. On this point, see also F. Gomez, ‘European Contract Law and Economic Welfare: A View from Law and Economics’ InDret April 2008; U. Mattei, ‘The issue of civil codification and legal scholarship:biases, strategies and developments’ 21 Hastings International & Comparative Law Review 1997-1998, 883; T. Wilhelmsson, ‘Private Law in the EU: Harmonised or Fragmented Europeanisation?’ European Review of Private Law 2002, 77-94. As set out in the Action Plan and subsequent communications, n 10 and 11 above. European Parliament resolution on European contract law of 4.9.2006. The European Commission has stated to be exploring the ‘opportuneness’ (sic) of such an optional instrument as an additional ’26th regime’ (nowadays 28th), which would leave existing national law untouched, in certain specific areas of contract law. Report from the Commission – First Annual Progress Report on European Contract Law and the Acquis Review, COM(2005) 456 final, paragraph 4.2. For an economic analysis and a proposal to use European Law as (then) 16th legal system as consequence descending from the subsidiarity principle, see R. van den Bergh, ‘Towards an Institutional Legal Framework for Regulatory Competition in Europe’ 2000 (53) Kyklos 435-466. See European Parliament resolution of 3 September 2008, http://www.europarl.europa.eu/sides/ getDoc.do?type=TA&reference=P6-TA-2008-0397&language=EN&ring=B6-2008-0374 and the statements of the Presidency of the European Commission in February 2009 http://register.consilium. europa.eu/pdf/en/09/st06/st06155.en09.pdf. All the groups are listed at http://www.copecl.org and include also the ‘Common Core Group’, the ‘Insurance Group’ and others. Website at http://www.sgecc.net. Website at http://www.acquis-group.org.

The Function of European Contract Law: An Economic Analysis

13

to the economic and social conditions in Europe’.24 The latter group’s method consists in ‘explor[ing] the existing EC law and elaborat[ing] its underlying principles’, including legislation, jurisprudence and legal literature, as well as the laws of the Member States implementing EC law.25 In order to have a more complete picture and, perhaps to abide by principles of good governance,26 an academic ‘Economic Impact Group’ (EIG) is also involved in the process leading to the formulation of the CFR. The EIG carries out an economic assessment of specific rules and principles of European Contract Law, as included in the DCFR. The economic comments will then hopefully be taken into account when the European institutions decide on enacting normative acts. Against the described background, the present contribution aims at setting in general terms the framework and methodology to provide an economic assessment as expected by the drafting groups and the European Institutions.

2. The economic focus A recurring theme permeates the discussion on European Contract Law: the reason put forward most often, albeit not exclusively, for the harmonisation or even unification of contract law across Europe is the risk that ‘the coexistence of national contract laws in the Member States directly or indirectly obstructs the functioning of the internal market.’27 Disparate national rules, and in particular different mandatory rules ‘may lead to higher transaction costs’ for suppliers and consumers engaging in cross-border trade across Europe. Similar market integration language has been employed by academics active in the comparative contract law field as well.28 It is clear that the main reason for framing the debate in terms of market integration is the need to justify and to give a legal basis to an eventual intervention of the European institutions, in the light of the current division of competences between the EC and its Member States.29 It cannot be, however, simply a matter of choice of words: real ‘obstacles’ must be identified in order to justify harmonisation, as the European Court of Justice itself

24

25 26

27 28

29

See the ‘Communication on European Contract Law: Joint Response of the Commission on European Contract Law and the Study Group on a European Civil Code’, March 2003 available at http:// www.sgecc.net/pages/en/home/137.joint_response_to_the_eu_commissions_communication_ on_european_contract_law.htm. See the Paper on Notion and Function of Contract, pages 3 and 4. See, for instance http://ec.europa.eu/governance/impact/index_en.htm on the necessity of an impact assessment prior to a regulatory intervention and DG Enterprise’s http://ec.europa.eu/enterprise/ regulation/better_regulation/index_en.htm. Commission’s 2001 Communication paragraph 23. See for example, PECL 2000: the facilitation of cross-border trade and the strengthening of the common market are indicated as the first two goals of their Principles. EC has competences limited to the areas conferred to it by the Treaty. Therefore, it needs to find a legal basis for action in the Treaty. Although art 308 or 65 EC Treaty have been suggested, the most promising legal basis seems to be art 95. Hence the stress on easing cross-border trade, eliminating transaction costs of trans-boundary contracts, in the market integration perspective. This has also been the legal basis of most of the sectoral interventions thus far, comprising 19 directives, some of them criticised for not having much to do with building an internal market.

14

Filomena Chirico

has highlighted.30 Hence, stakeholders and interested parties were consulted to find out whether the suspicion of the existence of obstacles was indeed grounded. Responses31 were, unfortunately, rather inconclusive, as far as real data and empirical evidence are concerned, while consisting, for most part, in opinions or statements of positions.32 It should be noted that, more recently, a privately conducted survey tried to identify any actual obstacles to cross-border trade and their magnitude.33 The results of that survey show that differences in legislation do seem to cause certain extra costs to market operators (especially related to obtaining foreign legal advice). However, the report also points out that such costs are not of great significance and should thus not be overemphasised. In general terms, the previous account highlights the constant economic emphasis put in the discussions of the matter, in particular through a continuous reference to the economic concept of ‘transaction costs’, more precisely cross-border ones. While such approach can be praised for paying attention to the economic impact of legislation (in this case: different national ones), it leaves the impression of being incomplete if not misleading. On the one hand, name-dropping economic concepts seems a too quick way out of the necessity to provide a solid basis for European intervention and to identify the function of the codification in progress. On the other hand, even if the invocation of ‘transaction costs’ can be interpreted as a true statement of faith in economic methodology, the approach is nevertheless incomplete under several respects. For example, while reducing the costs of complexity deriving from different legal systems has certain advantages, a complementary view of legal diversity points at its contribution to increasing quality of legal systems thanks to competition among national rules. Considerable improvement to the analysis would therefore be brought about by taking into account all the indications coming from economics and not only the transaction cost argument, as will be shown in more detail in the subsequent sections of this work. Besides the ubiquitous reference to transaction costs, the relevant texts laying down the function(s) of the CFR and of European Contract Law34 give the impression that, behind calls for market integration, better regulation and economic method, no clear vision emerges as to what legitimate function European Contract Law is expected to fulfil. A clear view of the function(s) of European Contract Law is crucial because it implies a choice of the guiding principles in the selection of the appropriate rules, of the normative criteria for the assessment of the proposed rules and of the methodological approach to the drafting of the DCFR. The aim of this paper is precisely to shed some light on this fundamental issue.

30

31

32

33

34

See the ECJ Tobacco cases of 2000 and 2006. The case concerned an EC-wide ban on Tobacco advertising justified on the ground that differences in national legislations affect interstate trade. Judgment of the Court of Justice 12 December 2006 in Case 380/03, Federal Republic of Germany v European Parliament and Council of the European Union. All documents are available on the EU Commission website at http://ec.europa.eu/consumers/ cons_int/safe_shop/fair_bus_pract/cont_law/index_en.htm. It can be observed that, generally, the position of businesses was rather of reluctance towards contract law harmonisation. Survey conducted in 2005 by the Oxford Institute of European and Comparative Law in collaboration with Clifford-Chance and published in S. Vogenauer / S. Weatherill, The Harmonisation of European Contract Law: Implications for European Private Laws, Business and Legal Practice (Oxford: Hart Publishing, 2006). Both in the statements by European institutions and academic statements. This will be discussed more in detail below.

The Function of European Contract Law: An Economic Analysis

15

3. The Function of European Contract Law in the DCFR The DCFR devotes a part of its introductory section to the discussion of its ‘aims and underlying values’, following the discussion on an earlier draft called ‘Notion and Function of Contract’. In its original version, there were explicit statements as to the function of contracts and of contract law. The function of contracts was indicated as being twofold: not only an economic one, but also capable of affecting individual lives of European citizens.35 The contract is the main tool for individuals to regulate their economic relationships and it is an implicit assumption in the decision to have an EC internal market36 and at the same time fundamental for the establishment of an ‘area of freedom, security and justice’.37 Paralleling the abovementioned twofold function of contracts, that document indicated as purposes of contract law:38 to contribute (1) to freedom and social security of EU citizens and (2) to enhancement of economic welfare. Contract law should therefore be governed by the principle of open market economy with free competition and ‘shall thereby secure an optimal supply of goods and services in the market’. Other functions specific to EC contract law39 were identified as the promotion of the internal market and elimination of obstacles to the four freedoms, the facilitation of cross-border contracts and creation of a European level playing field. The consultation run on this formulation has, however, led to substantial changes to the text. The current version does not refer to the ‘function’ of contracts and of contract law but to the ‘aims’ of European Private Law. Moreover, it appears to have been broadened substantially.40 The impressive list of ‘aims’ of European Private Law includes ‘at least’: justice, freedom, protection of human rights, economic welfare, solidarity and social responsibility, promotion of the Internal Market, preservation of cultural and linguistic pluralism. Moreover, rationality, legal certainty, predictability and efficiency are to be pursued too.41 The references to freedom (of contract) and justice seem to belong to classical legal theory. Interestingly for economic analysis, justice is intended as ‘corrective’, while ‘redistributive’ justice is considered not to be a concern for the DCFR, although, according to the text, this principle may still inspire certain (consumer) rules.42 As will be discussed further below, generally contract law is not very well suited to pursue redistributive policies.43 Freedom of contract is described as fundamental in that it leads to justice and just outcomes, with a reference to the concept of justice that seems to echo familiar economic concepts

35 36 37 38 39

40 41 42 43

Acquis group draft on notion and function of contract. Id, 6. Referring to art 61 et seq of the EC Treaty. Principles 4 and 5 in the discussion paper on notion and function of contract. The approach sketched above reflects the fundamental economic goals of general EC law: enhancement of economic welfare (through competition) and elimination of obstacles to cross-border trade, together with a seemingly non-economic reference to contributing to freedom and social security of citizens. DCFR, Sellier 2007, 10 et seq. DCFR Introduction, paragraph 22. Id, paragraph 24. For a critique of using European Contract Law for redistributive goal, see below in section 6 and Gomez, n 16 above.

Filomena Chirico

16

such as utility and welfare maximisation.44 Limitations to freedom are admissible but only in specific and justified contexts.45 A straightforward welfare criterion is also mentioned among the aims of the DCFR, generally pursued through the use of contract law default rules, but specifying also that the existence of market failures may justify an interference with parties’ agreement.46 However, despite claiming that ‘any legislative intervention’ should be checked against the welfare criterion, it is nevertheless acknowledged that other justifications, called ‘welfarist’ (ie rightsbased consumer protection), may replace promotion of efficiency as basis for intervention. Much less clear is the way contract law is expected to pursue other aims, such as protection of human rights, solidarity and social responsibility, and more general EU-specific aims, including contributing to the establishment of an area of freedom, security and justice and preserving linguistic and cultural diversity. Such long, broad and varied list of aims and values for European Contract Law leaves the reader with the flavour of a compromise to take onboard all the different viewpoints expressed during the drafting phase, without elaborating and taking position as to whether European contract law is really the best tool to pursue them all. Perhaps this can be ascribed to the desire to achieve a higher degree of ‘social acceptability’ for the DCFR. However, its being so all-encompassing somehow nullifies the guidance that is to be derived from such fundamental aims and values: it may translate into just leaving it to the drafters to cherry-pick the justification that best serves to explain their choice of a certain rule. Moreover, when it would occur that the different aims conflict with one other, the DCFR states that they will be balanced against each other in the context of the specific rules. However, in order to perform a balancing exercise, those drafting the rules need to have a normative meta-criterion to inspire them, otherwise the balancing runs the risk of being arbitrary. This work contends that using the economic method has the advantage of a clear metacriterion of welfare maximisation, and that this can help both to assess the effects of legal rules and to give guidance in the, admittedly more difficult, evaluation of their desirability. In particular, since any legal choice has consequences on welfare, it has been shown that any rule not inspired by such criterion may end up reducing instead of enhancing it.47 The economic method focuses on trade-offs: lawmakers may still want to enact the welfaredecreasing rule (after all, it is a policy decision), but they will likely need a strong counterbalancing argument to justify a rule with negative effects on society. The way in which the economic method can shed some light on the function of European Contract Law will be the subject of the next sections.

4. The economic method Economic analysis uses a consequentialist approach to assess legal rules. In contract law, for example, a key principle such as freedom of contract is not considered a fundamental tenet based on a certain philosophical or ethical precept, but rather in its economic function. In 44

45 46 47

The DCFR states for example in paragraph 25 of the Introduction that ‘As a rule, natural and legal persons are free both to decide whether or not to contract and to agree on the terms of their contract because in some situations, freedom of contract, without more, leads to justice.’ DCFR Introduction, paragraphs 25-28. DCFR Introduction, paragraphs 29-30. See L. Kaplow / S. Shavell, Fairness versus Welfare (Cambridge: Harvard University Press, 2002).

The Function of European Contract Law: An Economic Analysis

17

other words, freedom of contract is the prerequisite for the market to deliver the desired outcome, ie, statically, the allocation of resources to their best use and, dynamically, the incentive for investing into ways of increasing output and thus welfare.48 Economic analysis may support limits to freedom of contract,49 but such limitations need to be carefully justified. In principle, the DCFR seems to advocate a similar perspective, although the rhetoric used may be slightly confusing as to the methodology.50 The same kind of consequentialism advocates a focus on reasons and effects of Europeanisation of legal rules, rather than on statements of ‘rights’ or of consistency. This is not to say that rights and consistency are not important, but a more economic approach looks at them consequentially and treats them not as values or goals in themselves but considers the way in which they generate certain desirable consequences.51 There has been considerable discussion on the ‘correctness’ of such methodology. It is often contrasted to other ‘deontic’ approaches, such as those based on justice or fairness, focusing on the consistency of legal rules with those principles and on the ex post enforcement of pre-existing rights. While these approaches seem to have a stronger and more broadly acceptable ‘moral’ stance, they do not necessarily provide for a univocal and convincing method of assessment.52 The danger is that once certain rights have been asserted, the consequences of the rule thus established would not be considered relevant. By way of illustration, in contract law there is a lot of support for increasing consumer rights53 as expression of ‘fairness’. Granting additional rights may or may not have as a consequence an increase in consumers’ welfare, but it very often entails an alteration of the contractual conditions, most likely an increase in the price that consumers have to pay. Yet, this trade-off is hardly taken into account when introducing rules on consumer contracts, since the arguments are usually focusing on consumers’ status and on the ‘rights’ needed for a ‘higher’ level of protection. By contrast, an economic approach focuses on outcomes. If the outcome of a ‘fairer’ rule is that prices increase or products disappear, a consequentialist analysis suggests to check whether the totals surplus from cooperation is increased, ie whether such negative effects are counterbalanced by the improvements that the ‘fairer’ rule is intended to bring.54 The previous argument should however not be perceived as a claim that principles of fairness, justice and the like are irrelevant. As one author put it, ‘economic analysis of law

48

49 50

51

52

53

54

Allocative efficiency is maximised when parties have the freedom to bind themselves via a contract, provided that such contract does not have external effects. Legally imposed limitations of contractual freedom often result in sub-optimal allocation of resources. M. Trebilcock, The Limits of Freedom of Contract (Cambridge: Harvard University Press, 1997). In fact, the DCFR statements about freedom of contract refer it as ‘part of the fundamental rights of European citizens and enterprises’. If this amounts to a non-consequentialist approach, the legal rule and the recommendation of economic analysis may differ in concreto. For example, ‘rights’ may be seen as a way to structure debate and save on transaction costs while consistency as a way to ensure predictability and thus lower transaction costs and avoid distortions in incentives over time. J. Kraus, ‘Philosophy of Contract Law’, in J.L. Coleman / S. Shapiro (eds), The Oxford Handbook of Jurisprudence and Philosophy of Law (Oxford: Oxford University Press, 2002). See for example the statements by MEPs: http://www.epp-ed.eu/policies/imco/default_en.asp or the statements of the first 2009 EU Presidency at http://www.eu2009.cz/en/eu-policies/employment-socialpolicy-health-and-consumer-affairs-council/consumer-protection/consumer-protection-678. See an illustration of this in Gomez, n 16 above, 20.

Filomena Chirico

18

does not establish ideals’.55 A legal rule is also the product of other considerations, such as social acceptability. Yet it is important that lawmakers are aware of the consequences and of the trade-offs associated with pursuing a certain goal, so that they (and their constituencies) can make an informed decision. The economic method ‘helps developing alternative legal proposals, helps ascertaining their consequences and assesses which consequences best advance the established ideals’.56 In other words, since justice is a foundation of any society and fairness is a principle that helps preventing social conflicts,57 it seems more productive to put a consequentialist method to the service of justice and use it to achieve better and fairer outcomes. Moreover, it can be argued that a similar approach is not only advisable but actually necessary under EC law following the ECJ judgments in the abovementioned Tobacco Advertising and related cases.58 In these judgments, the Court clearly stated that approximation or unification of national rules is not to be regarded as a goal in itself, nor is consistency across national systems to be taken in itself as a guiding principle for European legislative intervention. Rather, the existence of obstacles to the functioning of the internal market, and the need to eliminate them, are to be used as criteria. To put it in more ‘economic’ terms, first the existence of an obstacle has to be demonstrated by identifying the actual costs associated with such obstruction; then the consequences – in terms of benefits and costs – of eliminating the impediment have to be assessed in order to substantiate the claim for harmonisation. A final clarification is worth mentioning: the previous arguments in favour of the use of the economic method and of a clear assessment of costs and trade-offs should not be interpreted as proscribing a regulatory intervention every time additional costs would be imposed on a private party. It may well be advisable from an economic point of view to increase the costs for a party if this results in an overall reduction of costs or an increase in the efficiency of transactions for society as a whole.

5. A framework for analysis Before setting out the economic method of analysing the function of European Contract Law, a preliminary remark seems necessary. It was hinted above that, in the general academic discussion about the function of European Contract Law and about the content of the CFR, a gap is noticeable in that the debate seems to have concentrated predominantly on finding a proper justification for having a ‘European’ instrument. This is certainly a necessary precondition for action, but it is not all that matters for the drafting of European Contract

55

56 57

58

N.L. Georgakopoulos, Principles and methods in law and economics (Cambridge: Cambridge University Press, 2005) 2. Ibid. After all, it is recognised also in the economic literature on contracts that despite the final allocation of resources following negotiations depends only on bargaining skills and should not be of particular concern, still a ‘fair’ outcome is preferable to an imbalanced one. N 30 above. In the sense that an economic approach should go even beyond the test performed in the Tobacco judgement, see also G. Tridimas / T. Tridimas, ‘The European Court of Justice and the Annulment of the Tobacco Advertisement’ European Journal of Law and Economics 2002, 171.

The Function of European Contract Law: An Economic Analysis

19

Law rules. What seems to have been somewhat neglected is the analysis of the function of contract law as such.59 In fact, analytically, the function of European Contract Law is split into two distinct, albeit connected, issues: (1) the function of contract law as such and (2) the function of a European contract law. The first issue belongs to more traditional legal theory, as developed – usually – in national scholarship, while the second has been rather the domain of comparative lawyers. However, with respect to European Contract Law and the preparation of a CFR, the first issue seems to have been treated as rather implicit. This can perhaps be explained by an assumption that it was not necessary to justify the existence of contract law at the outset, and that it was sufficient to refer to the importance of freedom of contract and the respect for the autonomy and will of contractual parties, which contract law is expected to protect. By contrast, the second issue, concerning the function and – by implication – the justification of a body of European rules, has been the subject of an intense debate that is still ongoing, and is frequently revitalised by the involvement of European institutions.60 To be sure, there is some evidence that academics have picked up this analytical imbalance already. For instance, some authors61 have questioned the view that the reduction of cross-border transaction costs is the main aim for ‘going European’ and thus the ultimate guide in the drafting of European Contract Law. Rather, these authors would aim for the creation of a Contract Law ‘of good quality’. Logically, for these authors, if the two goals – cost reduction and good quality of law – were pointing to different directions, then a rule ‘of good quality’ should be chosen rather than the one that allows costs savings. This work contends that the apparent conflict between these two aims can be avoided by properly clarifying what ‘good quality’ contract law is. ‘Good quality’ could be defined by reference to a ‘rights’ or ‘fairness’ approach: a good contract law would offer the best attribution of rights or the fairest balance of interests. On the basis of the methodological remarks made above, if good quality were to be understood in this way, then not only is there a methodological conflict (between this deontic criterion and the consequentialist ‘transaction costs’ approach) but the solution of such conflict is also difficult to envisage. By contrast, relying on the methodological perspective suggested above, ‘good quality’ could also be understood in consequentialist economic terms, the logical inconsistency would be remedied and even the ostensible conflict avoided. Going more in detail, in our framework, ‘cost reduction’ entails all the necessary assessment of the need of a European-level rule. This includes more than examining cross-border transaction costs, as we will see below. On the other hand, the evaluation of what a ‘good’ rule is pertains more directly to the question of which function contract law is supposed to accomplish, quite apart from the right legislative level. The contribution of economic analysis to European contract law can touch both issues and suggests ways to deal with the two ques-

59

60

61

A certain recognition of this issue can be found in the Acquis paper on notion and function of European Contract Law, discussed below. See for instance, the contributions to S. Grundmann / J. Stuyck, An academic Green paper on European contract law (The Hague: Kluwer, 2002), J.M. Smits (ed), The need for a European Contract Law. Empirical and Legal Perspectives (Groningen: Europa Law Publishing, 2005); Vogenauer / Weatherill, n 33 above. See Grundmann / Stuyck, n 60 above.

Filomena Chirico

20

tions: (1) which regulatory level is the most appropriate to enact certain contract law rules and (2) which rule is the ‘best’ one so as to fulfil the expected function of contract law.62

5.1. The function of Contract Law63 5.1.1. The general function

In economic terms, a contract is concluded when there is a mutual gain for the parties, that is when it produces an increase in welfare.64 Economics analyses contracts as tools at private parties’ disposal to regulate their relationship when certain risks are connected to an exchange. When there is a risk that a time gap between a promise and its performance may make performance more costly – or just less interesting – for the promisor, parties cannot rely that promises will be honoured and hence they will likely not enter into a contract at all.65 Contract law – and the whole legal system – are there to help making private commitments credible and thus encourage parties to enter into contracts in the first place.66 Therefore governmental enforcement of contracts is, together with the protection of ownership, seen as the basis of a free market economy.67 If the ultimate goal of social decision-making is to maximise social welfare,68 with respect to contracts this translates into maximisation of the number of contracts freely entered into by private parties. In other words, cooperation through contracts generates gains for society and should be encouraged.69 This fundamental point of economic analysis of contracts is probably what the DCFR itself hints at when referring to a welfare justification for obligations to cooperate, in addition to a ‘solidarity’ justification.70 There may, however, be a misunderstanding of it: the fact that in economic terms, cooperation – ie contracts – increases society’s welfare71 does not imply per se an obligation to cooperate in the legal sense. Such duty may well be economically justified in certain contexts, but an appropriate normative analysis is required to ground such conclusion. 62

63 64

65

66

67

68 69

70 71

It is certainly true that in some instances it may seem artificial to divide the two problems with regard to the same contract law principle or rule. However, for the purpose of the analysis of the (D)CFR, it is useful to keep in mind that the two issues are conceptually separate albeit connected, and can be better dealt with as if they were two variables of the same function. Rogier van Bijnen substantially contributed to the drafting of this part. See, for example E. Brousseau / J.-M. Glachant, The Economics of Contracts: Theories and Applications (Cambridge: Cambridge University Press, 2002). S. Shavell, Foundations of Economic Analysis of Law (Cambridge / Mass: Belknap Press, 2004) 297298. See R.C. Ellickson, Order Without Law: How Neighbors Settle Disputes (Cambridge / Mass: Harvard University Press, 1991) 167 et seq; J. McMillan / C.M. Woodruff, ‘Private Order under Dysfunctional Public Order’ 98 (2000) Michigan Law Review 2422 et seq. J. Stiglitz, Economics of the Public Sector, Norton 2000, 77; N. Mercuro / S.G. Medema, Economics and the Law (Princeton: Princeton University Press, 1997). Kaplow / Shavell, n 47 above. See R. Cooter / T. Ulen, Law and Economics (4th ed, Boston: Pearson Addison-Wesley, 2004) 195 et seq. DCFR Introduction, paragraph 32. This is in itself a positive statement, not a normative one.

The Function of European Contract Law: An Economic Analysis

21

Economic theory often employs the concept of ‘complete contracts’. A contract is complete when it (i) specifies obligations for the parties for every possible future state of the world and (ii) leaves no potential gain for trade unrealised.72 In a perfect market, contracts will be complete and efficient. Moreover, they would arise and be performed spontaneously, without the need of contract law. However, in reality perfect markets do not exist and complete contracts are just a theoretical utopia.73 Yet, even if contracts are bound not to be complete, the concept of complete contracts can still fulfil a function, similar to the paradigm of perfect competition in microeconomic theory, of a benchmark for the desirable allocation of resources. It is important to note here that the above does not imply the normative consequence that in the real world, real parties should draft contracts as complete as possible. Due to the presence of transaction costs, this may indeed be inefficient. Inefficiency may arise both ex ante, in the negotiation phase, and ex post, in performance and enforcement.74 Accordingly, parties face a trade-off between the costs of making a contract (more) complete and the risk of leaving a gap. Hence, the importance of good contract law, to foster cooperation between parties and prevent opportunistic behaviour that takes advantage of contractual incompleteness.75 It follows that a first conclusion about the function of contract law is that it aims at minimising market failures,76 so as to increase the number of welfare-increasing contracts.77 A conclusion similar to that just described seems to have been internalised in some of the statements concerning the ‘aims’ of the DCFR described above in Section 3. However, the same statements do not seem to take into account that this is only half of the story. Indeed, any intervention in the market comes at a cost. If market operators face trade-offs and are exposed to market failures, governmental intervention involves trade-offs too and may result in some kind of failure too. The function – and the challenge – of contract law is thus more complex than just correcting market failures. Contract law is (or should be) rather the result of a deeper analysis of the trade-offs involved, finding the right balance between the positive effects of reducing market failure and the costs associated with such intervention. It is worth noting here that, as mentioned above,78 the function of contract law should not include the (re)distribution of welfare. This is not to say that redistribution is not a desirable goal, as there are not only social and moral reasons for it but also good economic arguments in favour of redistribution in certain cases. However, it has been showed that

72

73

74 75

76

77

78

Note, however that definitions in the scholarship may differ. Compare I. Ayres / R.H. Gertner, ‘Strategic Contractual Inefficiency and the Optimal Choice of Legal Rules’ 101 (1992) Yale Law Journal 730; and Shavell, n 65 above, 292-293. For an account of different kinds of incompleteness, see E. Brousseau, ‘Contracts: from Bilateral Sets of Incentives to the Multi-Level Governance of Relations’, in E. Brousseau / J.-M. Glachant (ed), New Institutional Economics, A Guidebook (Cambridge: Cambridge University Press, 2008) 37-66. See Brousseau, n 73 above. See also Cooter / Ulen, n 69 above, 235; R. Posner, Economic Analysis of Law (5th ed, New York: Aspen, 1998) 108, etc. Cooter / Ulen, n 69 above, 218; to the extent market failure cannot be cured by parties themselves. Of course, fostering cooperation is not pursued exclusively through contract law, but many more legal and social institutions are relevant to this aim. Sub section 3.

Filomena Chirico

22

redistributing through contract law may just not work.79 Moreover, to the extent that it is actually feasible for contract law to achieve such result, there are convincing arguments that other means to achieve desirable redistributive goals are more cost-effective.80 To be sure, in a sense, contract law does have an impact on distribution: since contract law contributes to the reduction of the costs that the parties would otherwise face, it favours the conclusion of beneficial exchanges and creates surplus for society; by the same token, contract law avoids that non-beneficial exchanges occur, which could redistribute surplus in a non desirable way.81 The DCFR statement that redistribution is, in principle, not pursued is therefore to be welcomed, although the mentioned exceptions (granting rights to consumers) could represent an example where a redistributive aim ends up reducing total contractual surplus for society – and maybe even for consumers themselves.

5.1.2. Market failure

In the economic literature several types of market failure have been identified and analysed. When it comes to contract law, some of these types are relevant to the extent that they are responsible for contracts being incomplete: • externalities, when third-party interests are affected whilst they were not considered in the contract and eventually included in the price;82 • information asymmetry83 and irrationality84 because they may lead to the conclusion of contracts that would not otherwise have been entered into; • transaction costs85 that obstruct the conclusion of otherwise profitable contracts. In the light of the comments made above,86 it is worth specifying that transaction costs include the costs for finding the appropriate contractual partner, those for negotiating the most satisfactory agreement and those for enforcing and monitoring performance.

79

80

81 82 83

84

85

86

R. Craswell, ‘Passing on the Costs of Legal Rules: Efficiency and Distribution in Buyer-Seller Relationships’ 43 Stanford Law Review (1991) 361-398; L. Bernstein “Social Norms and Default Rules Analysis,” 2003 Southern California Interdisciplinary Law Journal, 59; Kaplow / Shavell, n 47 above, 33. Kaplow / Shavell, n 47 above, 33-34, for example, have shown that a direct welfare transfer through taxes is preferable since such a system is much more cost effective. For example, from a less informed or weaker party to a more informed or stronger one. Trebilcock, n 49 above, 58 et seq; Stiglitz, n 67 above, 80-81; Cooter / Ulen, n 69 above, 44-46. Stiglitz, n 67 above, 83-84; Cooter / Ulen, n 69 above, 47; I. Ayres / R.H. Gertner, ‘Filling gaps in incomplete contracts: an economic theory of default rules’ 87 Yale Law Journal (1989) 99-100. Cooter / Ulen, n 69 above, 218-219; J.D. Hanson / D.A. Kysar, ‘Taking Behavioralism Seriously: The Problem of Market Manipulation’ 74 New York University Law Review (1999) 632; C. Jolls / C.R. Sunstein, ‘Debiasing through Law’ University Chicago Law & Economics, Olin Working Paper No 225; R.B. Korobkin / T. Ulen, ‘Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics’ 88 California Law Review (2000) p. 1051. Stiglitz, n 67 above, 81-83; C. Patton / D. Sawicki, Basic Methods of Policy Analysis and Planning (2nd ed, Englewood Cliffs: Prentice Hall, 1993) 193; Cooter / Ulen, n 69 above, 220-225. Ie that not only cross-border transaction costs matter.

The Function of European Contract Law: An Economic Analysis

23

Although imperfect competition87 and market manipulation88 can also induce incomplete contracts, these failures are typically cured by competition law and financial market regulation respectively, and therefore it is not necessary to take them into account in the framework of contract law. In order to remedy the market failures listed above, regulatory action may be envisaged and, usually, contract law as a coherent body of rules is meant to address such problems. However, since market failures are of different kinds, also the regulatory response (type of rule, scope and breadth and so forth)89 has to be tailored to specific needs.

5.1.3. Regulatory failure

As mentioned above, intervening in contractual relations does not come without costs and in fact it may cause several types of government ‘failure’. First, mandatory contract law limits the freedom of parties to conclude the contracts they desire. Since freedom of contract generally maximises welfare, restricting such freedom implies almost by definition that certain contracts cannot be concluded.90 This is why the introduction of regulation, and a restriction in parties’ freedom, should be triggered only by the existence of a problem in the ordinary market mechanism. Second, introducing formalities – eg the mandatory transfer by a civil notary of real estate – in contract law can have a very useful function, but that also implies higher transaction costs for parties. These costs may very well be justified, but the point remains that they must be carefully evaluated.91 Third, in presence of multiple possible interpretations and outcomes of a rule of contract law, the chance of judicial error increases.92 Moreover, a judge adjudicating on a contractual dispute often has to base his or her decision on a certain amount of unverifiable information: the more unverifiable information the bigger the chances of judicial error.93 The possibility of judicial error creates uncertainty,94 which may either deter parties from entering into contracts, or impose additional transaction costs to draft more precise contracts so as to prevent such judicial errors.95 Fourth, closely 87 88 89 90

91 92

93

94

95

Cooter / Ulen, n 69 above, 35-37; Stiglitz, n 67 above, 77-79. Hanson / Kysar, n 84 above. See below the discussion of mandatory vs default rules, rules vs standards etc. See Ayres / Gertner (1989), n 83 above, note 15; C.J. Goetz / R. Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle: Some Notes on an Enforcement Model and a Theory of Efficient Breach’ 77 Columbia Law Review (1977) 554-594. Ayres / Gertner (1989), n 83 above, 123-125. A. Schwartz / J. Watson, ‘The Law and Economics of Costly Contracting’ 20 The Journal of Law, Economics, & Organization (2004) 11. R. Posner, “Is the Ninth Circuit Too Large? A Statistical Study of Judicial Quality”, 2000 Journal of Legal Studies at p. 752; A. Schwartz, ‘Relational Contracts in the Courts: An Analysis of Incomplete Agreements and Judicial Strategies’ 21 Journal of Legal Studies (1992) 282; G.K. Hadfield, ‘Judicial Competence and the Interpretation of Incomplete Contracts’ 23 Journal of Legal Studies (1994) 162; A. Schwartz / R. Scott, ‘Contract Theory and the Limits of Contract Law’ 113 Yale Law Journal (2003) 605-608. Posner (2000), n 93 above, 752-753; R. Scott, ‘The Case for Formalism in Relational Contract’ (2000) Northwestern University Law Review 859-860. Posner (1998), n 75 above, 564-565; Bernstein, n 79 above, 78; D. Charny, ‘Hypothetical Bargains: The Normative Structure of Contract Interpretation’ 89 Michigan Law Review (1991) 1820.

Filomena Chirico

24

related to the broader discussion on European Contract law, on which more will follow, having different sets of contract law in a single market, such as in the European Union, may on the one hand impose transaction costs on parties; on the other hand, it may also have a desirable effect, e.g. to stimulate experimentation and regulatory competition. Fifth, creating contract law and maintaining it comes at significant social cost, which should be worth the investment.96 While in national systems such costs have been incurred already, drafting European contract law requires to face them all. Finally, while contractual default rules are useful to reduce negotiating costs, at the same time, they are costly to contract around when not suitable to the specific contractual pair.97 Moreover, it has been argued that the existence of a certain default rule can make parties believe this rule is the best for them (due to the so-called status quo bias), even when deviating from it may in fact enhance their welfare.98

5.1.4. Mandatory versus default rules99

When drafting contract law, an essential choice is the one between mandatory and default rules.100 While parties are free to contract around a default rule, they cannot do that if a rule is mandatory. Mandatory rules suffer from regulatory failure, because they prevent parties from drafting contracts as they wish. The use of mandatory rules therefore always needs to be justified on the basis of the trade-offs it generates.101 In other words, a mandatory rule can be used (i) when the market failure it cures outweighs the regulatory failure it imposes and (ii) the market failure cannot be (as effectively) cured by default rules. As a rule of thumb, one can say that when a default rule can adequately cure market failure, a default rule should be preferred over a mandatory rule. The DCFR seems to adhere in principle to a general preference for default rules;102 however, the justification for the exceptional mandatory contract law rules contained in the DCFR has to be checked against the principles mentioned above.

5.1.5. Mandatory rules

Mandatory contract law rules are useful when protecting a disadvantaged (third) party. Default rules are often less suitable for this purpose, because the advantaged party often is also in a position to force the other party to contract around such a protective contract rule. Mandatory rules are often a good instrument to cure irrationality, information asymmetry and externalities. This can be established in two ways: (i) when a party needs protection 96 97 98

99

100 101 102

L. Kaplow, “Rules Versus Standards” 42 Duke Law Journal p. 557 (1992) Ayres / Gertner (1989), n 83 above. S.J. Schwab, ‘A Coasean experiment on contract presumptions’ 17 Journal of Legal Studies (1988) 237-268; R.B. Korobkin, ‘The Status Quo Bias and Contract Default Rules’ 83 Cornell Law Review (1998). Also denominated ‘facilitative’ rules. See, for instance W. Kerber / S. Grundmann, ‘An optional European Contract law code: Advantages and disadvantages’ European Journal of Law & Economic 2006, 215. I. Ayres, ‘Valuing Modern Contract Scholarship’ 112 Yale Law Journal (2003) 885. See Ayres / Gertner (1989), n 83 above, note 15; Goetz / Scott (1977), n 90 above. ‘A contract law provision is non-mandatory, unless it is stated to the contrary’.

The Function of European Contract Law: An Economic Analysis

25

against another party, giving the weaker party the right to invalidate those contract terms inducing the irrationality, information asymmetry and externalities and (ii) when a party needs protection against itself, contract terms inducing irrationality should be rendered void automatically. When drafting mandatory contract law, attention has to be paid to the distinction between rules and standards.103 The line may not be too easily drawn, but in general we can say that rules are strict norms, specifying more in detail the content of the law, while standards are more open norms, where a judge specifies the content of the law and develops the implementation of the standard in the specific case, after parties have acted.104

5.1.6. Default rules

Default rules are usually suitable for the purposes of reducing transaction costs. This is the general sense in which the DCFR refers to default rules as contributing to promoting economic welfare.105 The basic idea is that all parties for whom a certain default rule is desirable, can save transaction costs by not having to draft the rule themselves, while parties who consider the default undesirable are free to contract around it.106 The rule-standard dimension mentioned above is important also with respect to default rules. As a rule of thumb, a default rule is set in the way the majority wants (majoritarian default),107 while a default standard has to be applied by a judge in the way parties would have done themselves had they taken the time to regulate the issue in their contract (would-have-wanted standard).108

103

104 105 106

107

108

Kaplow (1992), n 96 above. H. B. Schaefer “Legal Rules and Standards”, German Working Papers in Law and Economics, Vol. 2002, No. 2, 2002, Available at SSRN: http://ssrn.com/abstract=999860; V. Fon and F. Parisi, “On the Optimal Specificity of Legal Rules”, Journal of Institutional Economics, 2007, p. 147-164; Y. Feldman, A. Harel “Social Norms, Self-Interest and Ambiguity of Legal Norms: An Experimental Analysis of the Rule vs. Standard Dilemma”, in Review of Law and Economics, 2008, p. 81 Kaplow (1992), n 96 above, 560. See the Introduction, as referred to above in Section 3. C.J. Goetz / R. Scott, ‘The Mitigation Principle: Toward a General Theory of Contractual Obligation’ 69 Virginia Law Review (1983) 971; C.J. Goetz / R. Scott, ‘The Limits of Expanded Choice: An Analysis of the Interactions between Express and Implied Contract Terms, 73 California Law Review (1985) 266; Ayres / Gertner (1992), n 72 above, 734; Charny, n 95 above, 1820. Goetz / Scott (1983), n 106 above, 971; I. Ayres, “Preliminary Thoughts on Optimal Tailoring of Contractual Rules”, 3 Southern California Interdisciplinary Law Journal 1 (1993); Korobkin, n 98 above, 613-614; Ayres / Gertner (1989), n 83 above, 114. Charny, n 95 above; Ayres (1993), n 107 above, 5-6; F. Easterbrook / D. Fischel, The economic structure of corporate law (Cambridge/Mass: Harvard University Press, 1991) 15, 34, 250; Kaplow (1992), n 96 above, 619. It should also be noted that the use of default standards is – due to high risk of government failure (in the form of judicial error) – highly questioned by some law and economics scholars, who state that the use of defaults should be minimized in order to encourage parties to draft as complete contracts as possible by themselves (formalistic approach). Schwartz / Scott, n 93 above, 594-608; Scott, n 94 above, 871; Posner (2000), n 93 above, 752; J. Kraus / S. Walt, The jurisprudential foundations of corporate and commercial law (Cambridge: Cambridge University Press, 2000) 198-199.

Filomena Chirico

26

While parties usually have to actively opt-out from default rules, this principle can also be turned around. Safe harbours are rules that only apply once parties actively have opted-in through the use of some predefined ‘magic words’.109 Safe harbours enable parties to include complex terms into their contracts at very low transaction costs. In exceptional circumstances default rules can also be used to reduce information asymmetry110 and irrationality.111 By setting a default rule in a way a party with information or rationality advantage does not want it, this party will be forced to contract around it and thereby revealing private information to the other party and/or giving to the other party the opportunity to make a rational decision (penalty defaults).112 When setting up a system of contractual defaults, a justified choice has to be made between majoritarian defaults, the would-have-wanted standard, the formalistic approach, safe harbours and penalty defaults.

5.2. The appropriate regulatory level – function of ‘European’ rules The economics of federalism provide indications as to the function of rules set up at the European level, rather than at lower ones.113 There is a significant body of scholarly contributions on how such economic tools can be useful in providing guidance for the drafting and the assessment of European Contract Law Principles.114 The basic case for Europeanisation should be based on the circumstance that the costs arising from the existence of different legal systems outweigh the benefits, after discounting the costs associated to the harmonisation intervention.115 It is important to bear in mind that divergence is not only a ‘cost’ for parties. Diverging legal systems may be the appropriate response to divergence in local preferences; moreover they may be useful in that they suggest different ways to deal with a certain problem. However, when diverging systems 109

110

111 112 113

114

115

Ayres (1993), n 107 above, 16; Goetz / Scott (1985), n 106 above, 282; R. Craswell, ‘Offer Acceptance and Efficient Reliance’ 48 Stanford Law Review (1996) 552; Goetz / Scott (1983), n 106 above, 1024; Ayres (2003), n 100 above, 899. Ayres / Gertner (1989), n 83 above, 91; Ayres (2003), n 100 above, 890-891; Korobkin, n 98 above, 618-619. See implicitly Ayres / Gertner (1989), n 83 above, 123-125. Ayres / Gertner (1989), n 83 above, 91 et seq. Inter alia: L.E. Ribstein / B.H. Kobayashi, ‘An Economic Analysis of Uniform State Laws’ 25 Journal of Legal Studies (1996) 131, at 138 et seq. Inter alia: H. Wagner, ‘Economic Analysis of Cross-Border Legal Uncertainty – The example of the European Union’, in Smits (ed), n 60 above; G. Wagner, ‘The Economics Of Harmonization: The Case Of Contract Law’ 39 Common Market Law Review (2002) 995-1023; see also the contributions by R. van den Bergh / G. de Geest, in Grundmann / Stuyck, n 60 above; Kerber / Grundmann, n 99 above, 215; J.M. Smits, Diversity of Contract Law and the European Internal Market Maastricht Faculty of Law Working Paper 2005/9; M. Hesselink, ‘Non-Mandatory Rules in European Contract Law’ European Review of Contract Law 2005, p. 44; N. Garoupa / A. Ogus, ‘A Strategic Interpretation of Legal Transplants’ 35 Journal of Legal Studies (2006) 339-363. Different costs are obviously associated with different types of regulatory intervention. See also F. Chirico / P. Larouche, ‘Conceptual divergence, functionalism and the economics of convergence’, in S. Prechal et al (eds), The Coherence of EU Law: The Search for Unity in Divergent Concepts (Oxford: Oxford University Press, 2007) 463-493.

The Function of European Contract Law: An Economic Analysis

27

communicate, costs may arise due to the presence of ‘failures’. It is interesting to note here that, following the distinction outlined above between market and governmental failures, the costs discussed in this section should, in principle, be characterised as ‘government’ failure, because they flow from the choices of (national) lawmakers. However, the way they present themselves is rather akin to market failures as above described. This ‘paradox’ is due to the fact that when we consider different jurisdictions and the case for harmonisation or federalisation, national legal systems are conceptually treated as if they were competing in an international market for legal rules, while the ‘central government’ in this case is the federal or the supranational one. The main determinants to be taken into consideration are: 1. Cross-border externalities: choices made in a certain jurisdiction impose costs to another jurisdiction, in which case the choice of the first jurisdiction is not based on a complete picture of costs and benefits (tradeoffs) involved. Although contract law seems usually not too prone to such kind of ‘failure’, nevertheless the relevance of externalities cannot be excluded at the outset. A partly different type of externality can be envisaged when a set of rules has been delineated to satisfy local preferences, eg a particular formality, but does not include consideration of the costs imposed on cross-border traders, which may find it very costly to comply with such formality. In such case, the externality does not affect other jurisdictions but is caused by the cross-border activity and affects different operators within the same jurisdiction. 2. Cross-border transaction costs: participants in trade incur a certain amount of costs when more than one legal system may apply to the transaction. First, they must acquire knowledge about the legal system in other jurisdictions; secondly, they must incur the costs necessary to draft contracts according to each legal system in which they are doing business; third, they must bear the costs of possible litigation under multiple legal regimes. The risks of unexpected changes in each of the legal systems concerned by the transaction also represent a cost for cross-border economic actors and so on.116 Cross-border sellers are pushed to adapt products, terms and conditions, etc, to the legal requirements of a number of jurisdictions, thereby increasing the cost of production and consequently the price. Higher prices and increased risks may eventually deter buyers (especially consumers) from purchases outside of their jurisdiction.117 In the dynamic perspective, on a macro-economic level, we might observe a reduction in international trade volume, level of investment, consumption and income and ultimately in the economic growth.118 The actual significance of these effects is however far from established. 3. Economies of scale: there might be occasions where economies of scale may justify the need of a uniform solution. This is especially the case when problems of complex technical nature can be better and more cheaply dealt with at centralised level, instead of having multiple players duplicating the same effort. 4. Advantages of diversity: the existence of different solutions in different systems has not only costs but also benefits. Differences in the law may just reflect different preferences, with the consequence that a change towards a harmonised solution may reduce citizens’ 116 117

118

On the costs of diversity, see Ribstein / Kobayashi, n 113 above, 138 et seq. ‘With respect to consumers, transaction cost analysis can reinforce rights-based arguments: the right of a person to be treated the same way irrespective of the legal system in question can be justified because it is deemed unacceptable that persons should bear the transaction costs associated with divergent legal systems.’ Chirico / Larouche, n 115 above. More extensively on this, see H. Wagner (2005), n 114 above.

28

Filomena Chirico

satisfaction.119 Decentralised solutions also bring about benefits in terms of experimentation.120 Moreover, when jurisdictions communicate with one another, even when the legal solution chosen by one jurisdiction would not be optimal, the existence of different approaches may still be beneficial. Without wanting to rehearse all the relevant literature,121 a process of regulatory competition or emulation may spur national lawmakers to improve their contract law to catch up with or maybe outperform a neighbouring jurisdiction. 5. Costs of harmonisation: the production of harmonised rules is costly, involving extensive background studies and discussions. Costs also arise because of the need of ‘developing new bureaucracies or demolishing old structures’.122 Other important costs are those incurred by economic (and legal) operators to adapt to the new rules. Adaptation costs also affect the legal system as such, since further changes might be needed to different areas of law not touched by the harmonisation exercise.123 Different costs are obviously associated with different types of centralisation initiatives and thus have to be specifically evaluated.124 A study of the impact of the costs of harmonisation in the contract law sector, has been performed for the directive on unfair practices125 and has shown that, due also to local enforcement, the cost-reduction function of harmonisation is bound to be much less effective than it is usually hoped. Most of the existing literature on the economics of Europeanisation of Contract Law has thus far focused mainly on finding – or criticising – the justification for a comprehensive contract law ‘code’ at European level, if it is to be optional or replace national laws and on the best instrument to achieve the stated goals. In some cases, attempts were made to push the discussion one level further, from the abstract analysis of the whole body of contract law to the application of the economic tools and reasoning to specific types of legal rules.126

119

120

121

122 123

124

125 126

This is not always true: legal rules may have been enacted without sufficient information and perpetuated by path dependence; or they may have been pushed though by pressure groups. In any event, it should not be excluded at the outset that a certain legal approach truly reflects citizens’ preferences. See F.A. Hayek, ‘Competition as a Discovery Procedure’, in F.A. Hayek (ed), New Studies in Philosophy, Politics, Economics and the History of Ideas (London: Routledge, 1978) 179. See inter alia, A. Ogus, ‘Competition Between National Legal Systems: A Contribution of Economic Analysis to Comparative Law’ 48 International Comparative Law Quarterly (1991) 405; van den Bergh, n 19 above, 435; D. Esty / D. Geradin, ‘Regulatory Competition and Co-opetition’, in D. Esty / D. Geradin (eds), Regulatory Competition and Economic Integration: Comparative Perspectives (Oxford: Oxford University Press, 2001); C.M. Radaelli, ‘The Puzzle of Regulatory Competition’ 24 Journal of Public Policy (2004) 1. H. Wagner (2005), n 114 above. It has to be taken into account also the possibility that divergent solution may be reproduced at the national or even infra-national level in the enforcement phase, if local judges give diverging interpretations. See the issues associated to: optional vs mandatory, opt-in vs opt-out, substantive rules vs private international law rules. The scope of application of the rules also affects the level of costs. Gomez, n 16 above. Kerber / Grundmann, n 99 above, differentiated between mandatory substantive rules, mandatory information rules and facilitative rules. A. Ogus has also differentiated between mandatory and

The Function of European Contract Law: An Economic Analysis

29

A further contribution to this discussion would be to perform this analysis with respect to specific provisions or principles,127 thereby assessing not only whether the rule or principle conforms to the indications of economic analysis of contract law, but also whether it is best placed as part of European Contract Law or is to be left at decentralised level.128 A few additional points are to be emphasised. A first remark is an empirical one: while cross-border transaction costs are theoretically a factor favouring harmonisation, in the concrete case of European Contract Law a serious estimate of such costs has not been so far performed and it is doubtful whether it could be properly carried out.129 A second observation concerns the fact that the efficient functioning of internal market does not only depend on contract law, but also on, inter alia, the rule of law, public regulation of the market. Hence the additional benefits from the harmonisation of contract law, in terms of enhancing the smooth functioning of the market, are likely to be lower than expected. The other stated ‘European’ aims of the DCFR, in particular that of contributing to the establishment of an area of freedom, security and justice, seem prima facie not to have a match in standard economic analysis. They could be interpreted as a reference to freedom of choice that goes beyond static short-term effects on welfare; or, they could also be seen as a way of reintroducing, with a European flavour, some of those distributional concerns that the DCFR drafters indicated elsewhere130 as being ‘less concerned with’. As mentioned already, however, economic analysis suggests scepticism about the opportunity to use contract law to address such issues.131

6. Conclusion The purpose of this paper has been that of addressing a fundamental but somewhat neglected issue regarding the discussion of the function of European Contract Law. At a time when both academics and political actors show their interest in drafting a comprehensive body of rules meant to be available to market operators across Europe,132 a thorough analysis of the aims of such legal instrument is fundamental and cannot be sufficiently addressed by an allencompassing list of values that European Contract Law is expected to pursue. The work tried to show how using an economic method can be of help at the core of European Contract Law, to identify its function(s) and to provide guidance as to the appropriateness of the rules set to become part of it. Such guidance includes both suggestions as to

127 128

129

130 131 132

default rules. See A. Ogus, ‘The Economic Basis of Legal Culture: Networks and Monopolization’ 22 Oxford Journal of Legal Studies (2002) 420. For example, the principles governing contractual remedies, the principle of good faith etc. Another useful contribution can be found in the economic approach to the well-known legal ‘subsidiarity principle’ to be used as a basis ‘for the optimal assignment of powers to different tiers of government in the multi-tier Union’. J. Pelkmans, ‘Subsidiarity between Law and Economics’ College of Europe Research Papers in Law 1/2005. H. Beale, ‘The European Commission’s Common Frame of Reference Project: a progress report’ European Review of Contract Law 2006, 303. DCFR introduction, paragraph 24. See supra. See for example, the metaphor of the ‘Blue button’ (the European blue flag) to be clicked on ecommerce website to trigger the applicability of European Contract Law and of the CFR.

Filomena Chirico

30

the appropriate regulatory level, going beyond the incomplete cross-border transaction costs argument, as well as indications as to what constitutes contract law of good quality. Given its broad scope, this paper cannot and does not go in the details of the specific DCFR provisions. This is nevertheless a worthy endeavour which the economists and lawyers involved in the DCFR CoPECL project133 are pursuing in further research. 133

As well as others not connected to it.

The Economic Function of Good Faith in European Contract Law Filomena Chirico*

Introduction Among legal scholars and practitioners, ‘good faith’ is often referred to as a fundamental principle, placed at the heart of the law and of all human interactions.1 It is usually intended as to include duties of fairness, honesty, loyalty, disclosure of information and cooperation with contractual counterparties.2 However the precise definition of good faith as well as its practical role in solving disputes is anything but uniform and uncontroversial. Restricting the observation to the contractual sphere, different jurisdictions have showed different attitudes to such principle and currently there seems to be no generally accepted approach. Against this background, the task of the groups3 in charge of drafting the Common Frame of Reference (CFR) for European Contract Law appears a daunting one with respect to the adoption of a uniform approach to good faith. It is maybe due to such difficulties that in the Draft CFR (DCFR) delivered at the end of 2007,4 a general principle of good faith applying to all situations during the contractual life does not appear as such but has been split into separate provisions, differently from previous drafts and from the so-called Lando principles.5

*

1 2

3

4

5

At the time this article was written, she was Assistant Professor and Senior member Tilburg Law and Economics Center (TILEC), Tilburg University, coordinator of the CoPECL Economic Impact Group. The author wishes to thank the participants to the Third Roundtable of the CoPECL Economic Impact Group in Venice, the participants to the Annual Conference of the Italian Society of Law and Economics in Bologna and the participants to the TILEC WIPs for their useful comments. Hugo Grotius’ De jure belli ac pacis is usually quoted in this connection. So-called ‘objective’ good faith. By contrast, ‘subjective’ good faith refers to the ignorance of certain facts. Subjective good faith is not the subject of this paper. The Study Group on a European Civil Code and the Research Group on EC Private Law (Acquis Group). Christian von Bar et al., eds., Draft Common Frame of Reference (DCFR), Interim outline edition (Munich: Sellier, 2008). The Interim outline edition was used as the basis for this contribution. In the meantime, the final DCFR is available (Christian von Bar et al., eds., Principles, Definitions and Model Rules of European Private Law – Draft Common Frame of Reference (DCFR), Outline edition (Munich: Sellier, 2009). Principles of European Contract Law, Ed. by: Ole Lando et al. Kluwer Law International.

Filomena Chirico

32

1. Good faith in the law Defining good faith is a very difficult endeavour. Moreover, definitions are bound to reflect the historical evolution and the function that is assigned to such principle in different legal systems. Broad references to duties of honesty, fairness and cooperation with the contractual counterparty are very common, while a more precise determination is usually left to a case by case assessment. It is often heard that “you recognise good faith when you see it” and it has been argued that good faith can only be understood by reference to bad faith.6 In international law, good faith is said to be a “fundamental principle from which the rule pacta sunt servanda and other legal rules distinctively and directly related to honesty, fairness and reasonableness are derived”. The actual application of the rules is to be determined “at any particular time by the compelling standards of honesty, fairness and reasonableness prevailing in the international community at that time”7 Despite the example of the above reference to the principle pacta sunt servanda as part of good faith, the function of such duty is usually said to be rather that of limiting party autonomy and freedom of contract.8 Interpreted in this way, good faith is immediately put in opposition to the function of contract law, as shaped by the principle of freedom, i. e. expected to maximise contractual parties’ satisfaction pursuant to their own preferences. The survey conducted by Whittaker and Zimmermann in 20009 shows the existence of a varied legal landscape in Europe. To sketch just a few approaches: the German Civil Code expressly refers to good faith in relation to contractual interpretation and performance, while pre-contractual responsibility is recognised when causing the other party’s reliance. Despite earlier attempts at introducing good faith, English Common Law10 does not recognise a general duty as such11 either in performance or in negotiations, except for the case of misrepresentation.12 However, even English Law offers (piecemeal) solutions to many of the issues of “unfairness” that are dealt with through recourse to good faith in other legal systems. In France, the Civil Code imposes good-faith performance and imposes to contractual parties obligations deriving from “equity”, in connection to other legal doctrines, such as the “abus de droit”.13 Duties to disclose and certain obligations of cooperation and loyalty complete the picture.14 Belgian law, inspired by the French one, has showed a somewhat more extensive 6

7 8 9

10

11

12

13

14

In the American scholarship, for example, Robert Summers “Good faith in general contract law and the sales provision of the Uniform Commercial Code”, 54 Va. L. Rev. 195 (1968): good faith as an “excluder”. O’Connor, Good Faith in International Law, Dartmouth Publishing Company. 1991. See also in the DCFR at II.–1:102. Good Faith in European Contract Law, Reinhard Zimmerman and Simon Whittaker (eds), Cambridge, Cambridge University Press, 2000. To be noted that other common law jurisdictions (for example the United States) have chosen different approaches. However, this is by no means the result of a monolithic and constant attitude of the English legal community, see the report by Zimmermann and Whittaker, cit. at footnote 9 page 42 et seq. Walford v Miles, see Nili Cohen chapter 2 in Beatson and Friedmann (eds) Good Faith and Fault in Contract Law, Clarendon Press 1995. It has been noted that in fact the original motto was “if it is contractual, it is fair” thus leaving no room for limitations of the parties’ contractual freedom on fairness bases. See Whittaker and Zimmermann, cit. at footnote 9 page 34. However, the function of such obligations is disputed among jurists.

The Economic Function of Good Faith in European Contract Law

33

recourse to good faith. The Italian Civil Code refers to good faith in the context of negotiations, interpretation and performance of contract, and imposes a duty of good faith on both debtor and creditor in relation to obligations in general. However, the judicial reliance on the principle of good faith to solve disputes is quite limited. The new Dutch Civil Code attaches to contracts also obligations arising from good faith and declares inapplicable all contractual arrangements contrary to good faith. Besides appearing in national legal systems, good faith has also been analysed in the context of international law and is contained in some international law instruments, for example in the UN Convention on Contracts for the International Sale of Goods,15 albeit not as a “traditional” duty imposed on contractual parties.16 Good faith has made its appearance also in European Law, notably through the oft-cited directive on Unfair Terms and Conditions17 which has inspired some of the DCFR provisions, meant to incorporate the Acquis. Through EU law, good faith has penetrated also legal systems previously showing resistance to the adoption of the concept as such,18 although the measure of success of such introduction is controversial.19 Broadly speaking, it seems that a principle that performance of a contract should be done in good faith is compatible with or at least conceivable in the majority of the legal systems, while there is much more controversy surrounding the imposition of a duty of good faith in negotiations. In fact, even when pre-contractual good faith is recognised, extra-contractual liability or equity are invoked, rather than standard contract law.20

2. Good faith in the DCFR The DCFR considers “Good Faith and Fair Dealing” as an “objective standard of conduct.”21 The subjective notion of good faith also appears in the DCFR, as a mental attitude characterised by an absence of knowledge.22 However such subjective notion will not be analysed in this paper. 15

16

17

18

19

20

21

22

A review of the case law concerning the interpretation of the concept is in Benedict Sheehy “Good Faith in the CISG: The Interpretation Problems of Art. 7” Pace International Law Review – Special Edition “Review of the Convention on Contracts for the International Sale of Goods” 20042005. As a result of a compromise between contrasting views over the duty of good faith and the role of the judges enforcing it. Council Directive 93/13/EEC of 5 April 1993 on Unfair Terms in Consumer Contracts, OJ L 95, 21.4.1993, p. 29-34. Of course, the absence of an express principle of good faith in a given legal system did not necessarily imply a different approach to the solution of contractual disputes. See G. Teubner “Legal irritants: Good Faith in British Law or How Unifying Law Ends Up in New Divergences”, in (1998) 61 Modern Law Rev. 11. For the French and American legal systems, see N. E. Nedzel, “A Comparative Study of Good Faith, Fair Dealing, And Precontractual Liability” (1997) 12 Tul. Eur. & Civ. L. F. 97 at 155. See Annex to the CFR on file with the author. Please note that at the time of writing, the accessible version of the Comments and Notes (July 2008) is still unofficial and may have been subsequently changed. Person “neither knew nor could reasonably be expected to know”. See Annex to CFR and provisions on representation, unjustified enrichment etc.

34

Filomena Chirico

The DCFR does not contain a real definition of good faith and fair dealing. However, in relation to the provision on performance,23 the (draft) comments to the DCFR refer to: “community standards of decency and fairness”, “due regard for the interests of the other party”, a requirement “not to act out of pure malice”. The use of an hendiadys, always combining “good faith” with “fair dealing”, seems to suggest the adoption of a more objective standard, based on known trade practices and to “fairness in fact” as opposed to internal motivation.24 The Lando Principles, as well as earlier drafts of what would become the DCFR did in fact contain a general clause concerning good faith in all stages of contractual life: II. – 1:105: Good faith and fair dealing (PECL 1:201) (1) A person has a duty to act in accordance with good faith and fair dealing in negotiating or concluding a contract or other juridical act and in exercising any right or performing any obligation arising from it. (2) The contract or juridical act may not exclude or limit this duty. (3) Breach of the duty does not give rise directly to the remedies for non-performance of an obligation but may preclude the person in breach from exercising or relying on a right which that person would otherwise have.

The published DCFR has abandoned this single and general formulation and has maintained a reference to “good faith” in a number of rules, each applying to a different stage of the contractual life, whenever good faith may be relevant. Accordingly, no general rule is proposed concerning common remedies for all instances of breach of the duty. In particular, good faith is mentioned in relation to the interpretation of the DCFR itself25 as well as of any contract.26 Good faith and fair dealing is, moreover, explicitly mentioned at the beginning of Book II as a limit to party autonomy.27 Good faith shall guide pre-contractual negotiations28 as well as performance of the obligations arising from the contract.29 Following, and arguably expanding, the current acquis,30 good faith is part of the two-step determination of unfairness of non-negotiated (or standard) terms.31 Good faith and fair dealing, unlikely the majority of the provisions in the DCFR, is indicated in II.–1:102: as one of the mandatory clauses that parties are not free to overrule. This general statement of principle is not accompanied by the specification of consequences and remedies for breach of such mandatory rule. The comments attached to this provision clarify that not all violations of mandatory rules imply cancellation of the contract and that the actual consequences for breach of good faith have to be searched in the specific provisions making reference to good faith.

23 24 25 26 27 28 29 30

31

Article III.–1:103. See Comments to article III.–1:103. I.–1:102. II.–8:102:, II.–9:101. II.–1:102. II.–3:301:, but see also provisions on mistake, fraud and misrepresentation. III.–1:103. In particular, the Council Directive 93/13/EEC of 5 April 1993 on Unfair Terms in Consumer Contracts, OJ L 95, 21.4.1993, p. 29-34. II.–9:404:, II.–9:405: and II.–9:406.

The Economic Function of Good Faith in European Contract Law

35

However, the general assertion of the mandatory nature of good faith is qualified, at least in the case of performance,32 by the (draft) comments attached to the text of the DCFR: the comments explain that courts are expected to use good faith to fill in contractual gaps, and not to “correct” the contract or make it more “fair” than what the parties have explicitly regulated.33 Good faith in the DCFR is said to be a duty, not an obligation, thus implying that ordinary remedies may not be all directly34 available: specific performance is not possible and bad faith on one side does not legitimate reciprocating, while recovery of damages is in principle available. In case of pre-contractual good faith, breaking off negotiations and other “bad faith” behaviours are a source of liability for “reliance damages”, i. e. the aggrieved party will receive compensation for expenses and losses incurred and work done (and in certain cases also for the lost opportunities) in reliance on the other party’s statement and behaviour as to the expected conclusion of the contract. With regards to performance and in exercise of rights good faith is still considered a “duty” and not a “true” obligation. Hence, under this construction, courts may not impose damages when the letter of the contract has been respected, although the clause may, in abstract terms, be contrary to good faith. In consequence of lack of good faith (i. e. of decency and fairness) in performance, a party may be precluded from taking advantage of certain contractual terms which would otherwise be available.

3. The economic function of the principle of good faith Good faith as a general duty has not been very often under the lenses of economic analysis scholarship, while it is easier to find economic analyses of more specific implementations of such “principle”. It seems reasonable to start the analysis with the observation that the function of the duty of good faith (in negotiation as well as in performance) has to be put in the context of the function of the whole of contract law.35 In very general economic terms, the final goal of contract law is to help market operators to maximise their own welfare (hence maximising the welfare of society). Contract enforcement by courts prevents mistrust of counterparties, discourages opportunistic behaviour and encourages trade. Contract law enables people to cooperate, encourages efficient disclosure of information, secures optimal commitment to performing as well as optimal reliance, minimises transaction costs, fosters enduring relationships.36 The duty of good faith, as a rule of contract law, is expected to serve the same economic purpose, similarly to other “general principles”, such as that pacta sunt servanda: in an economic view, both those “principles” are considered and assessed not as values in themselves but in their suitability to increase the value of exchanges and hence welfare. 32 33 34

35

36

Article III.–1:103. Comments to article III.–1:103. However, good faith may be the basis for an implied term which in turn gives rise to a specific obligation. See F. Chirico “The Function of European Contract Law. An Economic Analysis”, in ERCL n. 4 2009. See Cooter & Ulen Law and Economics, Pearson 2003 p. 195 et seq.

36

Filomena Chirico

Good faith can be considered an implied term pursuant to which parties will not take advantage of “the vulnerabilities created by the sequential character of contractual performance”.37 In other words, it is reasonable to assume that if contractual parties had thought about the possibility of being abused by the other one, they would have explicitly prohibited. As judge Posner put it in the Market Street Ass. case, the duty of good faith, is “not a duty of candor” and does not prevent form entering into a contract to purchase something that the counterparty undervalues.38 Economic analysis has emphasised in particular two issues in which good faith plays a role: optimal disclosure of information and prevention of opportunistic behaviour. Information disclosure seems to be a more peculiar problem for the pre-contractual phase,39 while the risk of opportunism is certainly present during the whole contractual relationship. In the same case referred to above, judge Posner stated that the concept of good faith is an approximation of “the terms the parties would have negotiated had they foreseen the circumstances that have given rise to their dispute”. Since parties want to minimise the costs of performance, a doctrine of good faith “is a reasonable measure to this end, interpolating it into the contract advances the parties’ joint goal”.40 Schäfer & Ott synthesise the normative evaluation of good faith by referring to the criterion of whether “good faith increases or decreases the value of a resource”41 and indicate the four conditions to make such assessment: presence of asymmetric information costs, the information is productive, there is a good faith premium, there is a risk of opportunism. Mackaay and Leblanc42 consider good faith as the opposite of opportunism. As opportunism is a very broad term encompassing a wide range of behaviours, the law needs an equally open ended and flexible instrument, which would be both able to accommodate the responses to opportunism, and compatible with the rule of law. Again, asymmetry in the position of parties is considered precondition for bad faith (or opportunism) as well as the undue advantage for one of the parties in excess of a certain level of tolerance.43 Russi44 compares the use of a good faith standard to determine the “fairness” of a behaviour in the individual case before the court (having regard to the preferences of the specific parties) with the standard to assess the “reasonableness” of a conduct having regard to the “average” behaviour. He concludes that the two standards are complementary with respect to good faith performance. 37 38

39

40 41 42

43 44

R. Posner, Economic analysis of the law, Aspen Pub. 2003, p. 95. Market Street Ass. See also the comment by Todd D. Rakoff “Good Faith in Contract Performance: Market Street Associates ltd. Partnership v. Frey”in 120 Harvard Law Review 1187 (2007). See also Wils, Wouter P. J. (1993), ‘Who Should Bear the Costs of Failed Negotiations? A Functional Inquiry into Precontractual Liability’, 4 Journal des Economistes et des Etudes Humaines, 93-134, and G. De Geest “Comparative law and economics and the design of optimal legal doctrines”, in Law and Economics in Civil Law Countries, Ed. by B. Deffains, T. Kirat, Routledge 2003. Judgment Market Street Associates. Hans-Bernd Schäfer, Claus Ott, The economic analysis of civil law, Edward Elgar 2004, page 385. E. Mackay and V. Leblanc “The economics of good faith in the civil law of contract”, paper Nancy 2003. See also E. Mackaay and S. Rousseau, L’analyse économique du droit, Dalloz (Paris) / Éditions Thémis (Montreal) 2008, page 379 et seq. Id., p. 11. Luigi Russi “Can good faith performance be unfair? An economic framework for understanding the problem” 29 Whittier Law Review 565.

The Economic Function of Good Faith in European Contract Law

37

4. Analysis of the DCFR rules The duty of good faith in the DCFR is considered as a general clause or standard, encompassing different behaviours, as compared to a specific rule directed at regulating a particular case. It is stated to be mandatory and applicable, with specificities, in different stages of the contractual life. In general, it can be noted that the text of the DCFR and of the comments betray the many difficulties encountered in devising a uniform rule for such a multi-faceted concept. Despite the use of the same terminology of “good faith and fair dealing” in different provisions, the function of the duty – hence its content – seem to be peculiar to the specific contractual situations envisaged (negotiations, performance, interpretation, determination of “unfairness” of clauses). Moreover, the relationship of good faith with other legal devices, such as the rules on fraud and misrepresentation and on long-term contracts, is not fully spelled out, thus leaving uncertainty as to the actual scope of application for the provisions on good faith. The following subsections will address more in detail from an economic perspective some contentious issues arising from the legal choices of the DCFR, as outlined above.

4.1. Mandatory nature II.–1:102: Party autonomy (1) Parties are free to make a contract or other juridical act and to determine its contents, subject to the rules on good faith and fair dealing and any other applicable mandatory rules. (…)

In the law and economics literature, it is widely claimed that good faith has an important role to play as long as it remains a facilitative or default rule, as opposed to a mandatory rule, in order to prevent opportunism.45 Accordingly, parties must be free to overrule the legal standard if they want so, since this will maximise their joint surplus. The reasoning behind the above claim starts from the acknowledgment that in contracts where performance takes place over a period of time, different contingencies may materialise, thus risking to frustrate the interests of the contractual parties. To eliminate such risk, parties could specify all such events ex ante. However, specifying all contingencies in a contract is very costly, which is the source of contractual incompleteness. In such situation, a legal provision containing an implied term of good faith will serve a useful economic purpose if it allows to cover a number of contingencies without the need for the parties to specify them. In other words, good faith helps parties to reduce the transaction costs connected to the contract.46 In this sense, a gap-filling rule concerning a duty of good faith is in principle, and subject to the choice of the appropriate design, desirable. Parties wanting something different from 45

46

See O. Williamson, for example, referring to a general clause to protect from “the hazards of contractual incompleteness”. Williamson, Oliver E., The Economic Institutions of Capitalism, New York, The Free Press, 1985. See also R. Posner, cit. at footnote 36. “ ‘Good faith’ is a compact reference to an implied undertaking not to take opportunistic advantage in a way that could not have been contemplated at the time of drafting, and which therefore was not resolved explicitly by the parties.” Kham & Nate’s Shoes No. 2, Inc. v. First Bank, 908 F.2d at 1357, quoted by judge Posner in Market Street Associates.

38

Filomena Chirico

what would normally be considered good faith behaviour (e. g. performance in the middle of the night or withdrawal from a long-term contract with a very short notice) would always have the option of introducing in the contract a specific clause in the desired direction. It is worth noting that the drafting of the DCFR and of the comments suggest that good faith is to remain in the background whenever the parties have “carefully regulated” their relationship. While this remark seems to correspond to a “facilitative” role for the duty of good faith, at the same time it seems to encourage the parties to spend time negotiating as many clauses as possible, thus implicitly denying the cost-saving function described above. Good faith could still serve an economic function if it is used to force information disclosure:47 when a party has very peculiar needs, such rule would force openness and disclosure, thereby reducing ex post enforcement costs. However, good faith in the DCFR seems set to prevail over explicit contractual clauses and constitute an effectively mandatory provision. From an economic perspective, such a regulatory choice would have to be justified on the basis of the existence of a “failure” in the contractual mechanism that cannot be corrected with other means. It is difficult to imagine what kind of failure could take place in this case. As mentioned above, beyond references to decency, morality and cooperation with the counterparty, it seems that usually the problem that the law tries to correct through a duty of good faith is an asymmetry in information which could generate a risk of opportunism and thus decrease the value of a contract for the parties (hence for society). However, non mandatory rules are usually enough to correct this kind of problems and in forcing disclosure of relevant information.48 In abstract terms, a mandatory duty of good faith would be justified in case of unavoidable contractual incompleteness, which is normally the case in reality, coupled with a structural risk of opportunism,49 which is probably restricted to particular circumstances rather than a general condition of contractual relationships. A tension seems to transpire out of the text of the DCFR which states in several occasions in the formal rules50 the mandatory nature of the fundamental duty of good faith, but then adds comments referring to its gap-filling function and to the respect of explicit contractual clauses. Given the different nature of the comments as compared to the black letter rules, it would be advisable to clarify the nature of good faith by including appropriate references already in the text of the rules or to avoid the mention of the mandatory character altogether. Another correcting device could perhaps be found in the rules on interpretation, where it is stated that judicial gap-filling inspired, inter alia, by good faith (to protect parties from “unacceptable risks”) does not apply when the gap was deliberate and parties had accepted

47

48 49

50

I. Ayres and R. Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 Yale L. J. 87 (1989). Id. See Williamson. Schäfer & Ott cit. at footnote 40, page 392 observe that opportunism is a market failure and that it may affect certain markets more than others (used cars, for example), but it is not clear whether in such context they would recommend a mandatory duty of good faith, prevailing over textual specifications. General principle cited above, repeated in the rule on good faith in negotiations and in good faith in performance: The duty may not be excluded or limited by contract.

The Economic Function of Good Faith in European Contract Law

39

the consequences.51 While it is not completely clear under which circumstances a gap can be presumed to be deliberate, it does seem to suggest that an express contractual clause would prevail over an implied duty of good faith.

4.2. Good faith in negotiations. II.–3:301: Negotiations contrary to good faith and fair dealing (1) A person is free to negotiate and is not liable for failure to reach an agreement. (2) A person who is engaged in negotiations has a duty to negotiate in accordance with good faith and fair dealing. This duty may not be excluded or limited by contract. (3) A person who has negotiated or broken off negotiations contrary to good faith and fair dealing is liable for any loss caused to the other party to the negotiations. (4) It is contrary to good faith and fair dealing, in particular, for a person to enter into or continue negotiations with no real intention of reaching an agreement with the other party.

Good faith in the pre-contractual phase seems to complete (and thus is different from) the provisions on fraud, misrepresentation or threat.52 It especially covers (a) breaking off negotiations and (b) negotiating without real intention of reaching an agreement. Such a provision is very contentious, being especially alien to the common law tradition.53 The idea that good faith may oblige a party to pursue negotiations till the end and to actually conclude the contract, collides with the general principle of party autonomy, as well as with the economic function of contracts of maximising parties’ joint surplus. At a very basic level, if over the course of negotiations, a party realises that the contract is not to his or her advantage, the legal system should not mandate the conclusion of such contract. In fact, at the core, the problem that a duty of good faith in negotiations is meant to correct seems to be that of protection of desirable reliance: parties may make investments and incur costs already while negotiating a contract, when they can rely on the expectation of a positive outcome of the negotiations.54 In an economic perspective, investments in reliance increase the value of a contract, hence the parties’ joint surplus. While reliance on performance of an already existing contract certainly deserves more protection, nevertheless it may be desirable that parties engage into value-enhancing investments already in the negotiation phase, when they can reasonably rely that a contract will be concluded.55 However, not every investment in reliance should be protected by the law, but only efficient reliance. The economic function of a duty of good faith in this phase would therefore be that of protecting only the reliance that is efficient. The more such early investments are necessary and the more relationship-specific they are, the more anticipated protection can be granted.

51

52 53 54

55

II.–9:101: Terms of a contract: (4) Paragraph (2) does not apply if the parties have deliberately left a matter unprovided for, accepting the consequences of so doing. See (draft) comments to article II.–3:301. See, for example, Teubner, cit. supra at footnote 19. W. Wils cit. supra at footnote 38 contended that losses incurred by one potential party should be reimbursed when the other party has misrepresented the chances to conclude the contract or the amount at stake and that this can be achieved through the imposition of a duty of honesty. See Bebchuk and Omri Ben-Shahar “Pre-Contractual Reliance” (2001) 30 J. Legal Stud. 423.

40

Filomena Chirico

The typical remedy in such case would be that of reliance damages, or negative interest, which is consistent with the statements contained in the (draft) comments to the DCFR. However, if the legal device of good faith is chosen to achieve this result, its content should be adequately tailored to cover only as much compensation as to ensure an efficient reliance.56 The comments to the DCFR will explain with illustrations in which cases breaking-off negotiations constitutes violation of good faith. For example, it may happen that already during negotiations and in order to show to the counterparty its ability and commitment, one of the parties delivers a detailed plan with substantial information which the other party then uses to obtain a cheaper contract with somebody else. In such case, we are faced with a relationship-specific investment intended to maximise the value of a contract. Such investment would deserve protection, otherwise potential parties may refrain from supplying all the necessary information for fear of opportunistic behaviour; the counterparty may, in turn, not have enough information to decide whether the transaction is advantageous. At the end, potentially advantageous contracts may not take place. However, to assess the actual risk of underinvestment in reliance, all incentives for the parties to invest have to be considered, including those deriving from pressure by competitors in the market, before concluding that legal intervention is necessary.57 A mechanical application of the duty of good faith without specific assessment of the concrete situations may therefore, be overinclusive. Negotiating with no intention to conclude a contract would also constitute a violation of good faith in the intentions of the drafters of the DCFR. In such case, the issue at stake for the parties is separating “serious” potential counterparties from the non serious ones.58 This feature may be difficult to assess by parties ex ante but it may be non observable even by courts ex post, which would render the legal protection moot. In fact, once again, much depends on the amount of relationship-specific investment that is required to maximise the value of the contract. For substitutable performance (i. e. provision of a good or service which can be obtained by many suppliers) a duty of good faith is redundant at best: competing with alternative suppliers gives incentives to the “serious” parties to signal their intent during negotiations. Conversely, for the case where the required characteristics or skills are scarce there is room for opportunistic behaviour (i. e. hiring of a specific professional who undertakes negotiations even when he is not interested in taking up the post). It can be argued, however, that in this type of cases, the level of sophistication and expertise of the parties could represent an adequate safeguard: the risk of failure has been sufficiently taken into account, reputational concerns play a significant role in more restricted communities etc.

56 57

58

Id. for examples of specific liability rules. Grosskopf and Medina “Regulating Contract Formation: Precontractual Reliance, Sunk Costs, and Market Structure”, 39 Conn. L. Rev. 1977 2006-2007 This case may be akin to Akerlof’s market for lemons. Akerlof, George A. (1970). “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism”. Quarterly Journal of Economics 84 (3): 488-500.

The Economic Function of Good Faith in European Contract Law

41

4.3. Optimal level of specification of the legal rule Leaving aside the more controversial case of good faith in negotiation, another aspect of good faith worth discussing is the optimal level of specification of such “duty”, i. e. whether it should be codified as a general principle59 leaving it to the courts to apply it in the concrete cases, or rather more specific rules should be drafted to govern in detail certain concrete situations of risk. Economic analysis of law suggests that the choice between “rules” and “standards” be dependent on the relative size of the various costs involved with formulation, interpretation and enforcement of the rule.60 A general principle has higher enforcement costs but better adaptability to circumstances.61 Moreover, vague clauses seem best able to reduce enforcement costs by discouraging parties from litigating over non substantial deviations from their wishes.62 On the other hand, with standards, the judicial process becomes more complicated and judicial errors are more frequent due to the difficulty of tailoring the general clause to the specific circumstances and the specific parties.63 Nevertheless, such errors should not be overemphasised either: judges develop professionalism and experience, they routinely rely on trade usages etc. The ability of parties to contract around rules or standards also varies64 and can influence the optimal design of a legal provision. The choice operated in the text of the DCFR seems mixed, probably reflecting once again the general quest for compromise by the drafters of the DCFR rather than a conscious choice between use of specific rules or of a general standard of good faith. Thus, in principle, a general standard is referred to, accompanied by additional provisions applicable in different phases of the contractual life. However, more precise rules have also been codified implementing – explicitly or implicitly – the general standard to regulate specific situations or with regard to specific contracts. Examples of application of the standard are given in the comments and, moreover, a set of more precise rules is explicitly advocated in the (draft) comments with respect to certain jurisdictions.65

59

60 61

62

63

64

65

As mentioned above, the principle is to be expressed in a default or facilitative rule, not as a mandatory duty. See Schäfer “Legal Rules and standards” available at http://ssrn.com/abstract=999860. Louis Kaplow “Rules versus Standards: An Economic Analysis” Duke Law Journal, Vol. 42, No. 3 (Dec., 1992), pp. 557-629, Jason Scott Johnston “Bargaining under Rules versus Standards”, Journal of Law, Economics and Organization, 1995, vol. 11, issue 2, pages 256-81, Parisi, Rules vs. Standards. Encyclopedia of public choice. For an application to contractual clauses, see Álvaro E. Bustos “Litigation and the Optimal Combination of Vague and Precise Clauses in Contracts”, http://ssrn.com/abstract=1153453. Richard A. Posner, The Law and Economics of Contract Interpretation, 83 Tex. L. Rev. 1581 (2005). Ian Ayres “Preliminary Thoughts on Optimal Tailoring of Contractual Rules”, 3 Southern California Interdisciplinary Law Journal 1 (1993), Arndt Christiansen and Wolfgang Kerber “Competition Policy With Optimally Differentiated Rules Instead Of “Per Se Rules Vs Rule Of Reason”.”, JCLE 2006. Comments to Article III.–1:103: See also below.

Filomena Chirico

42

When it comes to good faith, most legal systems seem to prefer use of a standard, rather than of specific rules.66 In history, however, the use of such standard by courts seems to have become widespread under certain circumstances, i. e. when the need arose to develop legal rules for new situations and problems, while its use subsequently shrunk as new legal institutions developed.67 Preference for the use of a more general standard in the context of the DCFR could be based on a series of considerations. On the one hand, a broader legal concept would be able to cover more instances of opportunistic behaviour68 as they arise, without the need to invest time and effort to draft precise rules trying to foresee the different forms that opportunism may take. On the other hand, a decentralised interpretation of the standard69 may make it more adaptable to the local situation and encourage creativity in the development of new legal institutions.

4.4. Harmonisation of the duty of good faith Perhaps even more than with other contract law provisions, harmonisation of good faith at the European level is very controversial among legal scholars. As recalled in the first part of this work, different legal systems have adopted divergent approaches to good faith, with the English one outright refusing a general clause in favour of specific rules. Despite the text of the DCFR contains some provisions referring to good faith, in the comments annexed70 we find express reservations as to the opportunity of extending a general duty of good faith to jurisdictions where it was not in use. The choice is, of course, left to the (future and hypothetical) European legislator, to whom, however, recommendation is made to add more specific provisions for the benefit of the judges not accustomed to the general principle. In the perspective of economic analysis, harmonisation of contract law in general seems to rest on a relatively weak basis71 and the same can be said about the provision on good faith and fair dealing. Additionally, such rule has problems of its own when it comes to the harmonisation issue. Even accepting the plausibility of the arguments in favour of a uniform European Contract Law, this could conceivably be achieved even without the inclusion of a general duty of good faith and fair dealing, which therefore needs a specific analysis. Unfortunately, it does not seem easy to reach a conclusion on the desirability of harmonisation of good faith. Besides the difficulty of assessing the expected costs associated to the choice of a uniform rule on good faith and the usual uncertainty as to the success of its 66

67 68 69 70 71

An alternative explanation in Hesselink “The concept of good faith”, in Towards A European Civil Code, ed. by A. S. Hartkamp et al., The Hague, Boston & London: Kluwer Law International, 2004, pp. 471-498. Id. See Mackaay and Leblanc cit supra at footnote 42. But see the comments in the following section concerning harmonisation of good faith. Available at the time of writing in draft version of July 2008. See, among others, F. Gómez “European Contract Law and Economic Welfare: A View from Law and Economics” in InDret January 2007, F. Chirico, cit. supra at footnote 35, Van den Bergh “Forced Harmonisation of European Contract Law”, in Grundmann and Stuyck (eds.), Academic Green paper on European contract law, Kluwer 2002.

The Economic Function of Good Faith in European Contract Law

43

implementation, additional doubts derive from the lack of clarity on a number of important points. The function(s) of good faith, as the above discussion has showed, can hardly be subsumed under a single heading, when all the instances in which it is relevant are considered (from negotiation to performance and so on).72 Moreover, the discussion in this work has taken an efficiency perspective. However, the DCFR rules may be expected (by its drafters, by the authorities or other stakeholders) to fulfil also other functions, with fairness and redistribution being one candidate.73 In conformity to established scholarship,74 economic analysis recommends caution and careful comparison of the relative efficiency75 of different legal instruments when it comes to redistributive goals. Using contract law for this purpose and especially harmonisation of the rules on good faith may have indirect costs, influence the determination of the price in contractual relationships in unintended ways and ultimately lead to welfare losses. More generally, the lack of clarity concerning the function of the DCFR and its future use as a legislative instrument at the European level76 adds to the difficulty of tackling the harmonisation issue and the mere fact that a good faith provision is generally included in national civil codes is not enough to automatically extend the same approach to a European enterprise.77

5. Conclusion This work attempted an economic analysis of the rules of the DCFR containing reference to the duty of good faith, which in traditional legal scholarship is considered to be a fundamental principle governing contractual relationships in different phases and in the majority of legal systems is considered a mandatory provision of any civil code. Probably reflecting the difficult compromises needed to formulate a rule harmonising different legal traditions with different preferences, the text of the DCFR shows an ambivalent approach to good faith: it reaffirms its central role in contract law, it refers to it in different contractual interactions but at the same time suggests that such mandatory rule is not so mandatory after all nor does it indicate clearly the available remedies. Going beyond the deference for such a “fundamental principle” at the heart all human interactions, an economic outlook on the actual function of this principle can offer a number of useful insights, by recommending to be candid in treating good faith in performance as a facilitative rather than mandatory rule; to prefer information disclosure rules rather than a sweeping good faith duty when it comes to protecting relationship-specific investment in the negotiation phase; to draft good faith as a broad standard, at least as far as contractual performance is concerned. 72

73 74 75

76 77

On the discussion about the multiple functions and the use of good faith, see Hesselink, M. “The concept of good faith”, cit. supra at footnote 64. See F. Chirico, cit. supra at footnote 35. See for example, A. Ogus, Costs and Cautionary Tales. Economic Insights for the Law, Hart 2006. In order to minimise the welfare costs connected to the use of one or the other legal tool to achieve the desired distributional goal. Tool box, model law, revision of the acquis etc. See F. Chirico, cit. supra at footnote 35. See Jan Smits “The Draft Common Frame Of Reference, Methodological Nationalism And The Way Forward”, in European Review of Contract Law 2008, pp. 271-281.

44

Filomena Chirico

In general, however, a compelling case for harmonising at the European level a duty of good faith cannot be made. A thoughtful decision as to the function and of the future use of the DCFR in the context of the European contract law is a first necessary step before possibly concluding that a duty of good faith has a useful role to play in a future European code of contracts.

Non-Discrimination in the Common Principles of European Contract Law Ann-Sophie Vandenberghe*

Introduction In this contribution, the non-discrimination rules of the Draft Common Frame of Reference (DCFR) of the Principles, Definitions and Model Rules of European Private law are analyzed from a law and economics perspective. The non-discrimination rule of the DCFR addresses the problem of discrimination in contracting – either in the choice of one’s potential contracting partner, or in the terms on which one is prepared to contract with particular contracting partners. The non-discrimination principle in general contract law necessarily interferes with the principle of freedom of contract. According to this latter principle, any merchant is free to deal as he may choose with any member of the public. Absolute freedom of contract implies the right to deal or not to deal with anyone for whatever reason, including their personal traits. It is not a question of motives or reasons for deciding to deal or not to deal; private persons are free to do either. In contrast, anti-discrimination laws attempt to prohibit certain motives, such as race, ethnic origin or sex, from affecting economic transactions. Discrimination is immoral in most contexts, so most of us agree that it should be counteracted. The remaining questions are how the problem of discrimination can be optimally addressed, what the optimal remedies against discrimination are and what limitations in scope of non-discrimination laws are appropriate. In section 1, an overview is given of the economic literature that explains discriminatory behavior in contracting. A clear analytical distinction is made between taste-based discrimination and statistical discrimination. Section 2 gives an overview of the economic literature on the legal prohibition to discriminate in contracting. Economic analysis deals with such questions as whether markets are effective in minimizing discrimination and what are optimal remedies against discrimination. The non-discrimination rules of the Draft Common Frame of Reference will be analyzed in section 3.

1. Economic explanation of discrimination In order to understand the causes of discrimination, the economic literature makes a distinction between two categories of discrimination: taste-based discrimination and statistical discrimination.

*

Rotterdam Institute of Law and Economics, Erasmus School of Law.

Ann-Sophie Vandenberghe

46

1.1. Taste-based discrimination 1.1.1. Occurrence

People have a taste for discrimination when they have preferences for not associating with members of certain groups.1 Some people do not like to associate with the members of racial, religious, ethnic or other groups different from their own. Sheer malevolence, irrationality or personal prejudice are factors that explain this taste in many cases.2 Therefore, this type of discrimination is also called ‘invidious’ or ‘animus’ discrimination. Taste-based discrimination occurs when an individual treats members of a group differently because of an aversion toward that group. Although there are pecuniary gains to trade between different groups, such trade imposes at the same time non-pecuniary psychological, but real, costs on those members of either group who dislike association with members of the other group.3 People who are willing to lose income in order to reduce association with members of other groups have a taste for discrimination. In some cases of discrimination the discriminatory seller is merely reflecting the tastes of buyers with whom the discriminator has a commercial relationship. The source of discrimination is then found on the buyer’s side of the market, where a group of buyers considers particular ways of association with members of another race or gender distasteful. A profit-maximizing seller – though not having himself a personal preference for discrimination – might be led to discriminate when a large part of his customers have both a taste for discrimination and attractive alternative opportunities to buy elsewhere. Consider for example, people who prefer not to fly in airplanes piloted by women or people who prefer not to associate with people from a different ethnic origin when they go out to pubs or clubs. Discrimination could also be the result of the prevalence of social norms for segregation. For example, within an ethnic group a social norm might prevail against mixing with people from another ethnic group. A seller who simply wishes to maximize his profits will be led to

1 2

3

Garry S. Becker, The Economics of Discrimination (Chicago: University of Chicago Press, 1973). According to Larry Alexander we discriminate against certain people and in favor of others because the satisfaction of our preferences leads us to do so. Preferences can be contextual, as where a person prefers a member of her own race as a spouse, but who is happy to work alongside members of other races or categorical, in that they show up in all contexts for choice from the most intimate to the most public. He distinguishes three types of preferences: (1) preferences based on biases, (2) preferences for particular types of people as reflections of role ideals and (3) personal aversions and attractions to particular types of people. Biases are premised on the belief that some types of people are morally worthier than others. A person is judged incorrectly to be of less moral worth and is treated accordingly. When such biases are not fully conscious, they are called ‘implicit bias’, an unconscious form of animus-based discrimination. The second category of preferences is based on a moral ideal that people with a particular trait should perform certain tasks and occupy certain social roles. The third category consists of preferences that are based on aversions or attractions to particular types of people but that are not biases because they are not based on judgments of differential moral worth. For example, some men may feel quit uncomfortable transacting corporate business with women and some women may feel quite uncomfortable with a male gynaecologist (Larry Alexander, ‘What Makes Wrongful Discrimination Wrong? Biases, Preferences, Stereotypes, and Proxies’ (1992) 141 University of Pennsylvania Law Review, 149). Richard Posner, Economic Analysis of Law (Boston: Little, Brown and Company, 1992) 651.

Non-Discrimination in the Common Principles of European Contract Law

47

comply with the social norm because non-compliance would be punished by informal sanctions imposed upon him by his own group.

1.1.2. Is taste-based discrimination inefficient?

Is the satisfaction of a discriminatory preference through a refusal to deal with minorities an inefficient practice? Economists often take the satisfaction of preferences as the appropriate goal of public policy. To that end, all preference satisfaction is counted as equally valuable. Inefficiencies would occur when the actor does not bear the costs of his preference satisfaction. However, prejudiced sellers who refuse to deal with members of another group – due to an unreasoned aversion against them – forego advantageous transactions with them in order to indulge their taste for discrimination. Hence, in competitive markets, discriminators forego the pecuniary gains from dealing with minorities and therefore they pay a ‘price’ to satisfy their preferences. From this perspective, there is nothing inefficient about a discriminatory refusal to deal. The fact that taste-based discrimination is not an inefficient practice does not mean that it should be lawful. A preference to discriminate is immoral in most contexts and the wealth effects can be quite large, so that it may be unjust on utilitarian grounds.4 The wealth effects are contingent on a number of factors such as the number of people among the favored and disfavored group and their relative economic status. When members of the majority systematically refuse to deal with members of the minority, the losses in terms of foregone pecuniary income from valuable transactions are higher for the minority than for the majority. To the extent that the members of the minority group constitute a smaller part of the economy, they are more dependent on trade with the members of the majority than vice versa. Hence, the number of advantageous transactions that are prevented by discrimination is likely to be higher for the members of the minority than for members of the majority. Finally, there are instances in which discriminatory behavior is unlikely to be efficient. This is the case for discriminatory harassment. Under sexual or racial harassment, physical or verbal aggression is used against persons. With harassment, the taste for discrimination is expressed in hostile, harmful behavior towards the victims of discrimination. Discriminatory harassment arises from interdependent negative utilities. Individual A harasses individual B because the resulting disutility to B confers utility on A. Coercion arising from interdependent negative utilities cannot increase the wealth of society and therefore cannot be an efficient act.5 From an economic perspective the case for regulating is much stronger for harassment than it is for alternative forms of discrimination, like the discriminatory refusal to deal with someone from another group.

4 5

Posner, n 3 above, 651; Posner, n 3 above, 662. Richard Posner, ‘An Economic Theory of the Criminal Law’ (1985) 85 Columbia Law Review 1193, 1197.

48

Ann-Sophie Vandenberghe

1.2. Statistical discrimination 1.2.1. Occurrence

Discrimination has a number of possible causes. Sheer malevolence, irrationality or personal prejudice are factors that explain discrimination in many cases. Another factor, however, that explains discrimination is the costs of information. It is not uncommon in markets for the seller’s expected costs of supplying the goods or services to vary across individual buyers. Sellers may set different terms of trade for similar transactions because particular parties present more risks or impose a greater cost than alternative contracting parties. Such endemic forms of differentiation are necessary for the efficient allocation of resources. However, it is often the case that sellers have limited information about the costs or risks characteristics of their individual consumers prior to contracting. With limited information, it might become privately rational for a seller to rely on various statistical proxies to judge unobserved characteristics. According to economists, a ‘good’ proxy is easily observable and predicts future costs accurately on average prior to contracting. To the extent that an easily observable fixed trait like sex or age is positively correlated with costs or risk characteristics, it is rational (and not invidious) for sellers to use the attribute as a proxy for the underlying characteristic with which it is correlated. This is called ‘statistical discrimination’. Proxy judgments are used in a variety of market transactions. In health insurance and pension insurance females pay higher premiums than males because they consume more health services on average (including pregnancy-related benefits) and live longer on average. In life insurance, on the other hand, females pay lower premiums because they live longer on average. Car insurance companies charge higher insurance premiums to the group of young, male drivers because they tend to have more car accidents than older and female drivers.

1.2.2. Is statistical discrimination inefficient?

The use of proxies in decision-making is usually considered to be an efficient practice. The economic rationale for statistical discrimination is to economize on information costs. Car insurance companies, for example, know little about their individual policyholders at the contract formation stage. The costs of making individual assessments of each customer are prohibitive. They therefore classify drivers into broad groups and set premiums according to the probability that the average driver will have an accident. Automobile insurance companies consider young males as a relevant group in determining driver insurance premiums because they tend to have more car accidents that older males or female drivers. However, given that group membership is almost always an imperfect predictor of unobserved characteristics, some individuals will be treated much worse (or better) than their true characteristics justify. For example, even though young male drivers are, on average, much more likely to be involved in an accident, there are some young men within that group who are even more reckless than average and some who are much less reckless than the group’s average. However, it might be too costly prior to contracting to distinguish a particular young driver from the average driver of the group to which he belongs. Young male drivers who are careful and responsible drivers will pay more for insurance than they would in a world with better information. Hence, insurance companies will make mistakes like overcharging safe young drivers and undercharging reckless drivers. Such mistakes are costly to insurers. In many cases the individuals know better than the insurance company what their true risks are.

Non-Discrimination in the Common Principles of European Contract Law

49

For example, the insured alone may know whether he drinks and drives. This asymmetrical information may induce only high-risk people to purchase insurance and low-risk people to purchase none. Yet, it does not make the use of sex or age as a proxy, crude as it is, inefficient or irrational.6 Insurance companies want to minimize the sum of the costs of errors and the costs of more individualized information. When the sum is minimized by using the proxy, they will use it, otherwise they will not. As long as the costs of mistakes are lower than the costs of gathering more individualized information, the policy of statistical discrimination maximizes profits and competition will reinforce the practice. Nonetheless, statistical discrimination against particular groups – though profit-maximizing for the party using it – may impose an external cost – that is, a cost to nonparties to the transaction between a given seller and buyer. Therefore, statistical discrimination may not be efficient overall. This argument has been invoked against statistical discrimination in labor markets. Donohue7 argues that statistical discrimination is likely to distort workers’ decisions with regard to investments in human capital. If a member of a group knows that he will be treated as though he possesses the average traits of all group members, he will have no incentive to make investments that would increase his productivity.

2. Economic analysis of the legal prohibition to discriminate Many articles in the field of law and economics analyze non-discrimination laws from an economic perspective. Again a clear distinction is made between the desirability of nondiscrimination laws to eradicate taste-based discrimination, on the one hand and to eliminate statistical discrimination, on the other hand. The major conclusions of this literature will be summarized in this section.

2.1. Taste-based discrimination 2.1.1. Legal remedies are unnecessary when markets are effective in minimizing discrimination

The conventional efficiency perspective on adopting legal remedies against discrimination is that they are unnecessary to constrain private acts of discrimination. They are not necessary because there are already economic forces at work in competitive markets that tend to minimize discrimination.8 How does competition undermine discrimination in markets for goods and services? 9 In a competitive market with many sellers, the intensity of the prejudice against minorities will vary. Some sellers will only have a mild or no prejudice against them. These sellers will not forgo as many advantageous transactions as their more prejudiced competitors. If sellers refuse to deal with some buyers, the discriminatory sellers will experi6 7

8

9

Posner, n 3 above, 339. J. Donohue, ‘Further Thoughts on Employment Discrimination Legislation: A Reply to Judge Posner’ (1987) 136 University of Pennsylvania Law Review 523, 532. Becker, n 1 above. Becker made the point that in a competitive labour market, under most circumstances, wage differences among groups erode over time. Posner, n 3 above, 652; R. Cooter, ‘Market Affirmative Action’ (1994) 31 San Diego Law Review 133, 141.

Ann-Sophie Vandenberghe

50

ence additional costs. Under conditions of perfect competition, all goods sell at costs, so discriminatory sellers will charge more than non-discriminatory sellers for the same good or service. Non-discriminatory buyers will purchase from sellers with the lowest price and this will enable the non-discriminatory sellers to increase their share of the market. The least prejudiced sellers will come to dominate the market because they sell at the lowest prices. Thus, perfect competition eliminates discriminatory sellers. To the extent that the discriminatory seller does not raise the price at which he sells – because doing so would induce buyers to go to non-discriminatory competitors – he must accept lesser profits than he would have obtained had he not been prejudiced. Hence, sellers who refuse to deal with minorities are foregoing profits and will, over time, be driven from the market by non-discriminatory sellers who are prepared to increase their profits through servicing minorities – profits that firms with discriminatory tastes are denying to themselves. Even when competitive markets tend to eliminate discrimination, one could ask the question why anyone would object to supplementing the market remedy with a legal remedy against discrimination. The answer given to that question is that all legal remedies are uncertain and costly. One concern is that many non-victims would bring unmeritorious suits that prompt defendants to settle so as to avoid the costs of litigation.

2.1.2. Legal remedies might be justified when markets are not effective in minimizing discrimination

Some qualifications are to be made with respect to the effectiveness of competitive markets in minimizing discrimination. First, discriminating firms might have imperfect abilities to adjust to all relevant information, and may therefore not respond to the powerful market incentives. Donohue has argued that anti-discrimination laws are efficient because they will tend to eliminate discriminators more quickly than the free market would.10 Economic theory predicts that competition will eliminate discrimination but only over time. Perfect competition might take too long to eliminate discriminatory practices. Second, because a firm that discriminates will have higher costs than one that does not, economic theory suggests that discrimination is most likely to persist when the owners or managers have the ability and the incentive to pursue a goal other than profit maximization. Firms with the luxury of not having to maximize profits in order to stay in business are those that have a least some degree of monopoly power in their product or service market. According to Epstein, the use of anti-discrimination provisions has powerful justification whenever practical or legal circumstances prevent the emergence of competitive markets.11 Third, it has been noted in section 1.1.1 above that a seller might be led to discriminate when a large part of his customers have both a taste for discrimination and attractive op10 11

J. Donohue, ‘Is Title VII Efficient?’ (1986) 134 University of Pennsylvania Law Review 1411. R. Epstein, Forbidden Grounds (Cambridge: Harvard University Press, 1992) 85. Epstein gives the example that in the conditions of earlier times, when travel was restricted to a few highways or rivers, frequently only one carrier was available or one inn at a given location. Both classes of entrepreneurs enjoyed genuine elements of monopoly power. It was understood that as an offset of the monopoly power, the innkeeper or carrier could not exclude any passenger or discriminate by price.

Non-Discrimination in the Common Principles of European Contract Law

51

portunities to buy elsewhere. These third parties are often not market actors in the ordinary sense – that is, they will not suffer competitive injury in markets if they indulge their discriminatory preferences. At least in many cases, the harm suffered by customers who indulge their prejudice – perhaps higher prices of some form or another – will not drive them out of any ‘market’, but instead, and more simply, make them somewhat poorer.12 Finally, an economic case can be made for regulating discriminatory harassment. Frequently it will not be the seller himself who harasses his customers, but his employees or other customers with whom he has a contractual relationship. Generally, it is not in the interest of the seller that incidents of harassment take place at his premises. The potential victims will refrain from dealing with the seller and he will lose the pecuniary benefits from dealing with them. Hence, we might expect that sellers would take private action against harassment in the form of announcing and implementing private anti-harassment policies.13 It might seem that the market would reduce harassment without any legal intervention. However, the costs to the seller of detecting and preventing all forms of harassment might be high. In the case of harassment that is conducted in secret, for example, there will be no remedial private action. When the expected costs of private punishment to the harasser are less than the costs that the harassment inflicts on the victims, the harasser will have succeeded in externalizing some of the costs of his misconduct. In such a case, an economic rationale exists for punishing the harasser with legal sanctions.14

2.1.3. Changing preferences as a solution to discrimination

Economic theory predicts that competition should, over time, erode the effects of discrimination, not by changing preferences, but by shifting productive resources to firms that are not constrained by an aversion to associating with minorities. To the extent that a legal prohibition forces particular transactions, such trade imposes non-pecuniary psychic, but real, costs on those members of either group who dislike association with members of the other group. These costs could be reduced by changing the preferences of discriminatory sellers in such a way that association with members of another group no longer imposes psychic costs upon sellers, in that way eliminating discrimination. Some people discriminate to satisfy a preference which is based on biases. Biases are premised on the belief that some types of people are morally worthier than others. A person is judged incorrectly to be of less moral worth and is treated accordingly. Such bias-based preferences are immoral. Our moral views and preferences seem to a significant extent to be inculcated and learned. To the extent that bias-based preferences are fully conscious, they should be eradicable relatively easy through moral education. Is such moral education possible through the law of non-discrimination? Where the non-discrimination norm over-rides the right to make autonomous choices, or to act on personal preferences, with respect to contracting partners or the terms on which one chooses to interact, it is unavoidably making

12

13

14

Cass. R. Sunstein, ‘Why Market Don’t Stop Discrimination’ (1991) 8 (2) Social Philosophy & Policy 22, 25. Such a policy might for example contain a rule that individuals who harass other customers will be denied access to the premises for a certain period of time. R. Posner, ‘An Economic Analysis of Sex Discrimination Laws’ (1989) 56 The University of Chicago Law Review 1311, 1318.

52

Ann-Sophie Vandenberghe

a moral judgment about the quality of preferences.15 When the law holds that discrimination is illegal and results in sanctions, individuals might learn the lesson that discrimination is wrong and they might alter their beliefs. The educative or preference-shaping role of law is to change preferences and values by inducing people to internalize norms.16 Once the nondiscrimination norm is internalized, it is obeyed out of a sense of guilt or shame. When the norm is internalized, many people would refrain from discriminating even when the nondiscrimination norm is merely symbolic and is seldom enforced formally. But not all legal instruments are equally appropriate in shaping preferences. The preference-shaping technology under private law is poor. The first reason for this is that private contract law is primarily designed as a system to foster the satisfaction of given, individual preferences through the enforcement of voluntary agreements, and not as a system to change preferences. Under private contract law, courts monitor the circumstances under which the exchange takes place to insure that force, fraud and incompetence are not involved. They usually do not plumb into the subjective preferences that contracting parties wish to satisfy through the exchange or by their refusal to deal nor do they normally decide on the desirability of the satisfaction of particular preferences. That decision is made autonomously by the private contract parties. The second reason why private law is a poor instrument to shape preferences is its sanctioning system. The way in which contract law remedies affect individual behavior is to treat preferences as exogenous and immutable but to shape the individual’s opportunities to give incentives for desired behavior. The remedies are not designed to affect the individual’s behaviour by changing the individual’s preferences. The primary sanction for rule violation in private contract law is to compensate the disadvantaged party for his losses. By their nature, compensatory damages do not directly aim at stopping violations of contract rules; they merely require the party who breaches the contract to bear the costs of his actions. Compensatory damages create a disincentive for discrimination, but they cannot be understood as an effort to eradicate discriminatory preferences. Eradicating discrimination would require a different system of sanctions. Criminal law, on the other hand, plays a potentially stronger role as a preference-shaping policy. Under the preference-shaping theory of criminal law, legal punishment of discrimination would be an expression of society’s condemnation of the discriminatory act and an effort to discourage preferences for discrimination.17 The public enforcement, moral condemnation and punishment of a criminal sanction, may be part of an overall social process of shaping people’s preferences. Still, the law may not be the most effective instrument in shaping preferences. Perhaps it would be better to leave the condemnation of discrimination to parents, schools, and intellectuals rather than to the law. Possible means by which society could conceivably conduct a societal preference-shaping policy towards adults include public education, public service announcements and statements by popular public figures. These people and organizations may be more effective than law in inducing people to internalize new values.

15

16 17

M. Trebilcock, The Limits of Freedom of Contract (Cambridge: Harvard University Press, 1993) 188. Cooter, n 9 above, 133, 167. K. Dau-Schmidt, ‘An Economic Analysis of the Criminal Law as a Preference-Shaping Policy’ (1990) 1 Duke Law Journal 1.

Non-Discrimination in the Common Principles of European Contract Law

53

2.2. Statistical discrimination Unlike taste-based discrimination, the operation of market forces will not eliminate statistical discrimination. It is inherent in the practice of statistical discrimination that the seller will make mistakes, but as long as the costs of these mistakes are lower than the costs of gathering more individualized information, the policy maximizes profits and competition will reinforce the practice. Hence, if one would want to eradicate statistical discrimination, this particular form of discrimination would need to be included in the definition of unlawful discrimination. In that way, the victims of statistical discrimination – those individual group members whose underlying characteristics are not adequately predicted by the proxy – will be protected. However, in the economic literature it has been argued that a legal ban on the use of discriminatory proxies is usually not the cheapest way to undo their effects. A ban will induce sellers to rely upon a more costly substitute for all customers. A cheaper solution than banning the use of discriminatory signals consists of augmenting the information flow about potential victims of statistical discrimination.18

2.2.1. A legal ban on statistical discrimination is a costly solution

In a world where information is available at low costs, the ideal would be that sellers make individual cost assessments of their customers. However, the law cannot, merely by outlawing a particular form of statistical discrimination, force sellers to judge every customer on an individual basis. The costs of obtaining information are too high in the real world. Sometimes the costs to ascertain individual qualifications would be indefinite. This is the case when the uncertainty is inherent and ineradicable. For example, when a pension insurer contracts with two 30-year old customers, one male and one female, he knows that the former has a shorter life expectancy than the latter, but he does not know and ordinarily could not discover whether this specific female will outlive this specific male. If differentiations are to be made, they must be made on a statistical basis. Deprived of the sex, race or ethnic origin proxy, some sellers will use other proxies to predict costs. It can be presumed that the next-best proxy is more costly than the banned proxy otherwise the seller would have used it in the first place. For example, when sex can no longer be used as a signal for life expectancy, firms will turn to alternative signals, such as lifestyle. This is a more costly signal than sex because the information costs of credibly learning about the lifestyle of all policy holders are higher than learning about someone’s sex. The increased costs need to be borne by some party and it is very likely that consumers as a group will bear these costs in the form of higher prices for the services or goods they buy. Moreover, when the prohibition of the sex and race proxies will lead to the substitution of other proxies, other people will be disadvantaged by the new proxy when it does not accurately predict their individual characteristics. The problem of statistical discrimination is shifted not solved. Some sellers, denied the use of certain forbidden proxies, will throw up their hands unable to distinguish between customers who are intrinsically costly and those who are less costly to serve. They will treat both groups indiscriminately. Insurers, for example, will treat both groups indiscriminately by charging uniform premiums and offering equal coverage. However, if insurers can no longer differentiate among risks groups on the most efficient basis, 18

Cooter, n 9 above, 133, 160.

Ann-Sophie Vandenberghe

54

social costs of insurance will increase because of ‘adverse selection’. Adverse selection arises in situations in which individuals who differ from each other in important ways selectively choose to enter into contracts. When insurers charge uniform premiums irrespective of the underlying risk, we might expect people whose risks are relatively high to be more likely than those whose risks are low to buy insurance. As a consequence, the insurer will receive more claims and will have to increase insurance rates (and tailor plans to place sharper limits on the amount of coverage that is available) than it otherwise would. The higher premiums, in turn, will deter some low risk individuals from buying insurance, even though they would be willing to pay lower premiums that the insurer would be willing to accept to cover them if it were allowed to identify them as the low-risk prospects that they are. When low-risk customers ‘leave’ the market, there will be more high risks in the insurance pool, which would raise the rates still further and quite possibly make the insurance unaffordable by those most in need of it. A legal ban on the use of efficient proxies would lead to an increase in prices and lower opportunities for insurance across the board.

2.2.2. Increasing the flow of information about potential victims of statistical discrimination is a cheaper solution

A cheaper solution than banning the use of discriminatory signals consists of augmenting the information flow about potential victims of discrimination.19 Recall that the major factor driving statistical discrimination on the basis of sex, race or ethnic origin, is the unavailability of more precise information on the cost characteristics of the individual seeking goods or services. Those individual customers with characteristics that justify better conditions could provide that information to the seller provided that such information could be produced and credibly communicated. The objection to this form of private remedy is that those individuals who make use of it will need to incur additional costs to inform the seller. This objection would be overcome where the state could provide additional information on individuals without charge. When governments help to provide decision makers with something that approximates complete information, discernable facts like race or gender will not be overemphasized and more obscure but relevant facts will loom larger. Moreover, it can be expected that the use of proxies in decision-making will decline over time. One of the most significant developments in the world of information technology has been the increased availability of information about individuals.20 Personal information that was once obscure can be revealed almost instantaneously via a Google search. Reputation tracking technologies are being used to track customer’s preferences and quirks, too. When such technologies increase the information on individual customers, the need to rely upon imperfect proxies becomes unnecessary.21 It has been argued that statistical discrimination should be banned because it is based on false stereotypes. It has for example been said that sex, race or ethnic origin proxies are based on overbroad generalizations or stereotypes. In this view, proxies are based (a) on a belief that 19 20

21

Cooter, n 9 above, 133, 160. L. J. Strahilevitz, ‘Reputation Nation: Law in an Era of Ubiquitous Personal Information’ (2007) Chicago Working Paper, no. 371, p.19. However, the information solution to the victims of statistical discrimination will make privacy laws and institutional arrangements that obscure information about individual’s reputations more problematic.

Non-Discrimination in the Common Principles of European Contract Law

55

members of a group have certain characteristics when in fact they do not, or (b) on a belief that many or most members of a group have certain characteristics when in fact only a few do, or (c) on a reliance on fairly accurate group-based generalizations when more accurate classifying devices are available. However, competition can teach a sharp lesson to firms who rely on false proxies that reflect social stereotypes, not accurate averages. An irrational proxy hurts its user. ‘Decision makers whose prosperity depends upon the accuracy of their perceptions are better situated than social critics or legislators to penetrate myths’, according to Cooter.22 Nevertheless, when the development of good proxies is costly, a single seller might not have the appropriate incentives to invest in obtaining more accurate information because there is a risk that competitors copy that proxy without having to pay for the costs of developing it. Insurers might under-invest in the development of better methods to place customers in more accurate risk classes if competitors can free-ride on their efforts. According to Posner ‘because of the difficulty of establishing property rights in information, people may have inadequate incentives to investigate even the average characteristics of the groups they deal with’.23 If this is the case, governments can play a welfare-enhancing role by developing more accurate proxies themselves and disseminating them among the sellers.

3. Economic analysis of the non-discrimination rules of the Draft Common Frame of Reference Principles, definitions and model rules of European Private Law (Draft Common Frame of Reference) Chapter 2: Non-Discrimination II.–2:101: Right not to be discriminated against A person has a right not to be discriminated against on the grounds of sex or ethnic or racial origin in relation to a contract or other juridical act the object of which is to provide access to, or supply, goods or services which are available to the public.

3.1. Only sex and ethnic or racial origin are adopted as protected grounds The drafters’ choice only to protect against discrimination on the grounds of sex or ethnic or racial origin is curious in the light of the EU’s proposal for a Council Directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age and sexual orientation.24 With the adoption of the Proposal, discrimination on the grounds of religion, disability, age and sexual orientation will be prohibited in all areas where the EU has prohibited discrimination on the grounds of sex, racial or ethnic origin, including the access to and supply of goods and services which are available to the public.

22 23 24

Cooter, n 10 above, 133, 160. Posner, n 3 above, 661 ff 26.5. Proposal for a Council Directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation, COM (2008) 426.

Ann-Sophie Vandenberghe

56

Analysis What might explain and justify the drafters’ choice only to protect against discrimination on the grounds of sex or ethnic or racial origin? An economic framework for thinking about the appropriate scope of the non-discrimination norm is to emphasize the costs of administering the non-discrimination norm as well as the social benefits from non-discrimination. The law cannot outlaw all types of discrimination as that would be too costly. The reach of law is limited by the costs of administering it. Under the normative stance of this approach, the law should only select for prohibition those broad types of discrimination that are most likely to cause a great amount of social harm given that the resources for enforcement are scarce. Or stated differently, the costs of administering non-discrimination laws are deemed to be warranted where non-discrimination laws would bring large social benefits compared to the state of affairs in the absence of non-discrimination laws. What is the state of affairs in Europe of the various groups? Non-discrimination laws could substantially improve the faith of females and racial and ethnic minorities. Race and sex are traits a high percentage of use will reflect biases, inaccurate stereotypes or shallow personal aversions. For the trait of age, in particular, this seems to be less true. Elderly people do not seem to be victims of pervasive ageism just as vicious and irrational as racism or sexism. This is not to deny the existence of false, denigrating stereotypes about the elderly, such as that they are bad drivers. But these misunderstandings about older people are unlikely to cause a great amount of social harm and have not led to their systematic exclusion from the access to and supply of goods and services. The same applies – but perhaps to a lesser extent – to the situation of disabled persons. The very fact that many people have disabled friends and relatives generates a substantial level of empathy rather than hostility towards them. Moreover, most cases of differential treatment on the grounds of age or disability in the supply of goods and services are based on costs differences in providing the goods and services, not on an irrational aversion against members of these groups. However, for the trait of ‘sexual orientation’ the situation is different. Homophobia is as vicious and irrational as sexism and racism. This could be a reason to add ‘sexual orientation’ as a forbidden ground for differential treatment in II.–2:101 of the draft.25

3.2. Only buyers have the right not to be discriminated against The draft (II.–2:101) prohibits sellers to discriminate against buyers on the ground of sex, race or ethnic origin, but it does not prohibit buyers to discriminate against sellers on those grounds.

Analysis The draft does not entitle sellers to a right not be discriminated against in the demand for their goods and services. Nevertheless, buyers might discriminate against sellers for the same reasons as sellers discriminate against buyers. Some buyers might have an aversion against 25

A decisive case for including sexual orientation as a protected ground in the supply of goods and services exists when it can be shown that homophobia has led to the systematic exclusion of homosexual individuals from the supply of good and services.

Non-Discrimination in the Common Principles of European Contract Law

57

dealing with members from other groups (taste-based discrimination) or an individual buyer might treat sellers who are membership of a particular group differently because he or she believes that group membership correlates with some attribute that is both relevant and more costly to observe than group members (statistical discrimination). Why are sellers not entitled to a claim for discrimination under II.–2:101? To answer that question, we could again apply the economic framework for thinking about the appropriate scope of the nondiscrimination norm. The law cannot outlaw all types of discrimination as that would be too costly. The reach of law is limited by the costs of administering it. Where the resources for enforcement are scarce, they should be used to address those forms of discrimination that have the highest expected losses. In this view, a single discriminatory buyer might not do as much harm as a large-scale seller who refuses to deal with many buyers on discriminatory grounds. This argument is not fully convincing because a discriminatory seller might be a single individual who is not in business (e.g. an individual who rents a room in his or her apartment) and a discriminatory buyer might be a large company who refuses to do business with sellers he dislikes. In addition, where many buyers of the majority group refuse to deal with sellers from other groups, the wealth effects on the sellers of the minority can be quite large. The drafter’s choice made in II.–2:101 to protect buyers but not sellers, has probably been based on their judgment that discriminated buyers are considered to be more deserving and more in need than discriminated sellers. II.–2:102: Meaning of discrimination (1) “Discrimination” means any conduct whereby, or situation where, on grounds such as those mentioned in the preceding Article: (a) one person is treated less favourably than another person is, has been or would be treated in a comparable situation; or (b) an apparently neutral provision, criterion or practice would place one group of persons at a particular disadvantage when compared to a different group of persons. (2) Discrimination also includes harassment on grounds such as those mentioned in the preceding Article. “Harassment” means unwanted conduct (including conduct of sexual nature) which violates a person’s dignity, particularly when such conduct creates an intimidating, hostile, degrading, humiliating or offensive environment, or which aims to do so. (3) Any instruction to discriminate also amounts to discrimination.

3.3. Definition of discrimination For the purpose of defining discrimination it is not decisive whether the disadvantage can be tightly connected to an intentionally, taste-based discriminatory act. Discrimination is understood to exist according to II.–2:102 when a person is disadvantaged on the basis of sex, race or ethnic origin, whether the reason for the conduct lies in taste or in statistical discrimination. It follows that the claim of discrimination is not merely for the prevention of certain irrational acts, or of prejudice, but for the elimination of disadvantages that are based on sex, race or ethnical origin. II.–2:101 in association with II.–2:102 condemns discrimination even if it is fully rational – that is, if it is based on an efficient proxy that predicts underlying characteristics accurately on average and is cheap to observe. However, while the Draft employs a somewhat over-inclusive definition of discrimination, it also allows for a broad category of exceptions in article II.–2:103.

Ann-Sophie Vandenberghe

58

II.–2:103: Exception Unequal treatment which is justified by a legitimate aim does not amount to discrimination if the means used to achieve that aim are appropriate and necessary.

3.4. Discrimination is not per se illegal The rules do not attempt to prohibit all forms of discrimination on the ground of sex, racial or ethnic grounds in the supply of goods and services which are available to the public. II.–2:103 acknowledges that there might be exceptions.

Analysis From an economic perspective, it can be said that, under a per se ban on discrimination (i.e. a rule that does not allow exceptions) the costs of trying discrimination cases would be reduced since the courts do not have to adjudicate the legitimacy of discrimination on a case by case basis. However, although the cost of trying discrimination cases are reduced by not bothering about exceptions, the result would be excessive punishment, leading to all sorts of serious social costs. It is important that lawmakers make clear in advance what types of discrimination are potentially justifiable in order not to create too much uncertainty for the parties who are subject to non-discrimination laws. In general, uncertainty about the application of legal standards can give parties economic incentives to ‘overcomply’ – that is, to modify their behavior to a greater extent than a legal rule requires.26 Unpredictability of legal rules may also impede economic initiative. Therefore, uncertainty about the legal standard used for determining justified discrimination, should be reduced as much as possible. This implies that legislators should clearly define what they consider to be a legitimate aim for justified discrimination and what factors will be taken in account when assessing the appropriateness and necessity of the means used.

3.5. Justified discrimination According to II.–2:103 unequal treatment may be justified by legitimate aims if the means applied to reach these aims are appropriate and necessary. According to the Comments, the aims are legitimate, if they constitute a protected value in a society, such as the need to protect privacy, decency, religion or cultural identity. The Comments also state that the exception test must be applied differently, i.e. more or less strictly, depending on the ground for discrimination. Hence, according to the Comments, it is justified, for example, for a woman to make an offer to rent two rooms in her apartment but only to female students for reasons of privacy and decency. But discrimination is not justified in the case that the woman makes an offer to rent rooms in her apartment only to white students because privacy should not be based or linked on the race of the roommates. In that way, the exception test introduced in II.–2:103 requires judges to engage in value judgments in deciding discrimination cases. It is 26

John E. Calfee and Richard Craswell, ‘Some Effects of Uncertainty on Compliance with Legal Standards’ (1984) 70 Virginia Law Review 965.

Non-Discrimination in the Common Principles of European Contract Law

59

said in the Comments that in exceptional cases (e.g. insurance contracts) certain economic factors can also provide justification for discrimination. But the major reason for allowing exceptions to the general non-discrimination rule is to allow voluntary affirmative action plans. Under affirmative action plans, members of the protected minority are given preferential treatment over the members of the majority.27 According to the Comments, affirmative action is not illegal discrimination against those dispreferred.

Analysis An important question is whether discrimination based on the use of race, ethnic origin or sex as a proxy for underlying characteristics in order to economize on information costs is justifiable under the legal standard for justified discrimination of article II.–2:103. An efficiency case for justifying statistical discrimination can be made when the costs of forbearing it are higher than the burdens imposed under it. If the costs to forego a particular form of discrimination (in order to comply with non-discrimination laws) are high compared to the burden that that particular discriminatory practice brings, it might be a reason to justify the discriminatory practice on efficiency grounds. What are the costs of prohibiting the use of sex, race or ethnical origin as a proxy for underlying characteristics? When the costs of making individual assessments are prohibitive, as they often are, prohibiting the use of the sex or race proxy will lead to the substitution of other proxies. And if the new proxies are less efficient than gender or racial or ethnical origin, – as they probably will be, because otherwise they would in all likelihood have been adopted without government prodding – prices for all customers will increase because the seller’s costs will be higher. In addition, legal suppression of efficient proxies could lead to an increase in error costs, leading to market inefficiency. To the extent that the prohibited proxy distorts the allocation of resources to those individuals of the discriminated group who have ‘desirable’ characteristics, the replacement of that proxy by a new proxy will distort the allocation of resources to those individuals who are disadvantaged by the new proxy. The problem is shifted, not solved. It could only be solved by requiring sellers to ignore statistical correlations between particular traits and unwanted behavior. But if statistical correlations are to be ignored and the costs of making individual assessment are too high, seller may stop trying to distinguish between high-cost and low-cost customers and use a uniform pricing system instead. Under a uniform pricing system, high cost customers benefit at the expense of low cost customers. Some of the buyers who would have bought the product or service when prices or contract terms were differentiated according to their cost characteristics, will stop buying the product, resulting in market inefficiency. These costs need to be balanced against the burdens placed upon the victims of the discriminatory proxy that is the subject of the legal conflict. The primary victims of a firm’s policy to use a proxy based on sex, race or ethnical origin are the members of these protected groups whose costs characteristics are not accurately predicted by the discriminatory proxy. II.–2:104: Remedies (1) If a person is discriminated against contrary to II.–2:101 (Right not to be discriminated against) then, without prejudice to any remedy which may be available under book VI (Non-contractual

27

It is also called ‘reverse discrimination.’

Ann-Sophie Vandenberghe

60

Liability for Damage Caused to Another), the remedies for non-performance of an obligation under Book III (including damages for economic and non-economic loss) are available. (2) Any remedy granted must be proportionate to the injury or anticipated injury; the dissuasive effect of remedies may be taken into account.

3.6. Damages for non-economic loss Violation of the non-discrimination rule of II.–2:101 gives rise to private contract law remedies. Paragraph (1) of Article II.–2:104 clarifies that the right to claim damages for loss includes non-economic loss. In the Comments it is stated that it may be extremely difficult in discrimination cases to prove that an economic loss has been suffered. Above all, it is said, discrimination violates human dignity.

Analysis Discrimination violates human dignity. The loss of dignity suffered by victims of discrimination is a paradigmatic instance of non-economic or non-pecuniary loss. Should the law allow for recovery of non-pecuniary loss? Compensation for non-pecuniary loss is optimal from an incentive perspective. It is necessary to give the supplier optimal incentives not to discriminate. In order to induce the supplier to consider the full costs of his or her discriminating actions, damages should reflect the total loss to the victim. However, the theory of optimal insurance suggests that non-pecuniary losses should not necessarily be compensated. Rich compensatory awards for dignitary loss tend to be passed on to the consumer-victim in higher product prices. In order to determine the optimal level of insurance, the question is asked what amount of insurance the average risk-averse victim voluntary would have purchased ex ante. It is found that a non-pecuniary loss is the sort of loss for which a victim would not insure at all ex ante. Hence there is a conflict between what is optimal from an insurance perspective (no compensation for non-pecuniary loss) and what is optimal from an incentive perspective (compensation for non-pecuniary loss). The theory of optimal insurance suggests that non-pecuniary loss should not be compensated, but lack of such compensation may affect the supplier’s incentives not to discriminate.28 This deficiency can be overcome in theory if the government imposes fines on sellers in the event of discrimination.29 This would lead to optimal insurance, i.e. no damage compensation for non-pecuniary loss and optimal incentives not to discriminate through the deterrent effect of the fine.

3.7. Dissuasive effect: punitive damages Paragraph (2) of Article II.–2:104 specifies that damages may have a punitive element with a view of having a dissuasive effect.

28 29

Samuel A. Rea, ‘Nonpecuniary Loss and Breach of Contract’(1982) 11 Journal of Legal Studies 35. Michael Spence, ‘Consumer Misperceptions, Product Failure and Product Liability’ (1977) 44 Review of Economic Studies 138

Non-Discrimination in the Common Principles of European Contract Law

61

Analysis Under the deterrence rationale, punitive damages are used to completely deter conduct that has no social value (i.e. the act is absolutely normatively proscribed). When damages are set merely at the level to compensate the victim for his losses (i.e. not at a punitive level), economic theory would lead one to expect the following variance in compliance with the nondiscrimination norm. If the inter-group associations brought about by such a law are slight, the cost of association even to prejudiced people will be low and they will not be willing to incur heavy costs in the form of damages for, or legal expenses of, resisting compliance in order to indulge their taste. If the costs of association to the discriminator are high, he might prefer not to comply with the non-discrimination rule and pay damages instead. In the latter case, the prospect of having to pay damages is not viewed as so threatening as to prevent the discriminatory act. However, if the purpose is to deter discriminatory acts completely, the sanctions for discrimination in contract need to be designed so that the discriminator is made worse off by his act; In order to completely deter the defendant to discriminate, punitive damages need to be set at the level where expected punishment is higher than the benefits to the discriminator from discrimination. It seems that the Comments want to achieve this result by saying that the dissuasive effect of the remedy needs to be taken into account. However, this view of contractual liability (i.e. that sanctions are designed to deter particular acts) deviates from the way standard remedies in contract law are normally conceived. The standard remedies for breach of contract (damages as the presumptive remedy; specific performance as an exceptional remedy) reflect that contract law in general has chosen to adopt a pricing regime rather than a sanctions regime.30 In this view of contractual liability as a pricing mechanism, breaches of contract are not absolutely normatively proscribed. In the present context, the question is whether discriminatory conduct should be absolutely deterred, because only then the use of punitive damages would be appropriate under the deterrence rationale. Above, it has been argued that some forms of discrimination are efficient and should not be deterred from an economic perspective. An example is the use of a cost-effective discriminatory proxy. When punitive damages would be applied in such case, there would be excessive punishment from an economic point of view. One solution would be to define discrimination and its exceptions in such a way as to exclude such examples. If the category that remains after the exception test is that of the forms of discrimination that are not efficient, penalty damages could be used to deter these forms of discrimination. If however, the choice is made for a rather narrow exception test, expected punishment cost should be set at a lower level, more specifically at a level that will not deter discrimination that is value-maximizing. Another role for punitive damages is to apply it in cases where the low probability of enforcement render it unlikely that the infringer (or breacher) will face ex ante the full expected social costs of his or her breach. The discriminator has a significant chance of escaping liability when damages merely compensate for the victim’s losses because individual losses may be insufficient to warrant the costs of suit ex post. Under the normal private law remedy of damages the discriminated party is entitled to compensation for his or her losses. For the calculation of the losses two measures can be used: the reliance measure or the expectation measure. Under a reliance measure, the discriminated party would be compensated for the costs he or she has made in reliance upon the conclusion of the contract, costs which are wasted in case the seller refuses to deal. Under an expectation measure the discriminated 30

R. Cooter (1984), ‘Prices and Sanctions’, 84 Columbia Law Review 1523.

Ann-Sophie Vandenberghe

62

party would be put in the position in which he or she would have been had the contract been performed. What would have been the advantages to the discriminated party in trading with this particular seller in comparison to his or her next-best alternative trading opportunity? It can be expected that the damages computed under the reliance or expectation measure will be rather small in the case of a discriminatory refusal to deal. It might then not be in the interest of the discriminated party to start a costly private law suit against the discriminator. When many victims of discrimination never bring suit, there is likely to be under-enforcement of, and under-compliance with the non-discrimination rule. To facilitate private enforcement in cases in which the damages are small and to induce discriminators to internalize the full costs of their discriminatory acts, damages could be grossed-up with a punitive multiplier in individual cases to offset the low probability that the victims of discrimination will bring suit. Calculation of the appropriate multiplier depends on the probability that the discriminator will be sued, which is itself a function of the size of the multiplier. II.–2:105: Burden of proof (1) If a person who considers himself or herself discriminated against on one of the grounds mentioned in II.–2:101 (Right not to be discriminated against) establishes, before a court or another competent authority, facts from which it may be presumed that there has been such discrimination, it falls on the other party to prove that there has been no such discrimination. (2) Paragraph (1) does not apply to proceedings in which it is for the court or another competent authority to investigate the facts of the case.

3.8. Presumption of discrimination In the Comments it is said that although the definition states that the unequal treatment must happen “on grounds” such as race or sex, no causal link between that reason and treatment is required because such a link would be difficult to prove for the person discriminated against. According to II.–2:105 the person who considers himself or herself discriminated against on such grounds must merely establish the fact from which it may be presumed that there has actually been discrimination. In that case, it falls on the other party to prove that there has been no such discrimination.

Analysis It seems that the rule of II.–2:105 permits the plaintiff to establish a prima facie case of discrimination with evidence that he was willing to accept the defendant’s offer but that the latter one refused to deal with him but not with someone from another sex, race or ethnic origin or with evidence that the defendant’s offer contained terms that were not included in the offer made to a person of a different sex, race or ethnic origin. But the rule does more: satisfying that burden of production of prima facie evidence creates a presumption of discrimination, meaning that if the defendant offers no evidence in rebuttal, the plaintiff’s claim of discrimination will be honored. The probability that the defendant refused to deal or was only prepared to contract on different terms because of the plaintiff’s sex, race or ethnic origin might seem not to be very high if the only evidence is as described. But this disregards the evidentiary significance of missing evidence. If the defendant, who after all made the decision not to deal with plaintiff or to offer different terms to him, maintains complete silence

Non-Discrimination in the Common Principles of European Contract Law

63

about the reason for his or her action, an inference of discrimination arises. If the reason was otherwise, he should have been able without great difficulty to produce some evidence of that. But reversing the burden of proof by introducing a presumption of discrimination may lead to a higher level of formalism in contracting. As an anticipatory measure, sellers may require their customers to sign documents that may be used later as evidence in the event the seller has to prove that he or she was not discriminating.

4. Conclusions What are the major conclusions from economic analysis for the DCFR on non-discrimination? Economic analysis determines the conditions under which markets will not eliminate discrimination in contracting. To eradicate discrimination, intervention is then needed. However, there is discussion about the appropriate form and type of intervention to eliminate preferences for discrimination. The DCFR entitles victims of discrimination in contracting to a private law remedy. Because the rules of contract law are in general not designed to change the subjective preferences that contract parties wish to satisfy through the exchange or refusal thereof, and general contract law remedies do not primarily aim at stopping particular behavior but rather aim at pricing it, it is worth investigating alternative policies, like moral education, to change discriminatory preferences, instead of, or in addition to, private law remedies. The Draft offers protection against discrimination on the basis of sex, race and ethnic origin. Other grounds like age, disability or sexual orientation are not included within the scope of protection. Given that the resources for enforcement are scarce, the law should only select for prohibition those broad types of discrimination that are most likely to cause a great amount of social harm. Under this economic approach towards the optimal scope of non-discrimination laws, differences in protection depending on the discriminatory ground may be warranted. Uncertainty exists about the application of the legal standard for justified discrimination of II.–2:103. An economic case can be made to include statistical discrimination as a form of justifiable discrimination under the legal standard of article II.–2:103. This conclusion follows from the acknowledgment of the costs imposed on society from forgoing statistical discrimination. Legal suppression of efficient proxies is likely to result in the imposition of substantial costs and may even result in harm to the prohibition’s intended beneficiaries. A less costly solution to help the victims of statistical discrimination – those individuals whose characteristics are inaccurately-predicted by the proxy – is to increase the flow of information on them – with the help of governments – so that reliance upon imperfect proxies becomes unnecessary. Damage compensation for non-economic losses as provided for in II.–2:104 is optimal from an incentive perspective but non-optimal from an insurance perspective. There is an appropriate role for punitive or aggravated damages as remedies for discrimination under II.–2:104 in cases where low probability of enforcement renders it unlikely that the discriminator will face ex ante the full expected social costs of his or her behavior.

2. Formation and Interpretation of Contracts

The Formation of Contracts in the Draft Common Frame of Reference – A Law and Economics Perspective Gerrit De Geest* & Mitja Kovač** Abstract What can lawmakers learn from 40 years of contract law and economics? This article surveys the economic literature on the formation of contracts, and derives some recommendations for lawmakers. It also comments on some articles of the Draft Common Frame of Reference. The main findings are: (1) In principle, the least cost information gatherer should produce and communicate information. (2) There should be no duty to reveal entrepreneurial information. (3) There should be no duty to be honest with respect to mere opinions or other non-falsifiable statements. (4) The difference between lying and concealing information is irrelevant. (5) The difference between intentionally and negligently misinforming is irrelevant, except for the determination of remedies and criminal sanctions. (6) Hence, there is no need for separate doctrines on mistake, fraud, misrepresentation or latent defect – all that is needed is a duty-to-inform doctrine. (7) ‘Consent theories’ or ‘will theories’ can never provide precise criteria for when a contract should be void. (8) The condition that the misinformed party would not have entered the contract otherwise is analytically imprecise and unnecessary. (9) To what extent sellers would spontaneously reveal the information on the market is generally irrelevant for lawmakers. (10) Less is more: parties get better informed when unimportant information is filtered away.

Introduction This article offers an economic comment on the part of the Draft Common Frame of Reference project that deals with the formation of contracts. It does so in two ways. First, it examines (in section 2) what can be learned – in general – from more than four decades of law and economics scholarship. Second, it analyzes whether the articles of the Draft Common Frame of Reference project that deal with informational issues make sense from an economic point of view (section 3).1

* **

1

Utrecht University (The Netherlands) and the University of Ljubljana (Slovenia). Washington University School of Law, St. Louis (MO), USA and Ghent University, Faculty of Law, Belgium. This article is the result of research for the Economic Impact Group of CoPECL. Earlier versions were presented at a workshop of the Economic Impact Group at Brussels, and at the annual conference of SECOLA at Amsterdam. Draft Common Frame of Reference (DCFR), prepared by the Study Group on a European Civil Code and the European Research Group on Existing EC Private Law (Acquis Group), based in part

68

Gerrit De Geest & Mitja Kovač

Law and economics is one of the most successful research programs of the last fifty years. Its purpose is to find out what good law is by analyzing incentive, risk and transaction cost effects of legal rules. The term ‘good’ can have two dimensions: good with respect to the content, and good with respect to the technical formulation. Most of the law and economics literature is focused on the first dimension. It tries to determine which legal rules have the most desirable effects, irrespective of how these legal rules are formulated. Yet, law and economics may also offer useful advice on how to improve the technical formulation of the rule.2 The reason is that lawmakers (legislators and judges) usually balance the advantages and disadvantages of alternative solutions – even though this balancing is often hidden behind a veil of fairness rhetoric. Law and economics tries to describe these advantages and disadvantages in a more accurate way. As a result, it may also accurately describe what lawmakers do, and hence, more accurately describe the law. Two caveats should be made. The first caveat is related to the scientific authority that should be attributed to our findings. Law and economics is strictly speaking not a theory, but a method for facilitating discussions on the law. Therefore, what we do is summarizing the most striking results that have been produced with this method.3 On some of these results a consensus has been reached; on others, scholars still disagree. As always, finding out to what extent a consensus has been reached is not easy, as journals usually publish new insights rather than opinions on existing ones. Therefore, the principles we draw from the literature only express a personal opinion on what should be concluded, rather than on what is generally concluded. The second caveat is related to the scope of this article. Since it is not possible to cover all aspects of the formation of contracts, we will focus on those that are linked to informational problems, like the duty to inform, mistake, fraud, and latent defects, and omit discussions on doctrines such as duress, capacity, timing of acceptance, revocability of offers, or liability for failed negotiations (though the latter has some informational aspects). Furthermore, we do not discuss the optimal remedies for breach of information duties.

1. Synthesis of law and economics scholarship on asymmetric information problems This section identifies thirteen principles and tries to compress them into a single general principle.

2

3

on a revised version of the Principles of European Contract Law (Articles Only Version, published as pdf file, version December 2007). G. De Geest, “Comparative Law and Economics and the Design of Optimal Doctrines”, in B. Deffains / T. Kirat (eds.), Law and Economics in Civil Law Countries (New York, JAI Elsevier, 2001) at 107-124. For a bibliography on the information problems in the pre-contractual stage, see B. Bouckaert / G. De Geest (eds.), Encyclopedia of Law and Economics (Cheltenham, Edward Elgar Publishing, 1999) nr. 4300, nr. 5510 and nr. 5110. For a synthesis see H. B. Schäfer / O. Claus, The Economic Analysis of Civil Law (Edward Elgar Publishing, 2004) and G. De Geest, Economische analyse van het contracten- en quasi-contractenrecht (Antwerpen, Maklu, 1994).

The Formation of Contracts

69

1.1. In principle, the least cost information gatherer should produce and communicate information The least cost information gatherer (a concept introduced by Kronman) is the party that can obtain information at lower costs than the other party.4 To illustrate, suppose that a dwelling gets humid during the winter. The seller of a house will usually find this out for free, as a by-product of living there. The buyer may also discover the same facts, but only by hiring a professional inspector. Since the latter implies substantial costs, the seller is the least cost information gatherer, and the cheapest way to get the information to the buyer consists of obliging the seller to tell what he already knows. Hence, legal systems can substantially reduce information costs by inducing the least cost information gatherer to produce this information and transmit it to the other party. Note that the costs of the least cost information gatherer are not always zero. For instance, discovering the side effects of a drug is usually very costly, though it is obvious that these costs are lower for a pharmaceutical company than for its consumers. In most cases, the seller is the least cost information gatherer. A first reason why information costs are usually lower for sellers is that they often acquire information as a by-product of owning and using a good (like in the house with the latent defect example). A second reason is that sellers may have economies of scale (like in the pharmaceutical firm example). Finally, sellers are more likely to be professionals than buyers (like in the case of an art gallery).5 While sellers are the least cost information gatherers in most cases, they are not the least cost information gatherers in all cases, and hence a rule that would always put the informational burden on the seller would not be a good one. In nearly all cases, every party will be the least cost information gatherer about its own needs. An exception is a fiduciary relationship or another relationship of trust, when a party is explicitly asked (and often paid) to give advice on the other party’s needs.

1.2. Information should not be communicated if the other party has (or should have) it already Transmitting information has a cost. This explains why legal systems do not oblige sellers to explicitly mention manifest defects. If the windows of a house are broken, the buyer will easily notice it, and obliging the seller to express in words what the buyer already knows would be a waste of time.

1.3. Information should not be communicated if the communication costs exceed the value of the information Most buyers are not interested in the name of the forest in which the trees grew from which the hardwood floor has been made. Hence, as a simple economic rule of thumb, information should only be produced and communicated if the information production costs plus the

4

5

A. T. Kronman, “Mistake, Disclosure, Information, and the Law of Contracts” (1978) 7 Journal of Legal Studies 1-34. J. M. Trebilcock, The Limits of Freedom of Contract (Harvard University Press, 1993) at 102-119.

70

Gerrit De Geest & Mitja Kovač

communication costs are lower than the value of the information to the buyer.6 This value may differ for different buyers. Some are interested in knowing the type of the trees (oak?),7 while others are not.

1.4. There should be no duty to reveal entrepreneurial information Some information (like expertise on works of art) is costly to produce, either directly or indirectly in the sense that it costs several years to obtain some expertise. Typical examples are the French Lorenzo Lotto painting case (in which a specialist in renaissance art discovered a Lorenzo Lotto painting)8 or the German Mozart Notebooks case (in which Mozart notebooks were discovered at a flee market).9 Such information may be called entrepreneurial information,10 in line with Schumpeter and other members of the Austrian school of economics11 who emphasized the importance of individuals who add information to the market. Entrepreneurial information is information that is costly to produce, impossible to protect via intellectual property rights, and valuable to other market players, so that without a right to keep that information secret, free rider problems would discourage the production of the information.12 The only way to give incentives to produce entrepreneurial information is to give a right-to-lie-and-conceal with respect to such information. Shavell has stressed that there may be an important difference between the seller’s and the buyer’s situation.13 If the seller discovers that a painting has been made by Van Gogh, he can make money by selling the painting as a Van Gogh. If the buyer discovers this, he is in a difficult situation, since when the information gets into the hands of the seller, he will incur the profit instead of the buyer.14 While Shavell’s insight is valuable, it does not follow that 6 7 8

9 10 11

12

13

14

B. Jovanovic, “Truthful Disclosure of Information” (1982) 13 Bell Journal of Economics 36-44. See, for example, Smith v Hughes (1871) LR 6 QB 596. Lorenzo Lotto case, decided by the tribunal de grande instance of Paris on 6th of March 1985, reported in J. Ghestin, “The Pre-contractual Obligation to Disclose Information, 1: French Report”, in D. Harris / D. Tallon (eds.), Contract Law Today, Anglo-French Comparisons (Clarendon Press Oxford, 1989). See also La Verrou by Fragonard case, Civ 1, 25 May 1992, Bull Civ I no. 165, where the court found that annulling the sale could be unfair to a bona fide purchaser whose effort lead to the discovery. AG Coburg, NJW 1993, 938. De Geest, n. 3 above. E. g., F. A. Hayek, “The Use of Knowledge in Society” (1945) 35 American Economic Review 519530. Kronman, n. 4 above. See also P. V. Goldberg, “Note on Price Information and Enforcement of the Expectation Interest”, in P. V. Goldberg (ed.), Readings in the Economics of Contract Law (Cambridge University Press, 1989) 80-83; H. Kötz, “Pre-contractual Duties of Disclosure: A Comparative and Economic Perspective” (2000) 9 European Journal of Law and Economics 5-19; J. Gordley, “Mistake in Contract Formation” (2004) 52 American Journal of Comparative Law 433-468 and R. Cooter / T. Ulen, Law and Economics (4th Edition, Addison Wesley Longman, 2004) 275-276. S. Shavell, “Acquisition and Disclosure of Information Prior to Sale” (1994) 25 Rand Journal of Economics 20-36. Since buyers by definition do not possess property rights in the goods they are seeking, they will, by revealing the information that raises the value of the good to seller, confer an advantage upon seller, allowing them to charge more for the good which is now worth more to the seller. Hence, in

The Formation of Contracts

71

rules should automatically be different for buyers, since sellers can have a similar problem when it comes to discovering negative information (like when a seller sells his shares on the stock market after having studied the future demand for the company’s products).

1.5. There should be no duty to be honest with respect to mere opinions or other non-falsifiable statements Parties should have a right to lie-and-conceal-information with respect to non-falsifiable statements.15 Examples of non-falsifiable statements are: ‘Pepsi-Cola tastes better then CocaCola’, ‘My Toyota is fantastic’, or ‘John is a great lawyer’. These statements cannot be falsified by directly observing the empirical reality, because there is no generally accepted definition of concepts like ‘fantastic’ or ‘great’, or because the statement is inherently subjective (‘It tastes better’). Of course people do not always tell the truth. To sell more cars, they may say that their cars are great, even if they do not really believe that. Even so, producing evidence to the contrary is usually problematic. Very often, the only chance to prove a misrepresentation of opinions is to have the other party’s confession. Yet penalties for confessing parties may discourage honesty ex post rather than encourage honesty ex ante.16 In addition, it would give a comparative disadvantage to sellers with a critical and open-minded attitude towards their own products because they would be sanctioned more often than non-critical sellers.17

1.6. The difference between lying and concealing information is irrelevant One of the most important findings of economic scholarship is the notion that to tell nothing is always to tell something. What that something is depends on the type of market. In some markets, consumers who get no information on the quality of a product will presume it is of an average quality. In other markets (like in Akerlof’s market for ‘lemons’),18 consumers will presume the lowest quality. Whichever presumption is made, the distinction between explicitly lying and just concealing information (by saying nothing) is less relevant than lawyers tend to believe. Both activities are intrinsically costly and wasteful (the liar invests in misleading through words,19 the concealer invests in non-detection, and in both cases the non-liar invests in detection).20 and both lead to inefficient allocations. Note also that any concealment can be transformed

15

16 17 18

19 20

order to induce buyers to acquire productive information they must be allowed to decide whether to disclose it or not; Shavell, id. at 30. G. De Geest / T. Jaspers / J. Siegers / A. S. Vandenberghe, “The right to lie: new law and economics versus Dutch labor law,” in G. De Geest / J. Siegers / R. Van Den Berg (eds.), Law and Economics and the Labour Market, New Horizons in Law and Economics (Edward Elgar, 1999) 34-55, at 38. De Geest, n. 3 above, at 179. De Geest / Jaspers / Siegers / Vandenberghe, n. 15 above, at 39. A. G. Akerlof, “The Market for Lemons: Qualitative Uncertainty and the Market Mechanism” (1970) 84 Quarterly Journal of Economics 488-500. R. A. Posner, Economic Analysis of Law (New York, Aspen, 1998), at 122. S. Shavell, Foundations of Economic Analysis of Law (The Belknap Press of Harvard University Press, 2004), at 329. See also S. Levmore, “Securities and Secrets: Insider Trading and the Law of Contracts” (1982) 68 Virginia Law Review 117-60 and E. R. Barnett, “Rational Bargaining Theory

72

Gerrit De Geest & Mitja Kovač

into an explicit lie by being asked to explicitly state that there is nothing that has been concealed!21

1.7. The difference between intentionally and negligently giving wrong information is irrelevant, except for the fact that extra-compensatory sanctions may be needed for to deter intentional misinforming This conclusion can be derived by analogy from the conclusions of the economic literature on tort law, where the rules for intentional wrongdoing and negligence do not differ, except for in the sense that criminal law (or other forms of extra compensatory sanctions) may be required to deter intentional torts.22

1.8. There is no need for separate doctrines on mistake, fraud/misrepresentation or latent defects besides a duty-to-inform doctrine While law and economics scholars have hardly made any statements on the optimal formulation of legal doctrines,23 the literature suggests that all asymmetric information problems can be solved by a duty-to-inform doctrine (which is a more refined fraud doctrine). The only other doctrine that is needed is a risk allocation doctrine for symmetrical informational shortcomings. If party A is imperfectly informed, there are three possibilities.24 First, A is the least cost producer of information. In this case, A should be sanctioned for not having produced this information. The simplest sanction is not to intervene in the contract. This does not require a separate doctrine. Second, B is the least cost producer of information. This can be solved by a duty-to-inform doctrine. Third, neither party is able to produce the information at a reasonable cost. In this case we have – from an analytical perspective – a pure risk problem. The question then is which party is in the best position to bear this risk – which may be answered by doctrines such as ‘impossibility’ (the analysis of which goes beyond this scope of this article).25

21 22

23 24 25

and Contract: Default Rules, Hypothetical Consent, the Duty to Disclose, and Fraud” (1992) 15 Harvard Journal of Law and Public Policy 783-803. De Geest, n. 3 above, at 182. For a synthesis of the conclusions of the economic literature on tort law see H. B. Schäfer, Tort Law: General , in B. Bouckaert / G. De Geest (eds.), Encyclopedia of Law and Economics, Volume II. Civil Law and Economics (Cheltenham, Edward Elgar, 2000); Shavell, n. 20 above, 175-289; R. Cooter / T. Ulen, n. 12 above, at 287-373. On this, see De Geest, n. 3 above. De Geest, n. 3 above. See for instance Art. II.–7:102 DCFR – Initial impossibility: “A contract is not invalid, in whole or in part, merely because at the time it is concluded performance of any obligation assumed is impossible, or because a party is not entitled to dispose of any assets to which the contract relates.” This could be integrated into a more general impossibility rule in that the solution is identical: the superior risk bearer should bear the risk (and the criteria for being the superior risk bearer are the same). Posner and Rosenfield identify the superior risk bearer as one who is either in better position to prevent the risk from materializing (precaution) or is better able to insure against the risk (superior insurer).

The Formation of Contracts

73

1.9. ‘Consent theories’ or ‘will theories’ can never provide precise criteria for when a contract should be valid According to these theories, a contract is valid if, and only if, it is based on the real ‘will’ or ‘consent’ of both parties.26 The contract is void when the consent of at least one of the parties is missing. At the time of contracting, however, parties never have perfect information. Ex post, they have more information than ex ante, and in this sense they are always mistaken. The fact that they had imperfect information should be no reason to avoid a contract because otherwise contracts would never be binding. The question is how to ensure that an optimal amount of information is produced in society (which generally requires that the party who can produce valuable information at least costs gets an incentive to do so). Since the optimal amount will generally be higher than zero information and less than complete information, all-or-nothing concepts like ‘will’ or ‘consent’ are intrinsically inapt to providing clear answers.

1.10. There is no need for a condition that the party would not have entered into the contract Legal systems often want to filter out minor informational distortions, and do this by requiring that ‘the mistaken party would not have entered into the contract with correct information’.27 Any introductory textbook in economics, however, will reveal that this condition has serious analytical shortcomings. First, in a perfectly competitive economy, the slightest detail will lead to another decision. Second, in less than perfect competition, details may not change the decision of most consumers, but there will always be consumers at the margin for whom it will matter. To understand why, just consider a standard demand curve with a downward slope. The slightest switch of this curve to the left (caused by the decreased perceived utility of the good) reduces the effectively traded quantity (though it may be hard to find out who those infra-marginal consumers are). Thirdly, even in a bilateral monopoly – where the question is not whether the product will be bought but rather how the surplus will be divided between seller and buyer – any detail may marginally change the negotiated price. Allowing the seller to subtly mislead the buyer in order to extract a slightly higher price is not socially desirable. In modern legal systems, there is no reason to remedy only the most serious information problems. The real concern of legal systems is probably twofold. First, legal systems may want to discourage litigation over small amounts. The first-best (and most systematic) instrument to

26

27

See R. A. Posner / A. M. Rosenfield, “Impossibility and Related Doctrines in Contract Law: An Economic Analysis” (1977) 6 Journal of Legal Studies 63-118. For a synthesis on superior risk bearer see Cooter / Ulen, n. 12 above, at 203 and Posner, n. 19 above, at 104. For a comparative treatise on consent and will theories see, K. Zweigert / H. Kötz, An Introduction to Comparative Law (Third Edition, Clarendon Press-Oxford, 1998) at 323-400. See, for illuminating exposition of those theories, W. Flume, Algemeiner Teil des Burgelichen Rechts, Das Rechstgeschäft (3rd Edition, Berlin-Heidelberg-Nova Lorque, 1979) at 435; J. Gordley / A. T. Von Mehren, An introduction to the Comparative Study of Private Law: Readings, Cases, Materials (Cambridge University Press, 2006) at 413 et seq. and R. Zimmerman, The Law of Obligations: Roman Foundations of the Civilian Tradition (Cape Town: Juta & Co., 1990). E. g., French Civil Code, art. 1116 (condition for fraud).

74

Gerrit De Geest & Mitja Kovač

obtain this result is through general litigation rules (that define minimum amounts). Second, legal systems may want to avoid that small informational imperfections are used as an excuse to exit a regretted deal. But this objective can be reached by granting damages only (which will be too low to be worth the costs of claiming them), and no right to avoid the contract.

1.11. To what extent sellers would spontaneously reveal the information on the market is irrelevant for law-makers, since legal rules may still be needed for the few that would not Several economists have shown that – under some conditions – market mechanisms may solve asymmetric information problems.28 For instance, in the model of Grossman, all sellers with above-quality products have an incentive to reveal their information (e. g. via warranties).29 As a result, the average-quality of the products of the other sellers gets lower, so that more sellers reveal their information, until all sellers are obliged to reveal their information. The discussion as to what extent markets solve information problems is less relevant for lawyers and policy makers than for economists, because the cases in which the market fails still need to be corrected. Countries with well-functioning markets can still benefit from market-improving rules, just like countries in which very few criminals plan to rob a bank may still need sanctions on bank robbery. Besides, the literature has identified a number of imperfect market solutions to information problems. Spence has shown that when accurate information on quality (e. g., how good a lawyer is) cannot be given, consumers react to elements that are statistically correlated to quality (e. g., the price of the lawyer’s car, which is statistically correlated to the lawyer’s economic success and hence the quality of her work).30 Through signaling, the market may solve some information problems. Still, while signaling behavior is better than providing no information at all, signaling remains a relatively inefficient and wasteful technique to provide information (e. g., lawyers may spend more on cars than they strictly would like to). Similarly, warranties are believed to be a remedy to some adverse selection problems.31 To illustrate, consider a product for which simple and precise (verifiable) statements can28

29

30 31

See e. g. S. J. Grossman / O. D. Hart, “Disclosure Laws and Takeover Bids” (1979) 35 Journal of Finance 323-324; P. R. Milgrom, “Good News and Bad News: Representation Theorems and Applications” (1981) 12 Bell Journal of Economics 380-391; S. Matthews / A. Postlewaite, “Quality Testing and Disclosure” (1985) 16 Rand Journal of Economics 328-340; J. Farrell, “Voluntary Disclosure: Robustness of the Unraveling Result, and Comments on Its Importance”, in R. Grieson (ed.), Antitrust and Regulation (Lexington books, 1986); J. M. Fishman / M. K. Hagerty, “The Optimal amount of Discretion to Allow in Disclosure” (1990) 105 Quarterly Journal of Economics 427-444 and H. P. Rubin, “Information Regulation (Including Regulation of Advertising)”, in B. Bouckaert / G. De Geest (eds.), Encyclopedia of Law and Economics (Cheltenham, Edward Elgar, 2000). J. S. Grossman, “The Informational Role of Warranties and Private Disclosure about Product Quality” (1981) 24 Journal of Law and Economics 461-483. See also M. Okuno-Fujiwara / A. Postlewaite / K. Sozumura, “Strategic Information Revelation” (1986) 57 Review of Economic Studies, 25-47 (analyzing the conditions under which voluntary disclosure leads to complete disclosure of information). A. S. Spence, “Job-Market Signaling” (1973) 87 Quarterly Journal of Economics 355-379. See e. g. Grossman, n. 29 above and J. K. Smith / R. L. Smith, “Contract Law, Mutual Mistake, and Incentives to Produce and Disclose Information” (1990) 19 Journal of Legal Studies 467-488.

The Formation of Contracts

75

not be made (e. g. what is good car? A low probability of breakdown is difficult to express in simple black/white statements). Warranties are indirect statements on the quality. They should be read as: “I inform you that this product has quality level x. To convince you that I am honest, I am willing to bear all costs of repair during x months.” Still, warranties are an imperfect solution at best, since they also make the seller also bear the losses that are caused by characteristics they cannot know (‘external factors’), and caused by the buyer’s (non-verifiable) moral hazard. It has also been argued that some contract types induce parties to self-select.32 For instance, a farm that would pay per harvested kilogram rather than a fixed wage will probably attract workers with an above average productivity and induce less productive workers not to apply (in other words: it would induce ‘agents’ to reveal ‘their type’ or ‘self-select’). These ‘hard contracts’ come again at a cost: they usually create more risk than optimal. While these second-best solutions do exist, legal rules may improve the market by facilitating first-best solutions (directly revealing the information).

1.12. The signing-without-reading problem (asymmetric information on the contract itself) leads to similar problems as asymmetric information on the characteristics of the transferred object A special form of information asymmetry consists in the drafter of the contract possessing more information on the contract than the party that simply signs it.33 Analytically, this specific type of information problem is similar to the general type. Hence, it should be possible to integrate rules that deal with transparency (on the contract itself) and those that deal with information (on the goods or services), except for when it comes to remedies.

1.13. Less is more: parties get better informed when unimportant information is filtered away Since Herbert Simon34 and Tverski and Kahnemann35 there is growing economic literature that tries to integrate psychological insights into the standard economic models. One of Simon’s central ideas is that human brains suffer from information overload. As a result,

32

33

34

35

For a synthesis and bibliography, see K. Werth, “Warranties”, in B. Bouckaert / G. De Geest (eds.), Encyclopedia of Law and Economics (Cheltenham, Edward Elgar, 2000). E. J. Stiglitz, “The Theory of Screening, Education and the Distribution of Income” (1975) 65 American Economic Review 283-300 and E. J. Stiglitz, Economics of the public sector (W. W. Norton & Company: New York, 1986). G. De Geest, “The Signing-Without-Reading Problem: An Analysis of the European Directive on Unfair Contract Terms”, in H. B. Schäfer / H. J. Lwowski (eds.), Konsequenzen wirtschaftsrechlicher Normen (Deutscher Universitätsverlag, Auflage 1, 2002) at 213-235. A. H. Simon, “Theories of decision-making in economics and behavioral science” (1959) 49 American Economic Review 253-283. A. Tverski / D. Kahneman, “Prospect Theory: An Analysis of Decision under Risk” (1979) 47 Econometrica 263-291. For a nice example of more recent (and applied) literature, see O. Bar-Gill, “Seduction by Plastic” (2004) 98 Northwestern University Law Review 1371.

76

Gerrit De Geest & Mitja Kovač

more information is not always better – ideally, information should be presented in a filtered way. Note that this applies to the signing-without-reading problem as well. A simple technique to induce consumers to read contracts is to make sure that there is less to read.36

1.14. Synthesis. Towards an optimal doctrine on a duty to inform (incl. mistake and fraud) The previous points can be summarized into the following outline for an optimal doctrine: • There should be a duty for party A to inform party B if all the following conditions are fulfilled: • A is the cheaper cost producer of this information; • the information is valuable to B (i. e. the value is higher than the information and communication costs); • it is unlikely that B possesses the information already; • the information is not entrepreneurial (entrepreneurial information is costly to produce and hard to be compensated for once it is revealed); • the information does not consist of mere opinions and other non-falsifiable statements.

2. Comment on the Draft Common Frame of Reference (DCFR) Is the Draft Common Frame of Reference in line with these recommendations? We briefly comment on (parts of) the articles that (directly or indirectly) deal with problems of asymmetric information.

2.1. Art. II.–3:101 DCFR: Duty to disclose information about goods and services (1) Before the conclusion of a contract for the supply of goods or services by a business to another person, the business has a duty to give to the other person such information concerning the goods or services to be supplied as the other person can reasonably expect, taking into account the standards of quality and performance which would be normal under the circumstances. (…)

This article correctly states that not all information has to be given. It implicitly applies the less-is-more principle: unnecessary information should be filtered away. It is also based on the notion that to tell nothing is to tell something, though it implicitly assumes that consumers will make the average quality presumption – which does not hold for all markets. The main shortcoming of the article is that its central criterion is tautological. What consumers can “reasonably expect” depends on what the law tells the sellers to do. Parties may reasonably expect that the law is applied, and therefore the law has to better define what can reasonably be expected. The article has the same precision as a (similarly tautological) tort law rule that would state that the injurer should take the precautions that the victim

36

De Geest, n. 33 above.

The Formation of Contracts

77

may reasonably expect. (The American Learned Hand formula is more precise by stating that precaution costs should be made when they are lower than the expected accident costs).37 A more precise criterion can be found in the comments: “It may be reasonable for a party to be expected to disclose information if the costs for the other party of seeking that information are disproportionately higher than the costs of providing the information…unless there are valid reasons for not disclosing this information”.38 In the first part of this quote we find Kronman’s least cost information gatherer principle (which is refined through the inclusion of communication costs into the formula). The second part suggests that there may be exceptions to the least cost information gatherer rule, without mentioning that the main exceptions are entrepreneurial and non-verifiable information. The article may be improved by moving the more precise statements from the comments to the main text of the article. The comments stat: “The provisions in this Chapter only deal with obligations to provide information. False and misleading information is not dealt with in this context.”39 As explained supra,40 there is no reason to make a distinction between not informing and falsely informing. The comments also argue that “… the level of disclosure that may reasonably be expected by a consumer when dealing with a business will ordinarily be higher than if the consumer were dealing with a private supplier.” This is economically correct. A business has economies of scale, which makes it more likely that it will be the least cost information gatherer.

2.2. Art. II.–3:102 DCFR: Specific duties for businesses marketing goods or services to consumers (1) Where a business is marketing goods or services to a consumer, the business must, so far as is practicable having regard to all the circumstances and the limitations of the communication medium employed, provide such material information as the average consumer needs in the given context to take an informed decision on whether to conclude a contract. (2) Where a business uses a commercial communication which gives the impression to consumers that it contains all the relevant information necessary to make a decision about concluding a contract, it must in fact contain all the relevant information. Where it is not already apparent from the context of the commercial communication, the information to be provided comprises: (a) the main characteristics of the goods or services, the identity and address, if relevant, of the business, the price, and any available right of withdrawal; (b) peculiarities related to payment, delivery, performance and complaint handling, if they depart from the requirements of professional diligence.”

The term “average consumer”41 may not always correctly express what is meant – at least not when “average” is taken literally. The implicit balancing is probably the following: if 37

38 39 40 41

United Sates v Caroll Towing Co. (1947) 159 F.2d 169, 173, 2nd Cir. ; A. R. Posner, “A Theory of Negligence” (1972) 1 Journal of Legal Studies 29-48. Comments to Art. II.–2:301 Common Frame of Reference. Id. See section A.2.6. The comments define “average consumer” as a consumer that “is reasonably well-informed and reasonably observant and circumspect, taking into account social, cultural and linguistic factors”, the comments (Paragraph 1) to Art. II.–2:302 Common Frame of Reference.

78

Gerrit De Geest & Mitja Kovač

more information is given than the average consumer requires, this may be helpful for some consumers with less than average knowledge, but it may harm the above-average consumers by creating an information overflow. The optimum amount of information is not necessarily the amount that the average consumer desires. To illustrate, suppose that 60 % of the consumers are experienced enough so that they do not need much information, while 40 % are inexperienced consumers who need to be informed about the basics. Does this mean that the business may neglect the needs of the inexperienced group? Paragraph (2) contains the idea that there is no need to communicate what is visible and what is already known.42 Furthermore, the comments correctly take communication costs into account when mentioning that “… the limitations of the communication medium employed are relevant. Thus, where the business is communicating with the consumer by telephone, it may be possible to provide fewer items of information than when using a website, or e-mail communication.”43

2.3. Art. II.–3:103 DCFR: Duty to provide information when concluding contract with a consumer who is at a particular disadvantage (1) In the case of transactions that place the consumer at a significant informational disadvantage because of the technical medium used for contracting, the physical distance between business and consumer, or the nature of the transaction, the business must, as appropriate in the circumstances, provide clear information about the main characteristics of the goods or services, the price including delivery charges, taxes and other costs, the address and identity of the business with which the consumer is transacting, the terms of the contract, the rights and obligations of both contracting parties, and any available right of withdrawal or redress procedures. (…) (2) Where more specific information duties are provided for specific situations, these take precedence over the general information duties under paragraph (1).

This article standardizes the informational duties of the (professional) seller – which creates legal certainty on the one hand, and reduces the consumer’s search costs on the other hand.44

42

43 44

See the Common Frame of Reference Art. II.–2:302 Paragraph (2): “the following information must be provided to the consumer where this is not already apparent from the context of the commercial transaction”. Comments (Paragraph 1) to Art. II.–2:302 Common Frame of Reference. By reducing overall transaction costs this will allow better matches between contracting parties, consequently enhancing allocative efficiency. On search costs see e. g. A. P. Diamond / E. Maskin, “An Equilibrium Analysis of Search and Breach of Contract” (1979) 10 Bell Journal of Economics 282-316 and Shavell, n. 20 above, at 327.

The Formation of Contracts

79

2.4. Art. II.–3:106 DCFR: Clarity and form of information (1) A duty to provide information imposed on a business under this Chapter is not fulfilled unless the information is clear and precise, and expressed in plain and intelligible language. (…) (3) In the case of contracts between a business and a consumer concluded at a distance, information about the main characteristics of the goods or services, the price including delivery charges, taxes and other costs, the address and identity of the business with which the consumer is transacting, the terms of the contract, the rights and obligations of both contracting parties, and any available redress procedures, as may be appropriate in the particular case, need to be confirmed in textual form on a durable medium at the time of conclusion of the contract. (…)

This article obliges the seller to minimize communication costs by choosing intelligible language, and makes it at the same time harder to mislead the consumer. It further standardizes informational duties, which has significant economic benefits.

2.5. Art II.–7:204 DCFR: Liability for loss caused by reliance on incorrect information (1) A party who has concluded a contract in reasonable reliance on incorrect information given by the other party in the course of negotiations has a right to damages for loss suffered as a result if the provider of the information: (a) believed the information to be incorrect or had no reasonable grounds for believing it to be correct; and (b) knew or could reasonably be expected to have known that the recipient would rely on the information in deciding whether or not to conclude the contract on the agreed terms. (…)

Condition (a) (believed the information to be incorrect or had no reasonable grounds to believe it to be correct) is illustrated (in the Comment B of the original Art. 4.106 PECL) with an example of a seller of livestock who had better information than the inspector who erroneously believed the livestock to be healthy.45 The seller is indeed the least cost information gatherer here and should provide the final information. Illustration 1 of the comments46 states that it is unreasonable to buy a practice without checking the accounts, and thus unreasonable to fully rely on the incorrect information the seller provided. This example is a case of what is called ‘joint care’ in the economic analysis of tort law: both parties need to exert an effort to acquire information. The article opts for ‘contributory negligence’ by holding the ‘victim’ fully responsible for the incorrect information, rather than for the ‘comparative negligence’ which is more common in modern tort law (which holds both parties responsible by sharing the loss). The economic literature has shown that in joint care situations, optimal incentives are provided by both contributory negligence and comparative negligence, though the former involves less administrative costs.47

45 46 47

Comment B to Art. II.–2:307 Common Frame of Reference. See illustration 1, Comment C to Art. II.–2:307 Common Frame of Reference. See e. g. M. W. Landes / R. A. Posner, The Economic Structure of Tort Law (Cambridge (MA): Harvard University Press, 1987) and C. Curran / D. Haddock, “An Economic Theory of Comparative Negligence” (1985) 14 Journal of Legal Studies 49-72.

80

Gerrit De Geest & Mitja Kovač

Condition (b) requires that the uninformed party would not have contracted with better information. As shown in section 1.10, there is no need for such a requirement.

2.6. Art. II.–7:201 DCFR: Mistake (1) A party may avoid a contract for mistake of fact or law existing when the contract was concluded if: (a) the party, but for the mistake, would not have concluded the contract or would have done so only on fundamentally different terms and the other party knew or could reasonably be expected to have known this; and (b) the other party; (i) caused the mistake; (ii) caused the contract to be concluded in mistake by leaving the mistaken party in error, contrary to good faith and fair dealing, when the other party knew or could reasonably be expected to have known of the mistake; (iii) caused the contract to be concluded in mistake by failing to comply with a precontractual information duty or a duty to make available a means of correcting input errors; or (iv) made the same mistake. (2) However a party may not avoid the contract for mistake if: (a) the mistake was inexcusable in the circumstances; or (b) the risk of the mistake was assumed, or in the circumstances should be borne, by that party.

Paragraph 1(a) contains again the criterion that the other party would not have contracted otherwise, or on fundamentally different terms. On this see supra section 1.10. Next, paragraph 1 requires that either the other party is the one to blame for the mistake (conditions (i), (ii) and (iii)), or that both made the mistake (condition (iv)). In the latter case, either the blame is on the victim, or it is a pure risk problem. If the blame for the mistake is on the victim of the mistake (the case of paragraph (2) (a)) because the mistake was ‘inexcusable’ (that is, it could have been avoided at reasonable costs), the victim is sanctioned for that mistake by not getting the right to avoid the contract. The article implicitly applies the least costs information gatherer principle of Kronman. While the text of the article does not explicitly mention that the blameworthiness of the parties is a matter of their relative information costs, Comment E to Art. II.–7:201 DCFR makes it clear that information costs are believed to be crucial: “But there are many cases where … one party has at little or no cost acquired information that the other obviously does not have and which is crucial to the contract. In this case the non-mistaken party should not be allowed to take advantage of the other’s misapprehension (…).” This is further illustrated with an example of a house with latent defects. While the DCFR article does not mention entrepreneurial information as an exception, Comment E to Art. II.–7:201 DCFR implicitly does so: “There are situations, however, in which a party should be permitted to take advantage of another’s ignorance. (…) Another is where one party has gone to considerable trouble to obtain the knowledge that the other does not have. If the party cannot use this hard-earned knowledge there would be no incentive to invest in obtaining it in the first place and both parties will be left worse off.” The comments also briefly deal with the possibility of a sequential bilateral care, when “the second party is aware that the first has made a mistake and it would take little trouble to point it out” (Comment I to Art. II.–7:201 DCFR). This is illustrated by a bid for construction job that forgets to take into account the costs of excavating rock, though the information given to bidders indicates that this will probably be necessary. In a law and economics framework, this is a sequential bilateral care case. The proposed solution is similar to the last-clear-

The Formation of Contracts

81

chance doctrine in tort law. While the economic literature does not consider this doctrine to be a first-best solution (the first-best is marginal cost liability), it is generally defended as a more practical second-best solution.48 The DCFR article is formulated in a vague way (by using the words “… and contrary to good faith and fair dealing”), presumably to allow for “flexibility” on when to use the least cost information gatherer criterion and when the exceptions. Supra we have argued that there is no need make the criterion vague: the only exceptions to the least cost information gatherer principle are entrepreneurial information and non-falsifiable information. If it is a pure risk problem, paragraph (2)(b) states that the first question is how the parties want to allocate the risk. In the absence of express contractual provisions, the question is who should bear the risk – in other words, who is the superior risk bearer. (One illustration of Comment J to Art. II.–7:201 DCFR deals with a chartered yacht that was destroyed before the contract was concluded. The chandler is believed to be one who should bear the risk, because he is a professional, with superior information on the yacht). Comment F to Art. II.–7:201 DCFR on shared mistake (symmetric lack of information) correctly states that the question becomes “… whether the contract was intended to allocate the risk of the loss caused by the facts turning out to be different.” If the contract did not explicitly allocate the risk, each party has the right to avoid the contract, which results in loss sharing between the parties. Again, this is correct from an economic point of view: when no party is the superior risk bearer, risk sharing is usually optimal.

2.7. Art. II.–7:205 DCFR: Fraud (1) A party may avoid a contract when the other party has induced the conclusion of the contract by fraudulent misrepresentation, whether by words or conduct, or fraudulent non-disclosure of any information which good faith and fair dealing, or any pre-contractual information duty, required that party to disclose. (…) (3) In determining whether good faith and fair dealing required a party to disclose particular information, regard should be had to all the circumstances, including: (a) whether the party had special expertise; (b) the cost to the party of acquiring the relevant information; (c) whether the other party could reasonably acquire the information by other means; and (d) the apparent importance of the information to the other party.

A first comment is that there is no need for a separate doctrine on fraud, when there is already a duty-to-inform doctrine and a risk doctrine (as argued in section 1.8). Paragraph (1) correctly applies the notion that to say nothing (‘non-disclosure’) is to say something. The condition mentioned under (a) – whether the party has special expertise – suggests again that the fundamental criterion is the least costs information gatherer principle, since information costs are a function of the available expertise. The same applies to condition (c), which introduces the costs structure of the other party into the formula. Condition (d) adds another economically relevant variable: the value of the information (which determines whether it was worthwhile to incur the communication costs). 48

See T. J. Miceli, Economics of the Law (Oxford, Oxford University Press, 1977) at 58-70.

82

Gerrit De Geest & Mitja Kovač

Condition (b) – which states that the costs of acquiring the information is relevant – seems to be inserted to allow for the exception of entrepreneurial information. Comment B to Art. II.–7:205 DCFR states that ‘mere sales talk’ is not fraud, as in the case when a lessor of a computer states that it is “the best of its size on the market”. Here the exception of the non-falsifiable information is added to the rule.

2.8. Art. II.–7:207 DCFR: Unfair exploitation (1) A party may avoid a contract if, at the time of the conclusion of the contract: (a) the party was dependent on or had a relationship of trust with the other party, was in economic distress or had urgent needs, was improvident, ignorant, inexperienced or lacking in bargaining skill (…)

As explained in section 2.1, a fiduciary relationship or another relationship of trust is special in that a party has a duty to give advice on the other party’s needs. This is an exception to the general case in which every party will be the least cost information gatherer about its own needs.

2.9. Art. II.–9:402 DCFR: Duty of transparency in terms not individually negotiated “(1) Terms which have not been individually negotiated must be drafted and communicated in plain, intelligible language.(…)”

This article (as well as article II.–9:408 DCFR) makes sense in the light of signing-withoutreading (and signing-without-understanding) problems, by obliging contract drafters to use optimally clear language, which reduces the other party’s reading costs.

Conclusion We have identified the principles that can be distilled from four decades of economic analysis of contract law and compared these to the principles that have been drafted for the Common Frame of Reference. We found that the latter often has an implicit economic logic. Yet, in several cases, the economic logic is more explicitly visible in the official comments, while the texts of the articles itself have an unnecessary degree of vagueness. Our analysis was not a political one, and hence we do not know whether vagueness is desirable for political reasons. The Draft Common Frame of Reference (and the PECL) is an enormous intellectual achievement. Even so, our analysis suggests that at least from a scientific viewpoint it must be possible to formulate the principles of European contract law in an even more accurate and simpler way.

Contract Interpretation – Interpretive Criteria Geerte Hesen* & Robert Hardy**

Introduction Contractual disputes are often founded in interpretation issues. Interpretation disputes arise either because parties have foregone to specify their rights and duties under the particular circumstances, or because they attach a different meaning to an element of their agreement. This is mainly due to the fact that, besides the ineptness in the use of words, most contractual transactions do not occur on the spot but over time: “contracts regulate the future and interpretive problems are bound to arise simply because the future is inherently unpredictable.”1 The problem of interpretation provides a central backdrop for the law of contract, which contains many rules and principles designed to address it.2 Compared to the economic analysis of contract law in general, however, little attention has been paid to the specific field of the economics of contract interpretation. The way a contract is interpreted relates to the actual law governing parties’ relationship. There are many factors that influence parties’ decision what to incorporate in their contract and in what form. One of those factors is the criteria proposed by the law to be followed by judges when called upon to resolve issues of contract interpretation. Economic theory can offer suggestions as to what constitutes efficient interpretative criteria. In this chapter, we use an economic approach to analyse the rules on contract interpretation of the Common Frame of Reference (the “CFR”). In Section 2, we address the issue of contract interpretation in general and link it to the economic issue of contractual incompleteness. In the third Section, we examine the different types of rules of interpretation and the incentives they create. In Section 4, the CFR rules on contract interpretation are scrutinized. This chapter ends with a conclusion.

* **

1

2

Associate, De Brauw Blackstone Westbroek; J. D., M. A., Ph. D., Maastricht University. Associate, Stibbe; J. D., Ph. D., Maastricht University; LL. M., Harvard Law School. This chapter was completed in June 2009 and does not incorporate any changes or amendments to the DCFR or CFR as of such date. Posner, R. A., ‘The Law and Economics of Contract Interpretation’, Texas Law Review 83 2005, p. 1583. Hermalin, B. E., Katz, A. W., Craswell, R., ‘Chapter on the Law & Economics of Contracts’ in: Polinsky, A. M., Shavell, S. (eds.), Handbook of Law and Economics I, North-Holland: Elsevier 2007.

Geerte Hesen & Robert Hardy

84

1. Incomplete Contracts and Contract Interpretation Almost everyone enters into contracts every day. From an economic perspective, a contract refers to any agreement under which two or more parties engage in mutual commitments concerning their behaviour.3 Within the legal context, a contract can generally be defined as a meeting of minds which creates effects in law. Parties contract in order to create a bigger transactional pie in a world where parties’ incentives are misaligned. The writing of contracts enables parties to hedge against market risks, and to coordinate the production of information, rights, duties and procedures. Contract lawyers seek to translate an economic deal into a legally enforceable written agreement. This comprises a planning role which involves a range of issues such as the resolution of ambiguities by suggesting express terms which describe in greater detail the deal, the allocation of risks for foreseeable and unforeseeable contingencies, the incorporation of legal and non-legal sanctions, and the introduction of mechanisms of dispute resolution. A contract performs the function of a governance mechanism: it comprises a clear record of the terms agreed upon between parties and the implied promise to adhere to these terms.

1.1. Incomplete Contracts Regardless of substantial planning efforts, contracts break down because real world contracts are invariably incomplete. In general, a contract is complete when it allocates the rights and obligations of parties efficiently across all possible future contingencies.4 A contract qualifies as incomplete if at least one of the following elements is not fulfilled: (a) the contract does not specify parties’ rights and obligations for all future contingencies; or (b) the contract does not exploit all gains from trade. The first element signifies that a contract fails to specify the obligations of parties under a certain state of the world which may or may not occur. The contract will contain a “gap” and courts will have to decide whether they will fill this gap or declare the contract void and unenforceable for matter of vagueness.5 The second element implies that the contract fails to efficiently allocate obligations across all states of the world. In other words, the terms that have been incorporated are not optimal from an economic 3

4

5

See also Brousseau, E., Glachant, J. (eds.), The Economics of Contracts: Theories and Applications, Cambridge: Cambridge University Press 2002. Eggleston, K., Posner, E., Zeckhauser, R., ‘The Design and Interpretation of Contracts: Why Complexity Matters’, Northwestern University Law Review 95(1) 2000, p. 100; Ayres, I, Gertner, R., ‘Strategic Contractual Inefficiency and the Optimal Choice of Legal Rules’, Yale Law Journal, 101 1992, p. 730; Schwartz, A., ‘Incomplete Contracts’ in: Newman, P. (ed.), The New Palgrave Dictionary of Economics and the Law, Volume II, Macmillan Reference Limited: London 1998, p. 277-278, Baird, D., ‘Self-Interest and Cooperation in Long-Term Contracts’, Journal of Legal Studies vol. XIX 1990, p. 583-596. See amongst others Schwartz, ‘Relational Contracts in the Courts: An Analysis of Incomplete Agreements and Judicial Strategies’, The Journal of Legal Studies, Vol. 21, No. 2, 1992, p. 271-318; Schwartz, A., Scott, R., ‘Contract Theory and the Limits of Contract Law’, Yale Law Journal (113) 2003, p. 540-596; Hadfield, G., ‘Judicial Competence and the Interpretation of Incomplete Contracts’, Journal of Legal Studies (23) 1994, p. 159-184; Ayres and Gertner 1992; Ayres, I., Gertner, R.,‘Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules’, Yale Law Journal 99(87) 1989, p. 87-130.

Contract Interpretation – Interpretive Criteria

85

point of view.6 Legal scholars tend to relate the incompleteness of a contract primarily with the first element, while economic scholars relate the incompleteness of a contract with the second element.7 The incompleteness of contracts is grounded in several different factors. Economic scholars attribute the incompleteness of contracts primarily to the fact that perfect markets – the concept upon which they base complete contracts – do not exist: markets “fail”. This failure to create maximum efficiency in the market is ascribed to factors such as bounded rationality, information asymmetries and transaction costs. Bounded rationality encompasses the thought that foresight is infinitely costly, and, therefore, as the economic literature on contractual interpretation emphasizes, the costs of foreseeing and providing for every possible contingency that may affect the costs of performance to either party over the life of the contract are prohibitive.8 Even in a setting of perfect foresight an interpretive problem may arise. This bounded rationality affects contracting parties as well as third parties. This implies that, while some variables might be perfectly well observable by the parties to a contract, these variables are difficult to measure and specify in such a manner that they can be proven to the satisfaction of the court. The variables are non-verifiable.9 Therefore, even if parties know which performance they desire and can personally observe whether this performance has occurred, the written contract is incomplete because performance needs to be specified in terms of an easily measurable proxy that can be readily observed by courts. Bounded rationality and information asymmetries raise transaction and enforcement costs. These costs include (a) the costs of gathering and processing information; (b) the costs related with negotiating, specifying and drafting an agreement; (c) the costs parties incur in order to ensure that each of them adheres to the agreement and fulfils his or her promise; and (d) the costs related to the enforcement of the contract by a third party, e. g., a court or arbitrator, including the uncertainty and error costs associated with such an enforcement.10

6

7 8

9

10

See amongst others Ayres and Gertner 1992; Eggleston et al 2000; Grossman, S., Hart, O., ‘The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration’, Journal of Political Economy (94)4 1986, p. 691-719; Hart, O., Moore, J., ‘Foundations of Incomplete Contracts’, Review of Economic Studies (66) 1999, p. 115-138; Williamson, O., The Economic Institutions of Capitalism: firms markets, relational contracting, New York: Free Press 1985, 450 p.; Maskin, E., Tirole, J., ‘Unforeseen Contingencies and Incomplete Contracts’, Review of Economic Studies (66) 1999, p. 83-114. Van Bijnen, R., Aanvullend Contractenrecht, Den Haag: Boom Juridische Uitgevers 2005, p. 87-88. See, for example, Richard A. Posner, Economic Analysis of Law 96 (6th ed. 2003); J. Tirole, ‘Incomplete Contracts: Where Do We Stand?’, Econometrica 67 1999, p. 741, 772. See amongst others Schwartz 1992; Triantis, G., ‘The Efficiency of Vague Contract Terms: A Response to the Schwartz-Scott Theory of U. C.C. Article 2’, Louisiana Law Review (62) 2002, p. 10681069, Williamson, O., The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, New York: Free Press 1985, 450 p; Bernheim, B., Whinston, M., ‘Incomplete contracts and strategic ambiguity’, The American Economic Review 1998 (88)4, p. 902-932; Maskin, E., ‘Incomplete Contracts: On Indescribable Contingencies and Incomplete Contracts’, European Economic Review 2002 (46), p. 725-733. Cf. Scott, R., Triantis, G., ‘Harnessing Litigation by Contract Design’, Yale Law Journal 115 (4), 2005-2006, p. 822-823, who refer to search and contracting costs as “front-end (transaction) costs”, and monitoring and enforcement costs as “back-end (enforcement) costs”; Schwartz 1998, p. 278279; Williamson 1985; Ayres and Gertner 1989, p. 92-93.

86

Geerte Hesen & Robert Hardy

Similar factors, such as transaction and enforcement costs and information asymmetries, have been used by legal scholars to explain why most contracts are incomplete.11 In particular, lawyers attribute the incompleteness specifically to inevitable ambiguities in ordinary language or to the bounded rationality on the part of the contracting parties. These factors thus undermine the ability of parties to construe complete contracts.12 As Scott states: “[P] arties write incomplete contracts either because (a) the resource costs of writing complete contingent contracts to solve contracting problems would exceed the expected gains or would exceed the costs to the state of creating useful defaults, or (b) the parties are unable to identify and foresee uncertain future conditions or are incapable of characterizing complex adaptations adequately.”13

1.2. Incompleteness, Interpretive Issues and Methods of Interpretation The issue of contract interpretation relates to the issue of incompleteness. When the contractual incompleteness is such that the contract remains silent on a particular issue, the parties may bring their case before court and the court will use its interpretive skills to attempt to fill the gap. In other situations, the incomplete contract may not be silent but rather unclear. Parties will have said either too little or too much with regard to a particular element of their agreement. In this case, courts will interpret the vague terms.14 The manner in which courts enforce incomplete contracts is essential to the effort that parties have to make in drafting their contracts. Depending on the manner in which courts interpret agreements, parties will exert more effort in their contract drafting or conversely rely on the court to supplement their agreement in case of an interpretive dispute. Two general methodologies that courts use to interpret contracts can be distinguished and determine the intentions of contracting parties with respect to their performance obligations: textualism and contextualism.15 Textualism refers to an objective approach to contract interpretation and contextualism to a subjective approach to contract interpretation. Some authors also employ the term “formalism”. Formalism can be projected on a scale where a high degree of formalism corresponds to textualism and a lesser degree of formalism to contextualism. Textualism is grounded in the idea of complete contracts, while contextualism is rooted in the idea of incomplete contracts. The theory of complete contracts assumes that parties have incorporated all necessary matters in their written agreement. Textualism relates to this theory as it holds that the words of the agreement govern its interpretation as opposed to inquiries into non-textual sources such as the intention of parties or surrounding context. 11

12

13

14 15

See amongst others Scott, R., ‘The Rise and Fall of Article 2’, Louisiana Law Review (62) 2002, p. 1038; Schwartz 1998, p. 277-282; Schwartz 1992, p. 278-281; Triantis 2002, p. 1071-1072. Other causes of market failure are imperfect competition, externalities and organizational failures. These factors, however, play a less prominent role in defining incompleteness (cf. Van Bijnen 2005, p. 117-137). For a more extensive discussion of these factors see amongst others Cooter, R., Ulen, T., Law & Economics, Pearson Addison Wesley: Boston 2003, p. 217-226. Scott, R. E. ‘The Case for Formalism in Relational Contract’, Northwestern University Law Review 94 2000, p. 847, 862. Ayres and Gertner 1989, 1992. The terms “contextualism” and “textualism” originate from the literature on statutory interpretation, where textualism is referred to as a formalist theory of interpretation.

Contract Interpretation – Interpretive Criteria

87

The contract in writing is presumed to be a complete and precise statement of the terms agreed between parties and if it is not included in the written document it is not part of the contract and vice versa; if it is in the written document, it is part of the contract. The whole of the agreement between parties is thus encompassed in the written document. This corresponds to the idea of completeness which signifies that the contract specifies the legal consequences of every possible state of the world. Completeness implies that the contract on face value should not contain any gaps. When a dispute concerning the interpretation of the terms of the contract does arise, the basis for an interpretive methodology embedded in textualism is to focus on the terms as expressed in the contract. The court will focus on the plain and ordinary meaning of the words as understood by ordinary and reasonable persons.16 Under such an objective approach to contract interpretation, courts must ascertain the intention of the parties from the words as expressed in the written agreement. The amount of information or base of materials that the court can take into account to ascertain what parties have meant in this case is restricted. An example is the “four corners” rule, a basic rule of contract interpretation in American law. The rule bars the parties to a written contract that is “clear on its face” from presenting evidence bearing on interpretation, which is to say “extrinsic” evidence – evidence outside the “four corners” of the written contract itself. And the judge alone determines what the contract means when no extrinsic evidence is presented, the theory being that he is a more competent interpreter of a document than a jury is. By limiting inputs into the interpretive process, proponents of textualism hope to reduce the risk of judicial error. Contextualism on the other hand, implies that a court will primarily inquire into the actual intention of the parties. The common or actual intention of the parties will prevail above the words as written in the contractual document and norms such as reasonableness and fairness play a strong role in determining parties’ agreements. In this case, the base of information or materials that the court can take into account in order to ascertain the actual intention of the parties is potentially unlimited. The court must discern what parties have actually meant, i. e. uncover the expectations of parties, by considering not only the written words of the contract but also other contextual evidence, which is used to interpret the scope of the contract. Courts must then be careful to give weight only to outward manifestations of intent and not to the secret intentions of one party. However, the task of determining what the contract is necessarily extends beyond the “four corners” of the written agreement. Rules on contract interpretation reflect a certain interpretive style by privileging certain materials and discounting others. Based on the above, these rules are termed formalistic when they confine the attention to a subset of materials that may or may not give rise to the same assumptions as would the whole of the materials. A rule that centres on the objective approach is, for example, the rule of parol evidence, which is common under common law. This rule provides that the written document integrates parties’ agreement and the agreement may not be contradicted or varied by other prior oral evidence. On the other hand, the subjective approach relates much more to the materials as a whole and an understanding based on the materials reasonably available.

16

Lake v. Simmons [1927] AC 487, Hayward v Norwich Ins. Ltd. [2001] Lloyd’s Rep. I. R. 410 (C. A.).

Geerte Hesen & Robert Hardy

88

2. Economic Consideration and Interpretive Strategies The goal of a doctrine of contractual interpretation is usually to minimize transaction costs. Because methods of reducing transaction costs are themselves costly, careful trade offs are necessary between the benefits of reducing transaction costs and costs of employing methods to do so. From an economic perspective, the objective and subjective approach to contract interpretation can be linked to the issue of optimal information gathering. The choice between the objective (textual) and subjective approach (contextual) to contract interpretation regulates the information transmission between parties and between the parties and the court. The interpretation of a contract is an ex post exercise, which implies that the court can only attempt to (partially) verify events which have taken place ex ante. Parties thus can often observe events that cannot be verified by the court. Events may be partially verifiable in the sense that a third party such as a court cannot observe these events with the similar precision as the parties to the contract. More in particular, doctrines relating to interpretation can be associated with the relationship between contracting parties, the third party enforcer and the extent to which burden of information acquisition is allocated between these two parties. A court’s interpretation of the contract is thus dependent upon the information available to it. This information set draws from three potential sources: the information embedded by the parties in the contract itself, the information presented by the parties to the court ex post, and the information available to the court based on its general knowledge and experience. Information communicated to the court by the parties is either based on a cooperative strategy in an attempt to maximize the returns from contracting or litigation, or on a non-cooperative strategy where each party individually decides what information to introduce. The introduction of information by the parties themselves is costly. Posner states that regarding the available information, a trade off takes place between type I errors and type II errors.17 A type I error implies that the court will enforce a term that is not part of a contract. A type II error on the other hand, implies that the third party adjudicator fails to enforce terms that are part of the contract. In particular, the interpretive regime of contracts consists of a function that specifies how the agent bestowed with the interpretive role, such as the court, understands possible sources of information. A fuller and broader context, in the sense of a broader base of materials, as under the subjective approach, can always be purchased. This will, however, happen at cost of time, trouble and the inference of incentives.18 The introduction of information to the court is costly, because it raises transaction costs for parties. However, the costs of ex post and ex ante information gathering interact, and more information can also help the court reach a more efficient decision from an economic perspective. In well-developed markets, courts should generally allow the context evidence to supplement express terms, but should generally not allow context evidence to override the plain meaning of express terms.19 As previously mentioned, an objective approach or more formalist approach to contract interpretation would be one that limits the influence of certain types of information in contrast to a subjective or less formal approach where all information available is considered.20

17 18 19

20

Posner 2005. Hermalin et al. 2007. Goetz, C., Scott, R., ‘The Limits of the Expanded Choice: An Analysis of the Interaction between Express and Implied Contract Terms’, California Law Review (73) 1985, p. 313. Hermalin et al. 2007.

Contract Interpretation – Interpretive Criteria

89

Given all these considerations, it is difficult to draw conclusions on which interpretive strategy should be preferred. When the assumptions on the costs and benefits of information acquisition are restricted, more specific conclusions may be drawn. Schwartz and Scott, for example, develop a model in which parties benefit when contract terms are interpreted correctly on average,21 and are quite risk-neutral to interpretive variance around the mean. In such a case it would be optimal for the court to make decisions on a minimum evidentiary basis while additional interpretative efforts are costly. Other authors on the other hand, assume that ex ante contract writing is costly and it is possible for parties to write at least some interpretive rules which may be applied without costs. In this case it is thus optimal for parties to leave at least some interpretive issues concerning contract terms to the courts. If contract-writing costs are sufficiently high, courts may override these terms in favour of another interpretation. Legal systems and harmonization instruments reflect different approaches that can be adopted and which enable the introduction of objective elements in to the interpretative process. The manner in which issues of interpretation are addressed and the types of legal rules that play a role in interpretive strategies, is largely dependent upon the legal system in which the dispute is brought before court. Common law jurisdictions generally take a more objective approach toward the interpretation of contracts. Civil law systems on the other hand, adhere to a rather subjective theory of contract interpretation. Civil law systems usually do not place limits on the kind of admissible evidence. In addition, dependent on the jurisdiction, courts may be more or less lenient to supply a term to remedy a contractual gap.

2.1. Types of Interpretive Strategies Apart from the two aforementioned main interpretive strategies, the textual or objective approach and contextual or subjective approach to contract interpretation, a range of different interpretive strategies exist which can be used to fill gaps or interpret vague terms. Economic analysis can help to identify the conditions under which various interpretive strategies approximate parties’ intentions and make courts and parties better off. Economists have proposed a number of interpretive strategies for courts.22 Some of the main strategies are discussed here. The regime of contract interpretation influences contracting parties’ behaviour in many respects. This concerns decisions to breach, advance precautions, mitigation of damages, the gathering of information, allocation of risks, et cetera. The doctrines concerning default rules for example relate to the relationship between the parties and third party enforcer, but are also concerned with the contracting parties’ informational relationship with each other.

2.1.1. Majoritarian Default Rule

The idea of a complete contract is closely related to the notion of default rules. These are rules which are used to fill a gap in a contract if parties have foregone to specify their actions under a particular contingency. The function of default rules is to provide guidance to third party enforcers in interpreting contract terms on which parties do not otherwise (clearly) 21 22

Schwartz and Scott 2003. Goetz and Scott 1985, p. 261-322.

90

Geerte Hesen & Robert Hardy

agree. A judge may choose to fill a gap in the contract with a default rule.23 Scholars in the field of law and economics literature have taken different approaches to the manner in which default rules should be set. One of these approaches has been introduced by Goetz and Scott who argue that the default rules should be set in such a manner that these minimize the cumulative transaction costs incurred by parties contracting around them.24 In the case where all parties face similar transaction costs and it is equally costly to contract around the terms, this implies terms that would be favored by a majority. In case of a majoritarian default, parties know that courts will apply a term that maximizes the probability of efficient trade. Accordingly, they would be more willing to enter a contract in the first place, despite high transaction costs, than they would under an alternative rule. In choosing a majoritarian default rule, the negative consequences of high transaction costs are reduced. In a system of a majoritarian default, the courts minimize transaction costs by choosing a mix of express and implied terms that most contracting parties would prefer. The majority of parties might want the court to use implied terms, because this saves contracting costs. Alternatively, a majority might believe that relying solely on express terms may be less reliable than relying on well established implied terms. An example of a majoritarian default rule is that contracting parties in a certain trade are bound by trade usages even when they are not familiar with them. Such a majoritarian default encourages parties in the particular area of trade to develop such usages and familiarize themselves quickly with them. The implied majoritarian default reduces the need for prolix documents. The majoritarian default rule acts like a public good in that it creates a standard set of terms. When an interpretation dispute arises, the majoritarian rule can be applied to the interpretation of express terms. In this sense, a plain meaning rule may also create an incentive for parties to learn the common meaning of words. This reduces the need for and costs of elaborate definition and explanation.25 The disadvantage of the majoritarian default theory is that empirics are necessary in order to define majoritarian rules. It is essential for the legislator and judges to uncover which rules are favoured by the majority of parties. This is difficult to achieve. In addition, it is necessary to determine when parties will contract around the majoritarian default. Goetz and Scott argue that default rules should be chosen to provide terms that would minimize cumulative transaction costs incurred by parties contracting around them.26 In the special case where all parties face similar contracting costs and it is equally costly to contract around all terms, terms would be favoured by the majority. If some terms on the other hand are costlier to contract out of or into than others, it is harder for parties to specify multifactor standards than simple rules. In this case, ceteris paribus, the default should be set to terms that are easiest for parties to escape.27 This is a transaction costs reducing approach to default rules. Moreover, Posner states that the responsibility for choosing default rules puts an unrealistically high informational burden on the courts.28 But although it is true that literalism places a lighter burden on courts, it does not mean that literalism is superior to the majoritarian approach. The choice between the two approaches is an empirical question about which we have no 23 24 25

26 27 28

Goetz and Scott 1985; Schwartz and Scott 2003; Ayres and Gertner 1992. Goetz and Scott 1985 Cohen, G. M., ‘Implied Terms and Interpretation in Contract Law’, in: Bouckaert,B., De Geest, G. (eds.), Encyclopedia of Law and Economics, vol. IV, Elgar, 2000, p. 78. Goetz and Scott 1985. Schwartz and Scott 2003. Schwartz 1998; Posner 2002.

Contract Interpretation – Interpretive Criteria

91

evidence. Schwartz stresses the complexity of the choices that the majoritarian approach requires courts to make, but he discounts the benefits to parties, who save on transaction costs to the extent that courts succeed, and can design their contracts to minimize risk to the extent that courts fail.29

2.1.2. Hypothetical Standard

Posner states that the goal of contract interpretation and manner in which courts should interpret contracts, is to save transaction costs.30 Therefore, the incentive should be to save transaction costs by completing a vague norm in the manner that parties would have agreed upon if they would have taken the effort to write a complete contract. Of course, it is difficult to discern which terms parties would have used. In this case, it is important to ascertain the actual intention of the parties, and the courts will use a broad base of materials in order to do so. This approach seems to relate very much to the subjective approach to contract interpretation. In order to distil the actual intention of the contracting parties, the court will indeed need to take into account the whole of the background materials or context in which the transaction is set. Only in this manner will the court be able to come within reach of the actual intention of parties ex ante. Account should be taken of the fact that in this case a type I error (see above) is more likely to arise, i. e. the court will risk enforcing a term that is not part of the contract. Benson argues in this respect that it is highly unlikely that anyone can know what rule parties to an actual contract with individual differences in bargaining power, information, risk-aversion, would have agreed to ex ante.31 In addition, legislators and judges, depending on their background and specialized knowledge will have different views on what parties would have agreed to if they had addressed the issue. Benson instead argues that default rules are in fact established by collective action and that it is very difficult to discover what percent of relevant contracting parties would have proposed the particular default rule if they had considered the issue.

2.1.3. Penalty Default

Strategic interpretation has as goal to reduce irrationality and information asymmetry.32 The strategy is to choose a “penalty default” – a meaning of a word that most parties to a particular contract would not use. This strategy, which would give parties an incentive to write a less ambiguous contract than they might otherwise, has two motivations. First, it discourages parties from externalizing the cost of interpreting the contract by way of the courts. If parties are clearer, courts will have less work to do. Second, it discourages parties from opportunistically concealing information from each other. If one party knows about the ambiguity of a particular word and prefers the majoritarian meaning, and the other party does not know about the ambiguity, then the first party would have no incentive to disclose 29 30 31 32

Schwartz 1998. Posner 2003. Benson, P. (eds.), A Theory of Contract Law, Cambridge: Cambridge University Press 2007. Schwartz and Scott 2003; Schwartz 1998.

Geerte Hesen & Robert Hardy

92

the ambiguity to the second, unless a penalty default rule held the informed party to the less favorable meaning. Ayres and Gertner state that textualism may be considered as a type of penalty default rule in the sense that it prevents strategic obfuscation by limiting judicial inquiries to the face of the text.33 The third party enforcer should not seek to consult contextual elements. Another example offered by Ayers and Gertner is the doctrine of contra proferentem, the doctrine under which ambiguous terms are interpreted against the interest of the drafting party. This rule serves as an incentive for the drafter to choose clear language and will disclose to the other party the meaning of the relevant terms. Penalty defaults thus are terms that operate as a penalty for non-disclosure. This can also be considered a type of information forcing approach.

2.1.4. Efficiency

Another strategy is for the court to enforce whatever term would be efficient in the particular case. One can derive this term by asking the question, supposing that transaction costs had been zero at the time of contracting, what would the parties have done? Buyer and seller would have anticipated their dispute about the meaning of a vague term, and either chosen a more precise term (if trade is still efficient) or not made a deal (if trade is not still efficient). What they would have done depends on the costs and values of the trade. The difference between this strategy and the majoritarian default is the difference between a standard and a rule. The court chooses whatever is efficient for the contract in dispute, rather than enforcing whatever term is efficient for the majority of parties who enter similar or identical contracts. This approach thus advocates choosing default rules for their substantive efficiency. This approach is especially appealing in settings with network externalities, adverse selection and bounded rationality in the contracting environment. Under these conditions parties are then reluctant to depart from familiar and widely used terms. Korobkin for example illustrates that cognitive and social-psychological factors induce parties to keep stick to default contract terms that they would actually be better off contracting around.34 Kahan and Klausner on the other hand have argued that that the network externalities which we find in the corporate and business environment regarding contracting practices, are due to the regular use of standardized forms.35 The value of these forms of course depends on the population of the users that have invested their expertise in the application. What rules are considered efficiency enforcing default rules is very context dependent. This perspective supposes that the third party enforcer can choose from a wide range of available defaults and imply whichever term is efficient in a particular situation. Efficiency effects may be created by regular use of standardized forms as Kahan and Klausner indeed argue. Another element however which plays a role here is that this perspective assumes that the court is able to enforce efficient terms. However, the court is also affected by bounded rationality, information asymmetries and cognitive biases which may lead to interpretive errors. In addition, the court would need to acquire both ex ante and ex post information on the contractual setting in order to determine the most efficient outcome or term at the time of the dispute and taking into account the turn of events which have lead to the dispute in 33 34 35

Ayres and Gertner 1992, 1989. Korobkin, R., ‘The Status Quo Bias and Contract Default Rules’, Cornell Law Review 83 1998. Kahan, M., Klausner, M. D., ‘Path Dependence in Corporate Contracting: Increasing Returns, Herd Behavior and Cognitive Biases’, Washington University Law Quarterly, 74(2) 1996.

Contract Interpretation – Interpretive Criteria

93

the first place. In short, high expectations are placed upon the third party enforcer under a strategy of efficiency.

3. Application to the CFR Rules on Interpretation The rules indicated in the proposed structure of the CFR appear to broadly reproduce those already contained in the Principles of European Contract Law. Article 8:101(1) of the CFR reveals that the “contract is to be interpreted according to the common intention of the parties even if this differs from the literal meaning of the words”. This is a subjective (or contextual) approach to contract interpretation. Paragraph (1) seems to follow the approach of the civil law systems where the common intention prevails over the plain meaning of words or the text within the four corners of the document. However, Article 8:101(3) of the CFR adopts a more objective approach: the contract is to be interpreted according to the meaning which a reasonable person would give to it if an intention cannot be established under Article 8:101(1) and (2). Article 8:105 also adopts a more objective approach: “Terms and expressions are to be interpreted in the light of the whole contract in which they appear.” We do not believe the subjective approach of Article 8:101(1) should have been chosen as the default. It seems to us that, if at all, firms in international transactions will choose the CFR to govern their contract. We agree with Schwartz and Scott that firms probably typically prefer courts to adhere as closely as is possible to the ordinary meanings of the words the parties used, i. e., in our opinion firms typically prefer the textualist approach.36 At the same time, the textualist theory will not suit all parties all of the time. Therefore, parties should have the possibility to contract around the textualist default. It should be noted, however, that ultimately the preference of firms between the objective and subjective approaches is an empirical question about which we have no real evidence. Article 8:101(2) of the CFR incorporates a penalty default rule: “If one party intended the contract, or a term or expression used in it, to have a particular meaning, and at the time of the conclusion of the contract the other party was aware, or could reasonably be expected to have been aware, of the first party’s intention, the contract is to be interpreted in the way intended by the first party.” Article 8:103 of the CFR can also be characterized as a penalty default rule: “Where there is doubt about the meaning of a contract term not individually negotiated, an interpretation of the term against the party who supplied it is to be preferred.” Article 8:102 establishes a matrix of relevant matters which may be taken into account when interpreting the contract, including the circumstances in which the contract was concluded (e. g., preliminary negotiations), the conduct of the parties, the interpretation which has already been given by the parties to terms or expressions, the meaning commonly given to terms or expressions in the branch of activity concerned, the nature and purpose of the contract, usages, and good faith and fair dealing. Good faith and fair dealing impose moral standards of conduct on parties, which may override the explicit terms of the contract if these do not satisfy requirements of decency, fairness, or reasonableness. Any attempt to provide a single, unifying definition of good faith and fair dealing becomes reductive. This is especially true for an instrument such as the CFR, as definitions of decency, fairness or reasonableness will be even more difficult to 36

Schwartz and Scott 2003.

94

Geerte Hesen & Robert Hardy

formulate in a multinational context. Parties’ conduct must thus be evaluated on a case by case basis by courts, which may result in judicial errors. Especially in the context of the CFR, judicial errors seem likely, since – assuming parties chose to have their contract governed by the CFR – national courts of, e. g., EU Member States would have to evaluate decency in, presumably, international contexts. In law and economics, a good faith and fair dealing obligation is considered an implied contractual term that prohibits opportunistic behavior. Opportunistic behavior occurs when a party behaves contrary to the other party’s understanding of the contract, but not necessarily contrary to the agreement’s explicit terms, leading to a transfer of wealth from the other party to the opportunistic party. We believe good faith and fair dealing should be a rule that parties are free to contract around. This is not explicitly mentioned in the CFR, and parties that would choose to apply the CFR to their contract, thus explicitly have to state this in their contract, which raises transaction costs. Critics of the doctrine of good faith and fair dealing in contract law have argued that the doctrine creates too much uncertainty. However, it is not certain that this argument holds for relational contracts in general or for international contracts in particular. In agreements that bind parties to repeat interactions over a long period of time, any uncertainty that results from the judicial construction of the good faith or fair dealing requirement may be overshadowed by the uncertain risks of entering into the relationship in the first place. In theory, parties can contract around any contingency, but their inability to predict the future places inherent limits on their ability to manage risks through contracting. The longer and more intense the relationship, the more difficult it is for parties to foresee potential problems. The chance that one party will act in bad faith raises the cost of entering into the contract in the first place, potentially rendering the cost of negotiations prohibitive. Therefore, the doctrine of good faith and fair dealing can act as insurance against these risks and reduce transaction costs. In relation to the other “relevant matters” of Article 8:102(1) of the CFR, we are of the opinion that courts, in principle, should use narrow evidentiary bases when interpreting agreements. Again, we assume that the CFR will primarily be used by firms in international transactions. On the other hand, courts should also comply with party requests to broaden the base that is applicable to them. This implication is at variance with Article 8:102(1). Article 8:104 of the CFR states that: “An interpretation which renders the terms of the contract lawful, or effective, is to be preferred to one which not.” Although we are of the opinion that this rule is generally useful, and corresponds with the preferences of most contracting parties, we do wish to state that the rules may carry with it the “risk” that parties will invest less resources in contract drafting. Finally, we believe Articles 8:104 (“Terms which have been individually negotiated take preference over those which have not”) and 8:107 (“Where a contract document is in two or more language versions none of which is stated to be authoritative, there is, in case of discrepancy between the versions, a preference for the interpretation according to the version in which the contract was originally drawn up”) contain rules that parties generally prefer. These rules are thus likely to lower transactions costs.

Contract Interpretation – Interpretive Criteria

95

Conclusion In this chapter, we have used an economic approach to analyse the rules on contract interpretation of the CFR. Our main findings are twofold. First, we do not think the subjective (or contextual) approach of 8:101(1) should have been chosen as the default. We assume that primarily firms in international transactions will choose the CFR to govern their contract. As explained, firms in our opinion typically prefer courts to adhere as closely as possible to the ordinary meanings of the words in the agreement. That is to say, firms typically prefer the textualist (or objective) approach. The textualist theory will not suit all contracting parties all of the time. Therefore, parties should have the possibility to contract around the textualist default. Second, we are critical of Article 8:102(1) of the CFR. In principle – assuming again that primarily firms in international transactions will choose the CFR to govern their contract –, courts should use narrow evidentiary bases when interpreting contract. However, courts should also comply with party requests to broaden the base that is applicable to them.

Judicial Control of Standard Terms and European Private Law Hans-Bernd Schäfer & Patrick C. Leyens

Abstract With the Draft Common Frame of Reference of European Private Law updated in 2009, the European Union is striving for a uniform legal framework for judicial control of standard terms. The draft offers a chance to point out perspectives for European private law from the view of law and economics without being burdened by the constraints of current national laws or European directives. These basic issue to be dealt with is the question for the regulatory purpose of judicial clause control, which is not to be found in superior market power on the part of the user of standard terms. Rather, the informational asymmetries which cannot or can only partially be overcome by market mechanisms, are the justification for judicial intervention in the private allocation of risks via standard terms. On this basis, perspectives for the future development of the European and national understanding of the law of standard terms will be presented: we will suggest a form of clause control limited to the positive contract value-information cost-relationship together with a control cap for business contracts, an equal standard for controlling consumer and business contracts, and, a differentiating approach towards preserving a clause by adjusting the excessive risk allocation to a permissible maximum (salvatory reduction).

Introduction The harmonisation of European private law is swiftly proceeding. In the beginning the harmonisation efforts were restricted to specialised areas. Directives regarding consumer protection soon followed and fundamentally influenced the national contract laws. The Directive on Unfair Terms in Consumer Contracts of 1993 (Unfair Terms Directive) requires the European Member States to adopt rules for the judicial control of standard terms.1 In Germany the Law on Standard Terms of 1976 was already in place and needed only marginal review.2 Its provisions were integrated into the second book of the German Civil Code (BGB) by the major Reform of the Law of Obligations in 2001.3 Recently, the European

1

2

3

Council Directive 93/13/EEC of 5 April 1993 on unfair terms in consumer contracts, O. J. L 95 of 21.04.1993, p. 29. Gesetz zur Regelung des Rechts der Allgemeinen Geschäftsbedingungen (AGB-Gesetz) vom 9.12.1976, BGBl. I p. 3317. Gesetz zur Modernisierung des Schuldrechts vom 26. November 2001, BGBl. I p. 3138.

98

Hans-Bernd Schäfer & Patrick C. Leyens

Commission has presented a proposal for the consolidation of the directives on consumer protection including the Unfair Terms Directive.4 The Draft Common Frame of Reference for a European Private Law (DCFR) of 2008 and updated in 2009 reaches beyond the limited area of consumer protection.5 It stands for a consolidated proposal to codify central tenets of European private law, including the rules on the judicial control of standard terms.6 The DCFR is conceived as optional, but it nevertheless strongly encourages discussion of the risks and opportunities of a general European civil law codification.7 This applies not only to the coherent treatment of normative questions. It also includes solutions to codification problems.8 The impetus of the DCFR is not limited to codifying currently applicable European law; it also considers the future development of private law in Europe. The DCFR thus offers an historic opportunity to strengthen the dialogue between law and economics. On the one hand, law and economics considerations can solidify the foundation for rule-making with respect to the further development of the DCFR. On the other hand, implementation by the national legislator is concerned. This article deals with basic problems and perspectives of the judicial control of standard terms and links them to concrete drafting recommendations. It explores the core questions of the judicial control of standard terms (cf. 1.): this includes the search for legal and political variables and the accurate investigation of the legal reasons for judicial clause control, which is not the market power of the user of standard terms, but the information asymmetry between the parties. On this foundation, concrete perspectives for the development of the European and national understanding of the judicial control of standard terms will be presented (cf. 2.): a clause control will be suggested that is restricted to a positive correlation of the contract value and the information costs. This implies a control cap for business contracts but, aside from that a (largely) equal treatment of consumer and business contracts. Finally, it will be suggested to prohibit the judicial adjustment of an excessive risk allocation to the permissible maximum (salvatory reduction) in areas of legal certainty but to allow it for cases of legal uncertainty.

1. Economic Foundations of Judicial Clause Control The necessity of legally regulating standard terms can hardly be questioned today. Nevertheless, it is clear that judicial control of such terms represents an intervention in private contractual autonomy. Within the legal and economic system of the European Union, which is based on private autonomy, such interventions must be justified and precisely delineated. To begin 4

5

6 7

8

Commission of the European Communities, Proposal for a Directive of the European Parliament and of the Council on Consumer Rights, Brussels, 8.10.2008, COM(2008) 614 final, 2008/0196 (COD). Study Group on a European Civil Code, Principles, definitions and model rules of European private law (Draft Common Frame of Reference), Interim Outline Edition, 2009, www.law-net.eu. DCFR Art. II.–9:401 to II.–9:410. See Ott / Schäfer, in: Vereinheitlichung des europäischen Vertragsrechts, in: Ott / Schäfer, eds., Vereinheitlichung und Diversität des europäischen Zivilrechts in transnationalen Wirtschaftsräumen, 2002, p. 203. Eidenmüller / Faust / Grigoleit / Jansen / Wagner / Zimmermann, 28 Oxford J. L. Stud. 659 (2008); German version in JZ 2008, 529. See also Beale, in: Schulze, ed., New Features in Contract Law, 2007, p. 343.

Judicial Control of Standard Terms and European Private Law

99

with, it will be shown that the legal-political starting point ought to be the equilibrium price, which reflects a distribution of contractual risks between the producer and the customer that minimizes the sum total of all costs of risk. This implies that the equilibrium price should include all contractual risks for which the producer is the cheapest cost avoider or cheapest risk taker, whereas all other risks should be borne by the costumer and consequently not be part of the equilibrium price. It will be shown that without judical control, the probability of reaching this equilibrium price is low. In specifying the regulatory reason for judicial clause control, we shall see that it is not superior market power, but rather entrenched information asymmetry which leads to detrimental deviations from the equilibrium price.

1.1. Standard Terms The search for the correct starting point for judicial clause control must deal with the ends to which standard terms are used. Standard terms serve the purpose of accelerating legal transactions. They allow the parties to focus on individual descriptions of performance and payment, i. e. on the essentialia negotii, which are the decisive factors in concluding any agreement.9 This acceleration of legal transactions, economically speaking, results in the reduction of transaction costs.10 This statement is founded on the realistic assumption that concluding a contract creates own costs. These negotiation costs have the effect, according to groundbreaking research by Coase, of negatively impacting the general economic welfare and reducing the commercially achievable surplus revenue of both contract parties.11 The legislator has set forth a minimum standard of fairness to be ensured by judicially controlling standard terms. According to the Coase-Theorem, contracts occurring under market conditions without transaction costs have the capability to optimise the allocation of scarce resources, irrespective of the content of contractual default rules. Parties can always find contractual terms which maximise their joint surplus. In this sense, the regulation of standard terms hinders judicial control of individually agreed-upon performance as well as the adequacy of the price. It places solely one particular form of concluding a contract under judicial review, namely the one-sided application of ancillary terms. The question arises, why standard terms imposed by one party will not lead to a contractual allocation of risk to the cheapest cost avoider. The argument of one-sidedness is certainly not a sufficient criterion for public intervention. Otherwise the state would have to regulate the quality of almost every product from bread to shoes to cars, which is usually fixed by the producer. Normally, market competition wipes out goods not meeting consumer preferences. The crucial question is: why does competition fail to wipe out one-sided standard

9 10

11

A. Fuchs, in: Ulmer / Brandner / Hensen, AGB-Recht, 10th ed., 2006, Vorb. § 307 Rn. 39. Kahan / Klausner, 83 Va. L. Rev. 713, 718 (1997). On more specific applications to agency costs Rakoff, 96 Harv. L. Rev. 1173, 1222 (1983). For an overview see Gillette, Standard Form Contracts, NYU Law & Economics Research Paper Series, Working Paper No. 09-18, April 2009 SSRN: http:// ssrn.com/abstract=1374990, p. 2, and Katz, in: The New Palgrave Dictionary of Economics and the Law, 1998, p. 502. Coase, The Problem of Social Cost, 3 J. L. & Econ. 1 (1960). On the Coase-Theorem Medema / Zerbe, in: Bouckaert/De Geest, eds., Encyclopedia of Law and Economics, vol. 1, 2000, p. 836. See also the “Hommage to Ronald Coase“ in Ménard, ed., Institutions, Contracts and Organizations, 2003, S. 37 ff., with contributions by North, Werin, Williamson and Ménard.

100

Hans-Bernd Schäfer & Patrick C. Leyens

form contracts, which shift most or all risks to the costumer, blow up the costs of risk reduction or insurance and reduce the joint surplus from the contract?

1.2. Market Power Judicial control is, as the prevailing view holds, orientated toward preventing one-sided risk allocations between parties. Intuitively, we may conclude that judicial control of one-sidedly applied standard terms is justified by the superior market power of one of the two contract parties.12 Typically, the producer or, in the language of the Unfair Terms Directive and §§ 305 et seq. of the German Civil Code (BGB), the user of the standard terms is in the stronger position and can dictate the conditions of the contract. Market power implies monopoly power in its most extreme form. Counter-intuitively, market and monopoly power do not offer a basis for judicial intervention in drafting standard terms. This is widely accepted in law and economics research.13 It is suggested here to compare the possible deviations from the equilibrium price in the context of the deployment of standard terms in both a monopoly market and a competitive market. It will be shown that there is no difference in the potential for the user of pre-formulated terms to shift risks onto the other contract party in either market situation, a monopoly market or a competitive market. This means that the reasoning underlying judicial control of pre-formulated terms does not depend on market power. In order to clarify this, we shall first focus on the two market situations of competition and monopoly. Preliminarily we exclude from our study additional factors which influence the determination of the equilibrium price, especially information asymmetries.

1.2.1. Competitive Market and Standard Terms

In a competitive market with perfect information, one-sided standard terms must reflect the cheapest cost avoider principle and lead to an efficient allocation of risk for both parties, producers and costumers. The parties know, while concluding their contract, that if certain circumstances occur, it could lead to an alteration in the course of the transaction or to the failure of negotiations, e. g. in the event of a defect of the sold product or a strike of workers. The realization of such transactional risks can impact the transaction goal or, if there is a fixed deadline, totally obstruct it. If both parties are equally able to recognise the possibility of such or other risks, they will distribute the respective consequences amongst each other pertaining to the transactional goal. In the spirit of accelerating contractual negotiations and thus reducing transaction costs, the parties will normally use standard terms. They will allocate risks to the party which is best able to avoid the costs involved (cheapest-cost-avoider principle). Unavoidable risks remain. In general, however, can be insured 12

13

This was the view of Kessler, 43 Col. L. Rev. 629 (1943). See also Rakoff, 96 Harv. L. Rev. 1173 (1983), and Slawson, 84 Harv. L. Rev. 529 (1971). For an overview Gillette, supra n. 10, p. 5, and Katz, supra n. 10, p. 502. See Adams, in: Neumann, ed., Ansprüche, Eigentums- und Verfügungsrechte, Arbeitstagung des Vereins für Socialpolitik, 1984; Bebchuk/Posner, 104 Mich. L. Rev. 827, 828 (2006); Behrens, Die ökonomischen Grundlagen des Rechts, 1986, S. 155 ff.; Kötz, JuS 2003, 209, 210; Trebilcock-Dewees, in: Burrows-Veljanovski, eds., The Economic Approach to Law, 1981, p. 93 et seq., 99 et seq. For a mathematical approach Schäfer, in: FS Ott, 2002, p. 279, 290.

Judicial Control of Standard Terms and European Private Law

101

against by one party at a lower cost compared to the other. Hence, the producer will be willing to bear the insurance costs if he is able to get more favourable rates. Accordingly, the producer will bear the insurance costs if he concludes a great number of similar contracts or if he is able to describe the product risks with greater certainty than the customer. If, for example, the producer can conclude an insurance policy for € 2 for a product valued at € 20 which the consumer is only able to insure at a cost of € 4 then it is in both parties’ interest if the producer is the one to conclude the insurance policy. The reverse result is also conceivable, especially if the risks have less to do with the product itself and more to do with the manner in which the purchaser intends uses it. In that case the customer will bear the insurance costs. Thus, we can conclude that the risk avoidance costs are included equally in the producer’s price and in the costumer’s willingness to pay. The distribution of risk is orientated toward the individual who is the cheapest cost avoider. If this is the producer, then he will carry the risks and they will become part of his marginal costs. In turn, they will become a part of the supply price. If, on the other hand the costumer is the cheapest cost avoider, he will accept the risk. The market price of the performance remains the same. However, the costs resulting from risk avoidance and possible insurance costs lead to a reduced willingness to pay and a reduction in demand. As a result, the maximum mutual (synallagmatic) yield is only achievable by contractual arrangements leading to an allocation of risks to the party that is in the best position to reduce the costs of the respective risks. Insofar as risk allocation is concerned, if both parties have complete information, there cannot evolve a difference between individually negotiated and pre-formulated contracts. In a competitive market, customers will always contract with producers with whom they can realise – either via an individually negotiated contract or via pre-formulated terms – an efficient risk allocation. Of course the equilibrium price of a good in a competitive market is lower, if producers shift all risks to costumers. But a low price cannot attract buyers who know that they are burdened with risks whose avoidance or insurance costs are higher for them than in an alternative risk allocation with a higher product price. In other words, contract drafts which do not allocate risks to the cheapest cost avoider will not be able to gain traction in a competitive market.14 This leads to a policy conclusion: The control of pre-formulated terms by courts can be counter-productive as long as the parties have (nearly) complete information regarding the risk situation. The parties’ specialised knowledge and learning effects will lead them to achieve the best possible risk allocation in their own reciprocal interest. A court, on the other hand, has the role of an external contract governance mechanism. In the best case, a court-guided risk allocation will refer to the information level of the respective parties. Here, the court will select the risk allocation which ought to have actually been adopted by the parties through their standard terms. However, the judicial decision will frequently deviate due to informational deficiencies of the court. This makes room for opportunistic lawsuits aimed at exploiting the courts’ informational deficits to the advantage of one party. Thus judicial control which implements an efficient risk allocation in the view of both parties would not reflect the hypothetical will of the parties, given that the costs involved in this process would reduce the total revenue surplus.

14

Katz, supra n. 10, p. 502.

102

Hans-Bernd Schäfer & Patrick C. Leyens

1.2.2. Monopoly Power and Standard Terms

The monopoly price argument has for a long time been firmly rooted in the debate surrounding the justification of judicial control of standard terms.15 In comparing the key factors of an economic surplus in a competitive market with those of a monopoly market, it will be shown why this view, today, must be seen as outdated. The monopolist is able to arbitrarily decide the cost of his services. He can demand a price that is higher than the competitive price and raise his producer profit at the costumers’ expense. In a monopoly market, no competition takes place insofar as contract terms are concerned since costumers are not able to migrate to other producers. From this the conclusion could be drawn that the monopolist will be able to impose disadvantageous terms upon his costumers without any repercussions. This conclusion, however, would simply be wrong, as we will see in light of the following example: let us assume that the monopolist sells his goods at the monopoly price of € 100. Under monopoly pricing, newly recognised risks can be dumped onto costumers via standard terms. Let us assume that the monopolist could fully eliminate a risk by spending € 2, while the costumer must spend € 20 to do the same. If the producer burdens the customer with this risk, the market price remains unaltered, at least superficially. The risk allocation, however, leads from the view of the customer to an indirect increase of the product’s price similar to the higher risk or insurance costs, i. e. equal to € 20. The above considerations vis-à-vis the equilibrium price, then, push the demand curve down by € 20. If the monopolist accepts the risk himself, then his production costs rise by € 2. The monopolist will include his production costs in the market price. In other words, the product will in the future be available for € 102. The demand curve remains grossly unchanged, since the costumer is protected against a newly recognised risk. By comparing the two possible modes of conduct, it becomes clear that the monopolist prevents himself from realizing profits which he would be able to realise if he was to accept the risk.16 Concretely, the monopolist suffers losses equivalent to the extent of the reduction in demand, which reflects the unnecessary risk costs of € 18 (20-2=18). By comparison with the equilibrium price, the use of pre-formulated terms which are, as the above example shows, unfair, leads to the loss of a synallagmatic profit which damages the profit in comparison with the equilibrium price monopolist. The monopolist gains less in terms of producer profit than the costumers lose as long as costumers are fully informed. Thus, we see that the monopolist will not, vis-à-vis the drafting of his standard terms, conduct himself differently than the producer who inhabits a competitive market. Even the monopolist will accept those risks which he can carry with lower cost. Conversely, he will pass on to the costumer the risks the latter is able to carry at a lower cost. This is the only behaviour in line with profit maximization. This result should not distract from the fact that the monopolist will exploit the customer by offering his products at a price higher than would be possible in a competitive market. However, he will nevertheless avoid risk allocations which contradict the cheapest cost avoider principle; indeed, unnecessary price increases will have a negative effect on demand and therefore also on his revenue potential. In the above example, he could carry the risk at € 2 while raising his sales price at the same time. 15

16

Kessler, 43 Col. L. Rev. 629 (1943), and the ‘reconstruction’ essay by Rakoff, 96 Harv. L. Rev. 1173 (1983). See also Slawson, 84 Harv. L. Rev. 529 (1971). Marotta-Wurgler, NYU Law & Economics Research Paper Series, Working Paper No. 05-11, August 2005, http://ssrn.com/abstract=799274, p. 2, with further references.

Judicial Control of Standard Terms and European Private Law

103

1.2.3. Competition Law, Antitrust Law and Clause Control

From the previous considerations we conclude that there is a central functional difference between the judicial control of standard terms and competition law as well as antitrust law.17 Competition law and antitrust law serve the purpose of eliminating or reducing the market failure as a result of the power to set prices systematically above the competitive equilibrium price. Those branches of the law focus on the creation and the abuse of market power. If market power leads to an obvious distortion of the equality of the parties, the law should exempt the party which has been treated unfairly from the obligations under the contract.18 In contrast, judicial control of standard terms is justified on the basis of market failures which are not caused by market power.19 We will use these considerations further below to develop a differentiating approach towards the role of judicial control of standard terms vis-à-vis general fairness control using principles as good faith, fair dealing or bonas mores.20 In the following section, the function of clause control and its justification on the basis of information asymmetries will be discussed in greater detail.

1.3. Informational Insecurity and Asymmetry An economically complete justification for judicial control of standard terms can be found by considering the consequences of information asymmetries between the parties. Information asymmetries can exist with respect to the possible reasons for, likelihood of and the extent and avoidance costs of a particular damage, as well as to the optimal contractual risk allocation.

1.3.1. Informational Deficits

Contracts are concluded in spite of informational deficits regarding the content and meaning of pre-formulated terms as well as the relative costs of reducing or insuring a contractual risk. This can be the result from the inability of consumers to process the information contained in the contract terms. Behavioural economics research contends that consumers typically price only a limited number of contract attributes and that the majority focuses on the same attributes.21 This gives room for the user of standard terms to opportunistically introduce ancillary terms on less salient attributes. It is a fact – above and beyond consumer protection – that recognising and comparing pre-formulated terms of various producers does not take place due to the excessive transaction costs involved therein when compared to the

17 18 19

20 21

See also A. Fuchs, supra n. 9, Vorb. § 307, Rn. 76. See also Wackerbarth, AcP 200 (2000) 45, 68. Ibid., p. 73 and in detail on the weaknesses of legal intervention in party autonomy on the basis of imbalances in market power Zöllner, AcP 196 (1996) 1, 15. Infra p. 17. Korobkin, 70 Univ. Chicago L. Rev. 1203 (2003). For the state of the discussion see Gillette, supra n. 10, p. 9.

104

Hans-Bernd Schäfer & Patrick C. Leyens

risk. The costs of evaluating standard terms are generally higher than the expected utility.22 However, this does not only apply to consumer contracts. Insofar as low contract values are concerned it is a general rule that the contents of pre-formulated terms are not recognised or compared. The economic consequences resulting from this lack of knowledge of standard terms can be explained by comparing two possible constellations. In the first constellation, the customers know the risk but assume that the producer’s pre-formulated terms will be fair to them and that they will only be burdened with the risks for which they themselves are the cheapest cost avoiders. If, however, against the customer’s expectations all risks, even those which can be borne by the producer more easily, are passed on to the customer, then the market develops as if all risks had been shifted to the cheapest cost avoider. This means that the product supply will eventually arrive at an equilibrium quantity and equilibrium price, which does not take into consideration that all or most risks have been allocated to the purchaser, i. e. product supply and demand will be too high. Specifically, customers will have a willingness to pay a price which does not discount all their risks. They will, thus, suffer an unexpected reduction of their customer surplus. The producer’s surplus, on the other hand, remains unchanged. In total, there will be a reduction in the general economic surplus. In the second constellation we assume that the customers know the risks but expect that they will be treated unfairly. The idea of a fully naïve consumer is particularly obsolete in our modern information age. Practically, the knowledge of the one-sidedness of risk allocation will penetrate a market, e. g. via modern means of communication or the media and the help of information intermediaries. Hence, it seems realistic to assume that consumers mistrust standard terms. Excessive risk avoidance costs thus become unavoidable for the buyer. The result is a suboptimal supply, meaning that the economy offers too low of a return for both producers and costumers. If in this situation an individual producer offers contract terms according to the cheapest cost avoider principle and charging a higher price, consumers will not trust him. It is too costly for them to clarify whether their mistrust is justified. The two above-described constellations are to be understood as extremes. Reality ought to lie in between these scenarios, e. g. in the event that customers at least partly know the risks but either underestimate their possible losses or the likelihood that the risk will occur.

1.3.2. Market Failure and Informational Deficits

Informational deficits, such as the ones described, can under certain conditions be overcome by information disclosure measures by the producer (signalling) or by information acquisition measures (screening) on the part of the customer. With regard to standard terms we observe a failure of these measures to overcome information asymmetries. This failure will be substantiated here insofar as the two above-discussed constellations are concerned: in the first constellation, as the customer assumes that he is being treated fairly by the producer, a continued worsening of risk allocation is to be expected. This is because the costumers cannot, due to the information costs, gain knowledge of pre-formulated terms. Thus, they are unable to react to potentially more favourable pre-formulated contract terms on the part of individual producers. As a consequence, both signalling and screening fail. 22

Adams, BB 1989, 781, 784. Same, Ökonomische Analyse des Gesetzes zur Regelung des Rechts der allgemeinen Geschäftsbedingungen (AGBG), in: Neumann, ed., Ansprüche, Eigentums- und Verfügungsrechte, Arbeitstagung des Vereins für Socialpolitik, 1984.

Judicial Control of Standard Terms and European Private Law

105

In this szenario, given that producers are likely to maximise their profit, they will behave in such a manner that their standard terms are always more disadvantageous than those of their competitors. By doing this, they ensure a higher producer surplus because they sell at the market price, which reflects a better distribution of risks but save costs, until their competitors adjust. Akerlof has shown that under these conditions a race to the bottom results, by the end of which the pre-formulated terms will have one-sidedly passed on all or nearly all risks upon the uninformed side.23 It is important, though, that at the end of the adjustment process, if we assume a competitive market, none of the producers will be able to achieve any profits by worsening their standard terms. This is because they will be offering their products at their marginal cost and the equilibrium price conforms to their marginal cost. They do, however, pass on significant costs onto the uninformed side which would, in a fully functional market without information asymmetries, be assumed by them and then passed on in the form of a higher price. The possible end of the development can be described as a ‘flea market’ with low prices which offer the costumer only a minimum of rights vis-à-vis the producer. Even in the second constellation, in which the costumer anticipates being treated unfairly by the standard terms of the producer, similar results occur. The initial consideration in this case is that the customer cannot gain knowledge of pre-formulated terms due to the high cost of doing so. In other words, screening fails. Producer signalling will fail as well because costumers assume that the standard terms will be unfair. They adjust their willingness to pay accordingly and producers cannot signal that their terms follow the cheapest cost avoider principle. Again the result is a “flea market”.

1.3.3. Welfare Benefits via Control of Standard Terms

Judicial control of standard terms under the above-named conditions is both a means of avoiding economically disadvantageous results as well as correcting the negative consequences of information asymmetries in individual cases. The decisive reason for the loss of welfare in the absence of judicial control is the (albeit rational) lack of knowledge of risk allocation by way of standard terms. A complete market failure in terms of a theoretical model, i. e. a market which altogether collapses, would for this reason only be caused in the rarest of cases. Costumers will not refrain from concluding contracts. Even if we assume relatively poorly drafted contracts, they cannot avoid purchasing certain items, e. g. day-to-day consumer goods. The markets would not disappear, but they allocate risks to those with the highest costs of handling them. This blows up the overall costs and markets become smaller in volume as a result. This partial market failure is the decisive argument for the legal and political decision to subject standard terms to judicial control. The absence of judicial control of standard terms would lead to a reduced total surplus of both costumers and producers in the market. Thus, it should be affirmed that the legal and political goal of standard terms is not just the protection of the contract party that is not adequately informed. Indeed, it is to ensure and support the continued functionality of markets. Judicial control counteracts the danger that informational deficits will lead to a (partial) market failure. Simultaneously, it ensures that the transactional advantages accruing to the economic surplus as a result of using standard terms remain intact. 23

Akerlof, 84 Q. J. Econ. 488 (1970).

106

Hans-Bernd Schäfer & Patrick C. Leyens

We have seen that protecting weaker market participants against exploitation by strong market participants does not explain the need for judicial control of standard terms.24 Unfair standard terms are used when terms are not controlled because, in competing on price, there is no other way to react except by reducing the costs of one’s own product. This market pressure impacts all producers with the same degree of intensity, e. g. small and medium-sized businesses, irrespective of whether they have any particular market power.

1.4. Failure of Market Control According to regulatory theory, interventions in the market are only to be advocated if it can be shown that alternative mechanisms of problem solving have failed. The previous sections have shown that producer incentives which arise out of incomplete information on the part of costumers lead to a race to the bottom concerning standard terms and that, for this reason, judicial control of pre-formulated terms is desirable. This assumption will be supported in the following sections with theoretical considerations as well as practical examples regarding the weakness of the existing market mechanisms.

1.4.1. Repeated Purchases and Learning Effects

By legally mandating minimum standards for standard terms, unproductive search costs on the customer side become superfluous. At the same time, learning effects are reduced on both sides of the contract. The producer’s search effort is aimed toward avoiding existing regulation via alternative means of risk allocation. This leads to distortions in the learning and searching processes which should be included in the political decision in favour of judicial control of standard terms.25 Information economics has reached the conclusion that information deficiencies dissipate over time and that an information-efficient market can come into existence even in the absence of regulation.26 This conclusion also applies to standard terms. Nevertheless, there are important differences which hinder the creation of an information-efficient market. Learning effects require that the relevant circumstances can be observed and evaluated. The consequences of risk allocation by standard terms are, however, only observable by the relative small number of individual costumers to whose detriment the risk has occurred. This impedes a rapid dissemination of information. Furthermore, the risk will in many cases only occur with a time delay. Especially in markets in which repeated purchases take place only rarely or after a long period of time, e. g. in the furniture of automobile markets, specific learning effects on the part of costumers will hardly be possible.27 This applies, in any case, to risks with a small probability of occurrence.

24 25

26 27

Explicitly on this point, see Adams, BB 1989, 781, 783. Kirchner, Kommentar zu Georg von Wangeheim und Sylvia Rückebeil, in: Eger / Schäfer, eds., Ökonomische Analyse der europäischen Zivilrechtsentwicklung, 2007. Schwartz / Wilde, 127 U. Pa. L. Rev. 630 (1979); same, 52 Rev. Econ. Stud. 251 (1985). For further examples from the tourism sector and construction businesses, see Adams, BB 1989, 781, 785.

Judicial Control of Standard Terms and European Private Law

107

1.4.2. Reputation of the Producers

In a competitive market in which risks and their allocation remain intransparent, producers will use all possibilities to reduce their costs by passing on risks to customers. The goal of this is to compete on price. There is an exception to this rule, however, namely when the producer is able to signal to the customer at the time of contract conclusion that the risk allocation is fair. Such a signal is, in accordance with the explanations above, normally not possible within the context of a single contract, but it is possible via building up reputation over time. Theoretically, it can be assumed that producers will on their own initiative attempt to correct the failure of the market by offering a risk allocation which is not inferior, but superior to that of their competitors. In practice, though, there are numerous examples that the reputation mechanism does not have (sufficient) traction in regard to the information problem connected to standard terms. The financial sector is, due to its high product and transactional complexity, especially suitable to illustrate the reach and limits of reputation as a guiding mechanism.28 Banks are, due to the characteristics of the products they offer, dependent to a high degree on a reputation of fair-dealing with customers. Financial products have strong credence characteristics which the customer is unable to judge.29 Accordingly, contracts for financial products only occur when there is a sufficient degree of trust in the drafting of the contractual offer and the accompanying pre-formulated terms. Likewise, numerous court judgements show that banks and insurance companies allocate or have allocated risks to their customers via standard terms which they could have been able to bear at a far smaller cost. Lastly, in the 1980s, a wave of lawsuits was brought before German courts against unfair practices toward banking customers. The Bundesverband deutscher Banken (Federal Association of German Banks) reacted in 1993 with a completely new version of the standard terms applicable to German private banking (AGB-Banken).30 This example shows that the reputation mechanism does indeed guide behaviour but that it by itself is not sufficient to deal with the dangers arising from unfairly formulated terms.

1.4.3. Organisations and Self-Regulation

The reason for the (partial) market failure is the above-noted impossibility of signalling the quality of standard terms, especially in markets which have a high intensity of competition. The producer side can react to this by coordinating its course of action, i. e. by imposing mutual obligations to maintain minimum standards. Through such a form of self-regulation due to the superior specialised knowledge of the actors, the results will often be superior to those which could have been achieved by legislation or judicial control.

28

29

30

For example, see the studies on credit card holders by Bar-Gill, 98 Nw. Univ. L. Rev. 1373 (2004) and Mann, 104 Mich. L. Rev. 899 (2006). On the various goods categories see Posner, Economic Analysis of Law, 6th ed., 2003, S. 113. See also Tietzel, in: Schäfer / Ott, Allokationseffizienz in der Rechtsordnung, 1989, S. 52, 53. Also instructive: Weise / Brandes / Eger / Kraft, Neue Mikroökonomie, 5th ed., 2004, S. 219. The history of the AGB-Banken goes back to the 1930ties. For more, Casper, in: Derleder / Knops / Bamberger, eds., Handbuch zum deutschen und europäischen Bankrecht, 2nd ed., 2004, § 3 para. 1.

108

Hans-Bernd Schäfer & Patrick C. Leyens

The precondition for functioning self-regulation is, however, that the participants have an appropriately high level of incentives to use their superior specialised knowledge. Additionally, they would need to be ready to accept private sector obligations. Examples of successful self-regulative efforts abound which have led to market-increasing revenue structures. To name some examples from Germany: In the 19th century, private organisations played a decisive role in preventing the breakdown in the market for steam engines, which at that time were extremely dangerous. Still today, the security of motor vehicles is ensured by an organization named the Technische Überwachungsverein (TÜV). In the financial sector, private efforts on the part of the banks have prevented loss of reputation. Long before it became mandated by European directives they established a deposit insurance system which is unique within the EU.31 Positive examples such as the previous one contrast with the time it took for the new version of the AGB-Banken, discussed above, to be in effect. The reasons for private sector regulatory efforts are multifaceted. The much-cited example of English takeover law shows that the concern about deep interventions by way of legislation is one important reason.32 Similarly, the readiness of the German private banks to establish their private deposit insurance fund can be seen at least in part as a reaction to the collapse of the Herstatt Bank in 1974 and the fear of legislatory intervention.33 It should be noted that unfair risk allocations by standard terms can have completely different effects depending on industry sectors. If only a few costumers are impacted by the consequences of the risk allocation, and therefore there is no danger of a significant decline in the number of contracts concluded or price discounts, then there are only minimal incentives for the private sector to become active. Thus, we can maintain that alternative guidance mechanisms will, in general, not be adequate to the task of replacing judicial control of standard terms.

2. Perspectives for European Private Law The judicial control of standard terms is, since the Unfair Terms Directive, an integral component of European private law. The DCFR for European Private Law, presented in 2008, supports this view.34 On the basis of the above considerations, the following sections will develop perspectives in regard to the extent and limits of clause control, the setting of the control standard regarding business to consumer and business to business contracts and the judicial adjustment of unfair terms. All these topics relate to both European and national law. Accordingly, we will address possible changes or amendments to the DCFR and also suggestions for its implementation in national laws.

31 32

33

34

For more, see Bigus/Leyens, Einlagensicherung und Anlegerentschädigung, 2008, p. 12. For more, see Davies, ZGR 2001, 268, 272. A comparative account is provided by Baum, RIW 2003, 421, 424. For an extensive overview, see Hoeren, Selbstregulierung im Banken und Versicherungsrecht, 1995, p. 199. For a brief retrospective of the Herstatt Bank case, Kaserer, VSWG 2000, 166. For more comprehensive information see Hoeren, supra n. 31, S. 86. For a detailed account see Pfeiffer, in: Schulze, ed., Common Frame of Reference and Existing EC Contract Law, 2008, p. 177.

Judicial Control of Standard Terms and European Private Law

109

2.1. Control of Standard Terms Judicial control of the clause contents is in the centre of the regulation of standard terms. The DCFR, similar to the Unfair Terms Directive, foresees that standard terms which contain a significant disadvantage are null and void.35

2.1.1. Scope of Judicial Control

Standard terms allow the parties to focus their negotiations on the essential terms of a deal (essentialia negotii).36 As a result, if the drafted clauses are fair, both parties can save on transaction costs. For an appropriate determination of the extent and limits of the competence of judges to engage in control of such clauses, we must investigate the regulatory reason for controlling clause contents. Legal scholars argue that the control of clause contents defends against dangers which result from the one-sided use of contractual freedom by the user of standard terms.37 The absence of negotiations is connected to a certain loss of accuracy, which in turn is reflective of a functional inequality of the parties.38 From an economic perspective, we can assert the following: the possibility of opportunistic market behaviour and one-sided imposition of disadvantages does not result from inequality or market power, but from information asymmetries.39 Economic gains resulting from the accelerative effect of standard terms are distorted by a race to the bottom which results from a competition for the most disadvantageous clauses. This reduces the joint surplus of the parties. The judicial control of standard terms is, therefore, necessary to protect the market against this kind of damnification by information asymmetries.

2.1.2. Drafting Recommendation

The current version of the DCFR has not yet been finalised as to the scope of the judicial control of standard terms. A parenthesis included in DCFR Art. II.– 9:403 DCFR suggests that the drafters are still deliberating whether judicial control should be expanded to indi-

35

36

37

38

39

Compare DCFR Art. II.– 9:403: “… a term (…) is unfair (…) if it significantly disadvantages the consumer, contrary to good faith and fair dealing.” (emphasis added) and Unfair Terms Directive, Art. 3 para. 1: “… unfair if, contrary to the requirement of good faith, it causes a significant imbalance in the parties’ rights and obligations, to the detriment of the consumer”. A. Fuchs, supra n. 9, Vorb. § 307 Rn. 39. Exceptions apply insofar as price determination is intransparent. BGH NJW 1976, 2345, 2346. For commentaries on the judicature see A. Fuchs, supra n. 9, Vorb. v. § 307 Rn. 26, Grüneberg, in: Palandt, BGB, 68th ed., 2009, Überbl. v. § 305 Rn. 3, 8. See Bork, Allgemeiner Teil des BGB, 2nd ed., 2006, Rn. 1744 (p. 639); Köhler, BGB Allgemeiner Teil, 31st ed., 2007, § 16 Rn. 1 (p. 259); Larenz/Wolf, Allgemeiner Teil des Bürgerlichen Rechts, 9th ed., 2004, § 42 Rn. 8, 29 (p. 756, 762), § 43 Rn. 1 (p. 771). Wackerbarth, AcP 200 (2000) 45, 64, 67, asserts that the inequality argument does not hold for cases of sufficiently transparent standard terms. A detailed account is provided by Fastrich, Richterliche Inhaltskontrolle im Privatrecht, 1992, p. 85. For details see supra p. 104.

110

Hans-Bernd Schäfer & Patrick C. Leyens

vidually negotiated terms of the contract.40 German law traditionally abstains from such far reaching competence to control beyond the scope of general principles as good faith or bonas mores.41 In other countries like France, for example, judicial competences are framed differently and the judicial competence to control contract terms expands to a general control of contractual fairness.42 It has been argued above that the judicial control of standard terms should be limited to cases of information asymmetries that can cause a market failure due to a rational ignorance with regard to the contents of standard terms. Conversely, the scope of judicial control should exclude cases where the contract contents are ignored by the other party although it would have been rational to acknowledge them. This leads us to a more differentiating understanding of the role of judicial clause control vis-à-vis judicial fairness control under general principles of private law as good faith, fair dealing or bonas mores. Clause control is thought to solve a specific problem of systemic market failure whilst general principles have a wider scope and serve to overcome unfair risk allocations in individual cases. The approach taken in this article seeks to separate cases of a systematic market failure caused by information asymmetries from singular failures of the parties or adhesive contracts. The unfairness test for standard terms follows a schematic pattern: Under DCFR Art. II.– 9:403 et seq. a term is unfair if it significantly disadvantages a consumer contrary to good faith or fair dealing. DCFR Art. II.–9:406 (I) clarifies that terms which are based on the provisions of applicable law are not subject to the unfairness test. The unfairness test thus measures the degree to which a standard term deviates from applicable law. A strong disadvantageous deviation renders the term void. A key for the understanding of the characteristics of the law of standard terms is DCFR Art. II.–9:406 (II) which explains that the unfairness test does not extend to the adequacy of the price. Accordingly, a term is void due to a disadvantageous deviation from the applicable law even if the price reflects the disadvantage. In other words, a disadvantage cannot be compensated by a price reduction. This is fully in line with the economic view taken in this article. Including the adequacy of the price to the unfairness test would aggravate the information problem instead of solving it. It is the very nature of the information problem that users of standard terms will reduce their costs by passing on risks to their customers with a view to offer a nominally lower price than their competitors. The consequence – as described above – is a race to the bottom. Falling prices are part of that development. It would therefore be inappropriate to extend the unfairness test to the adequacy of the price. Opposed to that, the fairness test under general principles like good faith, fair dealing or bonas mores does extend to the adequacy of the price. The fairness test under these principles does not follow a schematic pattern but includes all factors relevant for safeguarding a minimum level of contractual fairness. Accordingly, even a strong deviation from the applicable law can be upheld provided the minimum standard of contractual fairness is preserved, for example, by an adequate reduction of the price. Due to the inclusion of the adequacy of the 40

41 42

See DCFR Art. II.–9:403: Meaning of “unfair” in contracts between a business and a consumer. “In a contract between a business and a consumer, a term [which has not been individually negotiated] is unfair for the purposes of this Section if it is supplied by the business and if it significantly disadvantages the consumer, contrary to good faith and fair dealing.” See §§ 138, 242 German Civil Code (BGB). For a comprehensive overview Ranieri, Europäisches Obligationenrecht, 3rd ed., 2009, p. 371 et seq., 377.

Judicial Control of Standard Terms and European Private Law

111

price general principles are suitable for individual cases of superior market power of one party as well as adhesive contracts. For those cases good faith, fair dealing and bonas mores foresee a judicial competency to exempt the weaker party from the obligations under the contract instead of merely declaring void an ancillary term of the contract. The law on standard terms could only incompletely address the issues of those cases. To conclude, it is suggested here to delete the parenthesis of Art. II.– 9:403 in order to clarify that the specific rules for the judicial control of standard terms should be limited to terms which have not been individually negotiated.

2.2. Control Cap for Business Contracts Judicial control of standard terms is necessary to varying degrees. The decisive factor is to which degree of likelihood the standard terms will be acknowledged and to which degree that acknowledgement will influence the conduct of customers.

2.2.1. Contract Value and Information Costs

This offers some room for distinctions: insofar as everyday transactions are concerned, e. g. purchasing groceries or generally acquiring less expensive items, the information costs of acknowledging standard terms are higher than the expected gains. Therefore, it is rational not to acknowledge the standard terms. Conversely, when there is a very high transaction value, the importance of the pre-formulated risk allocation is obvious. Therefore, there is a comparably higher likelihood that the pre-formulated term will be acknowledged. Under these conditions, the danger of a race to the bottom, as far as the producers are concerned, and thus of a market failure, is virtually nonexistent. If a high contract value is at stake, there will be a high degree of alertness and willingness to acquire information, even at considerable cost. In such cases the regulatory reason for judicial control of standard terms cannot be rational ignorance of the pre-formulated terms. Even the much cited problem of the absence of ability to process information can be coped with by expending appropriate information costs including costs for the help of professional advisers. Professional advice can be necessary to understand the specialised language of a real estate contract or of financial products. It has been suggested by legal scholarship to juxtapose transaction costs and contract value pinpointing limiting judicial control to cases of information asymmetries.43 If there is a positive relationship, i. e. if the costs of gaining knowledge of the pre-formulated terms are, in light of the contract value, reasonable, then the clauses ought not to be controlled (and vice-versa). This proposal represents the very core of the above economic considerations as it tries to improve the degree of precision of judicial control of standard terms and its ability to address the market failure problem.

43

Leuschner, AcP (107 (2007) 491, 524.

112

Hans-Bernd Schäfer & Patrick C. Leyens

2.2.2. Drafting Recommendation

The purest form of implementation of this proposal would be a general clause, as foreseen in the DCFR44 as well as in German law45, together with a judicial assessment of the transaction value vis-à-vis information costs in each single case. An opposed model would be a fixed threshold, i. e. a defined range of judicial standard terms control. It is clear that a fixed threshold will not provide a continuous adaptation to market changes.46 It is also clear that the legislator in setting a fixed threshold refers to knowledge which as a rule is inferior to that of the involved private parties.47 Especially with a view to European harmonisation but also regarding national laws on judicial control of standard terms there are, however, strong arguments in favour of a fixed threshold. A fixed threshold provides legal certainty. A rather pragmatic but the perhaps most important argument is that European Court of Justice would hardly be prepared to handle a flood of requests for preliminary rulings by national courts. In their collaboration the English and the Scottish Law Commissions propose to limit judicial control of standard terms to contract values below a threshold of GBP 500.00048 (€ 559.440).49 There might be cases where the contract values vary but information costs remain unchanged. For example, a car rental that purchases cars for its fleet one after the other will have similar information costs as another car rental that purchases the whole fleet at once. It seems, however, that in the vast majority of cases the correlation between contract value and information costs is such that ignoring standard terms will be irrational above a certain threshold. Fixing a suitable threshold is a precarious task and should build upon more detailed information than currently available. Empirical research would be necessary and might well lead to the conclusion that a considerably higher threshold should be envisaged than the one proposed by the English and Scottish Law Commissions. A threshold fixed at a reasonably high contract value would alleviate the concerns, for example, against German law where it has been harshly criticised by politicians50 and by legal practice51 that the judicial control power expands to business contracts such as major M&A transactions.52 It is a separate question whether the cap should apply also to consumer 44 45 46

47

48

49 50

51

52

See DCFR Art. II.–9:403 et seq. See § 307 German Civil Code (BGB). On the weaknesses of regulatory control of standard terms Kötz, Gutachten A zum 50. DJT, 1974, p. A94. Conversely Leff, Am. Univ. L. Rev. 131 (1970), who argues that users of standard terms should have the option to submit potentially controversial terms to a bureaucratic agency for evaluation. The weaknesses of regulatory thresholds were discussed intensely before the enactment of the German Statute on Standard Terms of 1976. See Kötz, Gutachten A zum 50. DJT, 1974. Law Commission/Scottish Law Commission, Unfair Terms in Contracts, Report No. 292/Report No. 199, 2005, § 5.55. Exchange rate as of 30th October 2009. Within the deliberations on the reform of the law of obligations the Bundesrat (Federal Council of Germany) had warned that there is an increasing pressure for businesses to evade German law and to choose a foreign law that abstains from the judicial control of standard terms in business to business contracts; BT-Drucks. 14/6857, p. 17. For a recent exchange of views see the pleas against judicial control of standard terms in business contracts Berger, ZIP 2006, 2149, 2156, the counterplea by v. Westphalen, ZIP 2007, 149, 157, and the rejoinder by Lischek/Mahnken, ZIP 2007, 158, 164. Specifically on M&A transactions Leuschner, AcP 207 (2007) 491.

Judicial Control of Standard Terms and European Private Law

113

contracts. Caps are not unknown to consumer regulation in Europe. For example, the scope of the directive on credit agreements of consumers is limited to credit values that do not exceed € 75.000.53 However, with a view to the judicial control of standard terms in consumer contracts a cap seems inadequate. Consumers will seldom conclude contracts with high values. Ex ante there is a danger that consumers will treat high value transactions identical to small value transactions. Ex post there is a low probability of learning effects.54

2.3. Control Standard for Consumer and Business Contracts The German legislator has included business contracts to judicial control but until now the European Unfair Terms Directive has only made consumer contracts accessible for judicial review.55 The DCFR follows an intermediary course by addressing both types. However, it subjects each of them, respectively, to a different control standard. In dealing with consumer contracts, a “substantial disadvantage” is sufficient whilst insofar as business contracts are concerned, a “gross deviation from expected business practices” must be given.56 Presumably this differentiation is a compromise that seeks to accord to the different positions of those countries that abstain from controlling business contracts and those that include them in the judicial control competency.57 In the following, we will explain why this mandatory modification of the control standard ought to be abolished.

2.3.1. Differentiating Control Standards

In Germany, the origins of the differentiations between consumer and business contracts can be traced back to the jurisprudence of the Reichsgericht.58 The Court assumed that the exploitation of market monopolies (§ 138 BGB) ought to invalidate standard terms.59 The jurisprudence of the Reichsgericht is outdated since the Bundesgerichtshof 60 expanded control of clauses to violations of good faith (§ 242 BGB) and emphasised the weak legitimacy of 53

54

55 56

57 58

59

60

Directive 2008/48/EC of the European Parliament and of the Council of 23 April 2008 on credit agreements for consumers and repealing Council Directive 87/102/EEC, O. J. L 133/66 of 22.5.2008, Art. 2 para. 2 lit. c) On learning effects also supra p. 13. See also Gillette, supra n. 10, on differences between consumers and businesses. Unfair Terms Directive, Art. 1 para. 1. On consumer contracts, see DCFR Art. II.– 9:403: “a term (…) is unfair for (…) if it significantly disadvantages the consumer, contrary to good faith and fair dealing.” On business contracts, see DCFR Art. II.–9:405: “A term (…) is unfair (…) if (…) its use grossly deviates from good commercial practice, contrary to good faith and fair dealing” (emphasis added). A detailed account is provided by Ranieri, supra n. 40, p. 371 et seq. The Reichsgericht is the former German Federal Supreme Court in Civil Matters and the predecessor of the Bundesgerichtshof. See RGZ 62, 264, 266. For a detailed account see L. Raiser, Das Recht der allgemeinen Geschäftsbedingungen, 1935, p. 302. A comprehensive historic analysis is provided by Hofer, in: HistorischKritischer Kommentar, 2007, §§ 305-310, Rn. 9. For an overview of historical developments, see A. Fuchs, supra n. 9, Vorb. § 307 Rn. 15. Supra n. 54.

114

Hans-Bernd Schäfer & Patrick C. Leyens

one-sided standard terms from the perspective of party autonomy. From this perspective on the legal history it seems that German courts initially had not intended to make a differentiation by the addressee of the standard terms. The Statute on Standard Terms (AGBG) of 1976 largely served to codify this jurisprudence. Accordingly, standard terms in business to business contracts were still subject to judicial control. The reform of the law of obligations has changed nothing in this respect. It merely transferred the already extant rules into §§ 305 et seq. BGB. Today, the case law developed by courts for controlling the contents of clauses may be found in the catalogue of clauses with and without possibility of judicial evaluation, §§ 308, 309 BGB. These catalogues may, according to § 310 para. 1 BGB, only be directly applied to consumer contracts. The differentiation between business contracts which is expressed here may be considered ‘soft’, as the violation of one of the catalogues may, even in business transactions, indicate a disadvantage. The reasonableness of the disadvantage, however, is always – i. e. not only where the catalogue provides so – subject to judicial evaluation.61 The stipulations made by the Unfair Terms Directive were satisfied with the level of protection foreseen by the German Statute on Standard Terms which is applied (also) to consumer contracts. Special regulations on contents control of business contracts are not foreseen by European law. The DCFR and the differentiating control standards foreseen therein, namely a significant disadvantage (for consumer contracts) or a gross deviation from good commercial practice (for business contracts), open a new discussion on differentiating control standards.62 The justification for judicial control of standard terms, namely information asymmetries, indicates a transaction-specific differentiation of judicial control competencies, rather than one related to different types of addressees. Consumer and business contracts are presumably equally often accompanied by pre-formulated terms.63 For example, purchasing pencils or napkins for a cafeteria or light-bulbs for an office building are all standardised transactions – they all relate to goods which are used by consumers as well as businesses. The finding that the expected gain of reviewing the pre-formulated terms is less than the information cost involved in doing so will be made with roughly equal prohability for both kinds of transactions. It is therefore justified to include standard terms in business contracts in the scope of judicial control.64

2.3.2. Drafting Recommendation

A German court must take into consideration both the possible differences and similarities between consumer and business contracts. Modifications of the control standard, as far as business contracts are concerned, are only applicable for two reasons, both of which accord with economic considerations: first, the parties generally have more experience in dealing with standard terms. Second, business deals typically place high demands on fast legal transactions. The latter consideration included the legislator to expressly insert the stipulation

61 62 63 64

For more, see A. Fuchs, supra n. 9, § 307 Rn. 375. For the text of the DCFR see supra n. 52 Bunte, NJW 1987, 921, 923. For the controversy amongst German legal scholars see the references supra n. 47

Judicial Control of Standard Terms and European Private Law

115

into § 310 para. 1 BGB that the traditions and customs of business transactions are to be given due consideration.65 In contrast to German law, the DCFR suggests a binding modification of the control standard. Whenever businesses conduct themselves in accordance with consumer behaviour vis-à-vis their standard terms, the more stringent criterion of ‘gross’ rather than the loser ‘significant’ will lead to differing court decisions. The problem of information asymmetries would require a uniform treatment. Accordingly, the binding modifications of the control standard, as foreseen in the DCFR, ought to be removed. While it would be possible to have a business-specific interpretation of the criterion ‘gross deviation from expected business practices’, this would soon lead to difficulties if only because ‘gross’ is, as a result of its use in other contexts especially in ‘gross negligence’, already burdened with case law decisions which Member State courts would hardly be able to ignore while engaging in control of pre-formulated terms.66 Therefore, the surest path to an economically reasonable result is to change the wording of the DCFR and to use the term ‘significant’ for business contracts as well.

2.4. Prohibition of Salvatory Reduction As a final point we shall focus on what we call a salvatory reduction of clauses. Amongst German scholars, it is highly disputed whether a court should preserve a clause by adjusting its excessive risk allocation to a permissible maximum (salvatory reduction). This question is of equally high importance to both national and European private law.

2.4.1. Function of the Prohibition of Salvatory Reduction

According to the general rule of § 306 para. 2 BGB, if a clause is invalid, a contract’s contents shall be determined by the prevailing dispositive law. The law does not foresee a general judicial competency to adjust the contract terms. This has led to the assumption of a prohibition of salvatory reduction. The complete nullity of a clause can represent a considerable burden for its user. For example, a clause which foresees that loan interests may be raised if the lender “considers it necessary” gives a bank leeway which can no doubt be seen as an unreasonable disadvantage of the borrower.67 At the same time, however, it is clear that the banks can only keep the promises they make regarding their interest rates when they are able to cover their refinancing costs. In cases like these, it could be asked whether the clause ought to be reduced to its still-acceptable core which, though still a disadvantage, could nevertheless be salvaged. Up until now, salvatory reduction has been widely considered impermissible. This has led to courts engaging in all manners of legal contortion. Taking the example of the interest rate adjustment clause, the Bundesgerichtshof 68 has adopted an interpretation which implicitly connects the unspecified and too broad leeway of the bank to adjust interest rates specifi65 66 67

68

For more details, see A. Fuchs, supra n. 9, § 307 Rn. 372. Compare in general, DCFR, Annex 1: Definitions. See BGHZ 97, 212. For analysis see Kieninger, in: Münch. Komm. BGB, 5th ed., 2007, § 306 Rn. 16. See also Claussen, Bank- und Börsenrecht, 3rd ed., 2003, § 4 Rn. 20 (p. 97). Supra n. 54.

116

Hans-Bernd Schäfer & Patrick C. Leyens

cally and justifiably to changing circumstances vis-à-vis refinancing costs.69 Practically, this result does not differ from a salvatory reduction of the clause. The dogmatically twisting way to the desired result is, economically speaking, irrelevant. Even legally, one would need a microscope to be able to tell the difference.70 Salvatory reduction eliminates incentives for the user to attempt to formulate fair clauses from the start and it leads automatically to the use of unfair clauses becoming the dominant strategy. Only if the clause is totally null and void will the user suffer a penalty which creates an appropriate incentive to formulate a fair clause. Therefore, the prohibition of salvatory reduction is the appropriate rule as long as the users of standard terms know in advance which clause contents are legally effective, i. e. as long as there is legal certainty. In cases of legal uncertainty, however, the prohibition of salvatory reduction leads to overdeterrence. This finding can be illustrated as follows: If the disadvantage arising from the use of standard terms can be expressed by a scale of 1 to 10, it could be assumed, in the spirit of accelerating legal transactions, that 5 would represent arriving at the optimum contract draft. If the value is below 5, then it is too cautious. If it is greater than 5, then it is unreasonable. From the user’s point of view, the difficulty is to orientate the contract towards this scale in advance. For such risk allocations not yet having been the subject of court judgements, the values around 5 will represent a legal grey area. However, a clause closer to 10 than to 5 will clearly be unreasonable. If the user decides in favour of such a contract draft, then his opportunistic intent should be considered obvious and the prohibition of salvatory reduction should be applied fully for the sake of ensuring that the right incentives are established. The grey zone, however, is problematic, since any clause within this zone could be either reasonable or unreasonable in the court’s eye. The user of the clause, though, is unable to know this ex ante due to legal uncertainty. Within this grey zone, the result is that the user will opt for terms which place unreasonable risks on him. The negative impact on the economy due to overly cautious behaviour has been investigated with a view to civil liability, but can be applied analogously to the regulation of standard terms.71 Without discussing the conclusions of civil liability in great deal here, it is plain that the accelerative effect and the creation of right incentives cannot be optimally reconciled by legal rules which cause overly cautious behaviour.

2.4.2. Drafting Recommendation

For the future development of national and European private law, a differentiating rule appears appropriate which generally assumes the prohibition of salvatory reduction while at the same time offering an exception for contract drafts which are within the legal grey zone. Legal scholarship has debated suggestions which could be used in this context: clauses which “very explicitly or clearly and recognizably or to a quantitatively relevant extent” violate § 305 BGB ought to fall under the prohibition against salvatory reduction.72 On the other

69

BGHZ 97, 212, 216.

70

Kieninger, in: Münch. Komm. BGB, 5th ed., § 306 Rn. 16. Schäfer/Ott, supra n. 28, p. 113, 134. For a specific look at the liability of experts, Schäfer, AcP 202 (2002) 808. Kieninger, supra n. 67, § 306 Rn. 13. See already Basedow, AcP 182 (1982), 335, 356 et seq. See also Gottschalk, AcP 206 (2006) 555, 570.

71

72

Judicial Control of Standard Terms and European Private Law

117

hand, clauses of the sort which a reasonable observer could not, in advance, recognise as being unreasonable ought to be reduced to their still-acceptable minimum.73 In effect, results are achieved by the described method which approximate German court practices. We have already cited the example of the interest rate adjustment clause and the manner in which courts have dealt with it. There is a further example, namely separability of clauses, which has become popular, for example in rental law. In one famous case the landlord had passed on the risk of aesthetic repairs to the renter and, in the same clause, the risk of costs incurred for non-aesthetic repairs. Following the allegedly strict prohibition of salvatory reduction, the clause ought to have been declared entirely null and void. In fact, the court assumed separability and, as a result, declared the risk allocation valid concerning aesthetic repairs but invalid regarding non-aesthetic repairs (§ 307 BGB), due to the latter being an unreasonable disadvantage. For European law and national laws, there are numerous reasons to clearly advocate salvatory reduction for areas with insurmountable legal insecurity on the part of the user of standard terms from an ex ante perspective. It is necessary to recognise that the optimal use of leeway in terms of formulating clauses is also in the interest of the counterparty and especially in the interests of consumer protection. This is because the user of standard terms will always consider the cost of legal insecurity vis-à-vis a certain clause’s (lack of) validity. Consumers would not be well protected if the prohibition of salvatory reduction would cause users of standard terms to employ overly cautious pre-formulated terms, since this would lead them to offer their services at a higher price.74 The users of standard terms may try to exploit the elasticity of the law for their own profit. The users will choose to formulate their clauses in such a way that they fall within the grey zone of legal insecurity and they will, in any case, tend towards an unreasonable disadvantage. This can be countered by a subjective criterion, namely clear recognisability, which should be further developed by case law. Pre-formulated terms will, given a generalising view, be considered clearly and recognisably unreasonable if they are not impermissible just in one point, but rather in all facets. Especially those users of standard terms appearing frequently in the market, e. g. banks, insurers and building societies, are subject to virtually continuous judicial control. High expectations may justifiably, then, be placed on the clauses they use. With each judicial decision, the grey zone of legal insecurity becomes narrower.

3. Conclusions 3.1. Judicial control of standard terms is, economically and legally, a reasonable intervention in party autonomy. A better understanding of its goals and functioning reinforces attempts of harmonisation in European private law and a pragmatic interpretation of existing national law. 3.2. The purpose of clause control is not to avoid unreasonable disadvantages which result from market power. If customers have complete information about the allocation of risks even a monopolist will not pass on such risk to the costumer which he can bear at a lower 73 74

Kieniniger, ibid. For a more general thoughts on consumer protection vis-à-vis protection of the market see Schäfer, in: Grundmann, ed., Systembildung und Systemlücken in Kerngebieten des Europäischen Privatrechts, p. 549. For a comprehensive analysis see already Reich, Markt und Recht, 1977, p. 214.

118

Hans-Bernd Schäfer & Patrick C. Leyens

cost. Passing on risks against the cheapest cost avoider principle will result in a decrease of demand and a reduction of the producer surplus. 3.3. Judicial clause control serves to avoid a (partial) market failure which is the result of information asymmetries between the parties. The cost of acquiring information regarding the contents of standard terms routinely exceeds the anticipated gain. It is therefore rational to ignore the clause contents. As a consequence of this rational ignorance also a competition among issuers for the best terms will fail. The users of standard terms will rather engage in a competition for the most unfair terms (race to the bottom). 3.4. The market as a private mechanism cannot solve the problem as effectively as judicial clause control. Learning effects can be experienced only in industry sectors or markets with a high density of repeat purchases. In other industry sectors or markets the risk allocation will only become observable by few and only long after contract conclusion. Even among market participants such as banks, which place high value on their reputation, unfair contracts occur. The incentives of users to solve the problem by means of self-regulation are incomplete where the risk in question only rarely occurs and when neither demand reductions nor price reductions become noticeable. 3.5. On the basis of the preceding considerations we can assess perspectives for the drafting of the European private law and the interpretation of existing national law. The scope of specific rules for the judicial control of standard terms should not be extended to cases where there is no danger of a (partial) market failure. Accordingly, they should not be extended to individually negotiated clauses. The schematic pattern of the unfairness test is inappropriate for resolving fairness problems in individual cases that are not subject to an information problem, especially cases of inequality of the parties and contracts of adhesion. For those cases the general principles of good faith, fair dealing and bonas mores provide a judicial means to exempt the weaker party from the obligations under the contract instead of merely declaring the ancillary terms of the contract void. The unfairness test under general principles extends to the adequacy of the price, i. e. a disadvantage can be compensated by a price reduction. In contrast, the unfairness test applicable to the control of standard terms does not extend to the adequacy of the price. This difference is in line with economic considerations as a price control would aggravate the information problem instead of solving it. A price reduction might (partially) compensate for a disadvantageous risk allocation. However, it is the very nature of the information problem that users of standard terms pass on risks to offer at a nominally lower price than their competitors. Accordingly, extending the specific unfairness test to the adequacy of the price would jeopardise the effectiveness of judicial control of standard terms as a mechanism to overcome the problem of partial market failure. 3.6. Judicial control of standard terms should be limited to cases where in light of the value of the contract, ignorance of the standard terms is rational (contract value-information cost-relationship). Business contracts should not be controlled if the contract value exceeds a certain threshold. The English and the Scottish Law Commissions suggested excluding contracts with a value of more than GBP 500.000 (€ 559.440).

Judicial Control of Standard Terms and European Private Law

119

3.7. The dangers resulting from rational ignorance of standard terms are similar for businesses and consumers. The binding modification of the control standard as foreseen by the DCFR ought to be abolished. Judicial control should be transaction-specific and not addresseespecific. .

3.8. Judicial adjustment of clauses to the end that they accord with the greatest stillacceptable disadvantage for the other party (salvatory reduction) should be permitted for such cases where there is legal insecurity on the part of the user of standard terms. However, salvatory reduction should be prohibited for all cases where the unreasonability of the clause must be clear to the user of the standard terms. Only in this way the transaction cost efficiencies which result from the use of standard terms can be fully realised.

3. Performance of Contracts

Specific Performance, Damages and Unforeseen Contingencies in the Draft Common Frame of Reference Gerrit De Geest* Introduction If the promisor breaches her contract without a good excuse, the legal system has several choices. It can either oblige her to perform the contract as promised (specific performance), or oblige her to pay compensation instead (damages). Civil law countries are believed to prefer specific performance, while common law countries are believed to prefer damages.1 Even so, these differences tend to be smaller than often believed, especially if one takes a third remedy into account – substitute performance – which can be considered as a hybrid (Eisenberg, 2005), and which seems to be the most frequently used remedy in civil law systems as well (Lando and Rose, 2004). There is an extensive economic literature on whether specific performance is a better remedy than expectation damages (e.g., Kronman, 1978; Polinsky, 1983; Shavell, 1984; Ulen, 1984; Kornhauser, 1986). There is, however, no consensus in this literature. To some extent this may be due to the very nature of publishing: a paper is only publishable if it is novel, which means that it has to disagree on some points with the previous literature. Yet, the lack of consensus has two deeper causes. The first is that economic analysis has identified many different costs that are related to the choice of remedy for breach. The theoretical results show that many of these costs point in different directions. The second is that empirical data on these costs are not available. The result is a large degree of indeterminacy (Posner, 2003). Legislators and courts, however, cannot wait until these data are available. This holds also for the drafters of restatements (or ‘frames of references’) whose task it is to summarize the law in the most simple yet accurate form to reduce the learning costs of students and law practitioners. In De Geest (2009) I proposed a methodology to derive theoretical conclusions in the absence of empirical data,2 which then applied to the problem of to the optimal choice of remedies for breach of contract. The central idea is to apply the rule of Tinbergen, which states that N problems require N legal instruments. More specifically, I argued that incen*

1 2

One Brookings Drive, MO 63130-4899 St. Louis, USA. Phone +1-314-935-7839. E-mail: degeest@ wustl.edu. I would like to thank Michael Greenfield, John Haley, Scott Kieff, Camille Nelson, Anthony Ogus, Urs Schweizer, and participants of Third Roundtable of the Economic Impact Group (Venice, 23 June 2008) for comments. Eleonora Melato and Stuart Harrington provided valuable research assistance. See e.g., Zweigert and Kötz (1998) at pp. 470-485. Also Kornhauser (1986) came to the conclusion that the different viewpoints among law and economics scholarship are due to different empirical estimations rather than different theoretical viewpoints.

124

Gerrit De Geest

tives for efficient breach need to be given by a separate ‘efficient breach’ rule, instead of indirectly through expectation damages. There are traces of such an efficient breach rule in legal systems, in the form of broadly defined impossibility (like commercial impracticability) rules. Therefore, I argued (in line with White, 1988), that the doctrines on remedies and performance excuses should be integrated, because their main function is the same: to allow efficient breach and to prevent inefficient breach. In section 1, I summarize what I believe are the optimal rules, which serve as an economic benchmark for evaluating the rules of the Draft Common Frame of Reference. In sections 2, 3, 4 and 5, I discuss the Draft Common Frame of Reference. The last section concludes.

1. Economically optimal rule Before discussing the Draft Common Frame of Reference from an economic point of view, we need a ‘benchmark’, i.e., an economically ideal set of rules. This section summarizes the economic model rule derived from law and economics literature in De Geest (2009). As for the economic reasoning behind the rules, I will mention the main arguments. For a more extensive economic analysis, and more references to the economics literature, see De Geest (2009). Central in this model rule is the distinction between two zones: the zone in which breach is clearly efficient, and the zone in which performance is efficient. ‘Efficient breach’ means economically optimal breach. A breach is efficient when the performance costs got higher than the value of the good or service to the other party. To illustrate, consider a construction contract with a price of 100, an ex ante-estimated production cost of 50 to the builder, and a utility of 150 to the buyer. Suppose that the parties later discover that construction would cost 160 instead of 50. In this case breaching is economically optimal. Building the house would not create any wealth – to the contrary, it would make the parties as a whole 10 poorer. If the parties would have known the real cost structure at the time of contracting, they would never have agreed to build the house, because the seller would never have agreed to do the job for less than 160, while the buyer would never have been willing to pay more than 150. The efficient breach idea can conflict with the moral duty to keep one’s promise. The moral duty dictates that the builder needs to keep her promise to build the house, irrespective of the costs to do so.3 The efficient breach principle, on the other hand, dictates the breaking of bad promises. Note that ‘bad promise’ means ‘bad’ for the two parties taken together. It does not mean that the builder should have a right to breach whenever she would make a loss. For instance, if the production costs of the builder are 140, she should keep her promise from an economic point of view, even if she makes a loss of 40, because this loss is smaller than the buyer’s consumer surplus of 50. Note also that the efficient breach zone should also include the gray zone, in which it is not clear whether the performance costs exceed the utility. In reality, this may mean that the efficient breach point is not where the production cost becomes 150 in our example (the value to the buyer), but rather where they become 170 or 180. The reason is that every breach involves some extra transaction costs, in the form of communication costs, damage 3

At closer inspection the conflict may disappear because rational parties would never agree to a contract in which the performance cost exceeds the performance value. See Craswell (1989), and Shavell (2006).

Specific Performance, Damages and Unforeseen Contingencies

125

calculating costs, etc. In a sense, keeping one’s promise is always the simplest solution with respect to transaction costs. Therefore, the precise point is where the production costs exceed the value to the buyer plus the transaction cost of breaching. Only breaches that are clearly efficient should be allowed under the efficient breach principle. In the efficient breach zone, the remedy question largely becomes a risk issue, though incentives at earlier stages (to promise carefully and perform carefully) also play a role. Yet the starting principle in the efficient breach zone is that no damages are paid, for two reasons. The first is transaction costs: Calculating and paying damages always involves an additional transaction cost. The second is risk: Paying no damages means that both parties share the risk, because they each forego the benefit they expected from the contract. Risk sharing is generally the best solution if both parties are risk averse and insurance is not available. Nevertheless, damages should be paid in the efficient breach zone in four cases: (a) when the breach became necessary after the breacher’s defective performance, in order to give optimal incentives to perform well; (b) when the breach became necessary after negligent contracting, in order to give optimal incentives to contract in a careful way; (c) when the breach was the result of ‘good news’ that allowed the breacher to improve her situation, because this is the result dictated by standard risk analysis;4 or (d) when the breacher is clearly the superior risk bearer, or when the parties intended to allocate the risk to a specific party. In the efficient performance zone (where the production costs are lower than the utility), the main concern is to give incentives not to breach. This excludes weaker remedies like the reliance measure, and leaves a choice between three remaining remedies: specific performance, substitute performance, or expectation damages.5 Any of these three remedies would in principle do the job. As a matter of fact, none of them will ever be applied if the promisor is rational, since it is cheaper for her just to keep her promise than to be forced to perform, pay full compensation, or pay the full costs of a substitute performance. Therefore, the choice among these three is largely a matter of administrative costs. In terms of administrative costs, substitute performance seems to be the superior remedy (Lando and Rose, 2004). Specific performance should be allowed only for obligations to unique goods and services (for which there is no good substitute) and for obligations not to do something (which are easy to enforce specifically). Expectation damages should be used in the remaining cases (especially for late performance, and buyer’s breach). Yet even in this efficient performance zone there is a risk issue: Production costs can increase and values can decrease, even if the efficient breach point has not been reached. Optimal risk allocation is usually done in this zone through the choice of the contract type. For instance, spot contracts allocate the risk of increased production costs with the seller, while cost-plus contracts allocate it with the buyer (Polinsky, 1987). Nevertheless, contracts are never complete and some risks may not be foreseen, so court intervention in some extreme cases is economically justifiable (especially when the loss of one party is accompanied with a windfall for another party, Trimarchi, 1991).

4

5

Polinsky (1983), De Geest (2009). Allowing the promisor to breach without paying damages would create an additional and unnecessary windfall to the promisor and an artificial loss to the promisee. Substitute performance is often considered a form of damages, but it is a hybrid that has characteristics of both damages and specific performance. See Eisenberg (2005) and De Geest (2009).

126

Gerrit De Geest

2. The efficient breach rule: Contracts should not be performed in the efficient breach zone 2.1. Increased production costs DCFR III.–3:302 (3) states that specific performance cannot be enforced where performance would be impossible or unreasonably burdensome or expensive. The definition of the term ‘impossible’ is narrow, but the impracticability case is covered by the terms ‘unreasonably burdensome or expensive’. While it does not use the term ‘efficient breach’, this paragraph implicitly uses the notion in that it does not allow specific performance in the efficient breach zone, and as a result a rational promisor will always prefer to breach and pay expectation damages than to incur the higher performance costs.6 Note that substitute performance should also not be allowed in the efficient breach zone. From an economic viewpoint, the main issue is that there is no wasteful performance, neither by the promisor, nor in the form of an equally wasteful substitute performance by a third party. It is not clear though whether the DCFR allows substitute performance in this case. According to DCFR III.–3:302 (5), creditor cannot recover damages to the extent that she has increased the loss by insisting unreasonably on specific performance in circumstances where the creditor could have made a reasonable substitute transaction without significant effort or expense. In other words, specific performance is not possible when a substitute transaction is relatively easy to obtain. The relationship between sections (3) and (5) is, however, not clear. What if performance is unreasonably expensive to the promisor, and the promisee can easily obtain substitute performance from another party for whom performance is equally expensive? The comments of PECL do not discuss this case. Literally, the text seems to allow this, though the most likely scenario is that this specific case has not been considered by the drafters. From an economic point of view, substitute performance should not be allowed here, because it is equally wasteful. Another implicit example of the efficient breach notion is the requirement of an impediment to be insurmountable. DCFR III.–3:104 (Excuse due to an impediment, based on PECL 8:108) states that “[a] debtor’s non-performance of an obligation is excused if it is due to an impediment beyond the debtor’s control and if the debtor could not reasonably be expected to have avoided or overcome the impediment or its consequences.” The impediment must be “insurmountable” or “irresistible” (Lando and Beale, 2000, p. 381). The PECL comments further explain “[o]ne cannot expect the debtor to take precautions out of proportion to the risk,” for instance “the building of a virtual fortress” to prevent earthquake damage. In other words, whether something is insurmountable seems to be based on an implicit Learned Hand-type of balancing, with the precaution cost on the one hand, and the cost of non-performance on the other. In the text and the comments, no explicit distinction is made between precaution costs in a strict sense (which are made before the impediment occurs) and performance costs (which are made after the impediment occurs). Yet the criterion seems to be applicable to both: It must be too expensive to overcome the impediment by taking precaution ex ante (before the earthquake) as well as by making an extra effort ex post (after the earthquake). 6

The expectation measure creates optimal incentives for efficient breaches (given a fixed level of reliance). This is a well-known and generally accepted result of law and economics scholarship (Barton, 1972; Shavell, 1980).

Specific Performance, Damages and Unforeseen Contingencies

127

2.2. Decreased utility Breaching can become efficient not only because of increases in production costs, but also because of decreases in utility. Suppose that in our building contract example, where the production costs are 50, the utility decreases to 40 from 150. In this case, building the house is wasteful, and breaching is efficient. DCFR III.–3:301 states that where (i) the creditor has not yet performed the reciprocal obligation for which payment will be due and (ii) it is clear that the debtor in the monetary obligation will be unwilling to receive performance, the creditor may nonetheless proceed with performance and may recover payment. Two exceptions are when (a) the creditor could have made a reasonable substitute transaction without significant effort or expense, or (b) performance would be unreasonable in the circumstances. At first reading, this text seems to suggest that specific performance (i.e., demanding the full price after first continuing to build) is the rule and damages the exception. The comments of PECL 9:101 (on which the DCFR article is based) further explain the philosophy: “Under the principle of pacta sunt servanda the creditor is entitled to make its performance and thereby to earn the price for it. The debtor’s unwillingness to receive the creditor’s performance is therefore, as a rule, irrelevant.” (Lando and Beale, 2000, p. 391). Yet, the exceptions cover a wide number of cases. The first exception is when it is possible to find a substitute buyer, but only one who could not have been served otherwise. This condition ensures that the total number of sales would not be decreased by the buyer’s breach. Economically, this is the case of a seller or producer who works at maximum capacity.7 The second and probably more important exception is when the performance would be ‘unreasonable’. While no definition of reasonableness is given, this exception makes it possible to allow efficient breaches. The PECL comments state: “A typical example is where, before performance has begun, the debtor makes it clear that it no longer wants it. This situation can arise, for example, in construction contracts, other contracts for work, and especially long-term contracts.” (Lando and Beale, 2000, p. 392). In other words, the idea is that performance should not be enforced upon a party against its will. Therefore, if the debtor prefers efficient breaching (which she will if she is rational), there will be efficient breaching, except for in specific cases in which the promisee has an interest in performing the job.

3. Efficient breach zone: Should damages be paid? 3.1. In principle, no damages need to be paid in case of impossibility DCFR III.–3:701 (Right to damages) states that the creditor is entitled to damages for loss caused by the debtor’s non-performance of an obligation, unless the non-performance is excused. In other words, in the case of impossibility, no damages need to be paid in principle, which corresponds to the economically-optimal rule. The only cases in which damages need to be paid are those when the excuse is not accepted, for instance in cases of negligent contracting or negligent performance.

7

For a further discussion of the ‘lost volume seller’ problem in the context of the DCFR, see Ogus (2009).

128

Gerrit De Geest

3.2. Damages should be paid by the party who assumed the risk One can find this principle in the rules on initial impossibility, change of circumstances, and excuse due to an impediment. In the comment on PECL 4:102 (Initial impossibility, on which DCFR II–7:102 is based), it is argued that “there may be cases when a party should be treated as taking the risk of impossibility and therefore should not be entitled to avoid the contract. Specific performance of the contract will of course be impossible but the party which has taken the risk may be liable in damages for non-performance.” (PECL, p. 228) DCFR III.–1:110 requires as a condition for variation or termination by courts on change of circumstances in 3(c) that “the debtor did not assume, and cannot reasonably be regarded as having assumed the risk of that change in circumstance.” DCFR III.–3:104 (Excuse due to an impediment) requires that the promisor could not reasonably have been expected to have avoided or overcome the impediment or its consequences. In addition, it states that the non-performance is not excused if the debtor could reasonably be expected to have taken the impediment into account at the time when the obligation was incurred. The comments on PECL 8:108 (on which DCFR III.–3:104 is based) clarify that in this case “one may say either that the party affected took the risk or that it was at fault in not having foreseen it” (Lando and Beale, 2000, at p.380).

3.3. Damages for negligent contracting through the concept of foreseeability As explained in the previous paragraph, DCFR III.–3:104 also holds a debtor responsible for not having (reasonably) foreseen the impediment at the time of contracting. Here, the idea of negligent contracting is implicit. If the promisor should have taken the impediment into account at the time of contracting, she should not have made the promise in the first place. Negligent contracting should not be an excuse in order to create incentives not to contract negligently. Though specific performance is impossible, the promisor should pay damages. The DCFR nor the PECL are clear as to whether expectation damages or only reliance damages should be paid. From an economic viewpoint, reliance damages seem appropriate (De Geest, 2009). Similarly, the comment on PECL 4:102 on initial impossibility (on which DCFR II.– 7:102 is based) states that a party should be held liable if she should have known about the initial impossibility (Lando and Beale, 2000, p. 228). This is again an example of damages for negligent contracting.

3.4. Expectation damages in case of impossibility after bad or late performance DCFR III.–3:104 (1) states that the impediment must be beyond the promisor’s control as well. The comments on PECL 8:108 (on which the DCFR article is based) explain that the force majeure must have come without the fault of either party, which is the case if the impediment would not have occurred if the promisor had performed on time (Lando and Beale, 2000, p. 380). This implies that when the impossibility occurs after bad or late performance, expectation damages (the normal damages measure) need to be paid, which is in line with the economically optimal result.

Specific Performance, Damages and Unforeseen Contingencies

129

4. Efficient performance zone 4.1. Breaching is not allowed in the efficient performance zone As explained in section 2, a corollary of the efficient breach concept is that breaching should not be allowed when the promisor’s loss is lower than the promisee’s benefit. This corollary can be found in the requirement that the cost increase or utility decrease be excessive to be adapted or terminated by a court. Yet, this article does not attempt to define the exact conditions under which breach (i.e., termination) rather than adaptation is justified. Of course there is the general condition that performance will be “ruinous” for the debtor” (Lando and Beale, 2000, p. 324), or that there are “completely exorbitant costs for one of the parties,” (Lando and Beale, 2000, p. 325) which can in practice be a relative condition (ruinous or exorbitant compared to the value for the other party). The comment on PECL 6:111 explains that termination can be desirable when the court believes that simply modifying the contract (for instance by adapting the price) is not desirable (“the court can modify clauses of the contract but it cannot rewrite the entire contract,” Lando and Beale, 2000, p. 327). In summary, the efficient breach notion may be implicitly applied, yet this cannot be stated with certainty, because the drafters avoided more precise definitions and conditions.

4.2. Substitute performance, specific performance or damages? Above, I have explained that the choice of the remedy in the efficient performance zone is largely a matter of administrative costs. Lando and Rose (2004) have convincingly argued that substitute performance is superior in this respect. It avoids the many difficulties associated with supervising the specific performance by a non-motivated promisor. It also avoids computing expectation damages, which may be hard because is often requires subjective information (as argued by e.g., Ulen, 1984). The Draft CFR rightly seems to consider substitute performance as the first remedy. DCFR III.–3:302 argues that “[t]he creditor cannot recover damages for loss or a stipulated payment for non-performance to the extent that the creditor has increased the loss or the amount of the payment by insisting unreasonably on specific performance in circumstances where the creditor could have made a reasonable substitute transaction without significant effort or expense.” In addition it states that specific performance cannot be enforced where performance would be of such a personal character that it would be unreasonable to enforce it.8 Beside these cases, it leaves the creditor the choice of remedy (DCFR III.–3:101). Parties may also contractually exclude or restrict remedies (DCFR III.–3:105), unless it would be “contrary to good faith and fair dealing” to invoke these clauses. While this exception is vague, the principle of contractual choice makes sense from an economic point of view, since numerous economic models have shown that each remedy can make sense under specific conditions. 8

DCFR III.–3:302 (4) states that the creditor loses the right to enforce specific performance if performance is not requested within a reasonable time after the creditor has become, or could reasonably be expected to have become, aware of the non-performance. This paragraph makes economic sense because it creates an incentive to respond within a reasonable time. The duty to make a substitute transaction made within a reasonable time and in a reasonable manner (DCFR III.–3:706) makes economic sense as well.

130

Gerrit De Geest

4.3. Change of circumstances: risk sharing in the efficient performance zone While contracts should be performed in the efficient performance zone, adapting the terms to changed circumstances can make sense from a risk perspective (Trimarchi, 1991). DCFR III.–1:110 (which largely corresponds to PECL 6:111) subparagraph (2) allows courts to change the contract if “performance … becomes so onerous because of an exceptional change of circumstances that it would be manifestly unjust to hold the debtor to the obligation…” In that case, the court may vary the obligation in order to make it reasonable and equitable in the new circumstances.9 This is only possible, however, if “(a) the change of circumstances occurred after the time when the obligation was incurred; (b) the debtor did not at that time take into account, and could not reasonably be expected to have taken into account, the possibility or scale of that change of circumstances; (c) the debtor did not assume, and cannot reasonably be regarded as having assumed, the risk of that change of circumstances …” The comment on PECL 6:111 (on which DCFR III.–1:110 is based) clarifies the risk sharing idea behind this rule (Lando and Beale, 2000, p. 323). Courts in essence correct a disequilibrium. Illustrative examples include price adaptation after the closing of the Suez Canal and the erosion of contract prices by high inflation. As explained above, such court interventions can be justified on economic grounds.

5. The duty to renegotiate DCFR III.–1:110 (3)(d) holds that a court may change or terminate the contract only if “the debtor has attempted, reasonably and in good faith, to achieve by negotiation a reasonable and equitable adjustment of the terms regulating the obligation.” Is a duty-to-negotiate a good idea? Economic analysis has shown that such a duty has different effects than what people who are not familiar with game theory tend to believe. One should bear in mind that parties can renegotiate anyway. As a matter of fact, both parties have a strong incentive to do so, just like they have a strong incentive to settle cases in general, because settling is cheaper for them than going to trial. Moreover, finding out which party negotiated in good faith may be very hard for courts. The reason is that bargaining is in essence a divide 100 game. In this game, two players get $100 if they can agree upon how to divide it. All agreements are allowed (50/50, 60/40, 90/10, 10/90, even 0/100), but if they do not agree, they do not get anything. What will be the outcome of this game? Game theory reveals that a 50/50 solution is not guaranteed. If someone wants to get the largest part, she has to threaten to blow up the negotiations in case she does not get what she asks – just like in the chicken game where someone has to threaten to cause a crash if he wants to win. This makes it rational to appear irrational. Therefore, it is conceptually hard for courts to even define ‘irresponsible behavior’ in a negotiating process. Even to the extent that courts would define it as ‘overreaching’, i.e. as not accepting a 50/50 division, courts would need much information to find out the parties’ threat values. Because of this ‘nonverifiability’, courts tend to look at ‘signals’, i.e., easily observable facts that are believed to be associated with the negative behavior they want to discourage. Two such observable facts are the amount of time spent to negotiating, and which party stopped negotiating first. Yet these signals may unintentionally create a you-quit-first game. 9

Courts may also terminate the contract. See section 6.

Specific Performance, Damages and Unforeseen Contingencies

131

In this game, two men have been arrested, but the police officer does not know who is the sadist and who is the victim. Both keep saying that the other one is the sadist. To induce them to reveal the truth, the police officer places both men in one cell and tells them that they are not allowed to leave the cell until they reveal who is the sadist. Should one of them decide to leave the cell before an agreement is reached, that person will be considered as the sadist and the one who quits last will be considered as the innocent party. In the cell, each player tries to make the other quit first. Though any outcome is possible, chances are higher that the most vulnerable player will quit first – so that the legal system sanctions the wrong person. ‘Vulnerability’ in the context of contract renegotiations in the light of changed circumstances may mean having to lose most as time goes by without reaching a solution. Therefore, the likely effects of a duty-to-renegotiate are usually (a) that court decisions are delayed, because when a party comes to the conclusion that settlement is not possible (for instance, because parties have different expectations as to what the court will decide, so that there is no ‘settlement range’), she still has to wait until the mandatory renegotiation period is over before going to court; and (b) that parties are obliged to play a you-quit-first game in which they have to try to make the other party leave the negotiation table first, so that the other one is held responsible for the failure of the negotiations. Again, since renegotiation is always possible (it is legally impossible to make contracts ‘renegotiation proof’, considering any court would hold invalid a clause that disallows renegotiation), the relevant question is in the shadow of what parties have to negotiate. Renegotiations may take place (a) in the shadow of a third party decision, (b) in the shadow of the original contract (if the original contract remains in force as long as no new contract has been agreed to), or (c) in the shadow of termination (if each of the parties can terminate the contract). The DCFR opts for negotiations in the shadow of a court decision, which is good, but the legal duty to negotiate is pointless at best and harmful at worst.

Conclusions I have shown that the Draft Common Frame of Reference is largely in line with the results of the law and economics literature. While there is no explicit efficient breach rule in the DCFR, the idea is implicitly present in the rules on impossibility (DCFR III.–3:104), changed circumstances (DCFR III.–1:110), and the conditions for specific performance (DCFR III.–3:104). The more specific solutions in both the efficient breach zone and the efficient performance zone generally make economic sense as well. The main case in which the DCFR seems to be in conflict with the law and economics literature is the legal duty to renegotiate. Otherwise, the main weaknesses of the DCFR are the vague definitions of central concepts (like insurmountability and foreseeability) and the inarticulate conditions for contract adaptation versus termination. Therefore, the main message of law and economics scholarship is that the Draft Common Frame of Reference (and the PECL on which it is inspired) could have been structured and formulated in a much simpler way. Law and economics may also help to make the vague definitions more precise.

132

Gerrit De Geest

References Barton, J. (1972), “The Economic Basis of Damages for Breach of Contract”, 1 Journal of Legal Studies, 277-304. Craswell, Richard (1989), ‘Contract Law, Default Rules, and the Philosophy of Promising’, 88 Michigan Law Review, 489-529. De Geest, Gerrit (2009), How Indeterminate is Law and Economics after Four Decades?, working paper, Washington University School of Law. Eisenberg, Melvin A. (2005), ‘Actual and Virtual Specific Performance, the Theory of Efficient Breach, and the Indifference Principle in Contract Law’, 93 California Law Review, 975 ff. Kornhauser, Lewis A. (1986), ‘An Introduction to the Economic Analysis of Contract Remedies’, 57 Colorado Law Review, 683-725. Kronman, Anthony T. (1978), ‘Specific Performance’, 45 University of Chicago Law Review, 351-382. Lando, Henrik and Rose, Caspar (2004), ‘On the Enforcement of Specific Performance in Civil Law Countries’, 24 International Review of Law and Economics, 473-487. Lando, Ole and Beale, Hugh (eds.) (2000), Principles of European Contract Law. Parts I and II, The Hague: Kluwer Law International. Ogus, Anthony (2009), ‘Measure of Damages: Expectation, Reliance and Opportunity Cost’, in this volume. Polinsky, A. Mitchell (1983), ‘Risk Sharing through Breach of Contract Remedies’, 12 Journal of Legal Studies, 427-444. Polinsky, A. Mitchell (1987), ‘Fixed Price versus Spot Price Contracts: A Study in Risk Allocation’, 3 Journal of Law, Economics, and Organization, 27-46. Posner, Eric A. (2003), ‘Economic Analysis of Contract Law after Three Decades. Success or Failure?’, 112 Yale Law Journal 625 ff. Shavell, Steven (1980), ‘Damage Measures for Breach of Contract’, 11 Bell Journal of Economics, 466-490. Shavell, Steven (2006), “Is Breach of Contract Immoral?”, 56 Emory Law Journal, 439 ff. Trimarchi, Pietro (1991), ‘Commercial Impracticability in Contract Law: An Economic Analysis’, 11 International Review of Law and Economics, 63-82. Ulen, Thomas S. (1984), ‘The Efficiency of Specific Performance: Toward a Unified Theory of Contract Remedies’, 83 Michigan Law Review, 341-403. White, Michelle J. (1988), ‘Contract Breach and Contract Discharge due to Impossibility: A Unified Theory’, 17 Journal of Legal Studies, 353-376. Zweigert, Konrad and Kötz, Hein (1998), Introduction to Comparative Law, Oxford: Clarendon Press.

Measure of Damages: Expectation, Reliance and Opportunity Cost Anthony Ogus*

Introduction I interpret my task to be that of providing an economic commentary on those provisions in the Common Frame of Reference (CFR) which deal with the measure of damages for breach of contract. I avoid the language of “assessment” or “evaluation” because I consider it wrong to assume that the law of contract is concerned to implement exclusively economic goals. Indeed, the topic of measure of damages provides an excellent illustration of this. Most lawyers would consider that the primary objective of a damages award for breach of contract is to provide appropriate compensation for disappointed promisees, normally putting them in the position they would have been in if the contract had been performed (CFR, III.–3.702).1 Most economists have little interest in what the disappointed promisee should receive ex post the breach, since compensation is a transfer payment and, as such, has little significance for economic welfare generally. The key issue for them is what the breaching promisor should pay, since that has important implications for how promisors behave ex ante, and consequentially for trade and welfare more generally. In short, they are concerned with the capacity of the sanctions for breach of contract to induce economically appropriate behaviour. The purpose of this paper is, then, to explore and explain the economic perspective on damages for breach of contract and the extent to which its implications are compatible with the CFR principles which can be assumed to be based fundamentally on the compensatory goal. There is also a brief discussion of the question to what extent harmonisation of the law in this area may be justified.

1. The Economic Function of a Damages Award: Inducement of Efficient Performance In general, parties voluntarily entering an agreement expect gains from it. It follows that a contract is normally Pareto-efficient, provided that third parties are not adversely affected. If these conditions are satisfied, then, to achieve the envisaged welfare improvement, the contract has to be performed. The primary function of the contractual remedies is to induce performance of the contractual obligations and that should be secured if the sanction for breach imposes costs on the promisor greater than the costs of performing the obligations. To

*

1

Emeritus Professor, University of Manchester and Erasmus Professor of Fundamentals of Private Law, University of Rotterdam. I am grateful for the comments and suggestions of two anonymous referees. All references are to the December 2007 version of the draft Common Frame of Reference.

134

Anthony Ogus

achieve this effect, the size of the award of damages does not, in principle, matter, provided that it exceeds the cost of performance. The reasoning in the previous paragraph is, as was made explicit, based on the assumption that the contract generates gains for both parties. For this to arise the value (say V) of contract performance to the promisee must exceed the cost (say C) of performance to the promisor: clearly if C is greater than V, the promisor would demand a price which the promisee would not be prepared to pay. In some contractual situations, what will occur in the future, including the costs of performance and its consequences for the promisee, is known with such a degree of certainty that the efficient outcome is assured. If that is the case, then, the only important implication for the assessment of damages payable for breach is that the sum should exceed C. However, in the great majority of executory contracts there is some degree of uncertainty regarding circumstances occurring between the time of agreement and the time of performance. In particular, there may be uncertainty as to whether, at that second point of time, V will exceed C; and in consequence there is the possibility that performance of the contract will no longer be efficient. The important normative question then arises: should the law induce inefficient performance of the contract; or should it rather tolerate, perhaps even encourage, efficient breach?

2. Expectation Damages as the Default Remedy The so-called doctrine of ‘efficient breach’ has provoked criticism from a non-economic, legal perspective, reflecting the alleged moral basis of the law of contract.2 But the doctrine has deep roots: at the end of the nineteenth century Justice Holmes famously wrote: ‘The only universal consequence of a legally binding promise is that the law makes the promisor pay damages if the promised event does not come to pass. In every case it leaves him free from interference until the time for fulfilment has gone by, and therefore free to break his contract if he chooses’.3 While this may be a caricature of the law in practice,4 it nevertheless constitutes an important recognition that, in some circumstances, wrongdoing may generate economic advantages.5 Unstated in the Holmes dictum, but a crucial condition of the efficient breach doctrine, is the requirement that neither the victim of the breach, nor indeed any third party, should lose by the breach; in other words that the damages payable by the promisor should constitute perfect compensation for the promisee, so that, in effect, that party is indifferent 2

3 4

5

C. Fried, Contract as Promise: A Theory of Contractual Obligations (Harvard University Press, 1981), 113-123. See also: D. Friedmann, ‘The Efficiency Breach Fallacy’ (1989) 18 Journal of Legal Studies 1; F. Menetrez, ‘Consequentialism, Promissory Obligation, and the Theory of Efficient Breach’ (2000) 47 UCLA Law Review 859; F. Cuncannon, ‘The Case for Specific Performance as the Primary Remedy for Breach of Contract in New Zealand’ (2004) 35 Victoria University of Wellington Law Review 657. O.W. Holmes, ‘The Path of Law’ in O.W. Holmes, Collected Papers (Harcourt, 1920), 175. C.S. Warkol ‘Resolving the Paradox Between Legal Theory and Legal Fact: The Judicial Rejection of the Theory of Efficient Breach’ (1998) 20 Cardozo Law Review 321. I generalise on this in A. Ogus, Costs and Cautionary Tales: Economic Insights for the Law (Hart, 2006), chap. 7.

Measure of Damages:Expectation, Reliance and Opportunity Cost

135

between performance of the obligation and breach of it with damages. In short, the breach must be Pareto-efficient. The measure of damages which satisfies the condition correlates well with the key CFR principle (CFR, III.–3:702 cf PECL 9:502): ‘The general measure of damages is such sum as will put the creditor into the position in which the creditor would have been if the contract had been duly performed. Such damages cover loss suffered and gain of which deprived’. In the common law, this is referred to as expectation interest damages (Erfüllungsinteresse in German law) and its economic function is clear.6 It serves to ensure that the promisee is not made worse off by any breach. But that is not all: because it limits the sanction payable on breach to the value that the promisee attributes to the contract, it ensures that valuable resources are not spent on costly performance when such performance is not justified by the promisee’s expectations. Take the following example. A promises to supply certain machinery to B for € 100,000. The machinery when delivered to B is worth € 120,000 and, at the time of the making of the contract, the estimated cost of supply is € 90,000. After the contract, but before performance begins, the cost of the latter increases unexpectedly to € 140,000. Breach is now efficient since C exceeds V – the parties would not have made a contract if that information had been available at the time of formation, because to cover the promisor’s costs, a price of at least € 140,000 would have been asked, and B would not have been prepared to pay that amount. Consider, then, a variation on the example.7 Suppose that at the time of contract that, there is a 0.5 probability that the cost of supply will be €200,000 (Contingency A) and 0.5 probability that it will be € 100,000 (Contingency B). Since the average excepted cost (€ 150,000) exceeds the value of the machinery to the buyer (€ 120,000), the contract will not normally be made. However, if the seller is allowed to breach and pay expectation damages in the event of Contingency B, an efficient contact should be made. Assume that the negotiated price is € 115,000. Estimated ex ante, the seller make a profit of € 7,500 by performing the contract in Contingency A (0.5 x € 115,000-100,000). The estimated ex ante cost of Contingency B is only € 2,500 (0.5 x the difference between the value of the machinery to the seller (€ 120,000) and the contract price (€ 115,000). The estimated ex ante profit to the seller is thus € 5,000 (€ 7.500-2,500) and the contract will be made. Notice that arguments for compelling performance on the ground that A should take the risk of any increase in the cost of performance are not to the point, since on our definition of expectation damages B will be no better off with performance than with the damages award, and is not therefore disadvantaged by the outcome. The reasoning in this section thus leads to the proposition that awarding expectation damages for breach of contract creates appropriate incentives for the promisor for both efficient peformance and efficient breach. Of course, as I will explore in section 6, the parties

6

7

S. Shavell, Foundations of Economic Analysis of Law (Harvard University Press, 2003), chap. 13; B.E. Hermalin, A.W. Katz and R. Craswell, ‘Contract Law’ in A.M. Polinsky and S. Shavell (eds.), Handbook of Law and Economics (North Holland, 2006), chap. 1. I am indebted to an anonymous referee for this variation.

136

Anthony Ogus

could themselves, by inserting relevant provisions into the original contract, induce the same outcomes but, by providing the expectation measure as the default remedy, the law enables them to economise on transactions costs.

3. Dealing with Imperfect Monitoring and Enforcement If, for the reasons outlined in the last section, the economic goal of allocative efficiency justifies awarding expectation damages as the default remedy for breach of contract, we have now to take account of a variety of considerations which might explain some modifications to that rule. The first of these occurs because it is not always easy for promisees to monitor performance of the contract by promisors and some, particularly minor, breaches may escape detection. Even when a breach is detected, it does not follow that enforcement action of some kind will be taken since the promisee may wish to avoid the cost and the hassle involved. Imperfect enforcement affects the inducement characteristics of the damages remedy. If, for example, the promisor anticipates that there is a 0.75 probability that the promisee will not claim compensation, because either he will not detect breach or he will not pursue a claim, then liability costs of breach will be discounted by that amount and awarding only the expectation interest may generate incentives for inefficient breach. One might not expect courts to address this problem by routinely adding to their estimate of the expectation interest an amount reflecting imperfect enforcement: it would be almost impossible to determine ex post what were the probabilities of enforcement ex ante. It does, however, suggest that it might be appropriate for courts to err on the upper side of estimates. Moreover, the case for explicit punitive damages might be strong where the promisor has deliberately attempted to conceal the breach; punitive damages to deter such behaviour might then be acceptable.8 There is, however, no provision within the CFR which would enable a court to make such an award, apart from that (III.–3:710 cf PECL 9:509) allowing an agreed damages clause to be enforced (see below section 6).

4. Inducements for efficient promisee behaviour For efficient outcomes, account must be taken of the behaviour of the promisee, as well as that of the promisor.9 Different considerations and therefore analysis apply according to the time period during which the relevant behaviour might take place but, as we shall see, the common thread is to aim at what the parties would have agreed ex ante as the appropriate outcome, if they had been fully informed of the circumstances. 8

9

D. Farber, ‘Reassessing the Economic Efficiency of Compensatory Damages for Breach of Contract’ (1980) 66 Virginia Law Review 1443; B. Chapman and M.J. Trebilcock, ‘Punitive Damages: Divergence in Search of a Rationale’ (1989) 40 Alabama Law Review 741, 818-819. The need for punitive damages in a contractual context is rejected in J.M. Karpoff and J.R. Lott, ‘On the Determinants and Importance of Punitive Damages Awards’ (1999) 42 Journal of Law and Economics 527, but they assume that the promisor’s reputation constitutes a sufficient incentive for performance, which may not be the case. E.A. Posner, ‘Contract Remedies: Foreseeability, Precaution, Causation and Mitigation’ in B. Bouckaert and G. De Geest (eds), Encyclopedia of Law and Economics, Vol. III, (Edward Elgar, 2000), 162-178.

Measure of Damages:Expectation, Reliance and Opportunity Cost

137

4.1. Promisee behaviour post breach The promisee’s behaviour after the breach of contract has occurred can influence the size of the loss relative to the expectation value of the contract. Suppose it to be the case, for example, that he can purchase in the market an equivalent to the contractual undertaking. Assuming that this has become cheaper than performance by the promisor, whether direct or else by obtaining a substitute in the market, the purchase by the promisee is the efficient outcome because the parties would have opted for it ex ante (with consequential changes to the contract price) if the information had then been available. Limiting the expectation measure to the cost of purchasing the equivalent plus any difference in value between what is purchased and what was promised in the contract should, indeed, induce that outcome. In economic terms, that is, of course, the opportunity cost of the promisee’s bargain.10 This approach to quantification is familiar from the common law doctrine of mitigation11 and appears to be adopted in CFR III.–3:705 (cf PECL 9:505): ‘(1) The debtor is not liable for loss suffered by the creditor to the extent that the creditor could have reduced the loss by taking reasonable steps. (2) The creditor is entitled to recover any expenses reasonably incurred in attempting to reduce the loss.’ The same economic rationalisation can be made of the complementary principles CFR III.– 3:706 (cf PECL 9:506-7): ‘A creditor who has terminated the contractual relationship and made a substitute transaction within a reasonable time and in a reasonable manner may recover the difference between the contract price and the substitute transaction price as well as damages for any further loss’’ and CFR III.–3:707 (cf PECL 9:507): ‘Where the creditor has terminated the contractual relationship and has not made a substitute transaction but there is a current price for the performance contracted for, the creditor may recover the difference between the contract price and the price current at the time of termination as well as damages for any further loss.’

4.2. Promisee behaviour post contract but pre- performance or breach There are two principal ways in which promisee behaviour between the time of the contract and the date when performance is due can affect efficiency. First, there is the possibility of the promisee taking precautions which can avert, or reduce the risk of, breach. If the gain from such precautions, in terms of reduced risk of loss, exceeds the costs incurred, it is in the

10

11

R. Cooter and M. Eisenberg, ‘Damages for Breach of Contract’ (1985) 73 California Law Review 1435, 1440-1441. C.J. Goetz and R.E. Scott, ‘The Mitigation Principle: Toward a General Theory of Contractual Obligation’ (1983) 69 Virginia Law Review 967.

138

Anthony Ogus

interests of both parties that they should be taken.12 As a lover of opera, I know from bitter experience that there is a risk that one or more singers under contract with an opera company to sing a particular performance will be sick on the night, and some companies arrange cover with other singers to deal with the contingency. By parity of reasoning with the post-breach case, economic considerations suggest that the law ought to encourage promisees to take pre-breach precautions when they are efficient. Whether the CFR principles are capable of inducing this outcome is a matter of interpretation. CFR III.–705, quoted above, does not explicitly limit the reduction of loss to ‘reasonable steps’ taken after the breach, although the common law doctrine of mitigation (on which it seems to be based) is so limited. Then there is CFR III.–3:704 (cf PECL 9:504) which provides a rule of no-liability for losses for which the promisee is responsible: ‘The debtor is not liable for loss suffered by the creditor to the extent that the creditor contributed to the non-performance or its effects’. Literally construed, it could be said that a failure to cover for the promisor’s performance ‘contributed to the … effects’ of non-performance, but whether courts would be willing to adopt such an approach is by no means clear. The second possibility of inefficient promisee behaviour between contract and breach relates to reliance expenditure. In most contracts, the promisee engages in some expenditure prior to the date of performance of the contract, knowing that if the promisor duly performs, the value of the contract will thereby be increased. Often reliance expenditure, or a substantial proportion of it, is wasted if the contract is not performed. Given that there is always some risk of non-performance, and wasted reliance expenditure, in theory there is an optimal level of reliance, where the sum invested by the promisee prior to the anticipated performance by the promisor is proportionate to the risk of breach and what will be lost if that breach occurs.13 Unfortunately, the conventional remedy of expectation damages generates no incentives for the promisee to have regard to the optimal level of reliance, because whatever sum is invested will be reimbursed by the promisor on breach, given the aim of the award to put the promisee in the position he would have been in if the contract had been performed. It is theoretically possible to devise a formula to determine the optimal level of reliance based on the probability of breach at the time of making the contract,14 but in practice it is almost impossible to estimate this latter variable. Some commentators have concluded that there is no sanction for breach which can, at the same time, induce both efficient performance (or breach) and efficient reliance.15 However, if the goal is construed more loosely as constraining excessive reliance, rather than inducing efficient reliance, the pessimistic view can be avoided. In a recent paper, Eisenberg and McDonnell have argued that, in the real 12

13 14

15

R.D. Cooter, ‘Unity in Tort, Contract and Property: The Model of Precaution’ (1985) 73 California Law Review 1. S. Shavell, ‘Damage Measures for Breach of Contract’ (1980) 11 Bell Journal of Economics 466. See e.g. A.S. Edlin, ‘Cadillac Contracts and Up-Front Payments: Efficient Investment under Expectation Damages’ (1996) 12 Journal of Law, Economics and Organization 98. Shavell, above n 13, 483; P.G. Mahoney, ‘Contract Remedies: General’ in Bouckaert and De Geest, above n. 9, 123; and, for an experimental endorsement, see R. Sloof, E.Leuven, H. Oosterbeek and J. Sonnemans, ‘An Experimental Comparison of Reliance Levels Under Alternative Breach Remedies’ (2003) 34 Rand Journal of Economics 205.

Measure of Damages:Expectation, Reliance and Opportunity Cost

139

world, expectation damages rarely provide a complete reimbursement of reliance expenditure, and that therefore promisees incur significant risks by engaging in excessive reliance expenditure.16 Moreover, as I have argued elsewhere,17 the principle (to be considered in the next section) that only losses foreseeable by the promisor at the time of making the contract may be recovered (CFR III.–3:703) may act as a significant constraint.

4.3. Promisee behaviour pre-contract CFR III.–3:703 (cf PECL 9:503) provides that ‘the debtor is liable only for loss which the debtor foresaw or could reasonably be expected to have foreseen at the time of the conclusion of the contract as a likely result of the non-performance, unless the non-performance was intentional, reckless or grossly negligent’. The primary economic rationalisation of this principle relates to communication of information by the promisee to the promisor prior to the making of the contract.18 Information plays a key role in the planning of efficient contracts. Clearly, in deciding what resources to spend to avoid defective performance, the promisor must take into account the likely loss that such a performance would cause the promisee, and therefore the damages that potentially would be payable. The costs of the precautions, to the extent that they are economically justifiable, given the risk of defective performance, and any remaining potential liability costs, need to be reflected in the price, if the contract is to remain profitable. In the absence of any special information provided by the promisee, the promisor will base calculations on reasonably foreseeable losses to the promisee from defective performance. Where therefore the promisee would incur, on breach, an unforeseeably high level of losses, efficiency requires that information relating to those losses is communicated to the promisor prior to the making of the contract in order that the latter can make appropriate decisions regarding precautions and the contract price. That requirement is explicitly imposed by the second limb of the famous Hadley v Baxendale principle19 on which the CFR appears to be based.

5. Ex Ante Stipulation From a comparative law perspective, one of the most interesting and controversial issues arising in relation to damages in contract law is the extent to which ex ante agreements between the parties to stipulate the amounts payable for breach of contract should be enforceable. Perhaps surprisingly, arising from an historical concern that promisees will exploit promisors, 16

17

18

19

M.A. Eisenberg and B. McDonnell, ‘Expectation Damages and the Theory of Overreliance’ (2003) 54 Hastings Law Journal 1335 A. Ogus, ‘The Economic Basis of Damages for Breach of Contract: Inducement and Expectation’ in R. Cunnington and D. Saikov (eds.), Contract Damages: Domestic and International Perspectives (Hart, 2008) 125-138. L.A. Bebchuk and S. Shavell, ‘Information and the Scope of Liability for Breach of Contract: the Rule of Hadley vs Baxendale’ (1991) 7 Journal of Law Economics and Organization 284. (1854) 9 Ex 341.

140

Anthony Ogus

common law jurisdictions have tended to adopt a more restrictive approach than civil law jurisdictions.20 CFR III.–3:710 (cf PECL 9:509) provides a compromise between the two traditions: ‘(1) Where the contract provides that a debtor who fails to perform a contractual obligation is to pay a specified sum to the creditor for such non-performance, the creditor is entitled to that sum irrespective of the actual loss. (2) However, despite any agreement to the contrary, the specified sum may be reduced to a reasonable amount where it is grossly excessive in relation to the loss resulting from the non-performance and the other circumstances’. The starting point for the economic analysis is straightforward. Ex ante stipulation of the amount payable can be treated as part of an efficient contract. By prescribing the amount payable in advance, the parties avoid the transaction costs of ex post negotiation and litigation and by allocating the risk of losses arising from default generate the advantages of informational certainty concerning liability costs.21 These arguments are particularly important when the promisee has a substantial subjective interest in performance and when ex post judicial determination may therefore be difficult. They also apply even when the amount payable exceeds the promisee’s losses ex post and thus, in the language of the common law, is a penalty rather than liquidated damages. Penalty clauses may serve the additional purpose of enabling promisors to signal their reliability to the market, which may be particularly important when they are entrants and therefore need to establish a reputation.22 It follows that clause (1) has a strong economic justification. What then of clause (2)? We should first note that its application is not contingent on proof of any market failure, such as an impediment to consent resulting from a mistake, or of there being inadequate competition in the relevant market. The lack of contextual qualification, for example that it was in a standard form contract, also makes it inapt to deal with the ‘signing without reading’ problem.23 Much depends on the interpretation to be given to the requirement that, for judicial intervention, the specified sum must be ‘grossly excessive in relation to the loss resulting from the non-performance and the other circumstances’. If ‘loss’ is intended to refer to the sum which a court is likely to award in an objective assessment, the rule may constitute a serious impediment to efficient contracting, since the stipulated sum may reflect a very high subjective interest in performance which courts in practice would be reluctant to recognise. Note, too, that if the contract takes place within a competitive market, the promisee will have to pay an increased price for the agreed damages clause, since the promisor will have 20

21

22 23

See, e.g. A. Hatzis, ‘Having the cake and eating it too: efficient penalty clauses in common and civil contract law’ (2003) 22 International Review of Law and Economics 381; C. Calleros, ‘Punitive Damages, Liquidated Damages, and Clauses Pénales in Contract Actions: A Comparative Analysis of the American Common Law and the French Civil Code’ (2006) 32 Brooklyn International Law Journal 1. The literature supporting these propositions is summarised in G. De Geest and F. Wuyts, ‘Penalty Clauses and Liquidated Damages’ in Bouckaert and De Geest, n. 9 above, 141-161. For empirical verification, see L. Noll, ‘Vertragsstrafen im Experiment’ (2005) 34 Zeitschrift für das Wirtschaftsstudium 619. R.A. Posner, Economic Analysis of Law (Wolters Kluwer, 7th edn, 2007), 128. De Geest and Wuyts, above n. 21, 147-148.

Measure of Damages:Expectation, Reliance and Opportunity Cost

141

to cover the increased potential liability costs: effectively the promisor is acting as insurer for the promisee.24 Some further costs likely to arise from clause (2) should be noted.25 If the ability to stipulate liquidated damages clauses is constrained that will increase the costs of signalling reliability to the other party. It may also inefficiently reduce the care that promisors take to avoid breach, since they are relieved of some of the consequences of liability which they would have borne if the negotiated clause would have been enforceable. Commentators, particularly from common law jurisdictions, have striven to find economic justifications for the non-enforcement of penalty clauses. The arguments that their enforcement may inhibit efficient breach26 and motivate promisees to induce inefficient breaches of contracts27 are not very compelling, given that the remedy of specific performance has the same characteristics, and it is widely recognised that the latter can induce efficient outcomes, notably through bargaining.28 The assertion that penalty clauses may increase the amount of litigation29 is more plausible, if it be assumed that the higher the stakes, the less the probability of settlement.30 But litigation tends more often to result from uncertainty and informational failure, leading to over-optimism; and, on this basis, the rule which requires the court to distinguish between (enforceable) liquidated damages and (unenforceable) penalty clauses may generate more litigation than that permitting penalty clauses. In summary, while there is no consensus in the literature on the merits of the penalty clause rule, it is by no means clear that the principle which allows the courts to reduce ‘grossly excessive’ agreed damages is conducive to economic welfare.

6. Hypothetical Ex Ante Stipulation Of course, explicit provision in the contract for compensation may be costly and is unlikely to be adopted by large numbers of promisees, but the notion of what the parties would have agreed upon as appropriate compensation for a breach of contract in a hypothetical ex ante agreement may provide important guidance on what, for efficiency purposes, is the appropriate measure of damages to be awarded ex post.

6.1. Subjective value and non-pecuniary losses Take first the question of awards for so-called ‘non-economic loss’ authorised by CFR III.– 3:701(3) (cf PECL 9:501). Although most legal systems allow such compensation, there is 24

25 26 27

28 29

30

C. Goetz and R. Scott, ‘Liquidated Damages, Penalties and the Just Compensation Principle: Some Notes on the Enforcement Model of Efficient Breach’ (1977) 77 Columbia Law Review 554. I am grateful to an anonymous referee for alerting me to these points. R.A. Posner, Economic Analysis of Law (Little Brown, 2nd edn., 1977), 93. K. Clarkson, R.L. Miller and T.J. Muris, ‘Liquidated Damages v Penalties: Sense or Nonsense?’ (1978) 54 Wisconsin Law Review 351. A. Schwartz, ‘The Case for Specific Performance’ (1979) 89 Yale Law Journal 271. P. Rubin, ‘Unenforceable Contracts: Penalty Clauses and Specific Performance’ (1981) 10 Journal of Legal Studies 237. L.A. Bebchuk, ‘Litigation and Settlement under Imperfect Information’ (1984) 15 Rand Journal of Economics 404, 409.

142

Anthony Ogus

a significant divergence on how much should be awarded and for what types of losses.31 If the contract confers a significant consumer surplus on the promisee, then a (presumptively efficient) ex ante liquidated damages clause can stipulate a sum which reflects loss of the subjective, non-pecuniary benefits of performance.32 However, it is important to recognise that if this clause leads to the promisor taking additional precautions against breach or to an increase in the potential liability costs, these costs will be reflected in the contract price and, because of this, it may be cheaper for the promisee himself to deal with the subjective element in the loss. If that element is relatively small, or can be alleviated by other means, he may not be prepared to pay the increased price.33 Putting the same point in a more concrete form, supposing that the promisee purchased insurance cover against the losses from the breach of contract, would that cover, at an additional premium, have included non-pecuniary losses, or would those losses be, in effect, self-insured?34 Clearly the court can ask itself, hypothetically, if the parties had agreed on a liquidated damages clause, whether, in accordance with the above reasoning, it would have included compensation for non-pecuniary benefits. If the contract price is higher than that which is typically charged in the market, that may provide some hint as to how the question should be answered. But the absence of such an indication should not be treated as decisive. The reasoning and solution here is linked to what was said earlier on the impact of information on the pricing of the contract and the Hadley v Baxendale principle. If the existence and extent of the non-pecuniary loss were reasonably foreseeable at the time of the contract, the damages award should include it because the impact of such liability could have been reflected in the price. If they were not reasonably foreseeable, the price could not have been modified and award of the loss would be inappropriate. The above analysis also helps us to address another familiar issue arising from defective contractual performance, whether the damages should be assessed by reference to the diminution in value to the promisee or rather to the cost of curing the defect to provide what was promised.35 The problem only arises where the promisee has a non-financial interest in performance; in other cases, the award of the diminution of value, if accurately assessed, will create the appropriate incentives for efficient performance and efficient breach. Where the subjective value is significant, the cost of cure measure may constitute an appropriate alternative to the court attempting to assess the non-pecuniary benefit of performance.

6.2. Seller’s loss of volume When a buyer repudiates a sale of goods contract, should the seller’s damages be assessed by reference to the profit that would have been made (the ‘lost volume measure’), on the assumption that the number of profitable sales has been reduced? Or should it be assumed that 31

32

33

34 35

M. Whincup, Contract Law and Practice: The English System and Continental Comparisons (Kluwer, 1990), § 13.63. D.R. Harris, A.I. Ogus and J. Phillips ‘Contractual Remedies and the Consumer Surplus’ (1979) 95 Law Quarterly Review 581. D. Harris, D. Campbell, R. Halson, Remedies in Contract and Tort (Cambridge University Press, 2nd edn, 2006), 596-597. S. Rea, ‘Non-Pecuniary Loss and Breach of Contract’ (1982) 11 Journal of Legal Studies 35. T.J. Muris, ‘Cost of Completion or Diminution in Market Value: The Relevance of Subjective Value’ (1983) 12 Journal of Legal Studies 379.

Measure of Damages:Expectation, Reliance and Opportunity Cost

143

a substitute purchaser would be found, that the number of sales would not be reduced, and that therefore the seller should receive only the difference between the contract price and the market price at the time of breach, plus incidental expenses (the ‘market measure’)? The question has given rise to some debate.36 Applying the reasoning above and adopting a policy of inducing the promisee (here the seller) to take appropriate mitigating steps ex post the breach would lead us to relate the solution to the state of the market at the time of breach: if demand exceeds supply, the seller can be expected to find another buyer and the market measure should be awarded; if supply exceeds demand, no additional sale can be found and the lost volume measure is appropriate. Nevertheless, determining the conditions of the market in respect of particular sellers at the time of breach is not always easy and can render the application of the principle uncertain. An alternative is to ask how the parties did allocate, or would have allocated, the risks relating to market conditions. One possibility is that sellers would prefer not to respond to buyers’ defaults by mitigating the loss and finding an alternative sale. If buyers would be prepared to accept this, and the higher price which would ensue, given the increase in potential liability costs, that would suggest the appropriateness of the lost volume measure. But, of course, many buyers would not accept this, particularly since the amount of the sellers’ profit on the transaction is unlikely to be known to them. Another possibility is that both parties would treat the transaction primarily as an allocation in the risk of fluctuations in the price of the goods.37 The contract fixes the price in advance of the date of payment. If, on that date, the market price is lower than the contract price, sellers are protected against the buyer defaulting, while if the price is higher, buyers are protected against sellers defaulting. If both parties would have agreed to this characterisation of the contract, then the ‘market’ measure would be appropriate. In many transactions, particularly in well-functioning markets, this latter interpretation would seem to be the more likely approximation to the parties’ preferences, not the least because information on market prices is more readily available, and thus potential liability costs can be more easily predicted.

7. Harmonisation of the Law? The question of whether, and to what extent, there should be harmonisation of the law governing the assessment of damages for breach of contract can be related to discussions in the familiar law-and-economics literature on the harmonisation issue more generally.38 I do not wish here to rehearse the reasoning which leads to the general conclusion that harmonisation can be justified only where retaining national legal principles generates externalities which cannot be solved by other means, or where the savings in transactions costs clearly outweigh 36

37

38

C. Goetz and R.E. Scott, ‘Measuring Sellers’ Damages: The Lost Profits Puzzle’ (1979) 31 Stanford Law Review 323; V.P. Goldberg, ‘An Economic Analysis of the Lost-Volume Retail Seller’ (1984) 57 Southern California Law Review 283. Cf R.E. Scott, ‘The Case for Market Damages: Revisiting the Lost Profits Puzzle’ (1990) 57 University of Chicago Law Review 1155. See, e.g., R. Van den Bergh, ‘Subsidiarity as an Economic Demarcation Principle and the Emergence of European Private Law” (1998) 5 Maastricht Journal of European and Comparative Law 129; A. Ogus, ‘Competition between National Legal Systems: A Contribution of Economic Analysis to Comparative Law’ (1999) 48 International and Comparative Law Quarterly 405.

144

Anthony Ogus

the welfare losses which arise from overriding localised preferences. In the context of contract law, this latter contingency is unlikely, given that, if national principles are available as an option, the parties can select that which meets their preferences at relatively low cost by means of choice of law clauses. It follows that a set of principles which is likely to meet the preferences of most contracting parties across the European Union should, at most, be adopted only as a default code, leaving the parties free to select national principles if these better meet their preferences. As regards the assessment of damages, my earlier discussion reveals the importance of empowering the parities to select principles, for example, on the ready enforcement of agreed damages clauses, which generate appropriate incentives for efficient performance and efficient breach in the particular circumstances of their transaction. A default code can meet that concern. In a recent paper, Lesley Jane Smith strongly advocates harmonising the law of contractual damages for reasons of “consistency, fairness and equality”.39 Her reasoning is nevertheless mostly vague. In terms of fairness and economic “rights”, many would question her assertion that “a common position could readily be taken within what is otherwise an integrated (social-justice) market economy with common social and economic goals”.40 Indeed, the very diversity of legal solutions to questions of, for example compensation of non-pecuniary loss and the enforcement of penalty clauses suggests that there is no “common position”. Nor is it evident that, for the purposes of an integrated market, there should be such a position. Individual Member States are free to apply their own conceptions of distributional justice in relation to (say) income tax; why should it be different for contracts? A second argument that diversity “could lead to significant distortion of competition”41 fares no better if it is assumed that the contracting parties are adequately informed as to relevant legal solutions and can, at relatively low cost, negotiate to a mutually acceptable outcome. Freedom of choice with reference to legal rules, and their costs and benefits, in such a context leads to no more a “distortion of competition” than where there is freedom of choice relating to the agreed quality of the subject-matter of the transactions and the costs and benefits to which that quality gives rise. Of course, we know that in many contractual situations, notably in relation to consumer contracts, the assumptions of information and capacity to negotiate do not hold. That does suggest the appropriateness of limiting freedom of choice of law and legal principles in relevant transactions. Private international law does indeed inhibit the enforcement of choice of law clauses in contracts where that conflicts with public policy considerations imposed by domestic law.42 But for this purpose, there is no justification for requiring those considerations, for example, on defining the circumstances in which such protection is to be granted or the content of the mandatory principles of assessment of damages, to be the subject of harmonisation at a European level. As with income tax and distributional justice, these are matters best determined locally by politically accountable legislatures.

39

40 41 42

L.J. Smith, “The Eye of the Storm: On the Case for Harmonising Principles of Damages as a Remedy in Contract Law” (2006) 2 European Review of Contract Law 227, particularly at 234-243. Ibid at 235. Ibid at 237. See, e.g., Convention on the Law applicable to Contractual Obligations (Rome, 19 June 1980).

Measure of Damages:Expectation, Reliance and Opportunity Cost

145

Conclusions Harmonisation of the law governing the assessment of damages for breach of contract cannot, in my opinion, be justified. There are nevertheless arguments for a code providing a set of default principles or else guidelines to which national courts might refer for the purposes of interpreting their own law. As I anticipated in my introduction, I regard it as inappropriate to reach robust conclusions on the efficiency of the CFR principles for measuring damages for breach of contract. Apart from the possibility that these principles might, in part at least, reflect non-economic goals, account has also to be taken of their breadth. In that context, an important function of the economic appraisal will be to reveal how specific interpretations of the principles may be more or less consistent with efficiency goals. As I have shown, it is possible to interpret most of the principles in a way which renders them consistent with what economic analysis suggests should achieve efficient outcomes. There are, nevertheless, two potentially important criticisms which can be made of the principles from an economic perspective. First, they do not confer power to award damages exceeding the promisee’s losses and, on efficiency grounds, there may be a case for such awards where detection and enforcement by the promisee is particularly problematic. Secondly, the power conferred to reduce the amount stipulated in agreed damages clauses may, in some contexts, inhibit efficient contracting.

Remedies for Non-Performance: Chapter 3 of Book III of the DCFR from an Economist’s Perspective Urs Schweizer Introduction The DCFR1 is divided into seven Books, each of which is subdivided into Chapters. Book III deals with obligations and corresponding rights. It contains seven Chapters of which Chapter 3, entitled “Remedies for non-performance”, has seven sections. The present paper offers selective comments on some of the sections of Chapter 3. Yet, Section 2 (cure by debtor of non-conforming performance), Section 3 (right to enforce performance) and Section 4 (withholding performance) are left out entirely. Even of the remaining sections, the reader should not expect a systematic account from the present paper. The present paper does neither provide an overview of how standard law and economics theories with respect to contract law would view the proposals in the DCFR. Textbooks and handbooks on law and economics are widely available. The reader may find out and decide on his or her own which recommendations from that literature are specific enough to be directly confronted with particular provisions of the DCFR. Instead of being systematic in terms of topics covered, I have decided to be systematic rather with respect to the analytic approach. Well in line with rational choice and noncooperative games, economic incentives as generated by legal regimes will be addressed by the paper extensively. Uncommonly enough for an economist, however, the primary concern will be with compensatory properties of breach remedies. As it turns out, requirements to compensate and incentives are closely linked. Moreover, approaching incentives primarily from a compensatory perspective may be appealing to some readers, in particular to scholars of law. After all, expectation damages and other remedies aim directly at compensating the victims of breach. The compensatory perspective provides a convenient and powerful analytic tool. There are many topics not discussed by the present paper that would merit an analysis as well. But penetrating a subject along the above lines takes time and much remains to be done. Most of the topics actually covered have been brought to my attention by colleagues from the law department at the University of Bonn. The selection may be biased but, at least, it comes from legal experts. This, of course, does not imply that these experts also approve of what I have done with their questions. The paper’s content and organization is as follows. Section 1 provides the analytic tool that will be used throughout the paper. To visualize the compensatory properties of legal regimes, the creditor’s payoff is plotted as a function of the debtor’s performance or investment

1

The present paper refers to the interim outline edition of the DCFR (Draft Common Frame of Reference) edited by Bar et al. (2008).

148

Urs Schweizer

decision. The qualitative shape of this function will allow deriving the incentives generated by different regimes in a graphic way. In section 2, the tool is applied to the remedy of damages. The DCFR defines the general measure of damages for loss caused by non-performance of an obligation as such sum that will put the creditor as nearly as possible into the position in which the creditor would have been if the obligation had been duly performed. This provision requires damages to be calculated relative to a given obligation which, in Germany, is referred to as difference hypothesis. The main insight from section 2 will be that efficient incentives emerge whenever damages are awarded relative to an efficient obligation. Hence, over-reliance and related distortions must be due to the inefficiency of the underlying obligation or to less than full compensation. In section 3, the same tool is applied to the remedy of price reduction. The DCFR allows for price reduction in proportion to the decrease in value of what was received at the time made compared to the value of what would have been received by virtue of a conforming party. It is shown why this provision fails, not only, in terms of compensation but also of incentives. Determining the size of the price reduction by reference to a hypothetical bargaining situation, instead, would seem preferable from the economic perspective because efficient (ex post) performance incentives at least would be restored. Section 4 revisits the familiar over-reliance result. In a setting with costs of performance that are uncertain at the contracting stage, expectation damages are awarded relative to the creditor’s non-contingent obligation to perform. The inefficiency of this obligation is generating excessive reliance incentives which, however, may be corrected if damages are combined with the appropriate performance excuse. The DCFR excuses a debtor’s non-performance of an obligation if non-performance is due to an impediment beyond the debtor’s control and if the debtor could not reasonably be expected to have avoided or overcome the impediment or its consequences. This provision remains rather vague with respect to impracticability defences. Nonetheless, I show that an excuse based on non-profitability of performance would generally outperform the more familiar excuse based on inefficiency of performance. Such findings may be of help in drafting a more precise definition of the impediments admitted as a performance excuse. Section 5 discusses measures of damages for cases involving uncertain causation and losses of chances for which the DCFR misses to provide specific rules. The all-or-nothing approach still seems to be the prevailing arrangement. More recently, however, courts have started viewing the debtor’s deviation as having caused the loss of a chance and are awarding a quantum of damages below full recovery accordingly. The findings of this section support the emerging view. The all-or-nothing approach is shown to distort incentives as it misses the compensatory goal of damages rules. Amending the DCFR’s definition of damages measures for losses of chances may help to align what currently seems governed by rather disparate legal practice.2 Section 6 deals with the requirement to mitigate losses. In line with the DCFR’s provisions, the debtor is not liable for loss suffered by the creditor to the extent that the creditor could have reduced the loss by taking reasonable steps. In contrast to the subjects covered previously, the decisions of debtor and creditor must now be examined simultaneously. The incentives generated by the legal regime follow from the Nash equilibrium of the corresponding game. The Nash equilibrium is shown to be efficient in the absence of the requirement to mitigate. The requirement, however, may serve to provide efficient incentives even off the equilibrium path. 2

For an account of this issue, the reader may consult Mäsch (2004).

Remedies for Non-Performance

149

Section 7 addresses termination of contractual obligations which the DCFR regulates in meticulous detail. For illustration, the focus will be on delays of performance. If such a delay constitutes a non-fundamental non-performance the DCFR provides termination as a remedy for non-performance only after the creditor has allowed for an additional period of time of reasonable length for performance. Only if the delay constitutes a fundamental nonperformance the creditor has the right to terminate immediately. The analysis of immediate termination versus termination after an additional period of time leads to the conclusion that the trade-off will remain ambiguous and, hence, that the DCFR’s attempt to fine-tune the rules of termination lacks economic justification. Section 8, finally, provides an easily accessible summary of the main points. Some of the other sections may appear less easily accessible to non-economists, in spite of the fact that I have spent great effort keeping the analysis as simple as possible. In Sections 4 and 5 which probably are the most demanding ones in technical terms, I am abstaining from presenting general proofs but I rather illustrate the findings by reference to numerical examples. Unfortunately, even to understand these examples, the reader may have to refresh some of his math from high school which legal scholars generally tend to dislike. Yet, you cannot expect to benefit from the economic approach to law without getting acquainted with some of its basic tools.

1. Compensation and incentives Think of a debtor facing a (one-dimensional) decision x P X which affects both his and the creditor’s wealth position. His contractual obligation is to decide xo but he may consider deviating. The social surplus (sum of the two parties’ wealth positions) amounts to W(x) and attains its maximum at the efficient decision x* P X. The legal regime in place affects the distribution of the social surplus, W(x) = U(x) + V(x), where U(x) and V(x) denote the debtor’s and the creditor’s net share, respectively. The debtor aims at maximizing his private payoff U(x). The legal regime generates efficient incentives if the social surplus and his private payoff attain their maxima at the same decision. It generates excessive or weakly excessive incentives if his private optimum is in the range x* , x or in the range x* # x, respectively. Similarly, the legal regime generates insufficient or weakly insufficient incentives if the debtor’s private optimum is in the range x , x* or in the range x # x*, respectively. The debtor’s incentives will be derived from the shape of the creditor’s payoff V(x) as a function of the debtor’s decision. Figure 1 plots three possible shapes (a), (b) and (c) that correspond to different legal regimes. Details of these regimes will be explained in the next section. In the present section, it is shown which incentives follow from the qualitative properties of the three curves.

Urs Schweizer

150

Figure 1: Creditor’s payoff as a function of debtor’s decision To begin with, consider case (a). Suppose, first, the efficient decision is below the debtor’s obligation, x* # xo. Then the creditor’s payoff attains a minimum at the efficient decision. Put differently, the creditor is fully compensated for the debtor’s deviations from the efficient decision. Since the sum of the debtor’s and the creditor’s payoffs is highest at the efficient decision it follows that the debtor’s payoff must attain its maximum at the efficient decision as well. In other words, if x* # xo then the debtor has efficient incentives. If, second, the efficient decision is above the debtor’s obligation, xo , x*, then the debtor has insufficient incentives. In fact, since the creditor’s payoff increases but the social surplus decreases for decisions in the range above the efficient decision the debtor has no incentives to move into that range. Rather, since the creditor’s payoff has a positive derivative at the efficient decision whereas the social surplus’ derivative vanishes, the derivative of the debtor’s payoff function must be negative at the efficient decision and, hence, he has the incentive to keep his decision strictly below the efficient level. Incentives are insufficient indeed which fully settles case (a). Next, consider case (b). The creditor’s payoff attains its unique minimum at the debtor’s obligation. Therefore, if the obligation and the efficient decision are the same then the debtor has efficient incentives. If the efficient decision is higher then the debtor has insufficient incentives for the same reason as before. Similarly, the debtor has excessive incentives if the efficient decision is lower than the debtor’s obligation. Case (b) is fully settled. In case (c), finally, the debtor has always insufficient incentives because the creditor’s payoff is decreasing and has a negative slope everywhere. To summarize the above insights, if the creditor’s payoff attains a minimum at the efficient decision then the debtor has efficient incentives. If the creditor’s payoff is increasing in the range above the efficient decision and has a negative slope at the efficient decision then the debtor has insufficient incentives. If, finally, the creditor’s payoff is falling in the range below the efficient decision and has a negative slope at the efficient decision then the debtor has excessive incentives. It may seem uncommon to look at the payoff of one party if incentives of the other party are at stake. Yet, the method proves useful in subsequent sections of the paper. It rests on the fact that the requirement to compensate one party affects the incentives of the other

Remedies for Non-Performance

151

party. In particular, if the creditor is compensated or even over-compensated for the debtor’s deviations from the efficient decision then the debtor must necessarily be facing efficient incentives.

2. Damages for breach of contract The DCFR defines the general measure of damages for loss caused by non-performance of an obligation as such sum that will put the creditor as nearly as possible into the position in which the creditor would have been if the obligation had been duly performed (article III.–3:702). In other words, the creditor is compensated for harm caused by the debtor’s deviation from his obligation. Under this legal regime, the creditor’s payoff as a function of the debtor’s decision exhibits the following shape. The payoff functions of debtor and creditor before compensation are denoted as A(x) and B(x), respectively. To fix ideas, the creditor’s payoff B(x) is assumed to increase with the level x of the debtor’s decision. If the debtor had met his obligation, the creditor’s payoff would be B(xo) which, under the debtor’s actual deviation x ° xo , may be higher or lower. If he performs below his obligation, x , xo, the difference B(xo) – B(x) is positive and is equivalent to the harm suffered by the creditor from the deviation and, hence, the difference corresponds to the sum that will put the creditor exactly in the position in which the creditor would have been if the obligation had been duly performed. If, however, the debtor performs in excess of his obligation, xo , x, the difference B(xo) – B(x) is negative and the creditor actually enjoys a benefit from the debtor’s deviation. If classified as unjust enrichment the creditor may have to return the benefit. For simplicity, however, I assume that the creditor may keep such benefits for free. Damages owed by the debtor can then be summarized as D(x) = max[B(xo) – B(x), 0] for short. Including damages, the legal regime leads to payoff functions U(x) = A(x) – D(x) and V(x) = B(x) + D(x) of the debtor and the creditor, respectively. Since the creditor’s payoff B(x) increases with the debtor’s decision, in the range below the debtor’s obligation, the creditor suffers from harm which, however, she is fully compensated for whereas, in the range above the debtor’s obligation, she enjoys gains for free. Case (a) in Figure 1 corresponds to this situation. In case (a), the legal regime generates efficient incentives for the debtor provided that his obligation is at least as high as the efficient decision. As a first modification of the regime, imagine that the creditor is compensated in excess of her loss such that damages awarded amount to D(x) = (1 + λ) · max[B(xo) – B(x), 0] for some positive λ. In this case, the creditor’s payoff as a function of the debtor’s decision is as in case (b) of Figure 1 and, hence, the legal regime still generates efficient incentives but only if the debtor’s obligation consists of deciding efficiently, xo = x*. As a second modification, imagine that the creditor is compensated short of her loss such that damages awarded amount to

Urs Schweizer

152

D(x) = (1 – λ) · max [B(xo) – B(x), 0] only, again for some positive 1 , λ . In this case, the creditor’s payoff as a function of the debtor’s decision is as in case (c) of Figure 1 and, hence, this legal regime provides insufficient incentives, irrespective of the obligation’s relative position with respect to the efficient decision. To summarize: If damages are awarded fully in line with the DCFR’s definition then efficient incentives are generated whenever the debtor’s obligation is sufficiently high. If, however, the creditor is denied full recovery for reasons whatsoever the debtor’s incentives will be distorted downwards. If, finally, the creditor is compensated in excess of her loss the debtor is still facing efficient incentives provided that his obligation is efficient. From an economic perspective, the DCFR’s definition seems fine. Yet, in legal practice, it may prove difficult to award correct damages. Since the creditor may often fail to verify her true damages in front of courts, the legal system tends to award damages with a systematic downwards bias which, if anticipated, leads to insufficient incentives for the debtor quite generally.

3. Price reduction The DCFR allows for price reduction in proportion to the decrease in value of what was received at the time made compared to the value of what would have been received by virtue of a conforming party (article III.–3:601). Potential price reductions affect the debtor’s incentives to perform. To examine such incentives, I look again at the creditor’s payoff as a function of the debtor’s performance decision. In their contract, parties have agreed that the debtor should perform at level xo in exchange of which the creditor pays price po to the debtor. Nonetheless, the debtor may consider performing below his contractual obligation. Suppose, for reasons whatsoever, damages are denied but the creditor may still opt for a price reduction in line with the DCFR’s provisions.3 At actual performance x , xo, the reduction would follow from o

o

p – p(x) B(x ) – B(x) ________ = ___________ o o p

B(x )

such that, under DCFR’s legal regime, the creditor’s payoff amounts to B(xo) – po V(x) = B(x) – p(x) = _________ · B(x). B(xo) The shape of the creditor’s payoff as a function of the debtor’s performance depends on the sign of B(xo) – po. If parties have agreed on a price below the creditor’s valuation of due performance then the sign would be positive. This case is depicted as (c) in Figure 2. Since the creditor’s payoff is increasing everywhere it follows from the findings of section 1 that the debtor has insufficient incentives to perform. If, instead, parties have agreed on a price equal to the creditor’s valuation shape (a) in Figure 2 would emerge. In this case, the debtor has efficient performance incentives provided 3

The DCFR fails to define what exactly is meant by value. In the present paper, I interpret value as the creditor’s subjective willingness-to-pay. The drafters may have had value of a more objective nature in mind. Economists, in contrast, do not have an objective notion of value at hand.

Remedies for Non-Performance

153

that his contractual obligation is at least as high as the efficient decision. Notice, to ensure the creditor’s consent, an up-front payment would possibly be needed transferring the appropriate share of the surplus to her. If the price for due performance exceeds the creditor’s valuation then her payoff as a function of the debtor’s decision is as in case (b) of Figure 2. The debtor would still face efficient incentives provided that the contractual obligation is equal to the efficient performance decision. An up-front would payment, however, would definitely be needed to ensure the creditor’s consent.

Figure 2: Creditor’s payoff under price reduction The DCFR allows the price to be reduced in proportion of the decrease of the creditor’s valuation. Incentives may or may not be distorted as I have just shown. Another, potentially more attractive solution would have been to refer to a hypothetical bargaining situation as the DCFR actually does elsewhere. On termination, the recipient of any benefit by the other’s performance is obliged to return it. To the extent that the benefit is not transferable it is to be returned by paying its value (article III.–3:511). When no price was agreed the value of the benefit is the sum of money which a willing and capable provider and a willing and capable recipient, knowing of any non-conformity, would lawfully have agreed (article III.–3:513). Or the monetary value of an enrichment is defined as the sum of money which a provider and a recipient with a real intention of reaching an agreement would lawfully have agreed as its price (article VII.–5:103). Bargaining outcomes depend on the distribution of bargaining power. In our setting of price reduction, the actual bargaining power could be reconstructed from the contract. In fact, at the contracting stage, parties have agreed on the level of performance xo leading to a social surplus of W(xo) = A(xo) + B(xo). At the agreed price, the creditor’s fraction of the surplus under due performance amounts to

α=

o

o

B(x ) – p _________ o

W(x )

Urs Schweizer

154

If, at the performing stage, the debtor is performing below his contractual obligation, x , xo, the reduced price could alternatively be reconstructed from the hypothetical bargaining situation, based on the same bargaining power. The corresponding equation is B(x) – p(x) = α · W(x). The debtor’s payoff would amount to the residual (1 – α) · W(x). Since his payoff is proportional to the social surplus, the debtor would face efficient performance incentives quite generally in spite of the fact that the creditor falls short of full compensation. It seems preferable to construct price reduction by reference to the hypothetical bargaining situation. In terms of efficient performance incentives, this solution would outperform the one provided by the DCFR.

4. Over-reliance under expectation damages If contractual obligations are at their efficient level and if damages are granted such that the creditor fully recovers her loss caused by the debtor’s deviation from his (efficient) obligation then the debtor is facing efficient incentives quite generally (see Section 2). Therefore, if expectation damages are claimed to generate over-reliance,4 such distortion must be due to the inefficiency of the debtor’s obligation. To restore efficient incentives, impracticability defences may be of use.5 The DCFR offers the following provision. A debtor’s non-performance of an obligation is excused if it is due to an impediment beyond the debtor’s control and if the debtor could not reasonably be expected to have avoided or overcome the impediment or its consequences (article III. – 3:104). In the following, the method of section 1 is used to examine investment incentives under expectation damages alone and in combination with performance excuses. For simplicity, the analysis concentrates on a simple numerical example whose properties, however, extend far beyond.6 The debtor’s performance decision is assumed to be binary, either he performs or he does not. At the contracting stage, his costs of performance c are uncertain, being uniformly distributed in the interval [0,2] with density f(c) = 1/2. The creditor chooses the level of reliances. At expenditures y3 y2 K(y) = ___ + __ , 12 8 she can generate any benefit y from performance in the interval [0,2]. Ex post, it is efficient to perform if and only if the benefit exceeds the costs, c # y. The expected social surplus amounts to y2 ___ y3 1 · ∫ y(y – c) · dc – K(y) = __ W(y) = __ – o 2 8 12 and attains its maximum at reliances y* = 1. If the debtor fails to perform he owes expectation damages to the creditor such that she is exactly compensated for her loss due to non-performance. As a consequence, the debtor has efficient performance incentives. 4 5 6

Shavell (1980) was first to derive the over-reliance result formally. See Sykes (1990) or Wagner (1995) among others. For a general version of the analysis, the reader is referred to Schweizer (2008c).

Remedies for Non-Performance

155

To examine the creditor’s reliance incentives, the method of section 1 must be adapted accordingly. In particular, it now is the shape of the debtor’s payoff U(y) = W(y) – (y – p – K(y)) as a function of the creditor’s reliances that matters. Since the derivative of this payoff function is negative, U'(y) = (y – 2) / 2 , 0, it follows that the creditor has excessive incentives if the debtor is strictly liable for non-performance, well in line with Shavell’s over-reliance result. If, however, expectation damages are combined with performance excuses, efficient reliance incentives may possibly be restored. The exact condition under which the excuse is admitted turns out to matter. Authors from both sides of the Atlantic have argued in favour of admitting the defence as soon as performance fails to be efficient ex post.7 To examine reliance incentives under this interpretation of the doctrine, the shape of the debtor’s expected payoff as a function of reliances is at stake. If the price for performance as specified in the contract exceeds the actual benefit, y , p, but costs of performance do not exceed the price, c # p, then the debtor rather performs or pays expectation damages instead of invoking the defence. In fact, if costs of performance are low, c # y, the debtor performs whereas, if they are in the range y , c # p he declines to perform but pays expectation damages to the creditor instead. Only if costs are high, p , c, the debtor invokes the defence. Hence, for y , p, the debtor’s expected payoff amounts to

(

y2 1 · ∫ y(p – c) · dc + __ 1 · ∫ y(p – y) · dc = __ 1 · __ U(y) = __ – p · y + p2 o p 2 2 2 2

)

with derivative U’(y) = (y – p) / 2 , 0. If the price for performance as specified in the contract is lower than the benefit, p , y, then the seller performs if his costs are in the range up to the creditor’s benefit. If they exceed the benefit the debtor invokes the defence. Hence, for p , y, the debtor’s expected payoff amounts to

(

y2 1 · ∫ y(p – c) · dc = __ 1 · p · y – __ U(y) = __ o 2 2 2

)

1 · (p – y) , 0. with derivative U(y) = __ 2 It follows that, except at y = p, the derivative of the debtor’s payoff function is negative and, hence, the creditor’s reliance incentives are distorted upwards unless the price is equal to the benefit under efficient reliances, p = y* = 1. But even at price p = y* = 1, the creditor’s reliance incentives remain distorted upwards. In fact, the derivative of the creditor’s payoff function amounts to V'(y) = W'(y) – U'(y) = (– y2 + 3y – 2) / 4 and is positive in the range 1 , y , 2. It follows that a performance excuse based on the inefficiency of performance fails to restore efficient reliance incentives. For consistency reasons, Ackermann (2002) argues in favour of the defence to be admitted whenever the costs of performance exceed the price, p , c. In other words, the debtor’s duty is reduced to perform only if it remains profitable for him to do so. Under Ackermann’s interpretation, efficient reliance incentives would be restored as I now want to show. For benefits lower than the price, y , p, the debtor does not invoke the defence and, hence, the derivative of his payoff is U’(y) = (y – p) / 2 , 0 again. For benefits above the price, 7

Most recently, Posner (2009) and Köndgen (2008) have argued in favour of this interpretation.

Urs Schweizer

156

two cases must be distinguished. If costs of performance are lower than the price, the debtor performs. If, however, costs happen to be in the range p # c # y he threatens to invoke the defence in spite of the fact that performance would be efficient. In this range, parties are assumed to renegotiate with 0 , β , 1 denoting the fraction of the renegotiation surplus going to the debtor. It follows that, if p , y, then the debtor’s expected payoff amounts to β y p2 β 1 · ∫ p(p – c) · dc + __ U(y) = __ · ∫ p (y – c) · dc = __ + __ · (y – p)2 o 2 2 4 4 with derivative U'(y) = β · (y – p) / 2 . 0. Therefore, under Ackermann’s interpretation, the debtor’s expected payoff as a function of reliances attains its unique minimum at y = p. To generate efficient reliance incentives, the contract must specify a price equal to the creditor’s benefit under efficient reliances, p = 1 = y*. If it does then the creditor has efficient reliance incentives indeed.8 While the above analysis may provide guidance of how to interpret the impediment as an excuse of the debtor’s non-performance, a more precise foundation of the impracticability defence in the DCFR would still seem desirable.9 There exists yet another way out of the over-reliance dilemma. Parties may prevent overreliance by stipulating explicitly a specified sum to the creditor for non-performance in their contract. As it turns out, a sum amounting to “efficient” expectation damages, i. e. correct damages that were due if the creditor had invested efficiently (which she may claim even if she has not), provides efficient incentives for both parties.10 For this arrangement to work, it is important that the creditor remains entitled to the stipulated sum irrespective of her actual loss. The DCFR provides such entitlement in principle, but allows for reduction to a reasonable amount where the stipulated sum is grossly excessive in relation to the loss resulting from non-performance and the other circumstances (article III.–3:710). From the economic perspective, such dilution of the general principle seems difficult to justify as it may distort incentives for similar reasons as in the more familiar over-reliance result.11

5. Damages for losses of chances A doctor fails to diagnose a patient’s condition such that the correct treatment was postponed. Yet, even with timely diagnosis, the patient would have suffered from the same loss with positive probability. What quantum of damages (if any) should the doctor owe to the patient? 8 9

10 11

Notice, to ensure the creditor’s consent, up-front payments would probably be required. The DCFR contains further provisions dealing with excessive burdens for the debtor. A court may terminate the obligation if an exceptional change of circumstances occurred that made the obligation so onerous that it would be manifestly unjust to hold the debtor to the obligation (article III.–1:110). Moreover, specific performance cannot be enforced it performance would be unreasonably burdensome (article III.–3:302). It remains difficult, however, to reconcile the wording of the DCFR with the economic interpretation of the impracticability defence based on efficiency considerations. For details, see Cooter (1985). Schweizer (2008a) shows that it also distorts incentives if the creditor may claim expectation damages in excess of damages as stipulated in the contract.

Remedies for Non-Performance

157

The DCFR does not provide specific rules of how to award damages if uncertain causation and losses of chances are at stake. The drafters have missed the opportunity to provide guidance in a field which currently seems plagued by an array of rather disparate legal practice.12 Traditionally, the all-or-nothing approach has been the prevailing arrangement. In some cases, the patient must prove that the doctor’s deviation from his obligation has caused the loss. If the patient manages to do so she will be granted recovery of her full loss. In other cases, the doctor owes damages of full size unless he can prove that his deviation was not the cause of the patient’s loss. In the United States and some other countries, the preponderance standard applies. Full recovery is granted if it is more likely than not that the deviation has caused the loss. More recently, however, courts in Europe but also in the United States have started viewing the debtor’s deviation as having caused the loss of a chance and are awarding a quantum of damages below full recovery accordingly, for good reasons I think.13 Conceptually, the all-or-nothing approach if applied to cases involving uncertain causation and losses of chance seems to be flawed. All justifications I am aware of are based on the presumption that the loss is caused by exactly one entity, either by the debtor or by nature.14 Yet, as I have argued elsewhere,15 the interaction between debtor and nature may be of a more intricate nature such that the very question, which of the entities has caused the loss, need not make sense in general. The focus rather should be on the hypothetical situation of what would have happened if the debtor had met his obligation. But even modifying the all-or-nothing approach along such lines will fail to achieve the compensatory goal, let alone the incentive goal, of damages rules. In the present section, I am attempting to disentangle the issues at stake and to propose a general solution that would meet both the compensatory and the efficiency goal. For simplicity, the discussion is confined to a numerical example but the proposed solution extends far beyond.16 The debtor’s obligation consists of performing at level xo. But even if he had met his obligation, loss L would have occurred with positive probability, say, εo = 15 %. By actually performing below his obligation, x , xo, the probability for the same loss to occur has been increasing to, say, ε = 35 %. Yet, even if the debtor deviates, the loss would not have to occur for sure. Suppose, however, in the case at hand, the loss has actually occurred. This very fact may provide information, relevant for the solution of the case. Let α denote the probability, taking all available information into account, that the loss would have been avoided in the hypothetical situation. Under the all-or-nothing approach, some threshold τ, a parameter of legal policy, would be decisive whether full recovery is granted or not. Under the preponderance standard, e. g., the threshold would be 50 %. At 12 13

14 15 16

See Mäsch (2002) or Noah (2005). Most recently, in July 2008, Massachusetts’s highest court ruled that doctors can be held liable for negligence that reduces a patient’s chances of survival, even if the patient’s prospects for recovery was already less than 50 %. The court established a formula for juries to award damages proportionate to the reduced survival rate caused by the doctor’s negligence. Proportional recovery along similar lines has also been granted by other courts in the United States and in Europe before. Shavell (1985) makes this assumption explicitly. See Schweizer (2008b). For a more detailed discussion, the reader may consult Schweizer (2009).

Urs Schweizer

158

the end of the present section, it will be shown that the all-or-nothing approach misses the compensatory goal quite generally. In terms of the compensatory but also the incentive goal, there exists a more promising alternative, based on a legal principle widely accepted elsewhere. To make use of this principle, two events must be distinguished, as a thought experiment if not, for lack of observability, practically. In the first event, the same loss would have occurred even if the debtor had met his obligation whereas, in the second event, the loss would have been avoided. Notice, these events refer to the hypothetical situation where the debtor has met his obligation which, actually, he has not. In the first event, the legally correct solution would be to deny recovery, i. e. to award damages amounting to D = 0. In fact, in the first event, the debtor’s deviation obviously has not caused the loss. In the second event, however, the debtor’s deviation has caused the loss and, hence, legally correct damages would amount to D = L. Notice, if these two events happen to be observable, the proposed quantum of damages sticks with the all-or-nothing approach. If, for lack of observability, the two events cannot be distinguished the natural way out would be to still award legally correct damages but on average over what can be observed. Since the second event occurs with probability α, the proposed quantum of damages would amount to d = (1 – α) · 0 + α · L = α · L under the proposed scheme. At first glance, this rule may look as if damages where granted in proportion to some probability. Conceptually, however, the rule should rather be seen as taking averages over events for each of which the all-or-nothing approach is maintained. The remaining task consists of determining the probability α with which the loss would have been avoided in the hypothetical situation. Recall, the ex ante probability of the loss to occur under due performance was εo = 15 %. If this probability was derived from test series of truly identical cases then the situation is one of pure uncertainty in the following sense: the two (observable) facts that (i) the debtor has performed below his obligation and (ii) the loss has actually occurred do not allow updating beliefs. Hence, under pure uncertainty, the probability with which the loss would have been avoided in the hypothetical situation remains to be α = 1 – εo = 85 %. Average damages should be awarded in the quantum of 85 % of the full loss. The case of pure uncertainty seems particularly attractive as it requires relatively little information to implement average damages. Just the ex ante probability of the loss to occur must be known. Let me point out that pure uncertainty generates hypothetical windfall gains in the following sense. Suppose the debtor has performed below his obligation, the loss did actually not occur but would have occurred hypothetically. From the ex ante perspective, the creditor would enjoy such a hypothetical windfall gain with positive probability (1 – ε) · εo = 65 % · 15 % = 9,75 %. Since she could keep such windfall gains for free, her ex ante expected wealth position exceeds the one under due performance. Yet, what may look like over-compensation due to the damages measure is rather due to the presence of hypothetical windfall gains. Moreover, recall that over-compensation does not distort incentives provided that the debtor’s obligation is equal to the efficient performance level (see section 2).

Remedies for Non-Performance

159

If probabilities were derived from test series containing different types of cases uncertainty need no longer be pure in the above sense. Medical theory, e. g., may rule out that a patient is cured if the doctor has performed short of his obligation but would have remained sick if treated properly. To be sure, since a patient previously treated improperly is not the same patient any more, it may be difficult to verify such a theory empirically. Nonetheless, such theories may suggest ruling out hypothetical windfall gains. If hypothetical windfall gains can be ruled out then the observation that a loss has actually occurred requires updating of beliefs. To calculate the probability α in this case, as a thought experiment, four events must be distinguished. First, event e00 where the loss does occur neither in the actual nor in the hypothetical situation; second, event e01 where the creditor enjoys a hypothetical windfall gain; third, event e10 where the loss occurs under the debtor’s actual deviation but would have been avoided if he had met his obligation; and fourth, event e11 where the loss actually occurs but would also have occurred in the hypothetical situation. Let πij denote the ex ante probability of event eij. If windfall gains can be ruled out, then π01 = 0 and, hence, π00 + π01 = π00 = 1 – ε = 65 % must hold. Moreover, π01 + π11 = π11 = εo = 15 %. Finally, since π10 + π11 = ε = 35 % it follows that π10 = ε – εo = 35 % – 15 % = 20 %. Given that the loss has actually occurred, the conditional probability of being avoided in the hypothetical situation amounts to

ε – ε = ___ 20 % = 57 %. α = _____ ε o

35

Therefore, if hypothetical windfall gains can be ruled out average damages amount to 57 %17 of the loss only as compared to 85 % if the same probabilities reflect pure uncertainty. In either case, the creditor is at least as well off as if the debtor had met his obligation. The compensatory goal is met if legally correct damages on average over observable events are granted, no matter whether uncertainty is pure or not. As a consequence, the incentive goal is also met if the debtor’s obligation is fixed at the efficient level. To complete the picture, let me also briefly explore the compensatory properties of the all-or-nothing approach. In the case of pure uncertainty, the probability α = 1 – εo does not depend on the size of the debtor’s deviation. If α happens to exceed the threshold such that the creditor recovers her full loss, she enjoys over-compensation as compared to the situation where the debtor had met his obligation. If α is below the threshold, recovery is denied and the creditor is under-compensated for the deviation. In this case, the all-or-nothing approach definitely misses the compensatory goal of damages rules. If uncertainty fails to be pure such that hypothetical windfall gains can be ruled out then the creditor’s payoff as a function of the debtor’s level of performance is as follows. The probability of the loss being hypothetically avoided amounts to

α = α(x) =

ε_______ (x) – εo ε(x)

if the debtor performs at x below his obligation, x , xo. Since the probability ε(x) is decreasing in the level of performance, the probability α(x) must also be decreasing and, hence, there exists a unique performance level xτ at which the threshold is exactly met, α(xτ) = τ. For levels below xτ, the creditor fully recovers her loss. For levels above xτ, however, damages are denied. 17

Such damages would be equal to proportional damages as proposed by Shavell (1985) but, while Shavell refers to the probability of causation, I interpret them as averages of all-or-nothing damages.

Urs Schweizer

160

For such levels, the all-or-nothing approach misses the compensatory goal again. Figure 3 plots the creditor’s expected payoff as a function of the debtor’s level of performance.

Figure 3: Creditor’s payoff under the all-or-nothing approach The insights from section 1 allow, based on the shape of the creditor’s payoff as depicted in Figure 3, to derive the following incentive properties. If xτ , x* then the debtor has insufficient performance incentives whereas, if x* , xτ , he has performance incentives which are weakly excessive at least. Only if xτ = x* the debtor would be facing efficient incentives for sure. But this lucky outcome is quite unlikely. In fact, the threshold is a general policy variable such as, e. g., τ = 50 % under the preponderance standard. The critical performance level xτ, in contrast, depends on the parameter constellation and will heavily be case-dependent. Moreover, the critical performance level also depends on the debtor’s obligation. Therefore, in rare cases at most, the critical performance level will happen to be efficient. In all other cases, the all-or-nothing approach fails to meet the compensatory as well as the incentive goal if hypothetical windfall gains can be ruled out. In this sense, awarding correct legal damages seems to be the superior approach even if, for lack of observability, averages must be taken. Cases involving losses of chances, malpractice suits in particular, are currently obtaining much attention. The all-or-nothing approach seems misguided,18 from the incentive perspective for sure, but also in terms of compensatory goals of damages schemes. I have argued in favour of average damages. The law, so far, has provided little guidance. The DCFR, unfortunately, also remains silent and misses the opportunity to provide guidance in a field where it seems urgently needed.

18

Noah (2005) even refers to mathematical blunders in applying the loss-of-a-chance doctrine.

Remedies for Non-Performance

161

6. Reduction of loss The debtor is not liable for loss suffered by the creditor to the extent that the creditor could have reduced the loss by taking reasonable steps. The creditor is entitled to recover any expenses reasonably incurred in attempting to reduce the loss (article III.–3:705). In previous sctions, I have considered cases where the incentives of only one of the parties were at stake. In the section on over-reliance, the creditor’s reliance incentives were examined whereas in previous sections, the focus was on the debtor’s incentives. To evaluate the DCFR’s requirement of mitigating losses, the two parties’ incentives must be considered simultaneously. To this end, the model is extended as follows. Before potential compensation, the debtor’s and creditor’s payoff amount to A(x) and B(x, y), respectively, where x denotes the debtor’s performance decision and y denotes the creditor’s mitigation activity. The efficient decision profile (x*, y*) maximizes social surplus, W(x, y) = A(x) + B(x, y).19 To begin with, the requirement of mitigating losses is not taken into account. Suppose the contractual obligation of the debtor consists of performing at the efficient level, xo = x*. If he deviates, he owes the quantum D(x, y) = max[B(x*, y) – B(x, y), 0] of damages to the creditor. Notice, this quantum depends on the creditor’s mitigation activity. In the absence of the requirement to mitigate losses, incentives for the two parties are generated as follows. The creditor’s payoff including compensation amounts to V(x, y) = B(x, y) + D(x, y) $ B(x*, y) = V(x*, y) and, hence, she is fully compensated for deviations from the debtor’s efficient decision. The debtor in turn, by meeting his obligation, escapes liability. His payoff amounts to U(x*, y) = A(x*) = U(x*, y*), independent of the creditor’s mitigation activity. Since the two parties’ decision problems are interdependent, incentives are captured by the Nash equilibrium (xN, yN) of the game. By definition, the Nash strategies are mutually best responses. For the debtor, this means that U(xN, yN) $ U(xN, y) holds for any mitigation activity y of the creditor and, at the same time for the creditor, V(xN, yN) $ V(x, yN) holds for any performance decision x of the debtor. The legal regime generates efficient incentives for both parties if the Nash equilibrium is an efficient decision profile which it turns out to be. In fact, since U(xN, yN) $ U(x*, yN) = U(x*, y*) and V(xN, yN) $ V(xN, y*) $ V(x*, y*) it follows that the social surplus at the Nash equilibrium is at least as high as the social surplus at the efficient profile, W(xN, yN) $ W(x*, y*). Since the efficient profile maximizes social surplus, the surplus under the Nash equilibrium cannot be strictly higher and, hence, the two 19

In the fully general case, the debtor’s payoff would also depend on both parties’ decisions, A(x, y). In such a setting, a system of bilateral expectation damages as explored in Schweizer (2006) would generate efficient incentives for both parties.

162

Urs Schweizer

must be equal, W(xN, yN) = W(x*, y*). It follows that the Nash equilibrium is an efficient decision profile indeed and, hence, the legal regime of expectation damages generates efficient incentives for both parties, even in the absence of the requirement to mitigate losses. To ensure efficient incentives, the mitigation requirement seems superfluous. Yet, one role in terms of efficiency is left for the mitigation requirement because, off the equilibrium path, the creditor’s incentives may well be distorted. If the mitigation requirement is taken into account, the awarded quantum of damages D = D(x) no longer depends on the creditor’s actual mitigation activity but rather on the one which follows from the mitigation requirement. Including such damages, the creditor’s payoff amounts to V(x, y) = B(x, y) + D(x) and is equal, up to a term beyond the creditor’s influence, to the social surplus. It follows that the creditor’s best response is a socially best mitigation response, no matter whether the debtor has performed efficiently or not. Anticipating the creditor’s response, as it turns out, the debtor still has efficient incentives to perform.20 To sum up, efficient incentives would be generated even without requiring the creditor to mitigate losses. Nonetheless, mitigation requirements such as provided by the DCFR still serve a meaningful purpose as they generate efficient incentives on and off the equilibrium path.

7. Immediate termination vs. termination after an additional period of time A creditor may terminate if the debtor’s non-performance of a contractual obligation is fundamental (article III.–3:502). Moreover, a creditor may terminate in case of a delay in performance which is not in itself fundamental if the creditor gives a notice fixing an additional period of time of reasonable length for performance and the debtor does not perform within that period (article III.–3:503). Therefore, in cases of delayed performance, the DCFR attempts to fine-tune the remedy of termination according to whether the delay constitutes a fundamental non-performance or not. The DCFR defines a non-performance to be fundamental if it substantially deprives the creditor of what the creditor was entitled to expect under the contract … unless at the time of the conclusion of the contract the debtor did not foresee and could not reasonably be expected to have foreseen the result or if it is intentional or reckless and gives the creditor reason to believe that the debtor’s future performance cannot be relied on (article III. – 3:502). No doubt, this definition leaves discretion to the courts. In the following, I want to explore the performance incentives resulting from the above provisions of the DCFR. I consider the following sequence of decisions.21 At stage 1, the debtor decides whether to perform in full accordance with the terms regulating the obligation or to delay performance. If he duly performs the game ends and the net payoffs amount to uoand vofor the debtor and the creditor, respectively. The social surplus amounts to wo = uo + vo. If the debtor delays the game proceeds to stage 2. If the delay constitutes a fundamental non-performance, the creditor has the option to terminate and then to undertake a substitution transaction in which case the game ends and the two parties obtain net payoffs ϕs = us – ds and ψ s = vs + ds. These payoffs include the creditor’s damages claims ds for nonperformance. The social surplus amounts to ws = us + vs = ϕs + ψ s.

20 21

For details, see Schweizer (2005). The following discussion borrows from and extends the analysis of Schweizer (2007).

Remedies for Non-Performance

163

Whenever the creditor does not terminate but notifies instead the debtor to admit an additional period of time (or always if the delay does not constitute a fundamental non-performance) the game proceeds to stage 3. At stage 3, the debtor decides whether to duly perform though delayed in which case the game ends with net payoffs ϕd = ud – dd and ψ d = vd + dd for the debtor and the creditor, respectively. These payoffs include the creditor’s damages claims dd for delayed performance. The social surplus amounts to wd = ud + vd = ϕd + ψ d. If, at stage 3, the debtor still decides not to perform the creditor has the option to terminate and then to undertake a substitution transaction. In this case, the game ends with net payoffs ϕds = uds – dds and ψ ds = vds + dds. These payoffs include the creditor’s damages claims dds and the social surplus amounts to wds = uds + vds = ϕds + ψ ds. The following game tree summarizes the timing of decisions and the resulting payoffs.

delay 1: debtor

2: creditor

perform

[ ] uo vo

not perform

notify 3: debtor

terminate

[ ] ϕs ψs

[ ] ϕds ψ ds

perform

[ ] ϕd ψd

Two legal regimes F and NF are considered. In the version F, the game is as described above such that, at stage 2, the creditor has the option to terminate and to undertake a substitution transaction. In the version NF of the game, the creditor is denied the option to terminate at stage 2. The two versions F and NF of the game reflect the DCFR’s provision if the delay constitutes a fundamental non-performance (F) and a non-fundamental non-performance (NF), respectively. Both versions of the game have to be solved by backward induction. The two versions are then compared according to the social surplus of the nodes where the two games are predicted to end. A rule provides efficient incentives if the game ends in the node where the social surplus is at its maximum w* = max[wo, ws, wd, wds ]. If the debtor does not duly perform the creditor is granted damages in line with the DCFR.22 To begin with, suppose the compensatory goal is exactly met. Then the creditor’s payoffs are the same, no matter where the game ends, i. e.

ψ s = ψ d = ψ ds = vo must hold. In this case, the creditor would be indifferent whether he is given the option to terminate at stage 2 or not. Moreover, if she is given the option, she would not mind terminating if and only if termination is efficient. Under such behaviour, the debtor becomes residual claimant and, hence, he is facing efficient performance incentives. In other words, if the compensatory goal is exactly met then the distinction between fundamental and non-fundamental non-performance does not matter as both rules provide efficient performance incentives. 22

I am grateful to Thomas Ackermann who has suggested considering damages claims in combination with termination.

Urs Schweizer

164

The analysis becomes more intricate if, in some end nodes of the game, the creditor obtains a higher payoff than under due performance. As time has passed by, the creditor might be eager to terminate because some substitution transaction preferred by her may now be at her disposal. In such cases, she cannot claim damages but, nonetheless, may be better off as compared to the situation where the debtor had duly performed, i. e. ψ s = vs . vo and ψ ds = vds . vo may both hold. The creditor, however, is still assumed to be exactly compensated for delayed performance such that ψ d = vd + dd = vo remains valid. Notice, if it is efficient to perform at stage 1 then, as follows from our general principles laid down in section 2, the debtor has the incentive for actually performing at stage 1. In this case, efficient performance incentives emerge under both versions of the game. If, however, performance fails to be efficient at stage 1, several subcases must be distinguished. To reduce the number of subcases, let me further assume that, due to the creditor’s damages claims, the debtor is worse off under delayed performance as compared to delayed termination, i. e. ϕd = ud – dd , ϕds = uds. It follows from this assumption that the debtor would have the incentive still not to perform at stage 3 if stage 3 is reached in equilibrium. Suppose, first, that max[us, uds ] , uo holds. Then the debtor will perform in both versions of the game and the resulting surplus amounts to wo whether the creditor has the option or not to terminate at stage 2 and whether it is efficient or not to perform at stage 1. In this case, the two versions of the game are leading to the same outcome. On efficiency grounds, neither of them would be preferable. Suppose, second, that uo , max[us, uds ] and ϕs = vs , ψ ds = vds hold. Then the creditor does not exert the option to terminate at stage 2 and, hence, it does not matter whether she is given the option or not. In this case, too, neither version of the game would be preferable on efficiency grounds as both of them would lead to the same outcome. For the remaining parameter configurations, the following table summarizes the social surplus which results in equilibrium under the two versions of the game.

ϕd , ϕds = uds and vds , vs

configuration o

s

ds

NF

F

ds

s

comparison of surplus

w

wds . ws or wds , ws

wds

wo

wo , wds

uo , uds , us

wds

ws

wds , ws

uds , uo , us

wo

ws

wo , ws

(1)

u ,u ,u

w

(2)

us , uo , uds

(3) (4)

The proof is simple. For illustration, consider configuration (1). In the NF-version of the game, the debtor does not perform at stage 1 as his payoff uo under performance is lower than his payoff uds which is triggered if he does not perform at stage 1. In equilibrium, the social surplus amounts to wds as claimed in the above table. In the F-version of the game, since uo , us, the debtor prefers the creditor to terminate at stage 2 instead of performing at stage 1. The social surplus amounts to ws as claimed in the above table. If ws , wds then the NF-version is superior whereas if wds , ws then the F-version is superior. The equilibrium outcomes under configurations (2) to (4) can be derived similarly. In configuration (2), regime NF is superior whereas, in configurations (3) and (4), regime F is superior. To justify the DCFR’s provisions on termination from the economic perspective, a delay in performance would have to be classified as non-fundamental if the game NF outperforms

Remedies for Non-Performance

165

the game F whereas it should be classified as fundamental if the game F outperforms the game NF. Yet, it seems difficult to reconcile such a definition with the wording of the DCFR. As a final point, let me point out that, for the above analysis, it was implicitly assumed that the price for performance as specified in the contract does not depend on the regime NF or F in place. Rational parties, however, would react to a change in the regime. As a matter of fact, they might even be inclined to agree on a price sufficiently high such that damages awards make the creditor generally indifferent between all end nodes of the game (exact compensation), the advantage being that efficient performance incentives emerge no matter which regime is in place.23 I conclude from the above analysis that the DCFR’s attempt to fine-tune the rules of termination remains difficult to justify on purely economic grounds.

Conclusions The DCFR defines the general measure of damages in line with the difference hypothesis. The quantum awarded is in relation to the debtor’s contractual obligation such that the creditor is fully compensated for the debtor’s deviations from his obligation. If the obligation is fixed at its efficient level, the remedy generates efficient incentives even if the creditor is over-compensated. Under-compensation, however, distorts incentives downwards quite generally. Typically, the creditor has the burden of proof. Difficulties in verifying her claims in front of courts may be a source of systematic under-compensation. As a consequence, the creditor may have insufficient incentives to perform in spite of the compensatory goal underlying the general measure as provided by the DCFR. Price reduction as defined in the DCFR also leaves the creditor worse off than if the creditor had duly performed. As a consequence, even at an efficient obligation, the debtor may face insufficient performance incentives. Under-compensation would persist if, instead, price reduction were defined with respect to a hypothetical bargaining situation. Nonetheless, reference to a hypothetical bargaining situation – as the DCFR does if, on termination, the value of the benefit to be returned or the value of an enrichment is at stake – would seem preferable from the economic perspective as the creditor would obtain a fixed share of social surplus and, hence, would face efficient performance incentives at least. Due to an inefficient obligation, expectation damages may generate excessive reliance incentives. Performance excuses could serve as a countervailing force. If the impracticability defence were admitted based on non-profitability of performance, efficient incentives would be restored indeed. Yet, the DCFR’s requirement of an impediment beyond the debtor’s control does not seem specific enough to support the interpretation that is needed to restore efficient incentives. The DCFR also fails to be specific if the measure of damages concerns cases that involve uncertain causation and losses of chances. The traditional all-or-nothing approach fails in terms of both compensatory and incentive goals. As an alternative approach, the focus rather should be on the hypothetical situation of what would have happened if the debtor had met his obligation. If the outcome in the hypothetical situation remains uncertain, correct legal damages should still be awarded though, for lack of observability, on average over the ob-

23

For a more thorough discussion of this issue, the reader is referred to Stremitzer (2008).

166

Urs Schweizer

served event. The DCFR’s definition of the general measure of damages should be amended to cover such cases explicitly. The DCFR requires victims of breach to reduce losses. The Nash equilibrium of the corresponding game would be efficient even in the absence of this requirement. From this view, the requirement may appear superfluous. Yet, as it turns out, the requirement generates efficient incentives for the creditor even off the equilibrium path without distorting the debtor’s incentives along the equilibrium path. Finally, the DCFR attempts to fine-tune the rules of termination. In general, the creditor must allow for an additional period of time of reasonable length before she may invoke the remedy of termination. In other words, the creditor has to tolerate delayed performance. From the economic perspective, it remains unclear why this solution should be superior to the one that allows the creditor to terminate immediately if the debtor has failed to perform in full accordance with the terms regulating the obligation.

References Ackermann, T. (2002), “Die Nacherfüllungspflicht des Stückverkäufers,” Juristenzeitung 8, 378-385. Bar, C. von, Clive, E. and Schulte-Nölke, H. (eds.) (2008): Principles, Definitions and Model Rules of European Private Law – Draft Common Frame of Reference (DCFR). Sellier, european law publishers: Munich. Cooter, R. (1985), “Unity in tort, contract and property: the model of precaution,” California Law Review 73, 1-51. Grundmann, S. und A. Hoerning (2007), “Leistungsstörungsmodelle im Lichte der ökonomischen Theorie: nationales, europäisches und internationales Recht,” in: Eger, T. und Schäfer, H. B. (eds.): Ökonomische Analyse der europäischen Zivilrechtsentwicklung. Mohr Siebeck: Tübingen. Köndgen, J. (2008), “Die Entlastung des Schuldners wegen Unmöglichkeit der Leistung – Versuch einer Ehrenrettung des § 275 Abs. 2 BGB”, in: Eger, T. et al. (eds.): Internationalisierung des Rechts und seine ökonomische Analyse. Gabler: Wiesbaden. Mäsch, G. (2004): Chance und Schaden. Mohr Siebeck: Tübingen. Noah, L. (2005), “An inventory of mathematical blunders in applying the loss-of-a-chance doctrine,” Review of Litigation 24, 369-393. Posner, E. A. (2009), “Fault in Contract Law,” Michigan Law Review 107, 1431-1444. Schweizer, U. (2005), “Law and economics of obligations,” International Review of Law and Economics 25, 209-228. Schweizer, U. (2006), “Cooperative investments induced by contract law,” RAND Journal of Economics 37, 2006, 134-145. Schweizer, U. (2007), “Leistungsstörungsmodelle im Europäischen und Internationalen Vergleich,” in: Eger, T. und Schäfer, H.-B. (eds.): Ökonomische Analyse der europäischen Zivilrechtsentwicklung. Mohr Siebeck: Tübingen. Schweizer, U. (2008a), “Vertragsstrafen,” in: Eger, T. et al. (eds.), Internationalisierung des Rechts und seine ökonomische Analyse, 2008, 263-274. Wiesbaden: Gabler Edition Wissenschaft. Schweizer (2009), “Legal damages for losses of chances,” International Review of Law and Economics, 29, 153-160.

Remedies for Non-Performance

167

Schweizer (2008c), “Breach remedies, performance excuses, and investment incentives,” mimeo. Shavell, S. (1980), “Damage measures for breach of contract,” Bell Journal of Economics 11, 466-490. Shavell, S. (1985): “Uncertainty over causation and the determination of civil liability,” The Journal of Law and Economics 28, 587-609. Stremitzer, A. (2008): “Opportunistic Termination,” available at SSRN: ssrn.com/abstract=989557 Sykes, A. O. (1990), “The Doctrine of Commercial Impracticability in a Second-Best World,” Journal of Legal Studies 19, 43-94. Wagner, G. (1995), “In Defence of the Impossibility Defence”, Loyola University of Chicago Law Journal 27, 55-95.

4. Termination of Contracts

Termination and Compensation in Long-Term Distribution Contracts: An Economic Perspective of EU Law Fernando Gomez* Abstract** Termination and compensation after termination are two of the most relevant – albeit not the only ones – dimensions of the long-term contracts through which distribution chains are formed and structured. The EC rules that establish a regime of compensation after termination, in particular, the Commercial Agents Directive, have served as the inspiration for the envisaged rules in the DCFR of European Private Law. Those rules are analysed from an economic perspective, highlighting the importance of the open-ended nature of a relationship, termination as a disciplining mechanism against non-verifiable breach of distribution contracts, and the problem of specific investments. The contrast of the existing EC and the proposed DCFR rules with the theoretical findings of the Law and Economics literature dealing with the matter does not offer a promising view of their likely consequences for the contractual behaviour of the parties.

1. Introduction 1.1. The problematic nature of the matter A substantial part, if not most, goods and services in a modern economy reach consumers through distribution chains with a varying number of links, and a diverse formal legal shape of the contracts within the chain (franchise, selective distribution, etc.). There are many interesting contractual issues concerning those relationships, both from an economics1, and from a legal perspective. From the point of view of the legal intervention in such kinds of contractual relationships, no other question is as relevant as the issue of termination – and, eventually, compensation following it – of the relationship.

* **

1

Universitat Pompeu Fabra, Barcelona, Spain I am grateful to Stefan Grundmann, Karl Riesenhuber, Jules Stuyck, to two anonymous referees, and to participants at the COPECL workshop in Barcelona for comments on an earlier version of the paper, to the Spanish Ministry of Innovation and Science for financial support, under grants SEJ2007-60503 and SEJ2007-10041, and to Marian Gili for excellent research assistantship. The usual disclaimers apply. See, for instance, Roger D. Blair, and Francine Lafontaine, 2005, on franchising.

172

Fernando Gomez

I will use the term termination throughout the paper to mean the end of the contractual relationship due to the unilateral decision of one of the parties2, most typically, as we will see, the principal or manufacturer. When I use the term compensation after termination, I take it broadly, as to be equivalent to any kind of financial payment or other sort of monetary remedy between the parties conditioning, resulting, or otherwise related to the termination of the contract. Termination, obviously, provides the starting point or the background against which the existence and level of compensation has to be assessed. Reciprocally, termination cannot be properly understood irrespective of compensation. The imposition of the duty to compensate fixed at a certain – high – level can deter, totally or partially, one party from terminating the contract. Similarly, the absence of any duty to compensate, or compensation in an insufficient amount, can promote or induce termination in a wide variety of circumstances, not all of them desirable from the perspective of the law-maker, or from a social welfare perspective. Typically, compensation payable by one contracting party to the other is one of the key remedies against breach of contractual obligations, and provides one of the core instruments to create incentives for the decision to breach or not to breach those duties, as well as to regulate a variety of contractual behaviours and dimensions (specific investments post-contract, allocation of the risk of losses or future contingencies, and so on). Breach of contract appears thus, almost inextricably linked with the issue of whether compensation should be provided. Even in a broader sense, the legal decisions concerning compensation after termination are some of the most relevant mechanisms affecting many – if not most – dimensions and elements of the contract, and so their function, role, and consequences cannot be understood as a separate matter from the rest of the rules in place. For instance, it can be shown that the effort and investment of the franchisee depend crucially on the expected duration, probabilities of termination, and the consequences thereof 3. In fact, I do not take the ex post effects of the compensation mechanisms, if any, to be the most relevant issue to be analysed theoretically, nor the most important consequences in economic or practical terms. Bluntly said, money transfers among businesses are of little importance in themselves from the point of view of governing contractual relationships. It is the implications of those payments for the previous behaviour of the contracting parties that make the interesting and difficult – both theoretically and practically – part. Therefore, in what follows I will assume that the contracting parties behave prospectively, and that they know the rules dealing with compensation after termination, and that those rules contribute to shape the incentives to undertake all sorts of contractual behaviour: Drafting contract terms, making investments that increase – for the investing party or for the other contractual partner – the value of performance if it occurs, complying with, or breaching contractual obligations and duties. In all these, as well, I will assume, unless otherwise expressly relaxed, that the contracting parties are firms4, and that they are risk-neutral, that is, that they care only about the expected value of alternative courses of behavior, and not about the intrinsic

2

3 4

This use is consistent with the Draft Common Frame of Reference (henceforward, DCFR) in several provisions: art. III.–3:501, art. IV. D.–6:101-105, and more importantly, art. IV. E.–2:302-2:304. It must be noted, though, that the definition in III.–3:501 is entirely tautological. See, Roger D. Blair, and Francine Lafontaine, 2005, p. 266. This assumption will probably make my previous one of prospective and – at least legally – informed behavior of the contracting parties more palatable than if I was dealing with consumers.

Termination and Compensation in Long-Term Distribution Contracts

173

riskiness of each of them5. This assumption of risk neutrality obviously does not correspond with many observations of risk-averse behavior in firms, but I think it is tenable enough as the general framework, given that we are dealing here with general rules governing contracts, and not various particular contractual environments or economic sectors. I will, nevertheless, consider some consequences of risk-aversion in section 3.3.3 to fully understand the effects of a kind of compensation schemes for termination of a long-term distribution contract.

1.2. Crucial issues for understanding compensation after termination What I take to be the core of the matter in the legal treatment of long-term contracts in distribution chains, is how legal rules concerning termination and compensation after termination affect relevant conducts of the parties to a long-term distribution contract. In connection with these, what are the dimensions that touch upon those incentive effects more closely, and thus should shape a consequentialist analysis of such legal treatment? To anticipate, in a somewhat cursory fashion, what will be developed in more detail in section 3, a brief list could be as follows: First, that the open-ended nature of a relationship – that is, that termination can occur – is not indifferent for the incentives to cooperate with the contracting partner in achieving full exploitation of the contractual opportunities. Second, that termination can be a powerful tool to discipline, subtle and not easily verifiable before a Court or other external adjudicator, instances of breach of contractual duties by distributors. This disciplining or policing aptitude, however, may – at least theoretically – also be used by the other party to expropriate the distributor of the value resulting from relation-specific investments. Third, the fact that the protection of specific investments and reliance on implicit promises against the risk of imposed renegotiation or expropriation constitutes a possible rationale for imposing compensation, under some conditions. All those issues just enumerated provide, in my view, the necessary frame of reference to understand the most important effects of legal rules affecting on long-term contracts in a distribution chain, and eventually, to design some broad legal principles that can serve as building blocks for a more detailed and comprehensive legal regime.

2. EC Law basis for rules on long-term distribution contracts: the Commercial Agents Directive and the proposed rules in the DCFR It may seem strange to start a section dealing with EC Law rules touching upon the matter with a disclaimer of suspected relevance or, at least, suspected generality of the existing EC Law rules on compensation following termination. I think it is worth mentioning in the preliminaries of the summary of EC Law in the matter, the haphazard and problematic basis in the Acquis communautaire of the compensation for termination issue, at least from a general perspective. I am not referring to scarcity or lack of importance of the existing rules as such, which in fact is not the case. There are EC rules, and the rules establish a recognizable regime of 5

This assumption refers only to the parties themselves, and it does not imply that in choosing among different options about how to regulate contract behavior, the law-maker may prefer the rule creating less uncertainty in its interpretation by Courts, for instance.

174

Fernando Gomez

compensation after termination in their own scope of application. Artt. 17-19 of Directive 86/653/EEC, on self-employed commercial agents, address specifically, purposefully, and in detail monetary payments to the agent following termination of the agency relationship. In turn, art. 3 of Commission Regulation 1400/2002, on the application of Article 81 (3) of the Treaty to categories of vertical agreements and concerted practices in the motor vehicle sector, refer laterally to the issue of compensation after termination in connection with the duration of notice periods for termination of agreements of indefinite term. Art. 5 (b) of Commission Regulation 2790/1999, on the application of Article 81 (3) of the Treaty to categories of vertical agreements and concerted practices (block exemption regulation), also deals with non-compete obligations after the termination of the agreement, setting specific requirements for those covenants not to compete in order for the entire agreement to benefit of the block exemption. All these may not seem, at first blush, not so scant or inadequate in terms of an EC Law set of rules on which to build an analysis of compensation after termination and related issues. But there are powerful reasons to the contrary, I believe. On the one hand, the commercial agents Directive can be considered, to some extent, tainted as a general basis (not as a benchmark, however!) for a contractual regime of compensation after termination in long-term contracts. The reason lies primarily in its self-declared objective of protecting commercial agents vis-à-vis their principals, a goal of social protection which has been affirmed by the ECJ in the Bellone6, Ingmar7, and Honyvem8 cases, and which, to a non insignificant extent, approximates some of the underlying principles in the Directive to those of employment protection and labour-related legislation. There seems to be, at least in my view, a fundamental problem in grounding a contractual set of rules for firm behaviour and B2B firm contracting on a mandatory legal regime trying to improve the lot of one of the parties as a question of social policy. If one – as I would – thinks that general, that is, not circumstance-specific, legal rules to determine and guide contractual behaviour in a variety of economic sectors and social circumstances, ideally should promote -absent external effects towards third parties- the maximization of the contractual surplus available to the parties, a set of rules that purposefully departs from this approach to pursue social protection of one of the parties in a contractual interaction does not seem to be an appealing or promising starting point for attempting a process of generalization. The preceding remark does not absolutely preclude that the conceptual solutions underlying the rules in the Commercial Agents Directive – disgorgement of benefits rendered to the terminating party by the contractual partner; compensation for some types of specific investments – could have been worth exploring in the broad context of long-term distribution contracts, but with due consideration to side effects, alternatives, and the implications of theoretical and empirical findings in the economic literature on these matters. The DCFR, however, has not been cautious in the use of the rules in the Commercial Agents Directive on these issues. They not only have formed the basis for the rules on the entire range of long-term distribution contracts – commercial agency, franchising, distributorship – contemplated in the DCFR – cfr. IV. E.–1:101 and following – but have seen their scope expanded from the initial terms of the Commercial Agents Directive – cfr. IV. E.–2:304, which dramatically expands the scope of the termination restrictions, as they stand in art. 16 of the Commercial Agents Directive. 6 7 8

Case C-215/97, Barbara Bellone v. Yokohama SpA [1998] ECR I-2191. Case C-381/98, Ingmar GB Ltd v. Eaton Leonard Technologies Inc [2000] ECR I-09305. Case C-465/04, Honyvem Informazioni Commerciali Srl v. Mariella De Zotti [2006].

Termination and Compensation in Long-Term Distribution Contracts

175

As for the block exemption (both general and for the car distribution sector) rules, they do not appear either as a sound basis for developing an analysis whose ultimate purpose is to structure an optimal set of rules in Contract Law. Those rules dealing with the antitrust aspects of certain practices and clauses in vertical relationships provide an inescapable regulatory environment for the behaviour of suppliers and distributors in the European market, but they are not, and are not meant to be, a set of optimal rules regarding contractual behaviour. The reason for the scepticism lies not so much in the possible difference in goals separating Competition Law from Contract Law – as I hold the promotion of social welfare in the economic sense to provide the basic objective to both of them – , but for the reason that the rules contained in the block exemption regulations are not easily generalizable as abstract and general rules of contract Law. EC competition Law rules in this area of vertical agreements are increasingly context-specific (as optimal regulatory policies and rules should in principle be), and not clear-cut, universal mandates or prohibitions. They require ad hoc estimates of the particular effects of a given practice, clause or behaviour on the actual levels of competition in the affected markets, a task that requires a great deal of empirical assessment carried on in the context of the specific circumstances and the environment requiring the regulatory decision. Contract Law is, almost by definition, albeit not universally so, essentially horizontal across markets and economic environments. Although empirical studies of the real-world consequences of the rules and policy choices underlying general Contract Law are also necessary, the rules have to be drafted and operate at a more elevated level of abstraction than those of Competition Law, not the least by the heavy cost of changing the rules contained in centenary Civil and Commercial Codes. Finally, other sources in the existing EC Contract Law, which may touch on issues of breach and liability, although in diverse settings (Consumer Credit Directive, Sales Directive, Package Travel Directive) seem to me of limited use in the legal treatment of long-term distribution contracts, and in particular regarding both the issues of unverifiable breach of the obligations of a relational contract, and the treatment of specific investments, given that those European rules typically handle transactions in which the level of non-salvageable specific investments is typically low, and in which the specification of contractual duties and their breach pose no specially onerous burden for parties and Courts in terms of verifiability. In fact, the DCFR seems to have disregarded them completely in this respect. Undoubtedly, it is the 1986 Commercial Agents Directive9 the one that deals with contractual scenario that falls within the area of long-term contracts used for making goods and services available in different stages of a distribution chain and in different geographic markets. The Directive, however, applies only were a relationship between a principal and a commercial agent is at stake. A commercial agent is defined in the Directive as a selfemployed intermediary who has continuing authority to negotiate the sale or the purchase of goods on behalf of another person, hereinafter called the ‘principal’, or to negotiate and conclude such transactions on behalf of and in the name of that principal10. The DCFR contains a similar, if somewhat more general, notion in IV. E.–3:101.

19

10

Council Directive 86/653/EEC of 18 December 1986 in the coordination of the law of the Member States relating to self-employed commercial agents (OJ L382, 31/12/1986). Art. 1 (2) and 2.1 of the Directive clarify further its scope, by excluding from its reach: persons who, in their capacity as officers, are empowered to enter into commitments binding on a company or association; partners who are lawfully authorized to enter into commitments binding on his partners;

176

Fernando Gomez

This notion leaves outside the scope of the rules of the Directive, including those on termination and compensation, the bulk of contractual arrangements through which distribution chains are structured: distribution (exclusive, selective, and others) agreements, supply agreements, franchising, and so on. As I have already mentioned, however, the DCFR, to a large extent, makes this set of rules the model of all distribution contracts in what relates to termination and compensation. The 1986 Directive contains two main rules concerning termination11. First, art. 15 contains the principle of termination by notice of a contract of unlimited duration, and sets some minimum notice periods depending on the past length of the relationship. IV. E.–2:302 of the DCFR substantially generalizes this regime to whole set of contemplated distribution contracts. Second, art. 16 exempts from the mandatory – for the principal – requirements of art. 15 those cases in which there was failure to perform obligations – and not only in cases of fundamental breach – or where there are exceptional circumstances. IV. E.–2:304 DCFR introduces a more restrictive regime by forbidding any clause that allows the parties to terminate – without the notice periods foreseen in IV. E.–2:302, it should be understood – unless there is fundamental breach of contract by the other party – typically, the agent, the franchisee or the distributor. Artt. 17 thru 19 of the Commercial Agents Directive deal with remedial instruments following termination12 – as foreseen in art. 13 thru 16 – and oblige Member States to establish systems ensuring that a commercial agent, after termination of the agency contract,

11

12

a receiver, a receiver and manager, a liquidator or a trustee in bankruptcy; agents whose activities are unpaid; agents when they operate on commodity exchanges or in the commodity market. Article 15 1. Where an agency contract is concluded for an indefinite period either party may terminate it by notice. 2. The period of notice shall be one month for the first year of the contract, two months for the second year commenced, and three months for the third year commenced and subsequent years. The parties may not agree on shorter periods of notice. 3. Member States may fix the period of notice at four months for the fourth year of the contract, five months for the fifth year and six months for the sixth and subsequent years. They may decide that the parties may not agree to shorter periods. 4. If the parties agree on longer periods than those laid down in paragraphs 2 and 3, the period of notice to be observed by the principal must not be shorter than that to be observed by the commercial agent. 5. Unless otherwise agreed by the parties, the end of the period of notice must coincide with the end of a calendar month. 6. The provision of this Article shall apply to an agency contract for a fixed period where it is converted under Article 14 into an agency contract for an indefinite period, subject to the proviso that the earlier fixed period must be taken into account in the calculation of the period of notice. Article 16 Nothing in this Directive shall affect the application of the law of the Member States where the latter provides for the immediate termination of the agency contract: (a) because of the failure of one party to carry out all or part of his obligations; (b) where exceptional circumstances arise. On these remedies after termination of the commercial agency contract, see Joanna Goyder, 2005, p. 186. On the provisions of the different member states implementing the Directive, see Geert Bogaert, and Ulrich Lohmann, 1993.

Termination and Compensation in Long-Term Distribution Contracts

177

is either indemnified according to the rules broadly set out in art. 17.2, or compensated for losses or damages incurred in accordance with art. 17.313. The entitlement to indemnity or compensation shall also arise if the contract ends due to the commercial agent’s death14. The entitlement does not arise if termination was produced by the agent – unless that termination was justified by the behavior of the principal15, or by personal circumstances that reasonably prevent the agent from continuing the activity –, if termination by the principal was a legally justified response to the agent’s breach of contractual duties, or if the contract was assigned by the agent with the consent of the principal. Art. 17.2 provides that the agent is entitled to an indemnity if and to the extent that two conditions are met: (i) the agent has brought the principal new customers or has significantly increased the volume of business with existing customers and the principal continues to derive substantial benefits from the business with such customers; (ii) to pay the indemnity is equitable having regard to all the circumstances and, in particular, the commission lost by the commercial agent on the business transacted with such customers. This indemnity cannot exceed, in quantitative terms, the agent’s average annual remuneration over the preceding five years – or the entire length of the relationship if it was less than five years16. Art. 17.3 stipulates that the commercial agent is entitled to compensation for the damages suffered as a result of the termination of the relationship with the principal17. Such damages are considered to have occurred particularly in two set of circumstances: (i) when the agent has been deprived of the commission which proper performance of the agency contract would have procured, while providing the principal with substantial benefits linked to the activities of the agent; (ii) the agent was not able to amortize the costs and expenses incurred for the performance of the contract on the advice of the principal. The DCFR has been loosely inspired by this complex regime in setting several provisions, to the effect of simplifying – the compensation for damages option disappears, except to the extent it goes into damages for termination for inadequate notice – and clarifying the existing rules, and namely the following articles: i) IV. E.–2:303, probably inspired on the general principle in art. 17.2.c) of the Commercial Agents Directive, on damages for termination for inadequate notice, trying to ensure, essentially, that the agent obtains the benefit he or she would have enjoyed, had the notice period been complied with, and using the average benefit of past three years as the benchmark to assess that benefit; 13

14

15

16

17

The Commission has published in 1996 a report on the application of art. 17 in the internal laws of the Member States, and on the choices made by them between indemnity and compensation. The fact that compensation (including both indemnity and compensation in the strict sense used in the Directive) is to be paid in cases in which there is no influence on the behavior of the principal (like death, illness or retirement of the agent, or even bankruptcy of the principal) makes it clear that it does not pursue – at least exclusively – incentive goals. Again, a rationale of social protection of commercial agents as deserving a treatment not too dissimilar from employees, comes readily to mind. In case of breach of contract by the principal, moreover, the indemnity received according to art. 17.2 does not preclude the agent from seeking and obtaining damages for contract breach. The indemnity scheme adopted by the Directive has been directly inspired – copied from may be closer to the truth – by the system foreseen in art. 89 b) of the German Commercial Code (Handelsgesetzbuch), for the Handelsvertreter. On this provision, see Klaus Hopt, 2006, p. 393. The compensation scheme adopted in the Directive as an alternative to the indemnity scheme is based on the indemnité du prejudice subi of French Law.

178

Fernando Gomez

ii) IV. E.–2:305, which contemplates an indemnity for goodwill broadly consistent with art. 17.2. a) of the Commercial Agents Directive. It must be noted that this indemnity is not mandatory – except for commercial agency, under the complex regime proposed in IV. E.–3:312, in turn inspired by art. 17.2.b) of the Directive – but it would, unless otherwise agreed by the parties, be applicable also in favor of the party who had breached the contract, even fundamentally, something that the Commercial Agents Directive explicitly excluded in art. 18 (a).

3. The economic considerations behind termination and compensation Distribution chains are formed and structured in any developed economy through contracts that are not merely spot transactions, but that have a non trivial time horizon. On the upstream side, suppliers of components or inputs and manufacturers, and on the downstream, manufacturers and distributors at different levels, franchisors and franchisees, they all engage in contractual relationships varying in degree of legal formalization, scope, intensity of the interrelation, and initial term of duration. The contractual links between the different agents in the distribution chain (manufacturer, wholesaler, distributor, retailer, and consumer) and apart from the retailer-consumer relationships (not uniformly, though) to be long-term in duration and nature18. Though contracts within the distribution chains vary widely in actual commercial practice, they tend to share some important common features – with some exceptions, however, particularly regarding franchising: selectivity (the search for a contracting party is lengthy and costly, and entails the use of screening mechanisms or probationary periods); informality (many contracts are not in writing, or when they are in writing, their content is extremely sketchy), which determines a substantial and often almost entirely unconstrained interpretation and renegotiation by the contracting parties; the use of framework agreements, typically developed, specified or executed by more detailed or specific contracts for particular projects, invoices, or items; open-ended character (many contracts have no specified term of duration, or when they do, they are typically renewable automatically). All these features fit quite nicely with the most relevant dimensions of the relational contract notion, which has attracted considerable attention and effort from the Economic literature on contracts and organizations. For our task of understanding the role and effects of legal rules on long-term contracts in distribution chains, and building a desirable set of rules, several lessons can be drawn from the stock of analytical and empirical findings accumulated by the economic analysis of long-term contracts. Moreover, I believe that these lessons are the more relevant ones in order to address the questions surrounding the legal treatment of contracting in distribution chains. I am not implying that economic analysis alone can provide definitive answers to the underlying issues confronting the legal regime of long-term distribution contracts. Legal rules and institutions in isolation do not capture the full picture of cooperation in relational contracts. In fact, it has to be acknowledged that the big questions concerning the actual effects of contract Law on long-term contracts remain to some extent unanswered by economic theory in a fully convincing and comprehensive way. But I am extremely skeptical concerning the ability of 18

The long-term character is sometimes (not infrequently, in the supplier-large retailer relationship) the result of short-term contracts that are renewed at the end of each contractual period, over a long time horizon.

Termination and Compensation in Long-Term Distribution Contracts

179

legal rules and institutions in isolation to capture the full picture of cooperation in relational contracts, and not the least in distribution chains. Disregarding the variables and dimensions that economic analysis has unearthed as relevant for achieving efficient outcomes in relational contracts will be attempted by legal analysis at own risk. Inducing efficient behavior by the firms integrated in distribution chains seems to me the preferred goal that legal rules can, at best, approximate in this context. To focus exclusively on the “traditionally legal“ dimensions might induce a design of rules and legal instruments for contracts in distribution chains that interfere the goal of inducing efficient behavior, with the undesired result of reducing, rather than increasing, the level of cooperation in the market, and, ultimately, social well-being.

3.1. The importance of the open-ended nature of the relationship In distribution chains, if (a huge if, but that concerns relatively little the reach of Contract Law within distribution chains) consumers have full information on the goods and services provided to them through the distribution channels, and there is competition among chains, what is required for an efficient allocation of resources is that the different agents acting as parties in the various contract links in the chain cooperate to produce the adequate combination of attributes of the product or service: quantity and quality of inputs, of pre- and post-sale service, of advertising and market information effort, and so on. The goal would then be to maximize cooperation – or the contract surplus, to use a more precise, or jargon-laden, term – among the parties in the contracts in the distribution chain (supplier-manufacturer, manufacturer-distributor, franchisor-franchisee, and so on)19. Many naïve analysts may think that the fact that contracts in distribution channels are long-term, so that parties have much to gain from continuous and iterated cooperation to achieve the maximum available surplus, can be a powerful motivator or incentive instrument for the good. But as the game-theoretic analysis of economic interaction shows, if there are incentives not to cooperate in isolated or one-shot interaction, simple repetition of the game is not enough. The economic idea of how a long-term relationship can – abstracting from external forces such as the Law, or social norms – can sustain cooperation among parties with divergent desires, goes beyond the mere repetition of a one-shot interaction. Open-endedness is conceptually necessary for self-interested agents to cooperate in these settings. Cooperation remains unachievable when the parties know their relationship will last for a fixed and known length of time. But if the same interaction can be repeated an infinite number of periods or, more realistically, that it can be played an unknown – by the parties – number of periods, so the interaction can end at any round, with a given probability which the parties (or in this example, the distributor) are unable to control. In such a scenario, it is possible to show that if the parties care enough about the future (i. e., the discount factor at which they

19

The assumptions of competitive structure and full consumer information guarantee then that consumers would be the ultimate beneficiaries of achieving the best available contractual surplus in each link within the distribution chain.

180

Fernando Gomez

discount pay-offs ahead in the future is not too large), reciprocity strategies20 in the repeated interaction can lead to cooperative outcomes21. This powerful result that reciprocity can achieve cooperation has been crucial to the relevance of long-term contracts in recent economic theory, because it implies that even in the absence of other external mechanisms to achieve cooperation in economic exchange, most notably, even in the absence of complete, formal, and legally enforceable contracts, in long-term relationships, the parties themselves can attain, under certain conditions, satisfactory cooperative outcomes. But this result stresses the importance, in relationships in which, for different reasons22, the parties cannot, or do not want to, rely on legal remedies to induce cooperative behaviour, and to discipline uncooperative behaviour, of the open-ended nature of the contract. The last-period curse, makes apparent that the common practice in longterm contracting of not fixing a pre-specified duration, is not the product of inadvertence or naïvete, quite the contrary, it is a conscious choice intended to provide incentives through reciprocation and reputation. But for this mechanism to work effectively, it is necessary that the parties can make use of the disciplining or reciprocation strategies that serve to check the private incentives to deviate from cooperative behaviour, basically, termination of the contract at any time23. In fact, legal systems seem to acknowledge this function, by allowing parties to enter into open ended, potentially of infinite duration, relationships, but allowing them to retain the right to terminate at any time24.

20

21 22 23

24

Although many economists use terms such as trust, or reputation, in fact they refer mostly to something that is best captured by the term reciprocity. For instance, an infinitely repeated interaction, such as the one described in the text, is equivalent to one-shot two-players interactions among all members in a given society, where each placer is perfectly informed about the reputation (the past history of cooperation or defection), of each and every other player. This shows that reputation is, in conceptual terms, not very different from reciprocation. The two reciprocity strategies most used in the game-theoretic literature are the following: (i) The grim trigger strategy, that states for each player: Choose a cooperative action until the other players deviates from cooperation; once the other player has defected, play defect forever after. (ii) The Tit-for-tat strategy, that states for each player: Choose a cooperative action if the other player cooperated in the previous period; defect if the other player defected the previous player. See, Joel Watson, 2002. Some of them will be explored in the next subsection. Termination is, unfortunately, a double-edged weapon, because it can serve, as asserted in the text, as an indispensable part of a reciprocation strategy to check uncooperative actions by the other party. Termination, or the threat to terminate, however, in certain circumstances can also serve as an instrument of uncooperative behaviour to take advantage of the other party to appropriate a bigger share of the contractual surplus. This makes the legal regime of termination and its consequences particularly difficult and, to some extent, even potentially contradictory. Art. 15 of the Commercial Agents Directive is a good example of this attitude.

Termination and Compensation in Long-Term Distribution Contracts

181

3.2. Incomplete contracts and termination as a disciplining mechanism against non-verifiable breach of long-term contracts If parties could write perfectly complete contracts specifying all relevant actions25 for the parties, the pareto-optimal actions, and Courts could costlessly enforce them, of course, the short or long nature of the relationship, and the use or not of reciprocity strategies, is of no importance: Contract Law can force cooperation through the use of the appropriate legal sanction. When individuals and firms can write a complete contract that determines contractual behaviour in the whole set of possible contingencies, and the basis of each contractual determination can be verified by the legal enforcer, typically, a Court of Law, the role of the legal system would be (at least in the eyes of an economist) is essentially mechanistic: the role of the Law would be substantially equivalent to blind enforcement of each and every term in the complete contract set out by the parties. The problem for us, in the real world, is that the complete contract notion is a conceptual construct. Contracts in the real world are never complete, given the extremely exacting requirements that pareto-optimality, completeness of coverage, and verifiability of information pose on complete contracts. A number of reasons have been identified by economic theorists supporting the proposition that contractual incompleteness (and thus the preponderance of relational contracts) is the rule and not the exception: The prevalence of non-verifiable information concerning many relevant contractual behaviors; the inevitability of imperfect specification of actions that will take place in the future, and in all possible states of the world in the future; difficulties in measuring and evaluating the cooperativeness of contractual behaviors, given the multidimensionality of complexity of many of them; advantage, under certain circumstances, of internal (including reciprocity-based) motivators for cooperation over external mechanisms such as formal and legally enforceable contracts. Lawyers tend to add other factors to the long list of reasons that make drafting complete contracts chimerical in our imperfect world, basically the unavoidable ambiguities and uncertainties brought about by the use of ordinary language, the one in which contracts are written, and bounded rationality on the part of the parties, that prevents them from perfectly anticipating all future contingencies, or fully exploiting the gains from trade between the parties in all future states of the world26.

3.2.1. Incompleteness in long-term distribution contracts and the role of termination

If incomplete contracts are the rule in any ordinary setting of economic interaction, when long-term contracts, in distribution chains or in other areas, are the norm, the force and the extent to which the factors leading to incompleteness is even stronger and more apparent. Long-term relationships have, by their very nature, an extended time-horizon. The number, influence, complexity, and difficulty and cost in anticipation, of contingencies that can, in one way or the other, have an effect on the contractual outcomes, increases substantially. How can a contract adequately foresee a number of the future events that can affect the kind and quality of inputs, the conditions of demand and production functions, the contingencies 25

26

Actions that, by definition, would be pareto-optimal: If contracting partners jointly agree on them ex ante it can only be because they maximize joint welfare of the parties, that is, they are paretooptimal. Benjamin Hermalin, Avery Katz, and Richard Craswell, 2007.

182

Fernando Gomez

affecting the firm structure of the contractual partners, and so on? No way, is the short but truthful answer. This is why economists, when approaching long-term relationships, have routinely assumed that contracts between the parties are incomplete, that is, they do not specify the complete set of optimal actions by the contractual parties: Many of the relevant actions cannot be foreseen and specified when the contract is signed, and it is in the course of the ongoing relationship that the parties will adopt those actions, based upon the set of incentives arising from factors (personal, institutional) different from the formal contract and the legal rules in Contract Law. Some use the term relational contract to refer to this sort of relationship. Of course, a relational contract of this kind cannot be enforced purely as written27 by a Court or an arbiter. This does not mean that no contractual clause or no contractual behavior by the parties is able to be legally enforced. Some instances of undisputable breach of the contract – departures from the cooperative equilibrium in the long-term relationship – can be detected by the contractual partner, and verified in front of a Court (or there is a substantial probability that this can happen), and thus can be deterred by the use of legal remedies, such as specific performance and damages. For example, if the franchisee in a fast-food franchise chain cheats on the quality of the raw materials to save costs, it may be the case that the fall in quality is detected by consumers and by the franchisor (let’s think of repeated cases of food poisoning), so that the latter can sue for damages and obtain compensation. If (a big if, however) damages correspond to the decrease in value of the franchise chain due to the breach of contract by the franchisee, deterrence of the undesirable and non-cooperative contractual behavior is achievable using the ordinary contract remedies. Unfortunately, this is not always the case with a wide variety of cases of non-cooperative behavior within the contractual relationship. Many sorts of contractual behavior pose unsurmountable measurement problems, particularly when those behaviors are multi-dimensional. Maybe, concerning most of these, the contracting partner can observe the departure from expected behavior, or even only a signal that is correlated with a higher likelihood of noncooperative behavior of the other party. The chances that these non-complying contractual behaviors can be shown, in a sufficiently convincing manner, to a Court or other external adjudicator, is very low. How can a manufacturer demonstrate, satisfying the level of proof legally required, that the distributor provides helpful and adequate advice to prospective buyers, or that the personnel of the distributor is cooperative with respect to the manufacturer’s representatives, or that marketing effort through leaflets or other written material is satisfactory? Perhaps only with cameras or other recording devices. May be it is possible only for cases of the most egregious violations of the expected contractual efforts, giving rise to consumer complaints. Moreover, the cost of detecting and collecting evidence of these instances of breach, increase with the size, territorial, and product scope of the distribution chain, thus becoming more serious the more integrated and expansive the relevant consumer markets. The amount and scope of unverifiable breaches of contract would tend to increase not only with the chances that the breaching party would scape undetected or not subject to legal contractual remedies, but also with the chances that consumers would not detect or punish

27

In many occasions nothing is written, and the contractual intention has to be inferred from the behavior of the parties, prior or posterior to the initiation of the relationship. Very commonly, only something very sketchy and obligationally incomplete has been agreed, leaving enormous space for future adjustment of terms through more or less formal renegotiation, or to future decisions, again, more or less formal, by one party.

Termination and Compensation in Long-Term Distribution Contracts

183

the defecting distributor (percentage of non-repeat customers, informational deficiencies of consumers, and so on). It is true that parties to a long-term distribution contract can resort to several alternatives that can serve as – imperfect – substitutes of what would be perfect Court enforcement of a remedy against a perfectly verifiable breach of contract28.First, the contract can use, instead of a non-verifiable dimension, some veriafiable proxy of the desired contractual behavior, and thus make use of legal enforcement of this proxy, given that Courts could use legal remedies for its breach. Second, the contract can contain clauses that tend to decrease the benefits, and to increase the costs, of behavior of the other party deviating from the cooperative pattern. In other words, they can try to introduce clauses in the contract that serve as mechanisms facilitating its self-enforcing character: exclusive territories awarded to distributors, so they increasingly become the residual claimants of efforts in the promotion and service associated to the product; imposing certain suppliers of inputs on the distributor, so that cheating in quality is more costly; imposing minimum purchase requirements, etc. Of particular importance among the self-enforcing mechanisms alternative to legally formal enforcement, and specially relevant for the legal treatment of termination and compensation after termination, is the use of future quasi-rents for the distributor linked to the continuation of the relationship29 : If the terms of the contract are adjusted so that the distributor (or franchisee, or supplier of inputs, if one thinks of upstream chains from the perspective of the manufacturer) expects to earn quasi-rents on their investments in the distribution chain, if and only if the existing relationship goes on, the party has a powerful motive to stay into the contract, and thus to avoid the kind of negative behavior that can trigger the end of the relationship. One important source of quasi-rents (returns of an investment that are higher than the return in an alternative use) is the existence of relation-specific investments, and a crucial link to the continuation of the contract is a credible threat by the other party to terminate the relationship, if the deviating behavior takes place (although, remember, this fact cannot be shown with sufficient evidentiary force before a Court or other external adjudicator). This makes the proposed rule in the DCFR, IV. E.–2:304, containing a restrictive regime of termination in distribution contracts forbidding any clause that allows the parties to terminate unless there is fundamental breach of contract by the other party, particularly unwelcome and inefficient. In order to make this instrument of relation specific quasi-rents to work as an incentive mechanism to achieve cooperation in dimensions outside of what can be verified by a Court, preserving the effectiveness of the termination weapon is crucial. Thus, restricting termination as a remedy, or subjecting termination to a duty to compensate the other party, dillute the force of the threat to terminate. The terminating party would be less inclined to use termination, because, ceteris paribus, this has become a more costly option due to the adjoining obligation to compensate, or indemnify, the party (who, remember, at least in this hypothesis, is in breach, however non-verifiable this may be)30. The result is likely to be that the manufacturer deploys other – less efficient, arguably, or else he would have used 28

29

30

See, George Mathewson, and Ralph Winter, 1985, p. 503; Benjamin Klein, and Kevin Murphy, 1988, p. 265; Benjamin Klein, 1995, p. 9; Cándido Paz-Ares, 1997, and 2003, p. 31. The pioneering analyses along this line are Paul Rubin, 1978; Benjamin Klein, 1980; and Benjamin Klein, and Keith Leffler, 1981. Benjamin Klein, 1995, p. 28-30, underlines the importance of termination-at-will for the effectiveness of self-enforcing mechanisms, and how legal constraints – mandatory severance payments or compensation, or good cause requirements – severely limit this option.

184

Fernando Gomez

them from the start – means to check the non-cooperative behavior of the distributors, or eventually replacing, in a Coasean fashion, contract – distribution contracts – by authority – vertical integration of distribution inside the firm. In sum, considerations concerning the essentially incomplete nature of long-term distribution contracts, and the primary relevance of disciplining unverifiable breach of contractual duties by the distributor, would lead then, to a basic position against compensation after termination such contracts.

3.2.2. Empirical evidence on the effects of legal restrictions on termination

The previous subsection has summarized the economic arguments advanced by economists and economically oriented legal scholars, in favor of preserving the conditions of unconstrained – by legal requirements concerning good cause, or by the imposition of ex post compensation – exercise of termination at will in long-term distribution contracts. This view seems to be supported by the available empirical evidence on the effects of legal rules restricting termination31. This evidence refers to franchising32, but there does not seem to be a powerful reason to doubt that the main findings would not be applicable to other contractual arrangements in distribution chains sharing similar issues of controlling opportunism by distributor (and, as we will see in a moment, also by manufacturer). The first and best-known piece is Brickley, Dark, and Weisbach. They hypothesize that laws restricting franchisor termination rights will lead to less franchising and will avoid exploitation and abuse of franchisees by those franchisors who try to own (and thus, not share the profit with the franchisee) those units that, through franchisee’s sales effort, or through franchisee’s market discovery, turn out to be particularly profitable33/34. Other more recent studies that look at data on franchise terminations do not support or reject the efficiency or opportunistic explanation of terminations. Beales and Muris’ results 31

32

33

34

See, James Brickley, Frederick Dark, and Michael Weisbach, 1991, p. 101; John Beales III, and Timothy Muris, 1995, p. 157; Darrell Williams, 1996; Francine Lafontaine, and Kathryn Shaw, 2005; Jonathan Klick, Bruce Kobayashi, and Larry Ribstein, 2006; James Brickley, Sanjog Misra, and Lawrence Van Horn, 2006, p. 173. The reason for this, lies in the fact that the studies are based on the US experience, where State legislation interfering with termination at will has concentrated on franchise contracts. Moreover, it seems that franchise still play a major role in US distribution compared with other areas. Both explanations for termination by the franchisor (or by the manufacturer, or principal, more generally), the benevolent – discipline on non-verifiable breach by the other party – and the sinister (expropriation of value from specific investments by the “weaker” party) are consistent with the brute factual observation that it is principals, and not the other parties, who typically terminate the relationship. In a Spanish survey carried out by a business daily newspaper (Expansión, 9 December 1996), in 88 % of the cases termination is decided by the principal, in 8 % by the distributor, and in 4 % it is a joint decision (I have taken these figures from Cándido Paz-Ares, 2003, p. 32). A casual look at litigated cases also strongly points in the direction of the principal or manufacturer being the party behind most disputed cases of termination. In view of this, it seems that we need some more elaborated empirical analysis to test which theoretical explanation is empirically corroborated by facts. They focus their empirical analysis on differences across industries and argue that the effect of termination laws on the rate of franchising will be most pronounced in industries with mostly non-

Termination and Compensation in Long-Term Distribution Contracts

185

let imply that this factor is watered down by the franchisor’s interest in keeping a reputation in order to attract additional good franchisees. According to Williams, the desire to transfer the unit – very often, by the franchisee herself – and to end those units that were underperforming – due to inadequate franchisee’s performance or to a disadvantaged location – would be the main driving factor explaining termination rates of franchising contracts. Klick, Kobayashi and Ribstein argue that the disciplining effect of termination on franchisee’s non-cooperative behaviour seems to outweigh, in their empirical relevance, the opportunities for franchisor abuse and expropriation of value that termination at will may allow. In view of Lafontaine and Shaw, franchisor’s opportunism should not be considered an important factor behind the rate of termination because, contrary to what it would be expected, the more established franchising chains do not end up showing more company ownership (franchisor’s ownership) over time. Finally, Brickley, Misra, and Van Horn conclude that both the level of the investments by the franchisee, and the size and the experience of the franchisor, tend to increase contract duration35, which provides indirect evidence that the threat posed by opportunistic and exploitative behaviour on the part of franchisors does not seem to be a particularly worrying problem in reality36, or at least, to be sufficiently secondary not to show in the data. There is also a striking feature arising from some results in the empirical studies just summarized. It is the observation that legislation restricting termination at will increases, rather than decreases, the number of terminations, that is, that when the law sets some conditions – financial compensation or showing good cause for termination – for terminating a franchise contract, franchisors terminate more often, and not less, as could intuitively be expected. The explanation that has been forwarded for this counterintuitive empirical finding runs along the following lines37: unconstrained termination at will induces franchisors to be more forgiving of minor – even if verifiable – instances of breach by the franchisee. To be forgiving at the beginning is not too costly for the franchisor, given that he always retains the ability to terminate without any restriction, financial or otherwise, as soon as he observes that his benevolence has not been repaid with cooperative behaviour on the part of the franchisee. On the contrary, if the decision to terminate is legally constrained, the franchisor will terminate at the first occasion in which he can legally and costlessly – in terms of severance or compensation payment to the franchisee – do so. He will not forgive a first minor breach if there is sufficient evidence to show that termination would be deemed an acceptable punishment of franchisee’s breach. This would lead, then, to more terminations, rather than less, following legislation that makes termination more difficult and/or costly for the franchisor. However the soundness of the specifics of this “second chance” theory, the underlying facts would seem to give additional support to the basic position that, on economic grounds, opposes compensation – or other restrictions to its exercise – after termination of long-term

35

36

37

repeat business, where there is less potential for self-enforcement, and in the industries they classify as particularly subject to non-repeat customers (restaurants, hotels, and auto rental agencies). Roger Blair, and Francine Lafontaine, 2005, p. 259-260, also find that larger franchisors tend to offer longer contracts on average, than smaller ones. It is true, however, that they also find a positive effect of legal restrictions on franchise termination (in the state where the franchisor has its headquarters) on contract duration clauses: James Brickley, Sanjog Misra, and Lawrence Van Horn, 2006, p. 185. They hypothesize that this effect is due to the increased bargaining power such legislation gives franchisees upon termination of the contract, thus reducing the value of short term contracts for the franchisor. See, John Beales III, and Timothy Muris, 1995, p. 169; Cándido Paz-Ares, 2003, p. 52.

186

Fernando Gomez

distribution contracts. If they actually increase termination levels by manufacturers, and unless compensation is sufficient to restore the fall in welfare of distributors due to the increased terminations (a doubtfully plausible assumption), those legal rules could be aptly characterized as self-defeating: They harm those they intended to protect and benefit. From this section, thus, there seems to be a prima facie powerful case, both on theoretical and on empirical grounds, in favour of scepticism towards rules that impose compensation – in favour of the distributor, in practice, as data shows – after termination of a long-term distribution contracts. Let’s see if there are other considerations that allow a more optimistic view.

3.3. The relevance of specific investments The economic theory of contracts is concerned with the ways to ensure that in a voluntary economic interaction that protracts into the future efficiency is achieved on two abstract dimensions: trade, that is, the actual exchange of goods and services, so that they end up in the hands of whoever values them most; and investment, that is, committing resources in a productive way to enhance the surplus of the transaction, either by decreasing the costs of production, or by increasing the value of the goods and services38. The conditions and mechanisms to simultaneously induce efficient trade and efficient investment provide the core of modern economic contract theory39. Contract investments, in the economic literature on incomplete contracting, are typically assumed to be, to a varying degree, relation-specific. An investment is completely relationspecific if the value of the investment is entirely lost if trade between buyer and seller does not occur, that is, outside that particular contractual relationship. The lower the relationspecificity of the investment, the higher the external value of the resources invested. At the extreme, a completely non-specific investment produces the same value inside or outside the relationship.

38

39

Good introductions to such topics, in Patrick Schmitz, 2001, p. 1; Benjamin Hermalin, Avery Katz, and Richard Craswell, 2007. The problem is far less trivial that may look like from the outside, at least for a lawyer (who may just simply advise the parties: Hire a good lawyer and sign a detailed contract), given that uncertainty in the future can affect the value of performance, or the cost of production, or both, thus making trade possibly inefficient, and that the levels of investment are typically assumed to be non-contractible.

Termination and Compensation in Long-Term Distribution Contracts

187

3.3.1. The hold-up problem

The trouble40 with relation-specific investments is that they are very vulnerable to hold-up by the other party in the relationship41. Let’s think of the following example: A supplier of components, S, in a contractual relationship with a manufacturer, M, is deciding whether to invest in modifying its plant to adapt it better to the particular production requirements of the manufacturer. Let’s assume that the cost of the investment is 50, and it would reduce the supplier’s production cost in 100, so the investment is efficient. The investment is completely relation-specific, which means that in serving another manufacturer the new plant is worth 0. If the contract terms between S and M can be renegotiated42, M has an incentive to hold S up by threatening a termination of the relationship if the agreed contract terms are not modified in its favour. For instance, M could force S to accept a reduction in overall price of 90. S would accept it, because it would be obtaining 10 over production cost, which is better than nothing. S would regret having made the investment, because it invested 50 to get a return of just 10 (M is “expropriating” the remaining 90 of the internal – to the relationship – value of the investment). Not surprisingly, the incentives to invest by S are significantly diluted by the likelihood of hold-up. In fact, as long as it cannot capture a fraction of the full value of the investment that allows cost recovery, it will not invest. If S has all the bargaining power in the renegotiation stage, it will efficiently invest, as it would reap the full value of the investment. In fact, as we will briefly see, economic theorists have tried to devise clever ways in which the investing party enjoys bargaining power when entering the renegotiation phase. With a completely non-specific investment, however, such a hold-up situation would not arise, because the threat point of S in the renegotiation stage would have gone up by 100, because the investment has full value also serving another manufacturer. For intermediate degrees of specificity, the problem exists, albeit in a somewhat attenuated form compared to the complete specificity scenario. If renegotiation can hardly be ruled out in a spot contract, the opportunities for renegotiation in a long-term contract, distribution or otherwise, are almost infinite. In fact, the number, complexity, and unanticipated nature of many contingencies, almost naturally force the parties to a long-term contract to renegotiate the contract terms in the face of new infor40

41

42

Specificity of investments may not be the result of technological or other constraints, but a conscious – strategic – choice by the contracting parties who, by increasing specificity, reduce the value of outside alternatives to the existing contract, and thus making the existing relationship more attractive in relative terms – at a price, however, because the value of the investment is lost when the relationship, for whatever reasons, comes to an end. This has been explored in the setting of cooperative investments (the following subsection provides a characterization of this kind of investment) as a solution to the shortcomings of monetary instruments to simultaneously reward the investor for making cooperative investment, and punishing him for not investing: See, Matthew Ellman, 2006, p. 234. A good starting reference for the hold-up problem is Benjamin Klein, 1998. The problem was first spotted by Oliver Williamson, 1985, and shown formally by Oliver Hart, and John Moore, 1988, p. 755. Contract renegotiation is a mixed blessing for contracting parties: it allows ex post efficient trade – the Coase Theorem guarantees that outcome if renegotiation faces no transaction costs – but it also allows the possibility of hold-up, and thus reduces the incentive to invest. On this issue, see Christine Jolls, 1997, p. 203; Kevin Davies, 2006, p. 487.

188

Fernando Gomez

mation and new circumstances almost continuously. With this flexibility and ample room for renegotiation the chances of hold-up for the party making specific investments would rise, and thus the disincentive to invest to the efficient level. And some clever proposals at the interpretive level, such as an interpretation canon favoring the party making relation-specific investments43, cannot solve the problem in a general way.

3.3.2. Selfish vs. cooperative specific investments

Among specific investments, the literature in economic theory dealing with incomplete contracts has distinguished two pure types of such investments. Selfish investments are the ones that benefit the party making the investment, and not the other party: If the buyer makes the investment, it just increases the value of performance for the buyer, without decreasing production costs for the seller; if the seller makes it, it just decreases its production costs, without increasing the value of performance for the buyer. Cooperative investments44 are those that confer direct benefits to the other contracting party, and not to the party making the investment: If the buyer makes the investment, it just decreases production costs for the seller, without increasing the value of performance for the buyer; if the seller makes it, it just increases the value of performance for the buyer, without decreasing production costs for the seller. There are also hybrid investments, if they benefit both contracting parties, the investor and his partner, although we will disregard this complication in what follows45. For selfish investments, the solutions explored in the economic literature have evolved in two directions. First, to design mechanisms in an incomplete contract that can achieve the efficient outcome, both in terms of trade and in terms of specific investment. Most contributions explore ingenious procedures in the bargaining conditions in the renegotiation phase of the contract, so that the party making the investment receives the full value of the investment. Some opt for placing external conditions on the renegotiation phase46, others for the use of options and appropriate strike prices for their exercise47. Other approaches rely in contracts determining intermediate quantities to trade, so after renegotiation it turns out that the investing party receives in some cases more and in some cases less than the marginal social value of the investment, an if the quantity is adequately chosen in the contract, both effects may cancel out and in the end induce efficient levels of investment. These approaches, however, have only limited applicability to the design and operation of legal rules in the contract setting48. 43

44

45

46 47 48

Scott Baker, and Kimberly Krawiec, 2006, building upon the idea of interpretive rules protective of one of the contracting parties developed by Omri Ben-Shahar, 2004, p. 389. The first treatment of cooperative investments is W. Bentley MacLeod, and James Malcomson, 1993, p. 811. The standard treatment of these investments, in Yeon-Koo Che, and Tai-Yeong Chung, 1999, p. 84; Yeon-Koo Che, and Donald Hausch, 1999, p. 125. A very interesting recent contribution on contract remedies and specific investments, showing that the dismal result – concerning expectation damages – of Yeon-Koo Che, and Tai-Yeong Chung, 1999, is but a extreme case, Alexander Stremitzer, 2008. In fact, few papers explore hybrid investments: Yeon-Koo Che, and Donald Hausch, 1999, p. 125; Ilya Segal, and Michael Whinston, 2000, p. 603. Tai-Yeong Chung, 1991, p. 1031. Georg Nöldeke, and Klaus Schmidt, 1995, p. 163. Aaron Edlin, and Stefan Reichelstein, 1996, p. 478.

Termination and Compensation in Long-Term Distribution Contracts

189

The second strand of the literature on selfish investments specifically deals with the legal remedies against breach, and with their impact on the investment decision by the parties (also on the decision to perform or to breach the contract, but let us leave that aside). The pioneering contribution here is by Shavell 49. He addresses two different scenarios. The first one is the one in which the investing party is the potential victim of breach by the other party. He shows that expectation damages induce excessive specific investment by the potential victim of breach. The reason for this effect of overreliance lies in the fact that expectation damages fully insures the investing party against the possibility of losing the return on the investment, more than it is optimal from the point of view of the joint welfare of the parties: even when there should be (and there is) no trade, that is, when the contract should not be performed, the investing party gets the full return from the investment. Reliance damages perform even worse than expectation damages, that is, they induce even more overinvestment. These results have been to an important degree confirmed by experimental tests of contracting behavior in a controlled laboratory setting50. The second scenario appears when the investing party is the one who can take the decision to breach or to perform the contract. In this case, expectation damages induce efficient investment: the breaching party is the residual claimant of the value of the investment (the reduction in cost of production, for instance), because the damage award he has to pay (the value of performance to the other party) does not depend on the level of investment. Reliance damages also perform worse than expectation damages, although in a different direction than in the first scenario. Given that reliance damages generally induce too little performance with respect to the efficient level, the investing party will get the return on the specific investment less than what is optimal, and therefore the incentives to invest would be too low51. Shavell’s analysis in the first scenario (the investing party is not the one making the breach-perform choice) was extended along two different lines. First, that post-breach renegotiation by the parties does not alter the inefficient investment incentive under both expectation and reliance damages, and also the ranking of the two remedies: reliance damages are less attractive than expectation damages in order to provide less inefficient investment incentives52. Second, that with an appropriate instrument, one can transform the first scenario into the second one, that is, that the investing party is the one that can take the decision to breach or to perform. The instrument is a large up-front payment from the buyer to the seller (assuming the seller is the party who can invest), which ensures that the buyer would never want to breach – he would get performance for a price close to zero, because the up-front payment is sunk when the performance decision arrives. Then, any breach would come eventually from the investing party, who has efficient breach and investment incentives under the expectation damages remedy53. It is true, though, that using up-front payments can have problems of its own (basically liquidity problems on the part of the prospective buyers, or the non-investing party more generally), but it clearly shows how the investment problem can be solved, for selfish reliance expenditures, by concentrating the decision to breach, and the decision to invest, into one and the same hand.

49 50 51 52 53

Steven Shavell, 1980, p. 466. Randolph Sloof, Edwin Leuven, Hessel Oosterbeek, and Joep Sonnemans, 2003, p. 205. Steven Shavell, 1980, p. 484. William Rogerson, 1984, p. 39. Aaron Edlin, 1996, p. 98.

190

Fernando Gomez

The question is more complex still for cooperative (i. e. enhancing the other party’s payoff) specific investment. It can be shown that if the parties cannot commit not to renegotiate the contract, there is no incomplete contract, however complex, that can induce efficient incentives54. In fact, the dismal result is that a contract is no better for the parties than no contract at all.55 For these cooperative investments the role of legal remedies against breach has also been explored, both absent renegotiation, and when ex post renegotiation is feasible, usually considering that the investing party is not the one that can take the breach-perform decision56. In the first case, with no renegotiation, expectation damages perform very poorly, because as a remedy for breach it induces zero cooperative investment. Reliance damages, in turn, do much better according to Che and Chung: although at the price of some distortion – in the direction of excessive breach – in the breach-perform decision, they provide much better incentives for specific investments, and overall improves contractual surplus over expectation damages. With efficient renegotiation ex post, expectation damages continue to perform as poorly as before, but reliance damages can now achieve efficient incentives, both to perform and to incur cooperative specific expenditures. There is also a defence of expectation damages in this setting, albeit not ordinary expectation damages, but bilateral expectation damages. This implies that the party who can breach can be subject to paying expectation damages to the other party, the investor; but the latter can also be liable in front of the former if the level of investment falls short of the level determined in the contract. If this is the case, bilateral expectation damages do also induce efficient trade and efficient cooperative investments57. For this result to hold it is necessary not only that the actual level of investment can be verified before the Court (a condition for reliance damages to operate as remedy against breach, to be sure) but also that the parties can fix in the contract the efficient level of cooperative investment, which is a much implausible – though not impossible – assumption. Finally, in a recent paper,58 it has been argued that the extremely poor – zero cooperative investment – efficiency performance of expectation damages (the preferred remedy for breach of contract in Common Law, and also heavily used in Civil Law jurisdictions in many contexts) is due to an implicit assumption that the contract does not contain any threshold of performance, and that the Court cannot imply one either. If, on the contrary, the contracting parties or the Courts are able to compare performance with a verifiable legally binding threshold (over the relevant dimension of performance), expectation damages generally induce positive – albeit suboptimal – levels of cooperative investments, and even under certain conditions (those of the so-called maximum quality or Cadillac contracts59) they can even provide incentives for efficient investments of a cooperative nature. This paper

54 55

56 57 58 59

Yeon-Koo Che, and Donald Hausch, 1999, p. 125; Eric Maskin, and John Moore, 1999, p. 39. Alexander Stremitzer, 2008 shows that if the contract can be conditioned (i. e., Courts are able to verify) on whether performance is above or below a given threshold – of quality, as paramount example – under certain conditions first-best can be attained. Yeon-Koo Che, and Tai-Yeong Chung, 1999, p. 84. Urs Schweizer, 2006. Alexander Stremitzer, 2008. Aaron Edlin, 1996.

Termination and Compensation in Long-Term Distribution Contracts

191

also favors, when renegotiation is possible60, the use of an optional61 remedial regime for the non-investing party, consisting of the choice between specific performance and termination – with restitution of payments made, if any – if performance of the investing party falls below the legally enforceable threshold. Above the threshold, only specific performance would be available for either party.

3.3.3. Are the rules of the Commercial Agents Directive – and the proposed rules in DCFR based on them – sound in economic terms?

What lessons can be drawn from the last two subsections on specific investments for the understanding and application of rules on compensation after termination, such as those in art. 17 of the Commercial Agents Directive, or IV. E.–2:305 or IV. E.–3:312 DCFR? The first and obvious one for those who have struggled with the previous pages is that the treatment of specific investments is very complex, and that the legal rules dealing with them have to take into account the economic intricacies that pervade the entire area, particularly when confronting investment in long-term contracting. No simple or all-encompassing rules can function properly across the whole range of factors and circumstances that have appeared in the preceding analysis. We would need a diverse set of rules in order to deal with the different factual scenarios which, in turn, have different implications in terms of the economic consequences of alternative legal instruments. The second, and also quite obvious one, is that the rules on termination and the compensation rules after termination do affect the incentives of the contracting parties to undertake relation-specific investments: If, for instance, the distributor knows he will receive – let’ assume, full – compensation for the specific investments – let’s assume, of selfish nature – in cases in which the contract ends, for reasons such as death, illness or old age, or principal’s insolvency, he will invest excessively, compared to what would be optimal. In fact, when termination cannot be linked to behavior of the principal or manufacturer that could be described as breach of contractual duties – broadly understood, including opportunistic expropriation of value – , compensation after termination cannot perform a desirable incentive role, because it cannot restrain breach of those duties by the principal, because there is none, whereas it induces excessive specific investment by the agent or distributor. This policy may be justified, perhaps, on grounds of social protection of a truly disadvantaged group62, but

60

61

62

It is well known that absent renegotiation specific performance is likely to produce inefficient trade and thus undesirable results: Steven Shavell, 2004. The use of optional regimes has not often been considered in the literature, with the exception of Ronen Avraham, and Zijhong Liu, 2006; and Ronen Avraham, and Zijhong Liu, 2008 . I strongly doubt, however, that social protection can provide a solid basis for a general policy of compensating distributors in such cases. Distributors vary widely in many respects (size, wealth, future prospects) that are relevant for social protection policy so as to allow a one-size-fits all approach to the matter. Moreover, given that there is a contractual relationship between the parties, imposing social protection on one party, at the cost of the other – and not, the general taxpayer – may well lead to a worsening of contract terms in other respects: An increase in duties or rights that is not efficient, in the sense of increasing the surplus of the interaction, will imply a readjustment of the terms to the detriment of the beneficiary, the distributor in this case. This is the well known argument of passing-on of legal costs: Richard Craswell, 1991, p. 361.

192

Fernando Gomez

certainly not on economic grounds, because nothing is to be gained in efficiency terms by mandating compensation in those circumstances63. Through the kind and level of compensation, if any, following termination of the contract, we can influence the behavior of the distributor with respect to reliance expenditures. The previous analysis on incentives to make specific investments under alternative damage rules is relevant for capturing the likely consequences of the options we may consider as policy alternatives for the compensation issue. This is clear in case of early termination of a distribution contract for a definite time period, which could be, naturally and easily, even undisputedly, be considered a breach of the distribution contract by the principal or the manufacturer64. It is probably also the case of termination by the “stronger” party in a contract of unlimited duration, given that in effect what we have here is a decision not to trade any more in the future, to borrow the terminology of incomplete contracting. It is true that termination of this sort can serve very different purposes, as we have already pointed out several times before, some of those being benign, even commendable, others being exploitative. The problem is that, due to the very nature of the long-term relationship in the distribution context, except in the most egregious cases (of previous breach by the distributor, on the one side, or of hold-up by the manufacturer65 on the other) Courts would be unable to tell one kind of termination from the other, so they are likely to end up being subject to the same legal solution in terms of compensation following the decision by the manufacturer to terminate. What results from art. 17 of the Commercial Agents Directive under the light of the economic theory of specific investment and damage rules is, to a large extent, the impression of a very confused picture. The confusion, true, has diminished in the relevant rules of the DCFR, which have simplified and systematized the initial ones in the Commercial Agents Directive. But as the current legal basis is the Directive , I will essentially refer to it, so as to show how it should not be used as the basis for a future-oriented regime in the DCFR or in a similar instrument. The two schemes contemplated in the Commercial Agents Directive, indemnity and compensation, make no mention of the distinction between selfish and cooperative investments, which is very natural (legal rules do not need to use terms of art of disciplines such as economics; moreover, the notions in themselves have emerged with all clarity in Economic 63

64

65

Later in the text I briefly examine whether risk-aversion by the agents and the preference for protection against undesirable events such as death or sickness can provide a rationale explaining the compensation rules that art. 17 Commercial Agents Directive envisions for such a set of contingencies. I am disregarding here, and in fact, in the entire treatment of compensation after termination, the possibility of termination by the distributor and possible compensation to the principal or manufacturer. This seems to be a much rarer occurrence, at least according to reported cases and the Spanish survey cited above, note 47. Moreover, although there well may be specific investments on the part of manufacturer and principal, and the opportunities for hold-up by the other party, legal systems seem to have addressed this scenario through general rules of contract Law, and not special compensation provisions on termination. Probably, overall it makes sense to handle them through the rules against duress and on contract modification. See, Oren Bar-Gill, and Omri Ben-Shahar, 2004, p. 391; Steven Shavell, 2007. And for the former we can resort to ordinary law of contract breach – art. 18 a) of the Commercial Agents Directive excludes this case from the compensation schemes of art. 17 –, and for the latter, to the law of duress or to good faith requirements cautiously implemented by Courts.

Termination and Compensation in Long-Term Distribution Contracts

193

theory only in the past fifteen years or so). The problem is that they point, confusingly, at the same time in similar and in opposite directions. They mention, in different manner, though, factors that point at cooperative aspects of the investment, and at selfish ones, to elements related to the features of expectation damages (loss of commission or remuneration), and to others related to reliance damages (unamorticized costs and expenses). Moreover, the indemnity scheme hints at the use, instead of a measure of harm (be it at the reliance or at the expectation measure) of some sort of disgorgement of benefits acquired by the principal as a result of the new customers – or the increased volume of existing customers – brought by the efforts of the agent66. To have compensation after termination based on restitution or disgorgement of gains to the principal, instead of on liability for harm caused is by no means a matter of indifference for the functioning of the compensation scheme. The alternative between gain-based and harm-based compensation is not exclusive to this setting: Violation of intellectual property rights, such as copyright or patent, violation of the right of publicity, provide also good examples of scenarios in which legal rules and Courts have to choose between both approaches to estimate the adequate remedy. The choice has also been analyzed from an economic perspective, and in the economic theory of liability it is a well-established result that, if the legal system wants to give proper incentives through the imposition of liability, it is better to base it on harm to the other party – the victim – than to the gain accruing to the liable party. The reason is twofold. First, it is more efficient that the party who is the decision maker (who chooses between performance and breach, or between termination or continuation of the contract) faces the relevant curve – cost or harm function, in this case – of the other party, and not its own – the benefit or profit function, in this case67 – because otherwise he would exploit to his own advantage the superior information – compared to the other party, or the external adjudicator – on his own function that he would typically possess. Given that in our setting it is the principal who takes the decision that may cause harm – termination – he should be facing the cost of the agent resulting from the decision, and not his own profit function. Second, gain-based liability, or disgorgement of profits, is much more vulnerable to mistakes made by Courts or adjudicators in the calculation of the relevant amounts. Even very minor mistakes can produce extremely undesirable outcomes, whereas with harm the effects are much more attenuated68. In addition to these two general advantages of liability for harm vs. disgorgement of profit as instruments to induce efficient behaviour, there are other reasons disfavouring gain-based compensation in the particular circumstances of termination of a commercial agency contract (reasons that apply to a long-term distribution contract as well). First, the gain in new clients and increased volume of orders resulting from the effort invested by the agent – in marketing, sales and client management – is, to a large extent, uncertain and stochastic at the point in time of investing the effort. The agent, when making the specific investment, does not know for sure if the effort will pay off or not: Marketing and sales effort may be very successful, or they may fail, both outcomes being influenced by factors beyond the control of the agent – and the principal too. A given effort may bring a very large number of new clients, or none at all, as a result of random or non-controllable

66

67 68

See, for instance, Joanna Goyder, 2005, p. 186, interpreting the provision as a rule of restitution of goodwill attributable to the agent’s efforts. See, Donald Wittman, 1985, p. 173. See, A. Mitchell Polinsky, and Steven Shavell, 1994, p. 427.

194

Fernando Gomez

events, such as general economic conditions, market situation, evolution of clients’ tastes, and so on. The level of effort, on the contrary, is not subject to this uncertainty. The former implies that in our setting, gain-based compensation is inherently more risky than harm-based compensation. So, if the agent is risk-averse, for an identical expected amount of compensation, he would prefer the more certain remedy, in this case, the one based on the certain harm (the invested effort, for instance). Moreover, the realization of the random events may be such that very few new clients, and thus few gains, are observed; or the opposite, that many new clients and gains are observed. If the litigated cases are biased, for whatever reasons, in favour of one or the other set, the level of compensation would be then too low, or too high compared to the ideal one. Allow me now a brief excursus arising from this idea of the uncertainty associated with gain-based schemes. It is, in fact, the inherent riskiness of a gain-based compensation system what also makes it an unattractive safety net mechanism against circumstances that the agent may fear as threatening the continuation of the relationship, but are entirely independent of a voluntary decision by the principal. In case of termination due to death (art. 17.4 Commercial Agents Directive) or other circumstance – age, infirmity or illness – that reasonably prevents the agent from performing the activities (art. 18 (b) Commercial Agents Directive) indemnity or compensation is payable to the agent according to the Directive. It is true that the agent, if risk-averse, would like to be covered financially against the risk of those contingencies, which seriously and negatively affect the stream of income he receives. In short, he would like to be insured against such events. But awarding compensation based on the accrued clientele to the principal is an unattractive way to cover the risk that the agent faces. It is bad insurance, due to its intrinsic risk of the existence and amount of gains derived from the new clientele. And even the compensation option contained in art. 17 Commercial Agents Directive, as it also requires that the principal has obtained substantial benefits linked to the commercial agent’s activities, entails some – although more limited – element of risk in the compensation for the agent, again an imperfection from the point of view of insurance. For such contingencies (death, sickness) agents would be better-off if instead of being compensated ex post as the Directive provides, they could get the exact expected amount of the compensation during the life of the contract, and buying an insurance policy in the insurance market with that amount. This means that the “insurance” mandated by the Directive is inefficient, and parties would prefer to have a different insurance arrangement, a preference that the mandatory nature of the rules in the Directive makes impossible. The fact that the provision in art. IV. E.–2:305 DCFR is only mandatory for commercial agency, but not for other long-term distribution contract alleviates this problem, but simply alleviates, it does not eliminate the mistaken starting point. The second reason why a compensation system based on restitution of the gains from new clientele is an undesirable way to induce efficient specific investments is more subtle, and it is also independent of its uncertain nature. It has to do with the fact that we are dealing here with investments which are, at the same time, relation-specific and cooperative in the sense of benefiting the party not making the investment. It is clear that the increased stock of clients substantially benefits the principal69, thus making the investment to enhance this set a cooperative investment in nature. Moreover, if the investment is truly specific this 69

It is true that, depending on the way in which remuneration is determined – and art. 6 Commercial Agents Directive allows for multiple remuneration systems – the agent may also benefit – through increased commissions – from the new set of clients, but I disregard this complicating factor which would not substantially alter the qualitative nature of the reasoning.

Termination and Compensation in Long-Term Distribution Contracts

195

means that it has no value at all outside the particular relationship in which the investment takes place. In the jargon of incomplete contract theory, the value of the investment is zero if there is no trade between the contracting parties. In such a setting, a remedy in case of termination -such as restitution of profit from the new clientele – which is based upon the value of the investment after termination – a “no trade” situation – makes no sense, because the specific nature of the investment means that this outside value is zero. One would need to interpret, for it to make sense in a setting of cooperative specific investment, the gain-based indemnity not as the actual value – after termination – of the investment to enhance clientele, but as the ex ante expected value – in terms of revenue from increased clientele – of a given level of investment, let’s say, to put the gain-based remedy under the best possible light, the optimal level of investment. But this expected value of the optimal investment in increasing clientele is, for the purposes of the optimization problem of the agent – the one deciding on the investment – functionally equivalent to a form of liquidated damages clause, that is, a constant amount determined before the decision to invest is made and which is payable by the breaching – terminating – party. And in the setting of cooperative specific investments, at least according to Che and Chung70 reliance damages may be an attractive option, even outperforming the best – for the parties – available liquidated damages clause agreed by the parties beforehand. This is definitely the case when no threshold in performance is feasible (unverifiable breach).71 So if the gain-based remedy in our setting works as a liquidated damages term, and it can be shown that reliance damages is preferable to the best liquidated damages clause, it must be the case that reliance damages are superior to gain-based restitution in an scenario of cooperative specific investment. It is arguable, however, that the set of compensation schemes included in art. 17 Commercial Agents Directive – and also, in the text of the DCFR, the scheme in Art. IV. E.–2:305 DCFR, for all distribution contracts, and in Art. IV. E.–3:312 DCFR with mandatory character for commercial agency – are designed not for a setting of specific investment, but for one of general investment, that is, an investment whose value is portable, by the party benefiting from it, outside the particular relationship or contract in which the investment was made. One can think that this portability assumption is clear behind the gain-based indemnity in art. 17.2, and also in the indemnity for goodwill in Art. IV. E.–2:305: “… the principal continues to derive substantial benefits from the business with such customers”. We are here squarely in a situation of portable or general cooperative investment. And even the compensation option in art. 17.3 of the Directive may also implicitly be assuming this general character in the investment, as it also speaks – more neutrally and less explicitly, in temporal terms – of “… providing the principal with substantial benefits linked to the commercial agents activities,” as one of the circumstances in which termination occurs and damage to the agent is presumed. In other words, the principle lying behind the remedies in art. 17.2 Commercial Agents Directive and Art. IV. E.–2:305 and Art. IV. E.–3:312 DCFR, may be linked more closely to a rationale of avoiding the more extreme consequences of unjust enrichment by the principal when the investment by the agent is of generic nature, than to a logic of reducing incentives for hold-up detrimental to the agent. The fact that the principal can benefit from the new clients attracted by the agent’s investment does not automatically provide a sound basis for compensation in favour of the 70 71

Yeon-Koo Che, and Tai-Yeong Chung, 1999, p. 96. Alexander Stremitzer, 2008.

196

Fernando Gomez

agent, however. If a cooperative investment (benefiting the non-investing party) is entirely generic, the investing party will not invest at all if he is not compensated for the cost of the investment, given that, contrary to what happens with the case of the cooperative specific investment, here the investment does not increase the likelihood of trade between the parties, that is, the chances of continuation of the existing relationship: if the investment is specific, the value of the contract, compared to outside alternatives, increases for the principal, and thus the likelihood that the latter will not terminate the relationship; if the investment is generic, its value is perfectly portable, and thus the likelihood of termination by the principal does not decrease – quite the opposite – with the amount and return of the investment. This means that, absent compensation, the agent will invest zero if the investment is cooperative and generic. And given that the principal anticipates this outcome, he will know that he needs to compensate the agent in order to obtain the positive value of the investment. There is, then, no need to mandate compensation for such investments, given that it is in the interest of both parties that investment gets compensated. This is, for example, what happens in the employment relationship with covenants not to compete, which serve as implicit compensation – guarantee may be more precise here – for the employer to induce him to provide generic training to the employees, a theme that permeates the treatment of covenants not to compete. In fact, if transaction costs are sufficiently close to zero, as one could imagine is the case between the principal and the agent during the relationship, the Coase Theorem ensures that parties would achieve the optimal level of cooperative generic investment, using the compensation scheme (explicit or implicit, prior or posterior to the investment, during or at the expiration of the relationship) that is more convenient for the parties. The legally mandated system would only come as a less efficient – less preferred – surrogate to the one that the parties would have chosen absent the legal imposition – because if there is mandatory compensation ex post, this would crowd-out the other mechanisms that the parties may have crafted: The principal is not going to pay twice for the cost of the cooperative generic investment. The need for an ex post compensation system legally imposed upon the parties may arise if the agent is boundedly rational to the extent that he may misunderstand the nature or actual cost of the investment, or the reality of the agreed compensation system for the investment. But this is hard to assume as a good description of actual condition of agents, specially to assume it at the broad and general level of an abstract rule of private Law. A legal remedy may also be required in cases in which the principal reneges upon an implicit promise of compensating the agent for the investment. For instance, the parties may have a tacit understanding on a longer duration of the relationship, and increased commissions in that period, which may serve the agent to recover the cooperative investment made, and then the principal abruptly terminates the relationship without honouring the implicit compensation scheme. Here, a legal remedy to curtail such opportunistic behaviour by the principal is de rigueur, but such a remedy does not coincide with the capped indemnity of art. 17.2 Commercial Agents Directive and Art. IV. E.–3:312 DCFR, nor with the complex and convoluted compensation provision of art. 17.3 Commercial Agents Directive. In such circumstances, the principal should simply compensate the agent for the non-recovered cost of the cooperative and generic investment. All the previous observations strongly point in the direction that, when seen through the lenses of the economic perspective of incomplete contract theory, the compensation rules in the Commercial Agents Directive, and in the DCFR, look seriously flawed, more in need of reform or reinterpretation than allowing application as general rules in the framework of long-term distribution chains.

Termination and Compensation in Long-Term Distribution Contracts

197

This does not mean that it is never sensible to contemplate two – or more – compensation schemes as optional or default rules for parties to choose from, but the menu can hardly be the suspicious outcome of the EU law-maker paying respects to two divergent national legal traditions on the matter (the German and the French). One could refer to selfish specific (selfish generic investments pose no incentive problems) investments, and for these, expectation damages (that is, loss of profit, but for the agent, not the principal) perform better than reliance damages, particularly if coupled with a cap – like the one in art. 17.2 b), although the figure in this provision is arbitrary, which does not mean it is not sensible- to avoid the over-reliance effect of expectation damages over specific investments, when the investing party and the breaching party are different. The other could refer to cooperative investments, and for these a rule of compensating unamorticized expenditures that do increase benefits to the principal – the notion of expenses incurred by the agent at the principal’s advice is just a proxy for this, not always precise: the principal can give advice to the agent, or a distributor generally, on how to reduce own cost, due to more experience in the business- may be a more sensible solution. For cooperative generic investments, it may simply be a rule that ordains compensation of full investment costs when the implicit – evidenced in any form satisfying the preponderance of the evidence burden of proof standard – understanding for recovery of such investments has been broken by the principal or the manufacturer. As they now appear in the Directive, and in the various implementations in Member States’ legal systems, the two schemes significantly diverge from this model, inspired by the lessons from the economic theory of damages and specific investments. It is not only that the provisions in the Directive are confused and confusing, as I have argued before. A big divide between how the system looks like in theory, and how it works in practice, seems to exist. For instance, if one looks at the French system, in which the compensation scheme is based, in practice it amounts at payment of two years gross commission72, and has nothing to do either with substantial benefits for the principal, nor with non-amorticized costs. In fact, two years commissions amount simply to average or standardized expectation damages, which are particularly counterproductive for cooperative relation-specific investments. These practices are likely to survive, however altered, after the DCFR. Given these shortcomings, one should be extremely hesitant to adopt in the field of longterm distribution contracts the policy choices essentially contained in the DCFR. It does not only stick to the system of the Commercial Agents Directive when regulating commercial agency as such, but tsubstantially extends the application of the regime on compensation following termination contained in the Directive to other long-term contracts in distribution chains, even if only as a default regime for those other contract types, and not mandatory as in the Directive. Even if one wanted to pursue some explicit redistributionist policy in this area, one should be cautious about this extended application. Due to the contractual nature of the relationship between the parties, most if not all of the redistribution would be moot, due to readjustment in other terms that are not mandatory, while the incentives of the parties would have been inefficiently distorted. In sum, even if, as is the case, it is true that one cannot – and should not – radically rule out, on economic grounds, all compensation after termination in the area of distribution contracts, one would probably need to have a closer look at the relevant factors, which would not lead us to stick to the compensation schemes currently to be found in the Acquis. 72

See, Report on the Application of Article 17 of Council Directive on the Co-ordination of the Laws of the Member States Relating to Self-Employed Commercial Agents (86/653/EEC), of 23.7.1996, p. 16.

198

Fernando Gomez

Conclusions As I have argued in this paper, the issue of compensation following termination of a longterm distribution contract, is one of the crucial issues for the general legal understanding of such relationships, and is tightly correlated with other contractual aspects of distribution chains. Unfortunately, from the point of view of European Law, the existing basis in the Acquis (essentially, the Commercial Agents Directive) does not provide a solid and coherent starting point for a satisfactory legal treatment of this issue with respect to the entire field of distribution contracts. Their use in the relevant provisions of the DFCR is, consequently, not to be celebrated from an efficiency viewpoint. The contrast of those existing rules with the theoretical findings in the Law & Economics literature dealing with the matter does not offer a promising view of their likely consequences for the contractual behaviour of the parties. Moreover, the available empirical evidence (which is now noticeable, if still not comprehensive), specifically concerning franchising, seems to cast doubt on the merits of legislation restricting termination and imposing compensation after termination -this is only alleviated in the DCFR by not extending the mandatory nature of indemnity for goodwill to franchising and distributorship. It is, moreover, still more troublesome, that the DCFR has introduced a fresh an enhanced set of restrictions imposed on termination as a means to discipline non-verifiable instances of breach in a long-term relationship, by means of forbidding any clause allowing its use only in cases of fundamental – and, hence, verifiable by an external adjudicator – breach of contract. The preceding analysis does not universally imply hat the optimal legal regime is necessarily one in which no compensation at all is ever imposed as a remedy. The economic literature, and in particular the branch dealing with incomplete contracting, has identified several dimensions of long-term contracting -also in distribution chains, and may be with special- which play a role in this context. The nature and timing of the parties’ interaction, the role of reputation to control opportunities for non-cooperative behaviour, the incentives to make investments in the relationship, particularly relation-specific investments, all have a bearing on how to design a legal treatment for termination and, eventually, for a compensation scheme favouring the distributor when the principal has terminated the relationship. Translating this set of factors into abstract rules of Contract Law is, at least at this stage, too daunting a task to be undertaken with reasonable likelihood of success, and this seems to be perceptible in the rules contained in the DCFR.

References Philippe Aghion, and Patrick Bolton (1987), “Contracts as Barriers to Entry”, 77 American Economic Review, p. 388-400. Ronen Avraham, and Zijhong Liu (2006), “Incomplete Contracts with Asymmetric Information: Exclusive v. Optional Remedies” 8 American Law and Economic Review, p. 523-561. (2008), “Should Courts Ignore Ex-Post Information when Determining Contract Damages?”, Working Paper, Northwestern Universtity School of Law. Scott Baker, and Kimberly Krawiec (2006), “Incomplete Contracts in a Complete Contracts World”, Working Paper, University of North Carolina School of Law. Oren Bar-Gill, and Omri Ben-Shahar (2004), “The Law of Duress and the Economics of Credible Threats”, 33 Journal of Legal Studies, p. 391-430.

Termination and Compensation in Long-Term Distribution Contracts

199

Klaus Hopt, in Adolf Baumbach, and Klaus Hopt (2006), Handelsgesetzbuch Kommentar, 32nd ed., Beck, Munich. John Beales III, and Timothy Muris (1995), “The Foundations of Franchise Regulation: Issues and Evidence”, 2 Journal of Corporate Finance, p. 157-197. Omri Ben-Shahar (2004), “Agreeing to Disagree: Filling Gaps in Deliberately Incomplete Contracts”, Wisconsin Law Review, p. 389-428. Janet Bercovitz (2000), “An Analysis of the Contract Provisions in Business-Format Franchise Agreements”, Working Paper, Fuqua School of Business, Duke University. Massimo C. Bianca, and Stefan Grundmann (Eds.) (2002), EU Sales Directive Commentary, Intersentia, Antwerp-Oxford-New York. Roger D. Blair, and Francine Lafontaine (2005), The Economics of Franchising, Cambridge University Press, New York. Geert Bogaert, and Ulrich Lohmann (Eds.) (1993), Commercial Agency and Distribution Agreements. Law and Practice in the Member States of the EC and the EFTA, 2nd ed., Graham&Trotman/Martinus Nijhoff, London. James Brickley, Frederick Dark, and Michael Weisbach (1991), “The Economic Effects of Franchise Termination Laws”, 34 Journal of Law & Economics, p. 101-132. James Brickley, Sanjog Misra, and Lawrence Van Horn (2006), “Contract Duration: Evidence from Franchising”, 49 Journal of Law & Economics, p. 173-196. Yeon-Koo Che, and Tai-Yeong Chung (1999), “Contract Damages and Cooperative Investments”, 30 Rand Journal of Economics, p. 84-105. Yeon-Koo Che, and Donald Hausch (1999), “Cooperative Investments and the Value of Contracting”, 89 American Economic Review, p. 125-147. Tai-Yeong Chung (1991), “Incomplete Contracts, Specific Investment, and Risk Sharing”, 58 Review of Economic Studies, p. 1031-1042. (1992), “On the Social Optimality of Liquidated Damage Clauses: An Economic Analysis”, 8 Journal of Law, Economics & Organization, p. 280-305. (1998), “Commitment through Specific Investment in Contractual Relationship”, 31 Canadian Journal of Economics, p. 1057-1075. Richard Craswell (1991), “Passsing-On the Costs of Legal Rules: Efficiency and Distribution in Buyer-Seller Relationships”, 43 Stanford Law Review, p. 361-398. Kevin Davies (2006), “The Demand for Immutable Contracts: Another Look at the Law and Economics of Contract Modifications”, 81 New York University Law Review, p. 487-549. Aaron Edlin (1996), “Cadillac Contracts and Up-Front Payments: Efficient Investment Under Expectation Damages”, 12 Journal of Law, Economics & Organization, p. 98-118. Aaron Edlin, and Stefan Reichelstein (1996), “Holdups, Standard Breach Remedies, and Optimal Investment”, 86 American Economic Review, p. 478-501. Matthew Ellman (2006), “Specificity Revisited: The Role of Cross-Investments”, 22 Journal of Law, Economics and Organization, p. 234-257. Joanna Goyder (2005), EU Distribution Law, 4th ed., Hart Publishing, Oxford and Portland (OR). Oliver Hart, and John Moore (1988), “Incomplete Contracts and Renegotiation”, 56 Econometrica, p. 755-785. Benjamin Hermalin, Avery Katz, and Richard Craswell (2006), “Law and Economics of Contracts”, in A. Mitchell Polinsky, and Steven Shavell (Eds.), The Handbook of Law and Economics, North Holland.

200

Fernando Gomez

Christine Jolls (1997), “Contracts as Bilateral Commitments: A New Perspective on Contract Modification”, 26 Journal of Legal Studies, p. 203-237. Benjamin Klein (1980), “Transaction Cost Determinants of ‘Unfair’ Contractual Arrangements”, 70 American Economic Review, p. 356-362. Benjamin Klein (1995), “The Economics of Franchise Contracts”, 2 Journal of Corporate Finance, p. 9-37. (1998), “Hold-Up Problem”, in Peter Newman (Ed.) The New Palgrave Dictionary of Economics and the Law, Vol. II, MacMillan, New York-London. Benjamin Klein, and Keith Leffler (1981), “The Role of Market Forces in Assuring Contractual Performance”, 89 Journal of Political Economy, p. 615-641. Benjamin Klein, and Kevin Murphy (1988), “Vertical Restraints as Contract Enforcement Mechanisms“, 31 Journal of Law & Economics, p. 265-297. Jonathan Klick, Bruce Kobayashi, and Larry Ribstein (2006), “Incomplete Contracts and Opportunism in Franchising Arrangements: The Role of Termination Clauses”, Working Paper, George Mason University School of Law Francine Lafontaine, and Kathryn L. Shaw (2005), “Targeting managerial control: evidence from franchising”, 36 RAND Journal of Economics, p. 131-150. W. Bentley MacLeod, and James Malcomson (1993), “Investments, Holdup, and the Form of Market Contracts”, 83 American Economic Review, p. 811-837. Eric Maskin, and John Moore (1999), “Implementation and Renegotiation”, 66 Review of Economic Studies, p. 39-56. George Mathewson, and Ralph Winter (1985), “The Economics of Franchise Contracts”, 28 Journal of Law & Economics, p. 9-37. Georg Nöldeke, and Klaus Schmidt (1995), “Option Contracts and Renegotiation: A Solution to the Hold-Up Problem” 26 Rand Journal of Economics, p. 163-179. Cándido Paz-Ares (1997), “La indemnización por clientela en el contrato de concesión”, La Ley (1997), D-105 T. 2. (2003), “La terminación de los contratos de distribución“, 8 Advocatus. A. Mitchell Polinsky, and Steven Shavell (1994), “Should Liability be Based on the Harm to the Victim or the Gain to the Injurer?”, 10 Journal of Law, Economics & Organization, p. 427-446. William Rogerson (1984), “Efficient Reliance and Damage Measures for Breach of Contract”, 15 Rand Journal of Economics, p. 39-53. Paul Rubin (1978), “The Theory of the Firm and the Structure of the Franchise Contract”, 25 Journal of Law and Economics, p. 223-233. Ilya Segal, and Michael Whinston (2000), “Exclusive Contracts and Protection of Investments”, 31 RAND Journal of Economics, p. 603-633 . Patrick Schmitz (2001), “The Hold-Up Problem and Incomplete Contracts: A Survey of Recent Topics in Contract Theory”, 53 Bulletin of Economic Research, p. 3307-3378. Urs Schweizer (2006), “Cooperative Investments Induced by Contract Law”, 37 RAND Journal of Economics, p. 134-145. Steven Shavell (1980), “Damage Measures for Breach of Contract”, 11 Bell Journal of Economics, p. 466-490. (2004), Foundations of Economic Analysis of Law, Harvard University Press, Cambridge (MA)-London. (2007), “Contracts, Holdup and Legal Intervention”, 36 Journal of Legal Studies, p. 325-354.

Termination and Compensation in Long-Term Distribution Contracts

201

Randolph Sloof, Edwin Leuven, Hessel Oosterbeek, and Joep Sonnemans (2003), “An Experimental Comparison of Reliance Levels Under Alternative Breach Remedies”, 34 Rand Journal of Economics, p. 205-222. Alexander Stremitzer (2008), “Standard Breach Remedies, Quality Thresholds, and Cooperative Investments”, Columbia Law and Economics Working Paper No. 335. Joel Watson (2002), Introduction to Game Theory, W. W. Norton and Company, New York. Stephen Weatherill (2005), EU Consumer Law and Policy, Edward Elgar, Cheltenham. Darrell L. Williams (1996), “Franchise Contract Terminations: Is There Evidence of Franchisor Abuse?”, 10th Annual Proceedings of the Society of Franchising, Lincoln, International Center for Economic Franchise Studies, College of Business Administration, University of Nebraska. Oliver Williamson (1985), The Economic Institutions of Capitalism, The Free Press, New York. Donald Wittman (1985), “Should Compensation Be Based on Costs or Benefits?”, 5 International Review of Law & Economics, p. 173-185.

Decisions European Court of Justice Date 30.4.1998 9.11.2000 23.3.2006

Reference C-215/97 C-381/98 C-465/04

Parties Barbara Bellone v. Yokohama SpA Ingmar GB Ltd v. Eaton Leonard Technologies Inc Honyvem Informazioni Commerciali Srl v. Mariella De Zotti

Variation and Unilateral Termination of an Agreement in the Draft Common Frame of Reference Mitja Kovač*

Abstract European contract law and the derived Common Frame of Reference has recently been the subject of increasing attention and intense scholarly debate. This paper offers an economic evaluation of unilateral termination and variation of contracts, and provides an economic assessment of the related provisions concerning the variation and unilateral termination of agreements as drafted in the proposed Common Frame of Reference. The main propositions and conclusions are the following: (1) In principle, contracts should not be unilaterally terminable by any of the contracting parties; (2) efficiency under several preconditions requires permitting unilateral termination of open-ended contracts, as of fixed-term contracts where the right to unilaterally terminate is expressly provided for; (3) the operation of the efficient “breachexpectation damages” remedy is in open-ended contracts hindered by prohibitive assessment costs and potential infinite expectation damages award; (4) the amount of potential expectation damages always exceeding the value of outside options may prevent the operation of the efficient termination; (5) such unilateral termination of open-ended, continuous contracts where each of parties forgo their expected profits implies risk sharing; (6) unilateral termination enables parties to pursue a change in their market strategy and to enable flexible responses to changed market conditions; (7) the selfenforcing characteristic of unilateral termination implies that the hold-up problem is deterred by the threat of termination rather than by the threat of litigation; (8) in order to deter opportunism such termination should be awarded only if it is not due to socially harmful behavior; (9) a good faith standard for termination should be interpreted as a rule policing cheating, retaliation, opportunism, and other instances of socially harmful behavior which would, in economic terms, redistribute the risk already allocated by the agreement; (10) a certain advance notice period should be given in order to enable the non-terminating party to adapt to new, changed circumstances and to recoup initial investment; and finally (11) assessed provisions may be, from a scientific viewpoint, improved and elaborated further.

Introduction Should legal rules ever entitle an individual or rational party to vitiate or unilaterally terminate contractual relationships? This article offers an elaborated economic analysis of unilateral termination and variation of contracts, and provides an economic assessment of several articles of the Draft Common Frame of Reference dealing with the variation and termina-

*

University of Ljubljana, Faculty of Economics, Chair of Law, Kardeljeva ploscad 17, 1000 Ljubljana e-mail: [email protected].

204

Mitja Kovač

tion of agreements.1 However, a caveat related to the scope of this article should be made. Namely, since it is not possible to cover all aspects of the variation and unilateral termination of contracts, this article merely focuses on variation or termination by agreement or by notice, and omits the economic comment on the variation or termination by the court to a change of circumstances – which goes well beyond the scope of this article.2 Furthermore, this article does not discuss optimal compensation and optimal damage remedies in instances of unilateral variation or termination of continuous contracts, or a termination as a remedy for breach of contract. The use of open-ended contracts and of fixed-term contracts, with an expressly provided unilateral termination right, is widespread within societies and industries. Questions like whether contractually provided for unilateral terminations rights should be permitted or banned, or whether parties should be entitled to unilaterally terminate such contracts without any previously stipulated rights to such an effect are unanswered, where several efficiency based considerations have arisen. However, although contract termination is of equal importance as contract formation – which is regarded as the keystone in the current contract law’s analysis and teaching – and although it is one of the milestones of contractual relationships, it was analyzed less often than might be expected. There has been relatively little economic analysis of unilateral variation or termination of contracts, neither is there any substantial comparative analysis of such unilateral options.3 Moreover, economic theory besides the theory of efficient breach, efficient decision to terminate an agreement and the self-enforcing mechanism4 does not offer a much elaborated model rule or overwhelming justification of unilateral termination.5 This paper, by employing legal and economic analysis, overcomes doctrinal inconsistencies, and offers an additional explanatory framework for distinguishing unilateral terminations which should be enforced from those which should be banned. Throughout this paper the notion of termination means that one party puts an end to the open-ended, continuous or fixed-term contract pursuant to the power created either by the contract or by the law other than for its breach, i. e. where there is no uncured material

1

2

3

4

5

Draft Common Frame of Reference (DCFR), Interim Outline Edition, Sellier 2008; prepared by the Study Group on European Civil Code and the European Research Group on Existing EC Private Law (Acquis Group), based in part on a revised version of the Principles of European Contract Law. Article III.–I.110 DCFR on Variation or termination by court on a change of circumstances actually encompasses the problem of unforeseen contingencies, i. e. in legal scholarship known as the doctrine of commercial impracticability, frustration of purpose, imprevision, wegfall des geschäftsgrundlage or as hardship. For a synthesis see E. B. Hermalin, W. A. Katz, W. Avery and R. Craswell, “The Law and Economics of Contracts,” in Polinsky, A. Mitchell, Shavel, Steven (eds.), The Handbook of Law and Economics, Nort-Holland, 2008, at 122. On the threat of termination as a self enforcing mechanism see S. Macaulay, “Non-contractual Relations in Business,” 28 American Sociology Review 55, 1963; K. Benjamin and B. K. Leffler, “The Role of Market Forces in Assuring Contractual Performance,” 89 Journal of Political Economics 615, 1981, and K. Benjamin, K. Murphy, “Vertical Restraints as Contract Enforcement Mechanisms,” 31 Journal of Law & Economics 265, 1988. Yet, there is an extensive amount of literature regarding the unilateral termination of franchise relationships. For a synthesis see A. W. Dnes, “Franchise Contracts,” in Bouckaert, Boudewijn and De Geest, Gerrit (eds.), Encyclopedia of Law and Economics, Edward Elgar, 1999.

Variation and Unilateral Termination of an Agreement

205

breach by the other party.6 If the contract is terminated, all executory obligations are thereby discharged, but any right based upon prior breach or performance survives.7 The first part of paper (Section 1) offers a set of economic criteria for establishing when and under what conditions fixed-term and open-ended contracts should be unilaterally terminable. The second part (upon provided set of economic criteria) critically comments on some of the articles of the Draft Common Frame of Reference (Section 2). A conclusion is provided in the last section.

1. Variation and unilateral termination of contracts – economic assessment In the daily world of mutual transactions, parties continuously make contracts in order to warrant the future provision of goods and services, to mutually beneficially reallocate or share risks, or to alter the timing of the consumption due to the differences of opinion about the subsequent events.8 It is well established that except in the case of a spot market transaction, where parties manage reasonably well without formal contracting,9 contracting becomes valuable when there is a temporal element to their exchange or one party is unsure as to what her counterparty will do.10 In other words, in the modern world a contract enables trust, and thus fosters a mutual exchange, an efficient level of relation specific investment in contractual enterprise, facilitating the move of scare resources to its most valuable uses. However, it is also well established that without enforcement, a party would be able to appropriate funds that have been paid before contract performance, rendering the contract unworkable; a party might not deliver a promised good or perform a promised service; and the price cannot be fixed in advance, inducing a price holdup where a party might bargain opportunistically about the price of transaction.11 More importantly, this later holdup problem at the negotia6

7

8

9

10

11

As Ghosh points out, the natural question is how termination relates to breach of contract, which occurs if a party fails to satisfy an express or implied contractual promise. He argues that certain breaches are actually a form of termination; S. Ghosh, “Property Rules, Liability Rules, and Termination Rights: A Fresh Look at the Employment at Will Debate with Applications to Franchising and Family Law,” 75 Oregon Law Review 969, 1996. Hence, the unilateral termination as discussed in this article should be distinguished from the termination as the remedy for the breach of contracts. S. Shavell, Foundations of Economic Analysis of Law, Harvard University Press, 2004, at 296 et seq. See eg J. K. Arrow and G. Debreu, “Existence of an Equilibrium for a Competitive Economy,” 22 Econometrica 265, 1954. In absence of contract, the parties may be reluctant to trust each other to complete the agreed upon exchange at the stipulated time and thus loosing valuable exchanges; E. B. Hermalin, A. W. Katz and R. Craswell, “The Law and Economics of Contracts,” in Polinsky, A. Mitchell, Shavell, Steven (eds.), The Handbook of Law and Economics, North-Holland, 2008, at 122. For a discussion that contracts can be valuable even in non-exchange settings, when advance commitment enhances the value of a gift by enabling reliance by the beneficiary see e. g. R. Posner, “Gratuitous Promises in Economics and Law,” 6 Journal of Legal Studies 83, 1977; S. Shawell, “An Economic Analysis of Altruism and Deferred Gifts,” 20 Journal of Legal Studies 401, 1991. “… and reducing the value of the contract or discouraging the making of it altogether;” Shavell, supra note 8, at 297 et seq.

206

Mitja Kovač

tion stage for performance and payment will result in all manner of underinvestment in the contractual enterprise.12 Hence, it is widely accepted that contract enforcement is privately and socially desirable because it spurs production and trade. The obvious question which then appears is whether contracts shall be at all freely terminable and whether such an option would not completely destroy all provided incentives, would enable opportunistic behavior (including the holdup problem) and would actually destroy the operation of the market? Apparently, all previously listed arguments may be employed as the basis for severe criticism of a broad general rule of unconditional termination, where parties would be able to freely, at any moment, terminate a contract, which would (in the end result) undermine the function of the market and would be a source of stark inefficiencies. Hence, in general, contracts should not be freely and unilaterally terminable in case of the rise in costs of performance, change of preferences and other similar reasons. In such instances parties should simply breach the contract and pay expectation damages or any other restitution available under the law. Recall that expectation damages remedy for breach of contract provides efficient incentives to perform if and only if it is efficient to do so. As stated, if one would instead of the “breach – expectation damages” remedy apply an unconditional termination of contract, where all executory obligations are thereby discharged, obviously the operation of the market would then, due to all previous reasons, be seriously impaired. Yet, there are some additional economic arguments, needs and special contractual relationships, where such a unilateral termination right should be granted, since it is economically justified and efficient. In instances of contracts for an indefinite period of time (openended, continuous contracts) and fixed-term contracts with the contractually allocated right of unilateral termination, efficiency requires, fulfilling several preconditions, granting such a unilateral right to contracting parties. Moreover, while offering the exact framework for establishing the entitlement case, the exact content of such a unilateral right will be discussed and compressed into a single general principle.

1.1. Unilateral termination of long-term continuous contracts should be permitted This section addresses problems associated with the unilateral termination of long-term contracts of an indefinite period of time, when one of the parties engaged in such a contract seeks to terminate the contract unilaterally. Economists agree that full, contingent, costlessly enforceable contracts are not possible, and this holds especially for long-term, open ended contracts, which inevitably involve costly conflicts of interest which may prevent parties from achieving the desired results.13 For example, it is not trivial to assume that, as time passes, a party which has promised the delivery of goods or services may find that another better opportunity has arisen; the costs of performance have increased more than expected and so forth. Recall, that efficiency requires performance as long as its value exceeds its true costs, V > Cp. However, performance may 12

13

See eg E. O. Williamson, E. Oliver, Markets and Hierarchies: Analysis and Antitrust Implications, Free Pres New York, 1975; P. A. Grout, “Investment and Wages in the Absence of Binding Contracts: A Nash Bargaining Approach,” 52 Econometrica 449, 1984; B. Klein, R. G. Crawford and A. A. Alchian, “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process,” 21 Journal of Law & Economics 297, 1978. C. J. Goetz, R. E. Scott, “Principles of Relational Contracts,” 67 Vanderbilt Law Review 1089, 1981, at 1130.

Variation and Unilateral Termination of an Agreement

207

become disadvantageous for promisor when Cp > P, and so he would want to get rid of such a contract. In such circumstances expectation damages are certainly the most efficient remedy, since it ensures efficient performance of the contract, i. e. if and only if the contract is still efficient, discourages opportunistic behavior, assigns the risk on a superior risk bearer and economizes on transaction costs, since this is the rule parties would provide for if they would bargain for. Moreover, if it is efficient to breach the contract (since e. g. promisor’s cost of performance has increased to exceed the stipulated price, Cp* > P) the “efficient breach – expectation damages” remedy compensate the promisee for his expectations, provides incentives for promisor to perform as long as it is efficient to do so, provides contracting parties with incentives to take an efficient amount of reliance specific investments, deters opportunistic behavior, provides an efficient risk allocation and reduces transaction costs. Hence, the unprovided for and unconditional termination, where promisor puts an end to the open-ended, continuous contract, where there is no uncured material breach by the other party, and where all executory obligations are likewise discharged, might provide incentives for opportunistic behavior, deter relation specific investments, assign the risk on party which is not the superior risk bearer and in final consequence prevent contracting and the operation of the market. Is then the prohibition of unilateral termination of long-term continuous contracts the best option? Bearing in mind that the purpose of formal contracting is to make contracts essentially legally enforceable, would then contracting parties ever comply with contracts knowing that either party may walk away at will? I argue that this is not at all the case and that this argumentation holds for fixed-term contracts, whereas it does not hold for long-term open-ended contracts.14 However, it should be stressed that such a right to unilaterally terminate the long-term open-ended contracts should not be unconditional. In this respect one may delineate three distinctive risks, which may be inherent in any unilateral termination option. First, there is a significant probability that after some period of time contingencies might materialize which cause regret on the promisor’s side to have entered the agreement. Second, such a right may induce the promisor to strategically or opportunistically terminate a contract. Third, the termination might be induced by the promisee’s failure to achieve required counter performance. Regarding the first risk, assume for example, that parties negotiate a contract of indefinite duration and after, for example, 10 years changed circumstances cause either party to regret the agreement, since the value of initial contract V is lower than the value of an optional one, V*, yielding V* > V. In other words, during the time span of the hypothetically indefinite contract there might occur some outside options15 whose values exceed the value of service or goods of the original contract, and where an efficiency-minded contractor would thus regret the original agreement and would rather like to terminate it. Two rational contracting parties would hence engage in an open-ended, continuous transaction so long as the costs of doing so are lower then the transaction benefits.16 When the costs of doing so exceed the benefits parties should terminate the agreement. However, one should note that 14

15 16

As Goetz and Scott argue, the critically important variable in determining the optimal assignment of the risk of regret contingencies is whether the parties select an indefinite duration or a fixed-term contract; Goetz and Scott, supra note 13, at 1143. During the infinite time line they probably will occur. On the efficient termination of employment relationship see A. S. Vandenberghe, An Economic Analysis of Employment Law, Dissertation series, Utrecht University, 2004, at 186.

Mitja Kovač

208

such a unilateral termination may impose costs upon a non-terminating party, and since from a social point of view minimization of total costs is required,17 the termination would be efficient when the costs of continuation of the relationship to the terminating party are higher than the costs of termination for the non-terminating party. In other words, parties should thus continue trying to perform only if the expected benefits of performance, discounted by the probability of successful performance, p Bp, are greater than expected benefits of outside options, q Bo, which could be obtained from terminating.18 Hence, the efficient unilateral termination threshold may be stated as following: p Bp > q Bo As one might expect in such a continuous, relation context it is often economically impractical for the parties to specify and allocate risks of changed market conditions accurately in advance. Moreover, it may be argued that parties did not enter into such a relational, indefinite duration contract to be bound forever,19 and that such agreement actually permits either party to adjust to events that may cause regret over the original compensation agreement. Such a contractual setting provides, in relationships for indefinite periods of time, a flexible mechanism, which enables a response on the occurrence of more valuable outside options, and pursues a change in market strategy where termination of the original relationship is efficient. Hence, such a contract takes into account and enables the need for adaptation to changing circumstances, whereas fixed-term contracts of long duration may prevent such an efficient termination of relationship. One would thus argue that such an open-ended contract actually represents a trade-off between business flexibility and the security of fixedterm contracts.20 However, one might also argue that if a more valuable outside option arises, the party should, according to the efficient breach theory, breach the contract and pay expectation damages to the non-breaching party. Yet, one should note that since the contract is an open-ended one the administering costs incurred by a third party (judge) of establishing the amount of damages might well be prohibitive, or the potentially awarded damages might even be infinite. Namely, since the contract is an open-ended one, and since the expectation damages put the non-breaching party in the same position as if the contract would’ve been performed, then indeed damages might be infinite or at least very costly to access (or unverifiable). Hence, persistence on an efficient breach/expectation damages remedy may in essence destroy the business flexibility which such contracts tend to provide. Namely, since the expectation damages might be even infinite, a party would never be able to terminate such a contract which should be, according to efficient termination threshold, terminated.

17

18

19 20

The total costs consist of the costs to the terminating party to continue with the contract and of the costs to the non-terminating party of contract termination. Yet, as it will be discussed later on, in order to preserve incentives undistorted, economize on transaction costs and assign the risk on the most efficient risk bearer, such a right should not be unconditional. Unless expressly provided for. Which indeed provides security and thus enables relation specific investments, but which also hinders the flexibility to react on the changed market conditions, which altered the expected value of exchange and which in longer time periods necessarily occur.

Variation and Unilateral Termination of an Agreement

209

Simply, the amount of damages always exceeding the value of outside options might always prevent such an efficient termination.21 One should also note that in the instance of infinite damages parties prefer risk sharing, and permitting unilateral termination of open-ended, continuous contracts where each of parties forgo their expected profits, actually implies risk sharing. In addition, one should note that relational, open-ended contracts provide incentives for the efficient amount of relation-specific investments with another significant, but yet an informal instrument. Namely, the reputation effect in such long-term, open-ended relationships between two sophisticated parties may act as an instrument of pressure and, serving as a substitute for courts enforcement in fixed-term contracts, provides an incentive for the efficient amount of relation-specific investments.22 Moreover, in instances of such long-term, open-ended relationships, where due to prohibitively high costs of performance, measurement the breach of contract for underperformance may be very difficult to prove and parties would thus often have an incentive to renege on the transaction by holding up the other party in the sense of taking advantage of unspecified elements of agreement. In such circumstances Klein convincingly shows that such unilateral termination provision is actually an implicit self-enforcing contract, which by the threat of termination of the business relationship, rather than by the threat of litigation, deters opportunism and economizes on transaction costs.23 Furthermore, in such a long term open-ended contract, the risk of regret is assigned upon non-regretting party, since it represents the most plausible explanation for the selection of this particular transaction cost-saving method. As Goetz and Scott argue, the terminating party is in such a contract, in order to retain presumed risk allocation, entitled to terminate an indefinite duration contract in order to pursue a change in market strategy.24 However, in order to pursue with this strategy several efficiency-based preconditions should be fulfilled.

1.1.1. Optimal risk allocation

An unrestrained right to terminate might, as already noted, induce cheating, retaliation and instances of opportunistic behavior, where the promisor may due to socially harmful reasons 21

22

23

24

Actually, the efficient termination or efficient breach might not occur. Such potentially infinite expectation damages actually hinder the operation of efficient breach. Reputation effect in contracts for the indefinite period of time also enables the parties to develop mutual trust and, while getting to know each other, they are also able to incur relation-specific investments. Klein continues that such apparently one-sided terms may be crucial elements of minimum cost quality-policing arrangements, since given the difficulty of explicitly specifying and enforcing contractually every element of quality of service or of goods (e. g. in franchisee contracts) there is in such a relationship an incentive for individual opportunistic other party to cheat the promisor by supplying e. g. a lower quality of product than contracted for; B. Klein, “The Role of Incomplete Contracts in Self-Enforcing Relationships,” in E. Brousseau and J. M. Glachant (eds.), The Economics of Contracts, Theories and Applications, Cambridge University Press, 2002, at 358. If termination in fact does not represent the joint-maximizing outcome considering the interest of both parties, the terminated promisee would be free to bargain for a redivision of the joint contractual product in order to induce the disappointed promisor not to terminate; Goetz and Scott, supra note 13, at 1143.

210

Mitja Kovač

terminate such a contract. Such opportunistic behavior should be deterred, since it directly wastes scarce resources. In other words, opportunism and the other’s party prevention of opportunism wastes resources represent dead-weight social loss which should be avoided.25 In legal terms such opportunism is usually labeled as ‘bad faith behavior.’ Thus, one should introduce a necessary precondition that such a termination can be allowed only when it does not represent cheating or opportunistic behavior – i. e. when it is in good faith. This good faith standard for termination may thus be interpreted as a mechanism for policing cheating, retaliation, opportunism, and other instances of socially harmful behavior which would in economic terms redistribute the risk already allocated by the agreement.26 Indeed, also Goetz and Scott argue that in case of indefinite term contracts, the right of termination should be constrained only where necessary to discourage strategic behavior.27 Moreover, also the necessary recouping periods should be regarded as good faith requirements. Furthermore, such a rule which assigns the risk of opportunistic behavior upon the party who wants to terminate the contract provides efficient disincentives to engage in such a socially harmful activity. In addition, its self-enforcing characteristic implies that the opposite instance of the hold-up problem on the side of the promisee is deterred by the same threat of termination of the business relationship, rather than by the threat of litigation.28

1.1.2. Relation specific investments

A further important issue which should be resolved is the instance where the non-terminating promisee has incurred substantial relation specific investments. As a matter of fact, several conducted studies demonstrate that the investment specificity is one of the key determinants of contract duration.29 Moreover, they also note that many contracts permit termination at a certain date prior to the expiration of the contract, usually by means of unilateral options rather than contingent clauses.30 Furthermore, Joskow’s study shows that as specific invest-

25

26

27 28 29

30

Limitations on such an opportunism are always optimal when enforcement costs are negligible; E.W. Kitch, “The Law and Economics of Rights in Valuable Information,” 9 Journal of Legal Studies 683, 1980. Klein, Crawford and Alchian argue that such strategic behaviour which should be deterred has no socially optimizing effects, rather, it only distributes portions of an already allocated contractual pie; Klein, Crawford and Alchian, supra note 7, at 300. Hence Goetz and Scott define bad faith behaviour as intentional efforts to use contractually created discretion or other strategic opportunities to secure a larger share of the anticipated gains from existing contract; Goetz and Scott, supra note 13, at 1139. Ibid at 1144. Klein, supra note 18, at 358. See eg K. Croker and S. E. Masten, “Mitigating Contractual Hazards: Unilateral Options and Contract Length,” 19 RAND Journal of Economics 327, 1988; V. P. Goldberg and J. R. Erickson, “Quantity and Price Adjustment in Long-term Contracts: A Case Study of Petroleum Coke,” 30 Journal of Law & Economics 369, 1987. Ibid.

Variation and Unilateral Termination of an Agreement

211

ment becomes more important, the parties find it more advantageous to rely on longer-term contracts that specify the terms and conditions of repeated transactions ex ante.31 However, the essential role of relation-specific investments in motivating contracting is widely accepted in both law and economics. Indisputably, a party contemplating an exchange will be reluctant to make non-redeployable investments, or otherwise forego opportunities without reasonable confidence that the other party will uphold its end of the bargain when the time of performance arrives. Regarding the amount of relation-specific investments, Goetz and Scott suggest that in order to provide incentives for an efficient amount of the relation-specific investment the termination should be subject to recouping limitations.32 Such limitations, while deterring opportunism and preserving investment incentives, imply a reasonable duration of agreement so that the non terminating party, which has made substantial relation-specific investments may recoup his initial investment.33 In other words, in order to provide incentives for an efficient amount of relation-specific investment, the contract should not be terminable, for example, immediately after the conclusion of a contract. Such a period enables the party to recoup his relation-specific investments. However, the exact duration of such a recouping period should depend upon the specificity of facts, and should thus be decided on case-by-case basis. Moreover, such a recouping period may be due to purely technical reasons encompassed in the previously elaborated good faith standard. Furthermore, a sufficient advance termination notice period should be also considered in order to enable non-terminating party to secure other sources of supply, to adapt his production structure etc. In short, prior notification of the intent to terminate allows the non-terminating party to adjust to the changed situation, or to refrain from incurring further costs in reliance upon the continuation of original contract. If the notice period is contractually provided, then efficiency requires enforcement of this period. If nothing is stated then this advance notice period should be determined according to the nature and duration of the relationship and the level of relation-specific investments. Hence, the standard of the ‘reasonable notice period,’ which is often employed in legal literature and statutory provisions, should be interpreted as the time period necessary to recoup the non-terminating party’s relation-specific investments, to adjust to the changed situation and to refrain from incurring further costs in reliance upon the continuation of original contract.34 This efficiency-based time period would then obviously depend upon the specific facts of the relationship, and would thus differ from case to case. Such an implementation would thus require an empowering third party (judge) with the discretion to scrutinize in accordance with efficiency principle the actual duration of notice. Furthermore, Guriev and Kvassov also show in their model that a contract of indefinite duration that includes an option of unilateral termination with advance notice provides incentives for an efficient amount of relation-specific investments.35 Regarding the duration 31

32

33

34 35

P. L. Joskow, “Contract Duration and Relationship-Specific Investments: Empirical Evidence from Coal Markets,” 77 American Economic Review 168, 1987. They also argue that recoupment guarantess are not inefficient since if the objective of termination is only to increase the size of nonperformance penalty then there are other more efficient ways of doing so; Goetz and Scott, supra note 13, at 1145. E. Gellhorn, “Limitations on Contract Termination Rights – Franchise Cancellations,” 1967 Duke Law Journal 465, 1967, at 479 et seq. To secure other sources of supply, to adapt his production structure etc. See S. Guriev, D. Kvassov, “Contracting on Time,” SSRN Working Paper Series, 2003, Available at: http://ssrn.com/abstract=469180.

212

Mitja Kovač

of advance notice they argue that since the contract of indefinite duration is renegotiation proof, the advance notice time should be shorter than the duration of the relationship or the duration of the fixed-term contract.36 In addition, an informal reputation effect may act as an additional mechanism enabling relation-specific investments.

1.1.3. Economizing on transaction costs

Such a default rule allowing unilateral termination under severe preconditions is efficient and will also reduce transaction costs, since parties would contract for such a rule if they would bargain for. Otherwise, they would simply stipulate a fixed-term contract, of a long duration. In addition, regarding the issue of costs evolving from parties’ conflicts (monitoring costs, providing best efforts of other party etc.) Goetz and Scott point out that in continuous long-term contracts parties will seek to reduce such costs by establishing cost-effective monitoring or bonding mechanisms. Also, since monitoring is costly for both parties, those costs and consequently also the contract price may be improved by substituting reassurances of performance in the form of bonding provisions.37 Furthermore, the self-enforcing mechanism of unilateral termination for reasons other than opportunism, hold-up or failure of best efforts will be a cost-effective strategy only where changed conditions producing regret and create the prospect of substantial and continuing losses.38 Hence, as Goetz and Scott argue, this implies that the evidence of such exogenous conduct will be accessible to a terminated party seeking to establish a wrongful termination, and would thus also support a presumption of best efforts failure that the non-terminating party must rebut by establishing that termination was motivated either out of pure opportunism or by regret over changed conditions.39

1.2. Termination upon specific provision Frequently contracting parties write detailed, long-term contracts in which they leave themselves broad discretion to terminate the agreement on little or no notice at all, and, as reported, often make termination the exclusive remedy in the event of dissatisfaction with the other party’s performance.40 Yet, as in previous discussion, if one takes into account that the purpose of formal contracting is to make bargains enforceable, hence enabling relationspecific investments, the essential question is why would a party ever go through the trouble of specifying such a contract if another party would simply be able to walk away? Would not economic efficiency in such circumstances require overriding of such contracts? As it 36 37

38 39 40

Ibid at 19. As they have pointed out among the variety of possible bonding arrangements are capital contributions, covenants not to compete, self-imposed ethical standards for agents, and unilateral termination authorization. Goetz and Scott, supra note 13, at 1148. Ibid. Such contracts may be found among franchise contracts, equipment leases, distribution and advertising agreements, and software licenses; S. E. Masten, “Long-term Contracts and Short-Term Commitment: Price Determination for Heterogeneous Freight Transactions,” SSRN Working Paper Series, 2007, at 2. Available at http://ssrn.com/abstract=1010992.

Variation and Unilateral Termination of an Agreement

213

has been argued, there may be several significant justifications to enforce such a stipulated unilateral termination options. Although all previously advanced arguments in support of the right to – under established preconditions – unilaterally terminate a contract should govern also in cases of stipulated rights of unilateral termination some additional insights may be offered too. The first is related to the freedom of the contracting argument. As Epstein states, parties should be permitted to adopt this form of contractual provision if they so desire, since the principle behind this conclusion is that freedom of contract tends both to advance individual autonomy, and to promote the efficient operation of markets.41 Secondly, such a contractual provision should be respected as a rule of construction in response to the question of gaps in contract language.42 Moreover, such an explicit right to unilaterally terminate a contract actually enables the efficient termination of contract, enabling parties to pursue the change of their market strategy. Such an efficient, provided for termination actually permits either party to adjust to events which arise during the long-term duration of contractual relationship that may cause regret over the original compensation agreement, and hence spurs allocative efficiency. Furthermore, such contractual provision actually shows that parties have anticipated such a development and have explicitly, while stipulating e. g. the promisor’s right to unilaterally terminate a contract, assigned the risk of such regret upon the other party, i. e. the promisee. In addition, such a self-enforcing contract deters opportunistic behavior of the promisee, where the costs of monitoring are high and hence reduces transaction costs (costs of monitoring, litigation etc.). Last, but not the least, such a fixed-term contract of long duration with the specific unilateral termination provision combines, if necessary preconditions are fulfilled, beneficial features of indefinite and fixed-term contracts. Namely, such a contract at the same time provides protection to the promisor, alleviating the hold-up problem and securing an efficient amount of relation-specific investments. Also, by enabling the adaptation to changing market circumstances (efficient termination), circumvents the rigidity of fixed-term contracts. However, as in the previous discussion, in order to pursue this strategy several efficiency-based preconditions should be fulfilled, since such an unqualified right may (1) expose non-terminating party (promisee) to cheating, retaliation or other strategic attempts by the terminating party to extort additional compensation; and may (2) induce the promisor to strategically or opportunistically (the hold up problem) terminate a contract.

1.2.1. Optimal risk allocation

As previously discussed, the instance of unilateral termination of a contract for an indefinite period of time (no express provision), also in cases were unilateral termination right is provided for, it should be denied if the terminating party is acting in bad faith (cheating, retaliation, opportunism, etc.) in order to reallocate already allocated risks. Recall that such behavior should be deterred since, by representing pure dead-weight social loss, it wastes scarce resources directly and indirectly.43 In order to prevent it the risk of such behavior 41

42 43

R. A. Epstein, “In Defense of the Contract at Will,” 51 University of Chicago Law Review 947, 1984, at 951. Ibid. Opportunistic termination and cheating

214

Mitja Kovač

must be internalized and hence born by the party with the discretionary unilateral termination power. Yet, it should not be disregarded that the other party may be the most efficient risk bearer with regard to achieving the eventual best efforts standard or likewise. Hence, from an efficiency point of view, a provision of “unilateral termination in good faith” should be considered and should be used as a policing device to deter such socially harmful activities.44 The next important issue that has to be addressed is the efficient assignment of risk in the instance of possible regret contingencies when a party with the discretionary power may regret and want to pursue a change in his market strategy. Also in this case the unilateral termination provision actually assigns the risk of regret upon a party which is not entitled to unilateral termination. But, since the contract provides for such a shift of risk of regret contingencies, unilateral termination is efficient. Although, as noted, whenever the terminating party attempts to shift this risk of regret, such a unilateral termination right should be subjected to severe additional limitations (good faith, good cause, etc.).

1.2.2. Relation-specific investment

The role of relation-specific investments in motivating contracting is in law and economics widely accepted. Yet, as often advanced, a party contemplating an exchange will be reluctant to make non-redeployable investments without confidence that the other party will uphold its end of the bargain when the time of performance arrives. Hence, as previously discussed a unilateral right to terminate an agreement should be, in the case of substantial relationspecific investments, subjected to several qualifications, i. e. implying a reasonable duration of contractual relationship for non-terminating party to recoup his investment. However, this investment should be substantial and beyond mere ordinary efforts to perform. Furthermore, sufficient advance termination notice periods should be considered too in order to enable non-terminating party to secure other sources of supply, to adapt his production structure, etc. In short, prior notification of the intent to terminate allows the nonterminating party to adjust to the changed situation, or to refrain from incurring further costs in reliance upon the continuation of original contract. Generally, a contractually-specified notification period should govern. Yet, if not determined by the contract itself, than it should be subject to cautious third party evaluation45 and should be no less then t – ã 46 in which non-terminating party can avoid previously noted losses.47 Finally, as in previous discussion about infinite contracts, the informal reputation effect may act as an additional mechanism enabling relation-specific investments.

44

45

46

47

Such judicial or third party enforcement might be, in absence of the reputational considerations to deter opportunism, the only effective remedy. If the notice period is provided in contract than efficiency requires enforcing this period, if nothing is stated than this advance notice period should be determined regarding the nature and duration of the relationship and the level of relation specific investments. Where t represents time when one party wants to terminate and ã is the notice time. The exact duration should be decided upon the facts of the individual case. Costs incurred in reliance upon continuation of parties’ agreement.

Variation and Unilateral Termination of an Agreement

215

1.2.3. Reducing transaction costs

In instances where a particular, actual contractual performance is very costly to measure, the breach of contract for underperformance may be very difficult to prove to a third party enforcer (court). In such circumstances, as Klein points out, parties have often an incentive to renege on the transaction by holding up the other party in the sense of taking advantage of unspecified elements of agreement.48 If this is the case, Klein convincingly shows that such unilateral termination provision is actually an implicit self-enforcing contract, which by the threat of termination of the business relationship, rather than by the threat of litigation, deters such an opportunistic behavior and thus economizes on transaction costs.49 Furthermore, Goetz and Scoot as well interpret such unilateral termination provisions as bonding arrangements, which as means of enhancing assurances of performance reduces agency costs, and parties should thus freely select such a combination of monitoring and bonding arrangements (unilateral termination rights) that will optimize total agency costs.50 Recall that such a bonding provision, according to Goetz and Scott, by substituting reassurances of performance reduces agency costs thus improves contract price.51 In addition, Masten in his model shows that a price determination might be an additional argument for upholding such contractual provisions, since even if trade involves little or no relation-specific investments such a contract is a way to economize on the cost of pricing a series of heterogenous transactions, where contracting parties may gain from bundling multiple transactions over time to reduce the need to settle on a price for each transaction in a series.52

1.3. Termination of fixed-term contracts should be prohibited This section addresses the efficiency issues associated with the unilateral termination of fixedterm contracts, either of long or short duration, where there is also no contractual provision entitling unilateral termination. In such an agreement one of the engaged contractual parties seeks to terminate the fixed-term contract unilaterally. Obviously, in such a context when one party wishes to terminate and another to continue, questions arise as to the rights, consequences and the efficiency considerations of the respective parties’ actions.53 A contract enables trust and thus fosters mutual exchange, an efficient level of relationspecific investment in the contractual enterprise, facilitating the move of scarce resources to its most valuable uses – allocative efficiency. However, it is also well established that without 48 49

50 51 52 53

Klein, supra note 23, at 356 et seq. Klein continues that such an apparently one-sided terms may be crucial elements of minimum cost quality-policing arrangements, since given the difficulty of explicitly specifying and enforcing contractually every element of quality of service or of goods (e. g. in franchisee contracts) there is in such a relationship an incentive for individual opportunistic other party to cheat the promisor by supplying e. g. a lower quality of product than contracted for; Klein, supra note 23, at 358. Goetz, Scott, supra note 13, at 1134. Ibid at 1131. Masten, supra note 40, at 2 et seq. Of course the issues of failure of counter performance or of insufficient quality or likewise are here omitted since they are not instances of termination but of repudiation. On repudiation, however, see R. Craswell, “Insecurity, Repudiation, and Cure,” 19 Journal of Legal Studies 399, 1990.

216

Mitja Kovač

enforcement a party would be able to appropriate funds that have been paid before contract performance, rendering the contract unworkable; a party might not deliver a promised good or perform a promised service; and when the price cannot be fixed in advance, inducing a price holdup where a party might bargain opportunistically about the price of transaction – reducing the value of contract or discouraging it altogether.54 More importantly, this latter problem of holdup at the stage of negotiation for performance and for payment will, as stated, result in all manner of underinvestment in the contractual enterprise.55 Hence, it is widely accepted that contract enforcement is privately and socially desirable because it spurs production and trade. Obviously, a default rule which would enable/entitle contractual party to freely, at any time unilaterally terminate a fixed-term contract, would (1) completely destroy all provided incentives, (2) enable opportunistic behavior (the holdup problem), (3) increase transaction costs and might (4) actually destroy the majority of the advanced, non spotmarket transactions. In addition, the possible benefits of discharging inefficient contracts may be far better served by applying an efficient breach – an expectation damages remedy. In short, all previously noted enforcement arguments may be employed as the basis for severe criticism of a broad general rule of unconditional termination, where parties would be able freely, at any moment to terminate a fixed-term contract, which would in the end result undermine the function of the market, and would be a source of stark inefficiencies. If a party to such a fixed-term contract, due to changed circumstances, regrets and wants to change his/her market strategy, then efficiency requires denying any right to unilaterally terminate a contract. In such instances the party should rather breach the contract and pay expectation damages. Recall, that such a remedy will provide incentives to perform if and only if it is efficient to do so; it will preserve incentives for an efficient amount of relationspecific investments, precaution and mitigation of damages; it will assign the risk of regret on the least cost risk bearer; and it will deter opportunistic behaviour and will solve the hold-up problem. Also by providing the majoritarian default rule, which the parties would chose for if they would bargain for, reduce transaction costs, it would spur allocative efficiency and hence maximize social welfare.

1.3.1. Optimal risk allocation

By entering into a fixed-term contract each party accepts to bear the risk of regret. This decision to assume voluntarily the risk of uncertain or unforeseeable events is, according to Goetz and Scott, presumably motivated by the corresponding gains that the fixed term agreement offers in terms of optimal risk allocation.56 Their analysis suggests that since the risk of the party’s regret can be assigned only by specific agreement, the risk of regret is in a fixed-term contract assigned to the party who actually regrets his agreement.57 So the party in regret should according to ex ante allocation of risk bear it, and is thus the most efficient risk bearer. Providing otherwise would mean ex post shifting the risk upon the party who is not the most efficient risk bearer. The only exemption, as already noted, would be in the instance where parties would negotiate a broad unilateral termination clause.

54 55 56 57

Shavell, supra note 8, at 297 et seq. See e. g. Williamson, supra note 12. Goetz, Scott, supra note 13, at 1144. Id.

Variation and Unilateral Termination of an Agreement

217

However, an opposite default rule enabling such a unilateral termination of fixed-term contracts would open the doors to all kinds of opportunistic behavior and cheating, having serious inefficiency implications. Namely, such a rule would provide incentives for opportunistic termination of contracts where one party would simply, after changed market conditions, in order to reallocate already allocated risk upon party who was not compensated to bear it, unilaterally terminate the contract. Recall, that such behavior should be deterred since, by representing pure dead-weight social loss, wastes scarce resources directly and indirectly.58 Furthermore, such opportunism would prevent trust, deter contracting and would also in final instance impair the operation of the market altogether.59 In order to internalize the risk of opportunism and cheating efficiency requires that no unilateral termination should be, unless contractually otherwise provided for, ever granted and parties should either perform or breach the contract and pay expectation damages.

1.3.2. Relation-specific investment

The decision, implied in the fixed term contract, to voluntarily assume the risk of uncertain or unforeseeable events invoking regret is, according to Goetz and Scott, also motivated by the corresponding gains that such a fixed-term contract offers in terms of fostering relationspecific investments.60 In other words, such a provision actually enables relation-specific investments, since it is not trivial to assume that in the absence of such a provision parties would under-invest or stop altogether. In a world of unrestricted unilateral termination of all fixed-term contracts parties’ incentives to invest (rely) would be destroyed. As stated, a party contemplating an exchange would be reluctant to make any non-redeployable investments without confidence that the other party would uphold its end of the bargain when the time of performance arrives.

1.3.3. Reducing transaction costs

As follows from previous discussion all efficiency-based objections of such a general rule culminate in significant increase in transaction costs. By not providing a majoritarian default rule, which parties would choose for if they would bargain for (no unilateral termination in fixed-term contracts), such a legal system unnecessary increases transaction costs. In short, such a hypothetical rule would thus increase costs of enforcement, monitoring, negotiating, litigation, contracting and cooperation (preventing it altogether), and would endanger the operation of advance, non spot-market exchange. The opposite rule, which does not grant such a unilateral termination right in fixed-term contracts, by providing a default rule which parties if bargained would provided for reduces 58

59

60

The expenditures on opportunistic behavior and cheating as the expenditures to prevent such activities are pure dead-weight loss and should be thus deterred. In last resort there would be hardly any parties left willing to enter, unless in the spot-market exchanges, into such arrangements, where there would be exposed to opportunism and cheating. By providing the vulnerable party with the right to prevent such a unilateral termination, such an agreement requires the disappointed party to buy out the other’s difficult-to-value specific investments in order to secure a release from a now-regretted arrangement; Goetz, Scott, supra note 13, at 1144.

218

Mitja Kovač

transaction costs (negotiation, monitoring, enforcement, litigation etc.), enables cooperation and exchange. To sum up, efficiency requires denial of any unilateral right to terminate fixed-term contract, unless otherwise provided for.

1.4. Synthesis – towards an optimal unilateral termination doctrine After economically assessing the unilateral termination of contracts, several instructive conclusions and insights may be offered. Namely, economic examination shows that, in general, contracts should not be unilaterally terminable by any of the contracting parties. Providing otherwise would increase transaction costs, provide incentives for opportunistic behavior, destroy contracting and cooperation incentives, discourage relation-specific investments, assign the risk of termination upon the non-superior risk bearer, would be a source of serious inefficiencies, would produce pure dead-weight losses and would thus significantly decrease social welfare. This holds especially for fixed-term contracts where such an option or default rule would actually impair, besides the spot-market transactions, functioning of the market. Yet, despite these severe objections there are two, economically justified instances where such a unilateral termination may be permitted. Namely, in instances of open-ended (continuous) contracts, such a unilateral right to terminate contracts should be granted due to the prohibitive costs of awarding expectation damages, to share the risk, to enable parties to pursue a change in their market strategy, and to enable flexible responses on changed economic conditions. However, as it has been specified, such a right should not be unconditional. (1) In order to deter opportunism such termination should be awarded only if it is in “good faith,” i. e. when it is not due to socially harmful behavior (cheating, retaliation, recouping periods, etc.). (2) A certain advance notice period should be given in order to enable the non-terminating party to adapt to the new, changed circumstances, to find new sources of supply, to refrain from incurring further costs (reliance) in anticipation of further continuation of contractual relationship, and so forth. Hence, in order to reduce transaction costs, an efficiency-minded legislator or judicial official should consider adopting such an optimal model rule. Furthermore, such a right should be also granted even in instances of fixed-term contracts if such a right is expressly provided for contractually. Evidently, all the aforementioned necessary preconditions should necessarily be fulfilled. Previous points can be summarized into the following outlines for an optimal default on the unilateral termination of contracts: (1) In principle contracts should not be unilaterally terminable. (2) Fixed-term contracts can be unilaterally terminated only if it is expressly provided for. If unilateral termination is expressly provided for termination should be granted only: a) if it is in good faith ( not due to socially harmful behavior),and b) if the advance notice is given. (3) In order to enable parties to pursue a change in their market strategy, and to enable flexible responses on changed economic conditions an open-ended, continuous contract for indefinite period of time may be freely unilaterally terminable: a) if such unilateral termination is in good faith ( not due to socially harmful behavior), and b) if the advance notice is given.

Variation and Unilateral Termination of an Agreement

219

2. Comment on the Draft Common Frame of Reference This section comments critically on parts of some of the articles of the Draft Common Frame of Reference, and investigates whether they are in the line with proposed, economicallybased, recommendations.

2.1. Art. III.–I.108 DCFR: Variation or termination by agreement (1) A right, obligation or contractual relationship may be varied or terminated by agreement at any time. (2) Where the parties do not regulate the effects of termination, then: (a) it has prospective effect only and does not affect any right to damages, or a stipulated payment, for non-performance of any obligation performance of which was due before termination; (b) it does not affect any provision for the settlement of disputes or any other provision which is to operate even after termination; and (c) in the case of contractual obligation or relationship any restitutionary effects are regulated by the rules (…).

This article correctly states that the right, obligation or contractual obligation can be, by parties’ agreement, varied or terminated at any time. In other words, where parties have reached the agreement (mutual consensus) that a contract is terminated or varied then also efficiency requires enforcement of such an agreement. This implies that contracts should not be unilaterally terminable by any of the contracting parties. Recall, that this is one of the leading principles derived also from economic analysis.61 Obviously the crucial condition of this article is the party’s agreement upon which the right to terminate depends. It correctly standardizes the principles of party’s autonomy, free will and contractual freedom, bans unilateral terminations and hence reduces needless transaction costs, implies risk sharing (since both parties forgone their expected profits), provides incentives for efficient amount of relational investments and deters opportunistic unilateral terminations. All these aspects bring significant economic benefits.

2.2. Art. III.–1.109 DCFR: Variation or termination by notice (1) A right, obligation or contractual relationship may be varied or terminated by notice by either party where this is provided for by the terms regulating it. (2) Where, in case involving continuous or periodic performance of a contractual obligation, the terms of the contract do not say when the contractual relationship is to end or say that it will never end, it may be terminated by either party by giving a reasonable period of notice. If the performance or counter performance is to be made at regular intervals the reasonable period of notice is not less than the interval between performances or, if longer, between counter-performances. (…)

61

Such ban economizes on transaction costs, deters opportunism, hold-ups problems and cheating, assigns the risk of changed market conditions on superior risk bearer, enables trust and cooperation, enhances relation-specific investments and contracting, and thus actually enables functioning of the non-spot market transactions.

220

Mitja Kovač

The first paragraph of this article resembles the previous economic discussion on unilateral termination upon specific provision where efficiency requires permission of such an option. Namely, it has been argued that in instances where such a right, to unilaterally terminate even a fixed-term contract, is provided by parties’ agreement, efficiency requires enforcing such a contractually stipulated right. However, there are two major shortcomings of this paragraph. One is the lack of any policing mechanism for deterring opportunism, and the other being the inaccurate condition of the notice period which could be articulated more precisely. As previously stated, such a right should not be unconditional, since efficiency requires permitting such right only if it is not due to opportunism and cheating.62 Since opportunism and cheating directly waste scarce resources and hence represent pure dead-weight social loss, an efficiency-minded legislator and judicial official should deter them by prohibiting or by refusing enforcement of such opportunistic unilateral termination. Hence, a good faith or similar behaving standard, applied as a policing mechanism for socially harmful behavior, should govern the exercise of such an option. Yet, as previously emphasized, the first paragraph of the Art. III.-1.109 DCFR contains no general good faith requirement for unilateral termination, neither is there any requirement that party should act reasonably while employing such a unilateral right. Absence of such standard may however be, while providing incentives for retaliation, opportunism and cheating, a source of stark inefficiencies and should be thus open to criticism. However, it should be noted that according to the DCFR the general good faith standard might be applicable and should thus govern the article’s application.63 In such instance, a good faith requirement is part of article’s interpretation and should, as already emphasized, govern the exercise of such an option. Regarding the requirement of previous notice one may note that a more precise criterion might be provided. Although the notice period will always be of a variable duration, and will depend on the circumstances of the case, one may improve the criteria by stating that a notice period should be reasonable, clear and unambiguous. Obviously, such a requirement of notice period is in accordance with efficiency-based recommendations, and is of the utmost importance for providing incentives for efficient relation-specific investments, enables non-terminating party to recoup his investments, to adjust to the changed situation or to refrain from incurring further costs in reliance upon the continuation of the original contract. Hence, in order to improve the diction of the article even further the standard of “reasonable, clear and unambiguous notice” should be interpreted as the time period necessary to secure other sources of supply, to adapt the production to new circumstances or to refrain from incurring further costs in reliance upon the continuation of the original contract.64 Furthermore, this economic insight enables further implication on the specific classes of contracts or relationships, since it may be argued that the notice period of the, for example, employment contract might be identical or at least very similar across the whole sector of specific employment contracts. Such an application would reduce costs of judicial administering and would thus economize on transaction costs. 62 63

64

See Section 2.3. Article III.–1:103: Good faith and fair dealing states that: “a person has a duty to act in accordance with good faith and fair dealing in performing an obligation, in exercising a right to performance, in pursuing or defending a remedy for non-performance, or in exercising a right to terminate an obligation or contractual relationship.” This interpretation might be included either in the official comment of the related provision or into the article itself.

Variation and Unilateral Termination of an Agreement

221

The second paragraph of this article provides that continuous, open-ended contracts may be unilaterally terminable by either party by giving a reasonable period of notice. This paragraph correctly states that continuous contracts should be unilaterally terminable and implicitly embodies also the efficient unilateral termination threshold (p Bo > q Bo). Such a provision provides a flexible mechanism enables responding to the occurrence of more valuable outside options, and to pursue a change in market strategy where termination of the original relationship is efficient. Moreover, such a rule by providing cost-effective monitoring, bonding mechanism and by providing a term which parties would contract for if they would bargain for reduces transaction costs, allocates risk on the superior risk bearer, and thus enhances allocative efficiency. It also correctly pronounces a requirement of giving a reasonable period of notice and states that “if the performance or counter performance is to be made at regular intervals the reasonable period of notice is not less than the interval between performances or, if longer, between counter-performances.” Yet, as previously, this condition may be improved further by interpreting it as the time period necessary to secure other sources of supply, to adapt the production to new circumstances or to refrain from incurring further costs in reliance upon the continuation of the original contract. However, this second paragraph of Art. III.–1.109 DCFR contains in the first paragraph the same major shortcoming. Namely, it omits any good faith or similar behavior standard, which would be applied as a policing mechanism for socially harmful behavior, and which should govern the exercise of such an option. From an efficiency point of view, this lack of general provision policing the socially harmful behavior is source of inefficiencies, and thus provides a clear path for statutory improvement.

Conclusion This article summarized and synthesized the economic principles on variation and unilateral termination of contracts and compared them with related provisions of the new Draft Common Frame of Reference. Economic analysis shows that related provisions of the Draft Common Frame of Reference regulating the variation and unilateral termination of fixed-term as of open-ended, continuous contracts or as the expressly provided for unilateral termination of fixed-term contracts generally, tolerably approximates with the efficiency-based recommendations. However, due to the fact that such an unilateral right should not be unconditional the obvious lack of general policing device to deter opportunism and cheating might be source of inefficiencies and hence offers a significant room for improvement, where again a path for statutory improvement is clear. Moreover, the condition of advanced notice period is sometimes omitted and may be thus also a source of stark inefficiencies, or may be, when drafted, elaborated and improved even further. Hence, one may conclude that from the scientific perspective the principles could be drafted in an even more precise way.

222

Mitja Kovač

Bibliography Arrow, J. Kenneth, Debreu, Gerard, “Existence of an Equilibrium for a Competitive Economy,” 22 Econometrica 265, 1954. Bouckaert, Boudewijn and De Geest, Gerrit (eds.), “Encyclopedia of Law and Economics,” 1999. Brickley, A. James, Dark, H. Frederick, and Weisbach, Michael S., “The Economic Effects of Franchise Termination Laws,” 34 J. L. & Econ. 101, 1991. Caffey, A. Andrew, “Franchise Tremination and Nonrenewal Legislation: Recent Developments and Trends in Legislation,” 49 ABA Antitrust Journal 1343, 1981. Comino, Stefano, Nicolo, Antonio, and Tedeschi, Piero, “Termination Clauses in Parnetrships,” Working Paper Series, Universita degli Studi di Milano-Bicocca, 2006. Cooper, Russel, “Insurance, Flexibility, and Simple Contracts,” 691 Cowles Foundation Discussion Paper, 1985. Craswell, Richard, “Insecurity, Repudiation, and Cure,” 19 J. Leg. Stud. 399, 1990. Croker, Keith and Masten, E. Scott, “Mitigating Contractual Hazards: Unilateral Options and Contract Length,” 19 RAND J. Econ. 327, 1988. Dnes, W. Anthony, “Franchise Contracts,” in Bouckaert, Boudewijn and De Geest, Gerrit (eds.), “Encyclopedia of Law and Economics,” 1999. Dye, A. Ronald, “Costly Contract Contingencies,” 26 Int. Econ. Rev. 233, 1985. Epstein, A. Richard, “In Defense of the Contract at Will,” 51 U. Chi. L. Rev. 947, 1984. Gellhorn, Ernest, “Limitations on Contract Termination Rights – Franchise Cancellations,” 1967 Duke L. J. 465, 1967. Ghosh, Shubha, “Property Rules, Liability Rules, and Termination Rights: A Fresh Look at the Employment at Will Debate with Applications to Franchising and Family Law,” 75 Or. L. Rev. 969, 1996. Goetz, J. Charles and Scott, E. Robert, “Principles of Relational Contracts,” 67 Va. L. Rev. 1089, 1981. Goldberg, P. Vicor and Erickson, R. John, “Quantity and Price Adjustment in Long-term Contracts: A Case Study of Petroleum Coke,” 30 J. L. & Econ. 369, 1987. Gray, Jo Anna, “On Indexation and Contract Length,” 86 J. Pol. Econ. 1, 1978. Grout, A. Paul, “Investment and Wages in the Absence of Binding Contracts: A Nash Bargaining Approach,” 52 Econometrica 449, 1984. Guriev, Sergei and Kvassov, Dmitri, “Contracting on Time,” SSRN Working Paper Series, 2003, Available at: http://ssrn.com/abstract=469180. Harris, Milton and Holmstrom, Benght, “On the duration of agreements,” 28 Int. Econ. Rev. 389, 1987. Hermalin, E. Benjamin, Katz, W. Avery and Craswell, Richard, “The Law and Economics of Contracts,” in Polinsky, A. Mitchell, Shavel, Steven (eds.), “The Handbook of Law and Economics,” Nort-Holland, 2008. Hillman, A. Robert, “An Analysis of the Cessation of Contractual Relations,” 68 Cornell L. Rev. 617, 1982-83. Joskow, L. Paul, “Contract Duration and Relationship-Specific Investments: Empirical Evidence from Coal Markets,” 77 Am. Econ. Rev. 168, 1987. Kitch, W. Edmund, “The Law and Economics of Rights in Valuable Information,” 9 J. Legal. Stud. 683, 1980.

Variation and Unilateral Termination of an Agreement

223

Klein, Benjamin, Crawford, G. Robert and Alchian, Armen A., “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process,” 21 J. L. & Econ. 297, 1978. Klein, Benjamin and Leffler, B. Keith, “The Role of Market Forces in Assuring Contractual Performance,” 89 J. Pol. Econ. 615, 1981. Klein, Benjamin and Murphy, Kevin, “Vertical Restraints as Contract Enforcement Mechanisms,” 31 J. l. & Econ. 265, 1988. Klein, Benjamin, “The Role of Incomplete Contracts in Self-Enforcing Relationships,” in Brousseau, Eric and Jean-Michel Glachant (eds.), “The Economics of Contracts, Theories and Applications,” Cambridge University Press, 2002. Klick, Jonathan, Kobayashi, Bruce and Ribstein, Larry, “Incomplete Contracts and Opportunism in Franchising Arrangements: The Role of Termination Clauses,” Paper 61, American Law & Economics Association Annual Meetings, 2006. Macaulay, Stuart, “Non-contractual Relations in Business,” 28 Am. Soc. Rev. 55, 1963. Martin, L. Donald, “The Economics of Employment Termination Rights,” 20 J. L. & Econ. 187, 1977. Masten, E. Scott, “Long-term Contracts and Short-Term Commitment: Price Determination for Heterogeneous Freight Transactions,” SSRN Working Paper Series, 2007, at 2. Available at http://ssrn.com/abstract=1010992. Posner, Richard, “Gratuitous Promises in Economics and Law,” 6 Journal of Legal Studies 83, 1977. Shavell, Steven, “An Economic Analysis of Altruism and Deferred Gifts,” 20 Journal of Legal Studies 401, 1991. Shavell, Steven, “Foundations of Economic Analysis of Law,” Harvard University Press, 2004. Smith, L. Richard, “Franchise Regulation: An Economic Analysis of State Restrictions on Automobile Distribution,” 25 J. L. & Econ. 125, 1982. Triantis, J. Alexander and Triantis, G. George, “Timing Problems in Contract Breach Decision,” 41 J. L. & Econ. 163, 1998. Vandenberghe, Ann-Sophie, “An Economic Analysis of Employment Law,” Ph. D. dissertation series, Utrecht University, Faculty of Law, Economics and Governance, 2004. Williamson, E. Oliver, “Markets and Hierarchies: Analysis and Antitrust Implications,” Free Pres New York, 1975.

5. Specific Areas of Contract Law

Principles of European Insurance Contract Law: Law and Economic Insights Tomas Kontautas*

1. Methodological implications The reasons for unified insurance contract law appear to be purely economic. It is stated that the diversity of national rules creates obstacles for insurers to offer insurance cover abroad1. The insurers have to adapt their products to the requirements of the law of each Member State. The law and economics scholars will obviously refer to “transaction costs” problem the insurers face in such situation2. The Project Group on Restatement of European Insurance Contract Law (the Group; hereinafter the Restatement is called PEICL, which refers to the Principles of European Insurance Contract Law) took a comparative law approach to find a golden mean between the different national rules and to present the set of legal rules fitting to insurers which conduct business under different national regimes 3. Thus, it was understood by the Group that such golden mean would be a rational solution diminishing the transaction costs. Nevertheless, the problem of transaction costs should not be stressed as a single problem. Insurers of one Member State also want to be sure that legal rules in other Member States do not preclude the running of insurance business at least at the same level of confidence the national legal regime provides; also it is important that these rules are compatible with the general actuarial and economic principles of insurance. Therefore, PEICL, as the golden mean, should pass the same test the insurer would make for the legal rules of the other Member State. The paper tries to answer the question whether certain provisions of PEICL do not create obstacles for insurability (in a short or in a long run). Criteria of insurability characterise the risks dealt with by the insurers not only from legal (or social) bus also from actuarial and market point of view. Examples of such criteria were listed in a very systematic way by B. Berliner 4 :

*

1

2

3

4

Lecturer of the Civil Law and Civil Procedure Department, Faculty of Law, Vilnius University; partner and head of the Pan-Baltic Insurance Practice Group, Law Firm SORAINEN; tomas. [email protected]. European Economic and Social Committee, Opinion on “The European Insurance Contract” of 15 December 2004, CESE 1616/2004. Neoclassical definition – costs of trading across a market. Allen, D.W. Transactions costs. Encyclopedia of Law and Economics, 0740. http://encyclo.findlaw.com/0740book.pdf. Clarke, M.A.; Heiss, H. Towards a European Insurance Contract Law? Recent Developments in Brussels. Journal of Business Law, September 2006, p. 600-607. Innovating to Insure the Uninsurable. Swiss Re. Sigma. No. 4/2005, Zurich, 2005, p. 5.

Tomas Kontautas

228 Criterion

Characteristic

1.

Category

Risk/Uncertainty

Measurable

2.

Loss Occurrences

Independent

Maximum loss

Manageable

Average loss

Moderate

3. 4.

Actuarial

5.

Loss frequency

High

6.

Moral hazard, adverse selection

Not excessive

7.

Insurance premium

Adequate, affordable

8. Market-determined

Insurance cover limits

Acceptable

9.

Industry capacity

Sufficient

Public policy

Consistent with cover

Legal System

Permits the cover

10. 11.

Societal

The effects of legal rules on insurability of insurance risks are heavily stressed in the law and economics literature5. The readers familiar with the law and economics literature will notice that the criteria also include main topics analysed by the law and economics scholars, such as adverse selection6 and moral hazard7. It should be noted that this paper does not aim to give analysis of all legal rules, which are contained in PEICL so far, as such analysis could be the task of larger scientific exercise; this paper rather looks at the issues, which are the most problematic from the subjective point of view of the author of this paper.

2. Some general remarks Most legal scholars involved in drafting Common Frame of Reference (CFR) would like to see the rules of PEICL integrated into CFR. The logic behind is quite straightforward – rules on insurance contracts form a special part of contract law as legal rules governing other specific types of contracts, therefore special rules should be compatible with general ones. However, it seems that this point of view was not fully shared by the PEICL Group. Although the draft of PEICL of 17 December 2007 has the heading named “Common Frame of Reference, Chapter III, section IX”, text of PEICL is not integrated into CFR. Article 1:105 may serve as example – it establishes a rule that the questions arising from the insur-

5

6

7

Priest, G.L. The Current Insurance Crisis and Modern Tort Law. The Yale Law Journal, Vol. 6: 1425, 1987, p. 1521-1587; GILL, M. The Expansion of Liability and the Role of Insurance – Who‘s the Ckicken? The International Journal of Insurance Law. 1998, Vol. 6, p. 27-40; Faure M.G.; and Hartlief, T. Remedies for Expanding Liability. Oxford Journal of Legal Studies. Winter 1998, Vol. 18, p. 681706. Akerlof, George (1970), ‘The Market for Lemons: Quality Uncertainty and the Market Mechanism’, 84 Quarterly Journal of Economics; 488-500. Marshall, J. M.; Moral Hazard, The American Economic Review, 1979, Vol. 66, pp. 880-890; Chandler, S. J. (1996), ‘The Interaction of the Tort System and Liability Insurance Regulation: Understanding Moral Hazard’, 2 Connecticut Insurance Law Journal, 91-177.

Principles of European Insurance Contract Law

229

ance contract, which are not expressly settled in PEICL, are to be settled in conformity with the Principles of European Contract Law (rather than with other provisions of CFR!)8. It was mentioned that one of the reasons to have a uniform European Insurance Contract Law were transaction costs the insurers face while entering different markets with their products. However, insurers selling products abroad will often involve specialised entities (e.g. intermediaries, claim managers, loss adjusters etc) and will enter into contractual relations in other Member States not only with policyholders, but also with other legal or natural persons. Having general part of CFR as a general basis for relations outside the insurance contract and having PECL as the general part to the PEICL rules means dealing with civil contracts with two different legal regimes, and this obviously does not contribute to the decrease of the transaction costs but, to the contrary, will increase them.

3. Application of PEICL (Articles 1.101-1.105) The rules of PEICL are divided into three groups: 1. Rules which are absolutely mandatory and must not be deviated from; 2. Rules which may be deviated from to the detriment only of the insurer; 3. Rules which may be deviated from also to the detriment of the other parties to the insurance contractual relations. At the moment, the current version of the PEICL rules lacks exact reference, which rules fall within the scope of 1-3 above. As regards absolutely mandatory rules, there is a comment by the draftsmen that at the moment PEICL does not contain such rules. Regarding item 3 (rules from which the deviation to the detriment of the policyholder is allowed), PEICL follows the definition of “large risks insurance” already contained in the existing EU insurance directives, i.e. it is meant that parties to the insurance contract of large risks may deviate from the provisions of PEICL. Such an idea, taking into account a purely commercial nature of the parties and the contract, obviously is welcomed. Nevertheless, it raises certain doubts why parties to the large risks insurance agreement cannot freely choose all the terms according to which the risks will be insured, and why there is a need to have the list of absolutely mandatory rules for such parties and such insurance contracts. This is especially relevant for aviation and marine insurance, being already under special treatment in some Member States9, and in international practice governed by lex mercatoria

8

9

Also, some PEICL Group members in their publications related to PEICL clearly see the Principles of European Contract Law (PECL) as lex generalis to the rules of PEICL, saying that the need for a section of CFR dedicated to insurance contract law is very low. Ideas, stating that the future of CFR is not clear, are given to justify this. See Heiss, H. The Common Frame of Reference (CFR) of European Insurance Contract Law. In Materials of the Conference European Insurance Contract Law and the Common Frame of Reference (CFR). Trier, 21-22 January 2008; Basedow, J. The Common Frame of Reference and Insurance Contract Law. In The Future of European Contract Law. Alphen aan den Rijn, 2007. P. 154. E.g. in Germany Marine insurance is dealt by several articles of the Commercial Code (HDB), and not in the general Insurance Contract Law (VVG).

230

Tomas Kontautas

– e.g. standard insurance clauses developed by the London insurance market10, and it is very unclear how these standard clauses will be in line with the rules that PEICL Group would see as mandatory. Parties to the large risk insurance11 contracts have comparable bargaining power, usually they are specialists of certain areas of business activity (insurer – of insurance, ship owner – of maritime issues) and there are no consumer (policyholder) protection justifications to limit the freedom of contract. Taking into account the insurability criteria mentioned above, there are no public policy constraints (need for the outside body intervention) to limit the right to choose on what terms the insurance cover should be offered. Therefore, from the law and economics point of view parties to the large risk insurance contract should be allowed to deviate from PEICL, and PEICL should remain just only as the set of default rules for such parties.

4. General Rules (Articles 1.201-1.204) 4.1. Definition of Insurance Contract PEICL defines insurance contract as a “contract whereby one party, the insurer, undertakes to grant the coverage of a risk stipulated in the contract to another party, the policyholder, in exchange of a premium”. Although the main elements of insurance relations are reflected in the definition, PEICL does not include into definition the economic content of insurance relations. Understanding and stressing the economic nature of insurance in the definition of insurance is crucial for insurability of insurance risks, because this gives additional incentives for the courts to look not only to the specific insurer-policyholder relationship, but also to understand the context of such relationship (insurer-policyholder being a member of a group of persons participating in the risk distribution scheme managed by the insurer). Economic theories acknowledge two elements of insurance, namely transfer and distribution of risk12. The latter is not always clear and taken into account by the lawyers. Using insurance, the policyholder not only transfers the risk to the insurer, but also a part of the society (policyholders) through insurance expresses its solidarity with a single individual (policyholder) who has suffered damage. Insurance premiums paid by other policyholders are used to indemnify the victim. Emphasis on distribution of risk is of the utmost importance for interpretation of insurance contracts. Insurers often complain about recent tendencies in insurance markets of the 10

11

12

Pavliha, M. Lectures on Marine Insurance Law. The Course Outline. IMO International Maritime Law Institute. Malta, January 2004. http://www.fpp.edu/~mlas/slo/files/IMLI-arine%20Insurance%20 Law.pdf, p. 15; Marco, R.D. Aviation insurance. 2nd ed. London, 1989, p. 113. An insurance contract covering a large risk is defined in Article 5(d) of the First Council Directive 73/239/EEC of 24 July 1973 on the coordination of laws, regulations and administrative provisions relating to the taking-up and pursuit of the business of direct insurance other than life assurance. Directive as last amended by Directive 2005/68/EC. Definition refers to insurance of risk related to aircraft, ships, also some other risks of policyholders which carry business exceeding certain size determined in the Directive. Kimball, S. L. Reflections on the Meaning of Insurance. The International Journal of Insurance Law. 1995, Vol. 2, p. 227.

Principles of European Insurance Contract Law

231

world – the courts increasingly tend to expand the limits of insurance cover by the means of contract interpretation. Such example may be the doctrine of reasonable expectations of the policyholder prevailing in the USA. Based on this doctrine, not only contra proferentem rule is taken into account, but courts tend to see the reasonable expectations that the policyholder has had regarding the scope of the cover13. Cases when excessive (and sometines blind) focus on the interest of particular policyholder, disregarding the fact that this policyholder is a member of insurer’s insurance compensation scheme, are not rare in Europe as well. If judges bore in mind that groundless decision to indemnify a policyholder will be detrimental not only to the specific insurer but also to other policyholders who participate in the insurance compensation scheme of that specific insurer, there would be less cases where the policyholders are blindly protected using a “deep pocket” approach. Therefore, taking the above into account, it would be advisable to include not only the transfer, but also the distribution of risk into the definition of an insurance contract as reflection of economic nature of insurance relations and also the tool for more careful interpretation of insurance contract terms.

4.2. Insurable interest Traditionally, many textbooks on insurance law start the introduction to the insurance contract law with the explanation of the concept of insurable interest. Insurable interest is crucial for insurance contract law as this interest is a precondition for insurance contract relations to arise (insurance relations cannot be initiated if the policyholder or a person benefiting from insurance contract has no legal and pecuniary interest in insurance indemnity/benefit). Furthermore, insurable interest is crucial for such categories of indemnity insurance as double insurance, over-insurance, under-insurance, insurance indemnity etc. Last but not least, the insurable interest, which a person benefiting from insurance should have, serves as a means to distinguish insurance from other legal relations, at a first glance very similar to insurance, such as wagering etc. When the first initial draft of this paper was written in November 2007, the only available version of PEICL at that time was the wording of 2005 placed on CoPECL intranet site. At that time PEICL had provisions on insurable interest, and the initial draft of the paper criticised exclusion of insurable interest in the sphere of life insurance. The insurable interest was replaced with the rule met in continental law countries stating that under the insurance of fixed sums the person benefiting from insurance should not prove the existence of insurable interest – “in the insurance of fixed sums it is sufficient that the insured has consented to the contract”. The initial draft of the paper indicated that such approach was rather different from the one adopted by the common law countries. The laws of these countries (e.g. England) require the existence of insurable interest also in the contract of fixed sums. The common law has developed through ages the list of relations sufficient to conclude the existence of insurable interest (e.g. spouses, creditor – debtor etc.) 14. 13

14

Keeton, R. E. Insurance Law Rights at Variance With Policy Provisions. Harvard Law Review. Vol. 83, No. 5. 1970, p. 961-985; Jerry, R.H. Insurance, Contract, and the Doctrine of Reasonable Expectations. Connecticut Insurance Law Journal. Vol. 5:1, 1998. p. 21-57. Hansell, D. S. Introduction to insurance. London, 1996, p. 165.

232

Tomas Kontautas

It was stated in the draft that total exclusion of the need of insurable interest in the sphere of life insurance has undermined the essence of insurance (persons insure themselves in order not to face negative monetary consequences). It was argued that even in life insurance such consequences are identifiable. Fixed sum paid to the beneficiary when he reaches certain age is an additional income needed for a retired person. Fixed sum paid upon the death of the head of the family is an income the family needs when the head of the family has passed away. The benefit paid to the bank in case of death of the debtor is a form of repayment of the loan. Obviously in most cases it will not be possible to assess the exact monetary value of insurable interest in life assurance as in case with indemnity insurance. However, that does not undermine the two essential features of insurable interest – it must be legal and pecuniary. The recent draft of PEICL omits regulation of insurable interest at all, possibly due to the failure to combine continental and common law approach to this core category of the insurance contract law. Such exclusion is not appropriate for a legal text that aims to be an optional instrument in the sphere of insurance contract law. Missing rules on insurable interest raise the question whether the insurance contract is not at risk of becoming a contract of speculation, resulting into asuch negative external effect as moral hazard – gaining from insurance when there is no actual loss or other type of monetary inconvenience. There are some examples in early history of insurance – e.g. England of XVIII c., when making “insurance” contract, where the death of queen was indicated as the “insured event” for a person who had no interest in queen’s life, was met in practise rather often. Therefore, issues on insurable interest should appear in PEICL again and should be described more carefully, e.g. issues when the insured is a person lacking (or of limited) legal capacity (minors, mentally ill, or whose capacity is limited due to the abuse of alcohol, drugs etc.) should be also described. Since, these persons are especially vulnerable, moral hazard towards them is more likely even in extreme forms met in the history of crimes in the sphere of insurance (e.g. making an insurance agreement with the intention to benefit from the arranged death of these persons).

5. Applicant’s Precontractual Information Duties (Articles 2.101-1.205) If the policyholder has better knowledge about his risk level than the insurer, it is difficult for the latter to assess the risk level (high or low) of the particular policyholder. The insurer has to set the premium somewhere in between the premium for high-risk policyholders and low-risk policyholders. Consequently, high-risk policyholders will pay less and low-risk policyholders will pay more. Because of the rise in insurance premiums, low-risk policyholders will decide not to purchase insurance at all; they will leave the market. This phenomenon is called adverse selection. G. A. Akerlof was one of the first to analyse the problem of adverse selection and its presence in the insurance area15. Legal rules giving incentives for proper pre-contractual disclosure are one of the best tools to cope with the negative consequences of adverse selection. The authors of PEICL took continental approach to the duty of disclosure – under PEICL it is sufficient that the policyholder fills in the questionnaire prepared by the insurer. 15

Akerlof, G.; The Market for “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Quarterly Journal of Economics, 1970, pp. 488-500, see example about medical insurance given in pages 492-494.

Principles of European Insurance Contract Law

233

Traditional argument was used – the insurer as the professional should prepare thorough questionnaire which would enable him to extract all the needed information from the policyholder. To the contrary, English law imposes rather harsh duty on the policyholder to disclose all information which influences the judgment of the prudent insurer in fixing the premium or accepting the risk16, i.e. the policyholder should guess what exact information the insurer needs to make a decision to enter into contractual relations with this particular policyholder. Such approach, criticised by contemporary English scholars, had supporters in the camp of law and economics – e.g. H. Gravelle, who holds an opinion that the terms that appear harsh ex post may be efficient ex ante due to their deterrence effect17. Although continental and common law approaches seem to be radically opposite, the solution favouring continental approach chosen does not eliminate the problem of adverse selection that the insurers want to avoid so much. Say, the policyholder fills in the questionnaire and still knows some material information that is relevant to the insurer and not reflected in the questionnaire. Therefore, the insurer will insure high risk individual for a lower premium. Optimal solution would be the combination of continental and common law approaches. The disclosure should be recognised as effected when the policyholder has duly filled in the questionnaire of the insurer, unless the insurer proves that the policyholder has intentionally withheld the information which was not required in the questionnaire, and which would be considered by a prudent policyholder as material for insurer.

6. Precontractual Duties of the Insurer (Articles 2.201-1.203) Insurance contract law has been criticised for a long time due to the absence of exact scope of information which should be disclosed by the insurer to the policyholder, and effective legal measures for insurer’s non-disclosure. To the contrary, nearly in every jurisdiction policyholder’s non – disclosure is subject to harsh remedies which may be used by the insurer. PEICL, together with a duty to provide thorough information about the cover proposed (Articles 2.401), establishes insurer’s duty to warn the applicant about the inconsistencies of the cover offered with the requirements of the policyholder, of which insurer is or should be aware. The remedies available to the policyholder for insurer’s non-disclosure are full indemnification, unless the insurer acted without a fault, and termination of the contract. The “requirements of policyholder” that the insurer has to take into account raise the concerns of adverse selection. On the one hand, the policyholder has a duty to disclose only what is expressly asked by the insurer. On the other hand, PEICL is silent about the duty of the policyholder to disclose its requirements for insurance cover which are not reflected in the questionnaire. Possibly, the authors of PEICL believed that in any case the policyholder would have incentives to state his exact needs for insurance cover. However, this might not be the case. Say, the policyholder is aware that the insurance cover offered by the insurer is not in line with all his needs. If he discloses these requirements to the insurer, the policyholder risks to get express statement that such cover is not provided. If the policyholder chooses to 16 17

Mac Gillivray on Insurance Law, 9th ed., Sweet and Maxwell, London, 1997, p. 403. Gravelle, H.; Insurance Law and Adverse Selection, International Review of Law and Economics, 1991, Vol.11, pp. 23-45, at 25.

234

Tomas Kontautas

withhold such information, in case of excluded peril the policyholder still have a chance to raise arguments that he needed such cover, and the insurer failed to take this into account and warn that such cover is not provided. Consequently, the insurer may be forced to pay, in the form of damage compensation, for the risks he did not cover. Optimal solution to avoid such situation could be the clearly stated duty of the policyholder to disclose his exact needs for insurance cover to the insurer.

7. Brief Conclusions PEICL, as a compromise between the different legal rules of the Member States, is already an achievement. Comparative law approach used by the drafters is obviously a good one to find the right balance between the different rules. Nevertheless, it should be understood that insurance does not belong only to the sphere of law because its nature is purely economic. Efficient transfer and distribution of risk requires legal environment that does not undermine the general principles of insurability. The goal of the society is to keep as many insurance risks as possible, since insurance is one of the best tools developed by humankind in order to avoid negative monetary consequences of materialisation of risks. The method of law and economics, which pays due regard to the economic nature of insurance, could give some hints how to make legal rules regulating insurance contracts more efficient. In particular, the PEICL rules should be compatible with CFR in order to minimise transaction costs; they should not intervene into the areas where no special protection for policyholders is needed; they should not depart from core insurance concepts, such as insurable interest, and, finally, they should not aim at the blind protection of one contractual party, but rather at the fair balance of contractual rights of both parties to the insurance contract in order to provide efficient framework for creation and management of risk distribution schemes.

The DCFR and Consumer Protection: an Economic Assessment Hanneke A. Luth* & Katalin Cseres** This paper will discuss consumer protection in the DCFR. It will screen the DCFR and its model of consumer protection through the lens of economic analysis. The assessment will examine the general role of mandatory rules, which justification grounds they are based on and how this general model is reflected in the specific consumer rules of the DCFR. The paper explains the economic function of mandatory rules and subjects the DCFR to the criteria of economic analysis.

1. Background The DCFR has its origin in the European Commission’s initiatives to develop some kind of horizontal instrument that would comprise a single set of harmonized private law rules. The DCFR resembles another puzzle in the development of a European “schizophrenic contract law”1 being divided into a supply side oriented contract law and a demand side oriented consumer contract law. The Commission’s Communication on European Contract Law in 20012 was the beginning of this process of designing a European Civil Code. It was followed by an Action Plan towards A More Coherent European Contract Law.3 The Action Plan called for comments on three proposals: increasing the coherence of the acquis in the area of contract law more coherent, the promotion of the elaboration of EU-wide standard contract terms, and further examination of whether there is a need for a measure that is not limited to particular sectors, such as an ‘optional instrument.’4 Its principal proposal for improvement was to develop a Common Frame of Reference (CFR) which could then be used by the Commission in reviewing the existing acquis and drafting new legislation. The CFR would be a non-binding instrument. The Commission published a follow-up Communication in October 20045 that set out the 2003 Action Plan, in the light of the reactions from EU institutions, Member States and stakeholders. In this Communication the Commission proposed that the CFR should provide ‘fundamental principles, definitions and model rules’ which *

**

1

2 3 4 5

PhD candidate, Research assistant, Rotterdam Institute of Law and Economics, Erasmus University Rotterdam, Burgemeester Oudlaan 40, 3062 PA Rotterdam, The Netherlands, e-mail: [email protected]. Associate Professor of Law, Amsterdam Center for Law & Economics, University of Amsterdam, Roetersstraat 11, 1018 WB Amsterdam, The Netherlands, e-mail: [email protected]. Mattei, U. Efficiency and equal protection in the new European contract law: mandatory, default and enforcement rules, Virginia Journal of International Law, 1999, p. 546. COM (2001) 398 final. COM(2003) 68 final. Action Plan (2003) p. 2. COM (2004) 651 final.

236

Hanneke A. Luth & Katalin Cseres

could assist in the improvement of the existing acquis, and which might form the basis of an optional instrument if it were decided to create one. Model rules would form the bulk of the CFR, its main purpose being to serve as a kind of legislators’ guide or ‘tool box’ for proposals for future legal instruments in the area of contract laws. Thus the review of the acquis and the development of the CFR has been integrated. In December 2007 the Draft CFR, which has been prepared by academics, was published. The DCFR in many respects looks like a European Civil Code without family law, succession law and company law. It is clear that the Commission is not only proposing to improve the quality and structure of the present consumer acquis but at the same time it is aiming at the adoption of a “single, simple set of rights and obligations Europe-wide” a horizontal instrument which in fact equals a European Code of consumer or contract law. The DCFR contains principles, definitions and model rules of European private law as distilled from the acquis communautaire of contract law and the private laws of the Member States. It covers both general contract law rules as well as specific ones such as rules of consumer contracts.6 The two principal groups involved in the drafting of the DCFR are the Study Group on a European Civil Code (“Study Group”) and the Research Group on the Existing EC Private Law (“Acquis Group”). The two groups employ two different methodologies to identify and formulate legal principles common to the whole of Europe. The Study Group uses an “inclusive comparative law method”, based on national experiences but transcending them: the chosen rules are said to be those “best suited to the economic and social conditions in Europe”. The Acquis Group’s method consists in “explor[ing] the existing EC law and elaborat[ing] its underlying principles”, including legislation, jurisprudence and legal literature, as well as the laws of the Member States implementing EC law. The DCFR is a comprehensive, systematic, coherent body of law at one center European level and has a static form and thus resembles a contract law code.7 However, the original goal of the CFR seems to be fading away in the shadow of the present developments of European consumer contract law. In 2007 the Commission has adopted a Green Paper on the Review of the Consumer Acquis.8 The Review is intended to modernize EC consumer law by simplifying and improving the present regulatory framework. In October 2008 the Commission has adopted a Proposal for a Directive on Consumer Rights.9 This proposal merges four existing Directives on distance selling, doorstep selling, unfair contract terms and on sale of consumer goods and guarantees into one set of rules. The Commission claims that this new directive would update and modernize existing consumer rights, bringing them in line with technological change (m-commerce, online auctions) and strengthening provisions in the key areas where consumers have experienced problems in recent years – particularly in sales negotiated away from business premises (e.g. door to door selling). Once this new directive will have been adopted the content of the DCFR will become obsolete. This paper will discuss consumer protection in the DCFR by applying an economic analysis. The analysis, primarily focuses on the economic assessment of the DCFR’s consumer protection model. The assessment will examine the role mandatory rules have been awarded in the DCFR and which justification grounds they are based on. It will, first, analyse the 6 7 8

9

DCFR, Introduction, points 37, 41, 42. Hesselink, The CFR as a source of European private law, 2008, p. 4-5. Green Paper on the Review of the Consumer Acquis, COM (2006) 744 final. Available at: http:// ec.europa.eu/consumers/cons_int/safe_shop/acquis/green-paper_cons_acquis_en.pdf. COM(2008) 614 final, 2008/0196 (COD) Brussels, 8.10.2008.

The DCFR and Consumer Protection: an Economic Assessment

237

fundamental principles of the DCFR and then deal with the image of consumer. Then it will address more specific issues of consumer protection such as the notion of consumer, standard contract terms, information duties and the withdrawal rights.

2. Consumer protection and the DCFR 2.1. The role of consumer protection in the DCFR The DCFR contains principles, definitions and model rules of European private law as distilled from the acquis communautaire of contract law and the private laws of the Member States. It covers both general contract law rules as well as specific ones such as rules of consumer contracts.10 The coverage of DCFR is thus significantly broader and resembles a civil code than the acquis, which is limited to consumer contract law. Most of the EC legislation addresses specific issues of consumer law in the form of specific mandatory contract rules. This legislation followed a vertical, case-by-case approach based on minimum harmonization and relying on information disclosure as the core instrument of regulation. The link between the Community acquis of contract law and the DCFR is that the DCFR might be used to improve the existing acquis and thus it covers the eight EC Directives that are currently under review.11 It is, however, questionable how the DCFR, which primarily aims at designing default rules with its inherent principles and aims such as freedom of contract, could be used for developing and improving a body of law primarily based on mandatory rules and thus standing on the basis of intervening with the contracting parties’ autonomy and imposing restrictions on freedom of contract. Understanding the role of consumer protection in the DCFR therefore seems indispensable. This role can be the best captured by examining the balance the DCFR establishes between default and mandatory rules. Such a balance is a key aspect of a certain private law system that tries to establish an institutional framework for an efficient market. It should also be mentioned here that the DCFR has an important potential in the process of creating European private law. In the above described development of European private law inherent parts of contract law have been separated from each other. Contract law comprises of default rules, mandatory rules and enforcement rules.12 The wide coverage of the DCFR makes it in principle suitable to “re-unite” the oddly separated parts of private law in Europe and refocus on the one fundamental question: what should be the balance between default and mandatory rules in general contract law? Consumer protection is generally understood as the protection of weaker market parties by means of mandatory rules to adjust the environment where consumers bargain and conclude transactions. When social norms do not work and thus consumers are unable to discipline firm behaviour on the market by simple mechanisms of reputation and repeat sales and 10 11

12

DCFR, Introduction, points 37, 41, 42. These Directives are Directive 1993/13/EC on unfair contract terms, Directive 1999/44/EC on consumer sales and guarantees, Directive 1997/7/EC on distance selling, Directive 85/577/EC on doorstep selling, Directive 1998/6/EC on price indication, Directive 94/47/EC on timeshare, Directive 94/314/EC on package travel, Directive 98/27/EC on injunctions. Mattei, U. Efficiency and equal protection in the new European contract law: mandatory, default and enforcement rules, Virginia Journal of International Law, 1999, p. 538.

238

Hanneke A. Luth & Katalin Cseres

they are unable to effectively enforce their rights markets need an institutional framework composed of mandatory rules to intervene. Mandatory rules alter contractual terms and impose a restriction on freedom of contract. Mandatory rules require active state intervention and therefore they restrict the general principle of freedom of contract and intervene with free market mechanisms. Mandatory rules exclude particular deals and overrule individual preferences. They aim at redistributing the transaction costs between the contracting parties and restore the perceived imbalance in economic power between consumer and supplier. Mandatory rules are often subject to criticism on three points. Mandatory rules act as a compulsory insurance policy and they raise the prices of goods and thus create costs for both business and consumers. Mandatory rules decrease competition by excluding low-quality, cheap products from the market that could be to the benefit of less affluent consumers. Mandatory rules reduce individual incentives to take care of choosing their products or services.13 Mandatory rules are accepted regulatory means to overrule individuals’ own preferences where remedy market failures created by information asymmetry, irrationality or externalities. They can protect consumers against other parties or against themselves in situations of threats, pressure or ignorance. These rules allow consumers to invalidate contracts or make certain contracts automatically void. For example, product safety standards or interest rate ceilings in consumer credit contracts are provisions consumer cannot contract out. The question is which pressure, threat and ignorance should trigger intervention and whether mere disclosure rules or mandatory standards or strict mandatory rules should be legislated. The fundamental dilemma is to strike the “right” balance between default and mandatory rules. The relationship between mandatory and default rules is characteristic of a given private law system and illustrative of its underlying economic model. This is acknowledged by the DCFR in point 20 of the Introduction. The different models of consumer protection are determined by this fundamental relationship between state intervention and free market forces. The distribution of responsibilities, i.e. rights and obligations between the state, individual consumers, consumer organizations and lobby groups as well as suppliers and their organizations is illustrative of the different models of regulatory policies. These models stand on the basis of different degree of intervention and accordingly contain different combinations of market-conforming, market-complementary and market-corrective tools.14 This contribution will focus on the economic function of mandatory rules. Economic efficiency requires the smallest possible number of mandatory rules and restricts legislative choices to the unavoidable and cost-effective mandatory rules. The economic assessment of legal rules is concerned about the outcomes of legal rules and it focuses on the consequences of regulations on the behaviour of the affected parties. It introduces a cost-benefit analysis and poses quantitative questions about regulating through mandatory rules. Economic analysis offers quantitative answers to policy questions and tries to avoid the uncertainty associated with value judgments about the fair distribution of economic benefits and about determining relative deservingness. Economic analysis can allegedly provide objective and rational benchmarks for consumer law and enforcement by offering tools to explain what the expected effects of a certain trade practices are on final consumers. Obviously, such an economic assessment has its limitations in consumer protection as there might be overriding social interests that justify derogations from the economic calculation and because it is difficult to quantify the benefits of consumer regulation, for example, 13 14

Cartwright, p. 34-35. Reisch, L.A., Principles and visions of a new consumer policy, Journal of Consumer Policy, 27, 2004, p. 20, 23.

The DCFR and Consumer Protection: an Economic Assessment

239

when health and safety issues are at stake. Still, economic assessment draws the legislator’s attention to the financial consequences of imposing restrictions on freedom of contract, tries to drive legislative choices towards cost efficient solutions and to avoid excessive regulation.15 A valuable message from economic analysis is that consumer protection comes at a cost. Mandatory allocation of risk, mandatory disclosure duties and other obligations will create additional costs to producers and sellers, who will pass these costs on to consumers where possible. Moreover, providing excessive protection can undermine the individual responsibility of the consumer and may encourage careless behaviour. This contribution will screen the DCFR and its model of consumer protection through the lens of economic analysis. The paper explains the economic function of mandatory rules and subjects the DCFR to the criteria of economic analysis. Economic analysis will rely on insights from neo-classical economics and also on recent findings of behavioural economics.

2.2. The DCFR model of consumer protection The DCFR stands on the basis of a number of fundamental principles expressed as aims. The aims and the underlying values of the DCFR include among others freedom of contract, rationality, efficiency, economic welfare and minimum intervention. These principles may conflict with each other and balancing takes place in the individual cases of the model rules.16 The DCFR essentially stands on the basis of freedom of contract and it is less concerned with distributive justice goals. Still, when such restrictions can be justified such as in cases of inequality of information or lack of bargaining power may justify mandatory rules of content.17 The main question is which market failures are identified and what kind of mandatory rules are proposed in the DCFR. Furthermore, what are the guidelines of the DCFR to justify restrictions on the freedom of contract . First, the fundamental principles of the DCFR will be reviewed and then some specific consumer rules will be examined. The role of consumer protection in the DCFR is envisaged in its aims and underlying values and in the principles as explained in the introduction. Moreover, it contains certain specific mandatory contract rules, such as pre-contractual duties to inform and remedies, right of withdrawal, cooling-off period, unfair contract terms, consumer sales, consumer goods guarantees, consumer lease, mandate. Moreover, the notion of consumer is relevant to examine in order to see what the model of consumer protection is in the DCFR.

2.2.1. Fundamental principles

Consumer protection is not among the fundamental principles of the DCFR. However, the DCFR states that it is to be interpreted and applied in accordance with the aims and 15

16 17

Economic evaluation is a valuable guideline in making choices for regulatory interventions and enforcement institutions. Ramsay, I. Framework for regulation of the consumer marketplace, Journal of Consumer Policy 8 (1985) pp. 353-372. Trebilcock, M.J. Rethinking consumer protection policy in: Rickett, C.E.F., Telfer, T.G.W. (eds.), International perspectives of consumers’ access to justice, Cambridge University Press, 2003, pp. 68-98. DCFR, point 23. DCFR, points 24, 25.

240

Hanneke A. Luth & Katalin Cseres

principles on which the Member States and the EU stands. This includes among others the protection of consumers. The role of consumer protection in the DCFR is explained with regard to the core principle of freedom of contract. At first sight, this seems to correspond to how economics approaches consumer protection. Economics treats consumer protection as a subsidiary solution to correct market failures when the private law system of individual enforcement fails and competition law control is exhausted.18 Consumer protection plays a secondary role besides the primary framework composed of competition law and contract law, because it suffers from a relevant shortcoming.19 Mandatory rules overrule individual preferences and are based on paternalism.20 In economics welfare decreasing paternalistic interventions are justified if the gains outweigh the costs of the intervention. “If a paternalistic legal intervention is capable of creating benefits for people whose freedom of choice it restricts, then there is a prima facie case for intervention on efficiency grounds.”21 One such scenario could be when individuals cannot rationally react to adequate and full information. However, determining when individuals are not able to rationally react to adequate information and when the cost-benefit balance criterion is met is a difficult question.22 In the following the DCFR’s fundamental principles referring to consumer protection will be contrasted with economic analysis.

2.2.1.1. Restrictions on freedom of contract

The DCFR stands on the basis of freedom of contract, however, it does allow deviations from this general principle. The DCFR states that restrictions of freedom of contract through mandatory rules should be imposed if they can be justified in certain situations or types of contracts.23 The DCFR mentions that in order to impose restrictions on the freedom to determine the content of contracts the DCFR lists two justifiable grounds: inequality of information and lack of bargaining power.24 Economic analysis indeed acknowledges that law can increase the efficient use of resources by creating rules of conduct that correct market failures. Potential market failures could be lack of competition, entry barriers, product differentiation, information failures or 18 19

20

21 22

23 24

Ramsay (1985) p. 356. The function of competition law and contract law is to enable and regulate contracts, however, they effect different aspects of contracts. They form market institutions that have a number of efficiency functions. Contract law supplies default rules that enhance efficiency by reducing transactions costs of time and negotiation and it facilitates well-informed exchange of well-defined property rights and results in welfare-enhancing exchanges for the parties. It helps to eliminate opportunistic behaviour through enforcement rules. Mattei, U. Efficiency and equal protection in the new European contract law: mandatory, default and enforcement rules, Virginia Journal of International Law, 1999, p. 538. Paternalism is essentially the usurpation of one person’s choice of their own good by another person. D.Archard, Analysis, Vol. 50, No. 1 (Jan., 1990), pp. 36-42. Burrows 1998: 541. A.I. Ogus, regulatory paternalism:when is it justified? In: K. Hopt, E. Wymeersch, H. Kanda and H. Baum (eds.), Corporate Governance in Context: Corporations, States and Markets in Europe, Japan and the U.S (Oxford , Oxford University Press, 2006 p. 305-307. DCFR, Introduction, points 24, 25. DCFR, Introduction, point 27.

The DCFR and Consumer Protection: an Economic Assessment

241

third-part effects, the problem of externalities. Market power and entry barriers are issues that competition law deals with. Transaction costs, externalities, information deficits and cognitive dissonances are accepted arguments to justify intervention through mandatory rules. Economic analysis of consumer protection accepts information failures and failures of the institutional framework of enforcement as the most valid economic reasons to intervene and alter contractual conditions. It identifies the sources of information deficiencies and thus consumer harm either in the lack of competition or in the fact that information is unavailable or not costless or in uncertainty of individuals about the quality of product characteristics. Uncertainty can originate from incomplete or inadequate information. This can lead to information asymmetry problems, where one party possesses the information about a certain product or service characteristic whereas the other party does not. Information deficit on the consumer side is usually connected with consumers’ uncertainty about the quality of products, which cannot be assessed at the moment of purchase.25 When consumers cannot assess the quality of a certain product, other product characteristics like price will dominate the decision making process. Consumers’ uncertainty about the quality of experience or credence goods creates the danger of adverse selection or moral hazard and ultimately leads to quality deterioration.26 Neoclassical economics provides insights on how information affects the dynamics of markets, the determinants of bargaining and drives regulatory approaches of consumer protection to a cost-benefit based analysis. The assumptions of neoclassical economics, however, treat key aspects of consumer decision making as exogenous and as such can say little about the amount, the nature of and the way information should be framed and disclosed to consumers. Behavioural economics based on empirical research deals with endogenous aspects of consumer decision-making and has questioned present policies for consumer protection. Insights from neo-classical economics on information asymmetry, adverse selection, moral hazard, transaction costs as well as recent findings of behavioural economics about systematic cognitive errors of consumers’ decision making have relevant implications for designing new regulatory frameworks such as the DCFR.27 The empirical observations of behavioural economics exhibit institutional constraints on individual choice. The theory of bounded rationality argues that the capacity of the human mind to conceive and process complex information is relatively limited. Research in 25

26

27

Goods can be classified based on the consumer’s ability to determine quality into search goods, experience goods, and credence goods. Credence goods include the purchase of services offered by liberal professions, like lawyers or physicians. It can sometimes be even more difficult to ascertain the quality of these services. The consumer is forced to trust the supplier of the service as it is impossible to assess the quality of the good through the repeat purchase mechanism. Search goods quality is easily evaluated before purchasing the good, experience goods can be evaluated upon consumption, whereas the quality of credence goods will not be clear even after buying the product. Nelson, p. Information and Consumer Behavior, 78 Journal of Political Economy 2, 1970 pp. 311329; Darby, M.R. and Karni, E. Free Competition and the Optimal Amount of Fraud, 16 J. Law & Econ, 1973 pp. 167-88. Akerlof, G. The market for ‘lemons’: quality uncertainty and the market mechanism, 84 Q. J. Econ. 488, 1970 pp. 487-500. Akerlof, G. The market for ‘lemons’: quality uncertainty and the market mechanism, 84 Q. J. Econ. 488, 1970 pp. 487-500; Stigler, G.J. The economics of information, JPE 1961; Simon, H.A. A Behavioral Model of Rational Choice in: Models of man: social and rational: mathematical essays on rational human behaviour in a social setting, New York: Wiley 1957.

242

Hanneke A. Luth & Katalin Cseres

behavioural economics found that consumers may rely on heuristics instead of being guided by rationality.28 The new research showed that decision-makers systematically fail to deal with information even in situations where the market does not have difficulty in producing the socially optimal amount of information or in distributing it efficiently.29 In these situations consumer harm is a result of decision-making heuristics that cause consumers to underestimate benefits of searching and switching. For example, in case of endowment bias consumers value more what they have than what they might have or in case of hyperbolic discounting when they rationally do not weigh present gains against future benefits and put too much weight on the immediate. Framing biases originate from the specific ways objective information is provided.30 However, behavioural economics does not provide solid evidence to support a turn back to paternalistic regulation. The evidence is not yet conclusive with regard its impact in real world consumer markets. Still, its message is clear: information asymmetry cannot be solely offset by regulating the mandatory provision of information. The search costs of gathering and evaluating more information to improve a decision about buying a product or a service have to be weighed against the benefits of that added information.31 The search costs of acquiring and using information and the opportunity costs of time have to be weighed against the expected benefits. Neoclassical economics encourages more and better information to remedy market failures on the demand side, such as mandatory disclosure or third-part certification. However, more disclosure may conflict with competition policy as it might lead to collusion and it may not be as useful for consumers either. Behavioural economics suggests remedies aimed at framing effects and thus steer consumers’ choices towards welfare enhancing options.32 Paternalistic guidance towards certain options through framing the way information is provided could assist consumers to de-bias their decision-making and to channel their decisions to socially beneficial options.33 The DCFR does recognize that information failures are the main justifications to restrict freedom of contract and states that information on the characteristics of goods and services as well as the terms of the contracts could be deficient. The DCFR is, however, not aware of the costs of protection and in particular the possible consequences consumer protection regulation might have on the behaviour of respective parties. The DCFR acknowledges that consumer law is a more effective instrument of consumer protection when it prevents or deters unlawful conduct rather than provides a remedy for loss 28

29

30

31

32

33

Poiesz, Th.B.C. The free market illusion: psychological limitations of consumer choice in Market regulation: lessons from other disciplines, Ministry of Economic Affairs, 2004, The Hague, p. 17 Ulen, T.S. Information in the market economy – Cognitive errors and legal correctives in: Grundman, S., Kerber, W., Weatherhill, S. (eds.), Party autonomy and the role of information in the Internal Market, de Gruyter, Berlin, 2001, p. 98-99, 105. OECD, Roundtable on economics for consumer policy, Summary Report, DSTI/CP(2007)1/FINAL, p. 12. This is referred to as rational apathy, a term originating from political science literature explaining why people do not vote. R. Sunstein and R.H. Thaler, “Libertarian Paternalism Is Not An Oxymoron” (2003) 70 The University of Chicago Law Review, Fall, 1159. OECD Roundtable discussion on private remedies: class action/collective action; interface between private and public enforcement, United States of America DAF/COMP/WP3/WD(2006)34, p. 18

The DCFR and Consumer Protection: an Economic Assessment

243

or damage. The advantage of preventive measures such information disclosure is that they can avoid the social costs of compensating consumers after the event and that they focus on collective consumer interests, while remedial consumer law is aimed at the compensation of individuals. The DCFR is right that this can be achieved by primarily focusing on information provisions about price, quality, the nature and consequences of consumer transactions. However, the DCFR ignores the economic insights of information transmission that explains the optimal theory of duty to provide information. In sum, this theory accepts the duty to provide information when it can be provided by the cheapest cost producer of information, the information is sufficiently valueable as compared to its communication costs, it is unlikely that the other party already has that information, the information is not entrepreneurial and does not consist of mere opinions and non-verifiable statements.34 The DCFR, furthermore, does not elaborate on the situation when consumers are informed and market failures still occur. When consumers decision-making process fails despite the available information other remedies are needed that make decision-making optimal.

2.2.1.2. Economic efficiency and distributive justice

The welfare criterion of the DCFR is that of general economic welfare and efficiency. The DCFR aims at increasing general welfare by strengthening market forces and individual welfare while maintaining efficient solutions.35 Economists traditionally favour a total welfare standard on the basis that it generates the most for society as a whole and strives for the maximization of efficiency. The total welfare standard stands for allocating resources to those who value them most and it takes account of both allocative36 and of productive efficiency.37 It, furthermore, treats wealth distribution between consumers and producers neutrally. Economists consider the consumer welfare standard as arbitrarily favouring one group over another, at the same time impeding the maximization of efficiency, innovation, competitiveness and economic growth. The consumer welfare criterion lacks a firm foundation in economics. In welfare economics equal gains will yield equal increases in utility and these will have equal effects on social welfare. According to the consumer welfare standard utility transferred from consumers to producers will not improve total social welfare, although it will make someone better off. This standard discriminates between individuals in different interest groups as it assigns zero weight to seller-shareholder profits and actually disregards the fact that gains to sellers, producers and shareholders can be socially positive. As the consumer welfare approach considers wealth transfers from consumers to producers as being rather harmful than neutral, it is more critical of efficiency claims.38 34

35 36

37

38

Kronman, T.A. Mistake, disclosure, information and the law of contracts, Journal of Legal Studies, Vol.7, 1-34 (1978); De Geest, Kovač, The formation of contracts in the consumer acquis. DCFR, Introduction, point 29. Allocative efficiency is achieved when resources are distributed among alternative uses so that that the goods and services produces are those most highly valued by consumers. It ensures that goods and services are produced that consumers want as shown by their willingness to pay and it maximizes the consumption value of the existing stock of goods of social wealth. Production efficiency is attained when products and services are produced at the lowest possible cost using a minimum of resources under existing technology. Duhamel, Townley (2003) p. 18-19; Piaskoski, Finkelstein (2004) p. 280-281.

244

Hanneke A. Luth & Katalin Cseres

Contract law is an efficiency-oriented policy and therefore apt to target and promote the overall economic welfare of society instead of making value judgments on how such economic welfare should be distributed between different social groups. There are other public policies that are better suited to address the distribution of income on the basis of fairness and relative deservingness such as taxation or consumer protection.39 Moreover, while it could be argued that real world markets do not correlate with the theoretical assumptions of economic theory, a contract law focusing on pure efficiency arguments has an important virtue: efficiency is relatively objective and predictive as compared to equity. It avoids the uncertainty associated with value judgments about the fair distribution of economic benefits and about determining relative deservingness.40 While general contract law stands on the basis of a total welfare standard, with regard to consumer protection the consumer welfare standard seems logical. Consumer welfare is generally defined as the maximisation of consumer surplus, which is the part of total surplus given to consumers. This is realised through direct and explicit economic benefits received by the consumers of a particular product or service as measured by its price and quality. Consumer rules are not only based on economic efficiency but also equity is a fundamental principle. Consumer welfare is expressed in both economic and non-economic aspects within the realm of consumer protection and it has almost always a social justice component as well. The DCFR is concerned with corrective justice, but it is “less concerned with issues of ‘distributive justice’, but sometimes distributive or ‘welfarist’ concerns may also be reflected in the DCFR, for instance when it is decided that a consumer should always have certain rights.”41 By rejecting the aim of distributive justice the DCFR seems to correspond to what economic analysis says. The economics literature is critical to legal rules and institutions of consumer protection that follow a redistributive approach. The justification for mandatory rules has often been based on the belief that consumers are faced with disadvantages on the marketplace and that their unequal position can be ameliorated by different forms of regulation. In the legal literature mandatory rules of consumer protection are justified by values such as distributive justice or fairness. The legislative reaction to such situations of adhesion often has been the awarding of consumer rights through regulation, which aims at redistribution of power or resources. 42 The so-called power distribution43 or weakness 44 argument contends that it is necessary to protect weak consumers in order to reinforce their bargaining position the inequality of bargaining power between market parties can be remedied by redistribution through legal rules. This regulatory approach holds that the legislator can regulate the environment by means of mandatory rules and to redistribute costs between the contracting parties and to restore equality between them.45 It has, however, been proved that contract law is unfit to accommodate redistributive goals.46 Distributive justice concerns redistribution of power and resources on the basis of what is just instead of what is economically efficient. Its object is to shift wealth from one group 39 40 41 42 43 44 45 46

Farrell, Katz, The economics of welfare standards in antitrust, p. 9-10. Farrell, Katz (2006) p. 9. DCFR, Introduction, points 24. A.T.Kronman, Paternalism and the law of contracts, Yale Law Journal, April, 1983, pp. 770-772. Reich (1991) p. 262. Rekaiti, Van den Bergh (2000) p. 374. Rekaiti, Van den Bergh (2000) p. 373. Kaplow, Shavell (2002), Craswell (1981), G.Wagner (2007).

The DCFR and Consumer Protection: an Economic Assessment

245

to another by altering the terms under which individuals are allowed to contract in order to achieve fair division of wealth among the members of society. The inequality of bargaining power between traders and consumers as also mentioned in the DCFR has been used to justify distributive measures. However, such measures disregard the fact that consumers do not form a homogenous entity and thus such measures will have different distributional effects on different consumers benefiting some and making worse off others. Therefore, legislators should be cautious with imposing measures without taking account of their distributive effects and decide to whom and from whom distribution should take place. Such measures may not improve the position of consumers and prove ineffective or even perverse as a result of passing on the costs of protection to consumers as well as considering the costs of administration and enforcement. 47 It has often been argued that distributive effects can be better achieved by the tax system than detailed regulation of individual transactions.48

2.2.1.3. Price of consumer protection: distributional effects

The economics literature focuses on the consequences of legal rules, especially on the consequences for behaviour of the affected parties. A valuable message from economic analysis is that consumer protection comes at a cost.49 Mandatory allocation of risk, mandatory disclosure duties and other obligations will create additional costs to producers and sellers, who will pass these costs on to consumers where possible.50 A further concern of cost reluctant consumer protection is that it can undermine the individual responsibility of the consumer and may encourage careless behaviour. Moreover, it may cause consumer moral hazard problems, implying that consumers take advantage of certain obligations imposed on firms and professionals. An example of consumer moral hazard would be buying a certain good from a doorstep seller with no intention of keeping it longer than the cooling off period, but using it in the meantime.51 Consumer protection is not a zero-sum game; it has its price. Mandatory rules implementing consumer rights generate additional costs for business that will likely be passed on to consumers. This cost increase will have to be balanced with the increase in demand and supply of goods and services in order to prove that new rules are indeed necessary. The costs of a market failure have to be balanced with the costs of intervention and government failure and the impact of that remedy on the behaviour of affected entities should be estimated. These are costs of legislation, enforcement and compliance as well as potential indirect effects of intervention on consumer and business behaviour. For example, competition might 47

48 49

50

51

Craswell shows when efficiency and distributional concerns may converge depending on how consumers value certain legal rules. Craswell, R, Passing on the Costs of Legal Rules: Efficiency and Distribution in Buyer-Seller Relationships, 43 Stanford Law Review 361-398 (1991); Cartwright, 2001, p. 29. A.T. Kronman, Contract law and distributive justice, Yale Law Journal, April, 1980, p. 472. See for general discussion: Hartlief, T. Freedom and Protection in Contemporary Contract Law, 27 J. Cons. Pol. 2004, pp. 253-267. The extent to which this is possible depends on the price-elasticity of the demand for this particular product or service, see Cooter R, Ulen T. 1996. Law and Economics. New York: Harper Collins. 2nd ed., pp. 29-30. Rekaiti p. , R. Van den Bergh, Cooling-Off Periods in the Consumer Laws of the EC Member States. A Comparative Law and Economics Approach, 23 J. Cons. Pol., 2000, pp. 371-407, at 382-383.

246

Hanneke A. Luth & Katalin Cseres

be reduced as a result of decreasing the number of low-quality and cheap products or services on the market. Furthermore, the distributional impact of the costs and benefits among the different groups has to be considered. Intervention has its subsequent costs for governments and the implementation of new rules create compliance costs for business but also raise prices for consumers. As far as demand elasticity allows price increases business will pass on the costs of compliance i.e. increased protection to consumers. This means that consumers pay for their own protection. For example, an information disclosure imposed by the state to reduce search costs has different effect for different consumers. Such an intervention would raise firms’ average costs and accordingly leads to higher consumer prices. While this would reduce search costs for consumers who actively shop around, passive consumers would have to bear the increased costs without extra gains. The intervention is optimal only in case the gain to active consumers (reduced search costs minus the increased price) outweighs the losses to passive consumers and the administration of the law. 52 Two illustrations from the case-law of the ECJ. In Club-Tour53 the ECJ provided a broad interpretation of a package in Article 2 (1) of the Directive on package travel. The ECJ’s broad interpretation of package including holidays organized by travel agencies at the request and according to the specifications of a consumer or a defined group of consumers generates costs for travel agencies who have to take insurance to cover the risk of insolvency under Article 7 of the Directive. As a consequence many small travel agencies will have to leave the market not having the appropriate degree of creditworthiness for such an insurance or pass those costs through to consumers. In easyCar54 the ECJ was asked by The UK High Court whether Article 3 (2) of the Directive on distance contracts applied to car hire services. AG Stix-Hackl argued that such services operate with reservation as a precondition and thus would be unreasonably affected by the requirements of the Directive and would generate opportunity costs.55 The right of cancellation generates extra costs that eventually the consumer has to bear.56 Economics is concerned about the total welfare of both sellers and buyers without distinguishing between benefits to sellers or buyers. For non-economists such an approach ignores the distributional effects of gains and losses. Efficiency and distributional concerns might coincide when all buyers value the legal rule identically, but when different buyers value the rule differently then legal rules have different distributional effects on different subgroups of consumers. Craswell shows that the extent to which sellers’ are able to pass on costs or benefits to consumers affects the distributional goals of legal rules. The relevant question he poses is when a mandatory rule has to be imposed on what basis such a legal rule should be selected. Craswell argues that the distributional approach will coincide with the efficiency approach. As sellers pass on the costs of a new rule to consumers, consumers will benefit

52

53

54 55 56

Craswell, R, Passing on the Costs of Legal Rules: Efficiency and Distribution in Buyer-Seller Relationships, 43 Stanford Law Review 361-398 (1991); Schwartz, Wilde (1979) p. 668-671. Case C-400/00, Club-Tour, Viagens e Tursimo SA v. Alberto Carlos Lobo Gonçalves Garrido, [2002] ECR-I-4051. Case C-336/03 easyCar (UK) Ltd. V.Office of Fair Trading, [2005] ECR I-1947. Opinion of AG Stix-Hackl delivered on 11 November 2004, Case C-336/03, paras 59-68. Unberath H., Johnston A., The Double-Headed Aproach of the ECJ Concerning Consumer Protection, Common Market Law Review 44, 2007.

The DCFR and Consumer Protection: an Economic Assessment

247

when their direct benefits exceed the costs of the rule to the sellers. The crucial question is how much of their costs sellers can pass on to consumers.57 The pass on rate depends on the market structure, the elasticity of demand and on the fact that heterogeneous buyers value a certain legal rule differently. When consumers are heterogeneous the willingness to pay for higher priced product or service of marginal and infra-marginal consumers are decisive. Marginal consumers will stop buying the product or service when price increases and therefore they determine how high the price of the product or service will be. Those consumers who gain from a legal rule can compensate those who do not or even get harmed by the higher prices.58 Craswell concludes that when sellers can pass on much of their costs of a legal rule this indicates that consumers value the rule and benefit from the rule. When sellers cannot pass on much of their costs then consumers found the rule less attractive. When consumers are homogeneous then a comparison of the costs and benefits of the rule to sellers and consumers suffices to establish whether consumers benefit from the rule. When consumers are heterogeneous then the redistribution among consumers who benefit and who lose has to be examined. But even in this case it is the preferences of different consumer groups which are decisive and not the ability of sellers to pass on their costs.59 In sum, the DCFR could reflect this cost conscious approach in order to strike the balance between mandatory and default rules. This costs-benefit analysis provides an objective standard for the balancing of freedom of contract and its restrictions and makes justification transparent and evidence based. For example, the DCFR puts forward that interventions may promote efficiency when they target a market failure, for example inequality of information that makes the otherwise concluded agreement inefficient. Consumer protection rules, for example, can be seen not only as protective for the benefit of typically weaker parties but also as favourable for general welfare because they may lead to more competition and thus to a better functioning of markets … [ ] … Rules that, in relation to the making of a contract of a particular type or in a particular situation, require one party (typically a business) to provide the other (typically a consumer) with specified information about its nature, terms and effect, where such information is needed for a well-informed decision and is not otherwise readily available to that other party, can be justified as promoting efficiency in the relevant market. Indeed a legislator should consider whether this is the justification for the proposed intervention, or whether it is based on a welfarist notion that consumers simply should have the right concerned.60

The DCFR is unclear at this point: what is meant by efficiency promoting or welfarist mandatory rules? It should have explicitly put forward a gradual cost-benefit test for intervention: interventions through either information disclosure rules or by mandatory rules may in certain situations be welfare enhancing options when no other market based tools can remedy market failures. A clear benefit of information provision as opposed to for instance mandatory quality standards is that it will not lead to a decrease in choice and it stimulates the individual responsibility of the consumer. The choice for the form of intervention should 57 58 59 60

Craswell (1991) p. 361-366. Craswell (1991) p. 373-377. Craswell (1991) p. 398. DCFR, Introduction, point 30.

248

Hanneke A. Luth & Katalin Cseres

be clearly linked to the underlying costs and benefits of those choices when it is decided whether mandatory rules of substance or mere information disclosure suffice or even more interventionist substantive consumer rights should be awarded. Furthermore, the DCFR puts forward that if intervention is justified it should be limited to a minimum. The DCFR explains that when the form of intervention is decided information disclosure and the right of withdrawal should be preferred if the market failure can be remedied in such a way. Should such information disclosure prove insufficient then “the interference with freedom of contract should be the minimum that will solve the problem while providing the other party (e. g. the business seller) with sufficient guidance to be able to arrange its affairs efficiently.”61 The DCFR does not explain how intervention can be kept to the minimum. Economics offers a benchmark by comparing the costs and benefits of market failures with those of government failures. Intervention through remedying market failures has its costs of compliance, rule formulation, enforcement and indirect effects on other market behaviours. Mackaay argues that public legal rules should substitute private rules where that option leads to costs savings. He puts forward the cost minimisation formula developed by Wittman. He argues that the role of contract law is to minimize costs of the parties and of the courts writing contracts and the costs of inefficient behaviour arising from poorly written or incomplete contracts.62 In other words, economics asks whether regulation has a cost advantage over parties’ self-protection. Regulation often aims to restrain opportunism, but pointing to this opportunism might suffice as well instead of direct regulatory measures. The DCFR continues that similarly with contract terms: it must be asked whether it is necessary to make a particular term mandatory or whether a flexible test such as ‘fairness’ would suffice to protect the weaker party. A fairness test may allow certain terms to be used providing these are clearly brought home to the consumer or other party before the contract is made. The fairness test thus interferes less with the parties’ freedom of contract than making a particular term mandatory would do.63

While standards are more open norms than rules and therefore intervene less with market mechanisms, criteria such as fairness and good faith are too vague to form a solid benchmark for regulation. Fairness test might seem flexible vis-à-vis direct regulation of contractual terms but it is a broad and general standard which can lead to diverging outcomes in national legal systems as a result of different implementation methods and different interpretations. The same applies to the principle of good faith. For example, in Article II.–1:102: on party autonomy the DCFR states that “(1) Parties are free to make a contract or other juridical act and to determine its contents, subject to the rules on good faith and fair dealing and any other applicable mandatory rules.” Alternatively, in Article II.–9:404: The meaning of “unfair” in contracts between a business and a consumer is explained

61 62

63

DCFR, Introduction, point 28. Wittman, D.A. Economic foundations of law and organization, Cambridge, Cambridge University Press, 2006. DCFR, point 28.

The DCFR and Consumer Protection: an Economic Assessment

249

In a contract between a business and a consumer, a term [which has not been individually negotiated] is unfair for the purposes of this Section if it is supplied by the business and if it significantly disadvantages the consumer, contrary to good faith and fair dealing.

Good faith and fairness provide no valuable criteria for intervention: how should one decide when a transaction as a result of intervention is fair and how fair a consumer transaction should be? A fairness argument based intervention relieves consumers from their own responsibility to take care of their transactions and reduce their incentives to take care of the consequences. The DCFR’s statement on party autonomy is therefore in striking contrast with its underlying value of freedom of contract and economic efficiency. Party autonomy based on good faith and fair dealing has a strong paternalistic flavour without a cost-benefit ratio. Moreover, such general standards can have significant disadvantages in the implementation phase as compared to specific legislation. Specific, more detailed legislation can have a better direct effect on market participants’ behaviour than general rules, which might be difficult and costly to interpret. General clauses require a strong enforcement mechanism in order to operate properly and to have any influence on the working of the market. Otherwise the strong protection offered by legislation is unenforceable and can outlaw detailed national legislation without providing a sufficient replacement. While these general standards might prove less restrictive of market processes, at the same time they can incur high monitoring and enforcement costs.

2.3. The image of the consumer in the DCFR The normative dimension of consumer protection becomes clear through the “image” and the concept of consumer adopted in a specific legal system or piece of law. The DCFR refers to consumers as the “typically weaker parties”.64 Further, the DCFR refers to the average consumer in Article II.–3:102: on specific duties for businesses marketing goods or services to consumers and to the vulnerable consumer in section II.–3:103: on duty to provide information when concluding contract with a consumer who is at a particular disadvantage. In Article II.–3:102 the DCFR in this provision obliges business to the disclosure of information “as the average consumer needs in the given context to take an informed decision on whether to conclude a contract”. The average consumer is an unclear and indefinite notion and it lacks a statistical basis.65 Therefore, it gives little guidance for choosing an overall benchmark and it is open to largely diverging interpretations in national jurisdictions. A model of expected consumer behaviour is significant as it is the underlying basis for regulatory choices. As there is no further explanation of the average consumer in the DCFR the case-law of the ECJ and the Directive on Unfair commercial practices66 can serve as a background for discussion. 64

65

66

“Consumer protection rules, for example, can be seen not only as protective for the benefit of typically weaker parties but also as favourable for general welfare because they may lead to more competition and thus to a better functioning of markets.” DCFR, Introduction, point 30. Gomez, F. The unfair commercial practices Directive: a law and economics perspective, ERCL 1/2006, p. 26-27; Gomez, F. the harmonization of contract law through European rules: a law and economics perspective, ERCL 2/2008, p. 113. Directive 2005/29/EC on Unfair Commercial Practices.

250

Hanneke A. Luth & Katalin Cseres

The Directive on Unfair commercial practices also takes the benchmark of the average consumer. The Directive refers to the average consumer when the assessment of unfairness (misleading or aggresssive) of commercial practices has to be decided. A practice deems to be unfair when either it is contrary to the requirements of professional diligence or when it it materially distorts or is likely to materially distort the economic behaviour with regard to the product of the average consumer whom it reaches or to whom it is addressed. In other words, the perceptions of the average consumer is decisive in which practice deems to be misleading or aggressive.67 The Directive defines the average consumer as a reasonably well-informed and reasonably observant and circumspect person, taking into account social, cultural and linguistic factors, as interpreted by the Court of Justice. The Directive emphasizes that the average consumer test is not a statistical test. National courts and authorities will have to exercise their own faculty of judgment, having regard to the case-law of the Court of Justice, to determine the typical reaction of the average consumer in a given case.68 This definition is still too vague and leaves wide margin of discretion to national jurisdictions. The case-law of the ECJ governed by the image of the responsible consumer, while the directives refer to the confident consumer. The notion of the average consumer in EC law can be traced back to the ECJ’s judgment in Gut Springheide69 and has developed through a number cases70 of the ECJ and has been defined as a reasonably well informed and reasonably observant and circumspect person. The ECJ placed a great deal of confidence in consumers’ ability to process information and therefore it has given preference to rules that require information disclosure as part of intervention in the market. The Court’s perception of the consumer’s abilities to process information was clearly expressed in a number of cases concerning free movement of goods under Article 28 EC. For example, in Pall Corp,71 Clinique,72 Mars73 and Procureur de la Republique v. X.,74 the ECJ has condemned national rules for alleged consumer protection as being over-regulatory by arguing that the consumer can

67

68

69

70

71 72 73

74

Article 5 (2) of Directive 2005/29/EC Recital 16 and Article 8 further explains that commercial practices are aggressive when they significantly impair or is likely to significantly impair the average consumer’s freedom of choice or conduct with regard to the product and thereby causes him or is likely to cause him to take a transactional decision that he would not have taken otherwise. Recital 18 of Directive 2005/29/EC on unfair business-to-consumer commercial practices in the internal market. Case C-210/96 Gut Springenheide GmbH and Rudolf Tusky v Oberkreisdirektor des Kreises Steinfurt – Amt für Lebensmittelüberwachung [1998] ECR I-4657, paras 31, 32. The Court clearly stated how it perceives the consumer. It argued that “… in order to determine whether a particular description, trade mark, promotional description or statement is misleading, it is necessary to take into account the presumed expectations of an average consumer, who is reasonably well informed and reasonably observant and circumspect.” Case C-470/93, Mars, [1995] ECR I-1923, para 24; Case 220/98, Estée Lauder Cosmetics, [2000] ECR I-117, para 27; Case 303/97, Verbraucherschutzverein eV v.Sektkellerei G.C. Kessler GmbH und Co, [1999] ECR I-513. Case C-238/89 Pall Corp. v p. J. Dahlhausen & Co. [1990], ECR I-4827. Case C-1315/92 Verband Sozialer Wettbewerb eV v. Clinique Laboratories SNC, [1994] ECR I-317. Case C-470/93 Verein gegen Unwesen in Handel und Gewerbe Köln eV v. Mars GmbH [1995] ECR I-1923. Case C-373/90 Procureur de la Republique v. X [1992] ECR I-131.

The DCFR and Consumer Protection: an Economic Assessment

251

take care of his own interests. In these cases the ECJ relied on the “reasonably circumspect consumer” who is able to process information and make informed choices.75 The economics literature primarily relies on the rational choice theory. Rational choice theory is a coherent theory of human decision, which starts from the presumption that consumers have transitive preferences and seek to maximize the utility that they derive from those preferences. Choice is considered to be rational when it is deliberative and consistent and reasonably well-suited to the attainment of the goals of their choices.76 Individuals are assumed to be the best judges of their own welfare, often referred to as consumer sovereignty.77 Consumers are sovereign in the sense of being able to define their needs concerning goods and services, to send a message of their needs to the market and to the producers and to satisfy those needs at a reasonable price and by choosing good quality.78 Consumer sovereignty is regarded as the right and the power to decide about producers’ success or failure through the exercise of consumers’ freedom of choice. Consumer sovereignty is a normative standard against which the performance of markets can be judged. The market rules should guarantee that consumer preferences are the ultimate control of the process of production.79 Consumers know their own preferences and needs the best, while regulators may not be able to protect them properly. Social norms such as reputation and repeat sales are regarded as an

75

76

77

78

79

In GB-INNO the Court was not convinced that a Luxembourg law was protecting consumers against confusion by prohibiting special offers of reduced prices stating the duration of the offer and the specification of the previous price. The repression of consumer information in the interest of consumers was clearly rejected by the Court. In Yves Rocher the Court affirmed the relevance of market or product-related information. It condemned a provision of the German law on unfair competition prohibiting individual price comparisons, except where these were eye-catching. The Court stated that “… the prohibition in question goes beyond the requirements of the objectives pursued, in that it affects advertising which is not at all misleading and contains prices actually charged, which can be of considerable use in that it enables the consumer to make his choice in full knowledge of the facts.” Case C C-126/91 Schutzverband gegen Unwesen in der Wirtschaft v. Y.Rocher GmbH [1993] ECR I-2361, para 17. In Clinique a German law prohibited the use of the name “Clinique” on the grounds that it can mislead and confuse consumers so as to believe that it is a medical product and not a cosmetic one. The Court again found that the alleged consumer confusion did not justify the effects of the rule, namely the impediment on trade and the restriction of market communication. Ulen, T.S. Rational choice theory in law and economics, in: Bouckaert, B. and De Geest, G. (eds.), Encyclopedia of Law and Economics, Volume I. The History and Methodology of Law and Economics, Cheltenham, Edward Elgar, 2000, p. 790-794. Veljanovski contends that consumer sovereignty and the assumption of stable and given preferences prove the normative basis of economics despite many disclaimers that it is value free. Veljanovski (1980) p. 163. Consumer sovereignty is actually a “set of societal arrangements” that makes the economy respond to the messages consumers send out to other market participants about their needs. Lande, Averitt (1997) p. 715. Kerber, W., Vanberg, V. Constitutional aspects of party autonomy and its limits – the perspective of constitutional economics in: Grundman, S. Kerber, W., Weatherhill, S. (eds.), Party autonomy and the role of information in the Internal Market, de Gruyter, Berlin, 2001, p. 54-55. Lande, R., Averitt, N. Consumer sovereignty: a unified theory of antitrust and consumer protection law, 65 Antitrust L. J. 713, 1997, p. 715.

252

Hanneke A. Luth & Katalin Cseres

effective market policing tool while government protection would create substantial costs for producers and eventually for consumers.80 Neoclassical economics starts from a number of assumptions such as market players have stable set of preferences and they make consistent and rational choices in order to maximize their own welfare. However, recent empirical findings of behavioural economics challenge these assumptions on the basis of examining what people actually do how consumers analyse, interpret and use product and service information. Consumer preferences seem to fluctuate depending on the situation in which they have to make their decisions. Individuals lack the ability to build constant and reasoned preferences because they are influenced by the context where they seem exhibit certain cognitive errors related to time or memory or simple miscalculation. Consumer behaviour is context dependent and the form, context, quantity and substance of information have an impact on the ability of individuals to assess that information.81 Consumers will only look for and process a certain amount of information. As a consequence individuals fail to maximize their welfare under specific circumstances. As a consequence individuals fail to maximize their welfare under specific circumstances82 and they take short cuts when making decisions leading to choices that might be inconsistent with promoting their own welfare.83 The theory of bounded rationality argues that the capacity of the human mind to conceive and process complex information is relatively limited. Consumers may rely on heuristics instead of being guided by rationality and they fail to deal with information even in situations where the market does not have difficulty in producing the socially optimal amount of information or in distributing it efficiently.84 In these situations consumer harm is a result of behavioural biases.85 These biases can take many forms such as misunderstanding small probabilities, pseudocertainty, hyperbolic discounting, overconfidence, default bias, decision-conflict as a result of information overload. In case of inertia people are unable to process complex information and take irrational decisions. Or the oversupply of information may be counterproductive and may deteriorate market transparency, a situation referred to as “confusopoly”.86 Even when comparative information is available to consumers this inertia may be explained by computational difficulties, perceptions that search costs are high, or by possibly misplaced

80

81

82

83 84

85

86

C.F. Camerer, The economics of competition and consumer policies, presentation at the Global Forum on Competition, OECD, 2008. R Smith, S King, Does competition law adequately protect consumers?, ECLR 2007, 28 (7), p. 416 Korobkin, R. B. and T. S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 Cal. L Rev 4, (2000), pp 1077. Herbert A. Simon, pp 261, 270-271 (1957). Ulen, T.S. Information in the market economy – Cognitive errors and legal correctives in: Grundman, S., Kerber, W., Weatherhill, S. (eds.), Party autonomy and the role of information in the Internal Market, de Gruyter, Berlin, 2001, p. 98-99, 105. OECD, Roundtable on economics for consumer policy, Summary Report, DSTI/CP(2007)1/FINAL, p. 11. J Gans, “The Road to Confusopoly,” available on the ACCC conference Website at http://www. accc.gov.au/content/index.phtml/itemId/658141/fromItemId/3765.

The DCFR and Consumer Protection: an Economic Assessment

253

trust in consumers’ present supplier.87 Framing biases call the attention to the specific ways objective information is provided. The implications for consumer protection differ in the case of neoclassical economics and behavioural economics. While recent research in behavioral economics has identified a variety of decision-making errors that may expand the scope of paternalistic regulation, it does not provide solid evidence to support a turn back to paternalistic regulation. 88 The evidence is not yet conclusive with regard its impact in real world consumer markets. The costs and benefits of such paternalistic regulation in the long-run do not justify more government intervention but rather reaffirm the findings of neoclassical economics and rational choice theory. Paternalistic regulations based on the findings of behavioural economics may lessen the incentive to engage in learning and developing rational behaviour or even intensify irrational behaviour by introducing moral hazard. Consumers become aware of their biases and learn from their mistakes. The learning effects increase the more standardized the product and service and the more frequently the transactions are exercised. While firms can still intervene in various ways with the learning process of consumers by for example bundling products and services, creating artificial non-standardization or multidimensionality a more cautious regulatory approach.89 The question is whether cognitive biases impede competition and prevent market forces to perform efficiently even when consumers are sub-optimally or uninformed. The answer to this question significantly influences the need to introduce market correcting regulatory measures.90 Both the European Commission and the ECJ has clearly placed the emphasis on an “information-related model of consumer behaviour” who can facilitate the improvement of market integration. Thus the average consumer test perceives the consumer as an active, critical and well informed person capable of protecting himself by processing the marketrelated information and by searching for goods and services. While this test corresponds to the rational choice theory of neoclassical economics, a definite model of consumer behaviour that could serve as a benchmark is absent both in EC law and in the DCFR. What an average consumer in the given national context is will differ substantially across Member States and across markets and therefore lead to different national perceptions on a broad scale between paternalistic and rational approach. Besides the average consumer the DCFR in Article II.–3:103 refers to consumers who are at a significant informational disadvantage because of the technical medium used for 87

88

89

90

C. Wilson, C. Waddams, “Irrationality in Consumers’ Switching Decisions: when more firms may mean less benefit, CCP Working Paper 05/04, ESRC Centre for Competition Policy, University of East Anglia, 2005. C. Jolls et al., A Behavioral Approach to Law and Economics, 50 STAN. L. REV. 1471, 1545 (1998); R. B. Korobkin, T. S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 CAL. L. REV. 1051, 1059 (2000)Thaler RH: Toward a Positive Theory of Consumer Choice (1980) Journal of Economic Behavior and Organization 1 39-80. J.D. Wright, Behavioural law and economics, paternalism, and consumer contracts: an empirical perspective, NYU Journal of Law &Liberty, Vol. 2, No. 3, p. 472-473 Wright has investigated credit card markets, standard contract terms and allocation of shelf space in supermarkets and found no empirical evidence to support the presence of behavioural biases. F. Gomez, The harmonization of contract law through European rules: a law and economics perspective, ERCL 2/2008, p. 113. F. Gomez, The unfair commercial practices Directive: a law and economics perspective, ERCL 1/2006, pp. 12-14.

254

Hanneke A. Luth & Katalin Cseres

contracting, the physical distance between business and consumer, or the nature of the transaction. This notion raises two relevant questions. First of all, what is the underlying idea of differentiating between the average and other more vulnerable consumers and impose further information disclosure obligations on business. What kind of legal, social or economic argument demonstrates who is vulnerable and what kind of measure is necessary to rectify such situations? Second, how should business decide whether it faces consumers at a particular disadvantage? Does enforcement of this particular provision depends on the nature of the product and service at stake, the age, education, mental state or other criteria are decisive? Consumers at a particular informational disadvantage may refer to the notion of vulnerable consumers as known in EC law. The Directive on unfair commercial practices contains provisions aimed at preventing the exploitation of consumers whose characteristics make them particularly vulnerable to unfair commercial practices. Where a commercial practice is specifically aimed at a particular group of consumers, such as children, it is desirable that the impact of the commercial practice be assessed from the perspective of the average member of that group. The Directive argues that when certain characteristics such as age, physical or mental infirmity or credulity make consumers particularly susceptible to a commercial practice or to the underlying product and the economic behaviour only of such consumers is likely to be distorted by the practice in a way that the trader can reasonably foresee, it is appropriate to ensure that they are adequately protected by assessing the practice from the perspective of the average member of that group. This test of the vulnerable consumer has been heavily criticized for lacking practical and logical foundation.91 If vulnerability lies in the increased difficulty of obtaining and processing information then what kind of remedy can make vulnerable consumers to make an informed choice? Can the disclosure of more information be a remedy? Or is the way information is disclosed a proper tool? From an economics point of view such a differentiation lacks justification and raises the costs of such rule for both business and consumers without providing balancing benefits. Instead of the present distinction of an average and a disadvantaged consumer one could implement a distinction of consumer behaviour, which is more well-founded in economics: the distinction between rational consumers and bounded rational consumers. This distinction takes account of both the cost-benefit advantages of adopting a rational behaviour model, where consumers act in their best self-interest as well as of behavioural biases when consumers systematically depart from the rational model and when regulations might be ineffective. The dilemma, however, with regard to this distinction is the following. While to the extent that the cognitive errors identified by behavioral research lead people not to behave in their own best interests, paternalism may prove useful. But, to the extent that paternalism prevents people from behaving in their own best interests, paternalism may prove costly.92 Accordingly, the dilemma is how to help boundedly rational consumers to

91

92

Stuyck, J. E. Terryn, T. van Dyck, Confidence through fairness? The new Directive on unfair business-to-consumer commercial practices in the internal market Common Market Law Review (2006) 43 p. 121, 151; R.Incardona, C.Poncibó, The average consumer, the unfair commercial practices directive and the cognitive revolution, Journal of Consumer Policy, 2007, p. 29. Camerer, C., S. Issacharoff, G. Loewenstein, T.O’Donoghue en M. Rabin, 2003, Regulation for Conservatives: Behavioral Economics and the Case for «Asymmetric Paternalism», University of Pennsylvania Law Review, 151: 1211-1254. pp. 1211-1212.

The DCFR and Consumer Protection: an Economic Assessment

255

avoid making costly mistakes, while at the same time causing little or no harm in terms of minimizing costs to rational people. One solution would be to replace the notion of the average consumer by the interpretation of the ECJ, namely the reasonably observant and circumspect person or even more explicitly referring to the rational behaviour model. A clear behavioural model is essential to serve as a starting point for rule and policy making in order to explain what the legal rule envisages as expected behaviour from market actors. Furthermore, the DCFR could make use of behavioural economics on bounded rationality. The main message of this stream of research is that the assistance of (vulnerable) consumers can be provided by determining how information disclosure takes place. Framing the information is a valuable remedy. For example, providing information about contract terms in minority languages. Another solution would be to focus on the situations the regulatory measure addresses instead of on the persons. Such a solution is to implement a regulation, which corresponds to what Camerer et al call asymmetrically paternalistic: “A regulation is asymmetrically paternalistic if it creates large benefits for those who make errors, while imposing little or no harm on those who are fully rational. Such regulations are relatively harmless to those who reliably make decisions in their best interest, while at the same time advantageous to those making suboptimal choices.”93 Such a solution should also take account of the distribution of costs between the parties. When a certain contractual situation is likely to raise disproportionately the costs of consumers and sellers can remedy that situation at lower costs then they should bear the cost burden. For example, when we examine the above mentioned two provisions of the DCFR then their contents seem to largely coincide and thus the question is why a second consumer “Leitbild” of disadvantaged consumers is necessary. The difference between the two provisions lies in the more detailed information on possible hidden costs such as taxes and more detailed information on the terms of the contracts as well as providing reasonable time for the consumers to process this information. Information about hidden costs, enforcement, withdrawal, redress and terms of the contracts seem reasonable for rational consumers as well. Therefore, one single list of relevant pieces of information could suffice. The particular kind of contract could be decisive whether more or differently framed information is necessary, for example in case of distance contracts. This makes it easier for business to deal with the same set of disclosure obligations and eliminates the need to examine in every case whether they contract with disadvantaged or rational consumers. Such a list can take into account the debate about how much information consumers can process and about the costs and benefits of providing information in specific settings as regulations must take such costs into account.

3. Harmonization of mandatory rules and the DCFR Although the DCFR is not specifically connected to EC consumer law, it is a product of comparisons of national private laws of the EU Member States and EC law and as such based on a similar harmonization goal as EC law.94 Moreover, it is to be interpreted and applied in 93

94

Camerer, C., S. Issacharoff, G. Loewenstein, T.O’Donoghue en M. Rabin, 2003, Regulation for Conservatives: Behavioral Economics and the Case for «Asymmetric Paternalism», University of Pennsylvania Law Review, 151: 1211-1254, p. 1212. DCFR, Introduction point 21.

256

Hanneke A. Luth & Katalin Cseres

a manner consistent with the aims and principles on which not only the laws of the Member States but also the European Union are based, including among others the aim of the creation of an open internal market with free and fair competition and free movement of goods, persons, services and capital between Member States, and the protection of consumers.95 Accordingly, the question of the optimal level of intervention rises with regard to the DCFR as well. Whether a centralized legal system such as the DCFR or the decentralized legal systems of the Member States are more efficient to regulate and enforce consumer law. Neo-classical economics offers a number of criteria to judge whether centralisation or decentralisation is more successful in achieving the objectives of the proposed action. Economic reasons in favour of centralisation are: the danger of destructive competition between legislators i.e. “race to the bottom”, the need to internalise externalities across legal orders, the achievement of economies of scale and transaction cost savings through extending the size of jurisdiction. Externalities may be a reason to centralise decision-making. Negative externalities occur when Member States issue policy decisions that have adverse external effects for other states. Pollution is an obvious example. When scale economies are important, like in the case of national defence, centralisation may again be a better solution. Transaction costs in the case of different rules may be high, while in the case of uniform rules the search costs of information could be saved. These costs might be important for private firms operating in interstate commerce, but the same might not hold for consumers. Uniform rules can guarantee more stable and predictable jurisprudence and may significantly contribute to legal certainty.96 Economic reasons in favour of decentralisation are: the need to cope with information asymmetries between the centralised and decentralised actors i.e. the European Commission and Member States, the possibility to gather information about the costs and benefits of alternative legal rules, which is generated by competition between legal orders and diverging preferences in the respective regions or other social differences that influence the effectiveness of regulation. Information asymmetries arise because centralised actors have an information disadvantage compared to decentralised actors with respect to the firms they have to control. As firms may be unwilling to reveal information needed for central agencies there is a possibility of providing false information. Therefore, the information obtained might have to be checked and cross-checked against information from competitors, consumers and official sources. Local authorities will have a better overview of the market and thus the firms to be controlled than a supranational authority. Competition between different legal orders can serve as a learning process. Trial and error is very useful to find the best solutions to complex legal problems. Different rules imply dif95

96

DCFR, Introduction, 35. EU-specific aims. The DCFR is to be interpreted and applied in a manner consistent with the aims and principles on which not only the laws of the Member States but also the European Union are based, including the aim of establishing an area of freedom, security and justice, and the creation of an open internal market with free and fair competition and free movement of goods, persons, services and capital between Member States, and the protection of consumers and of others in need of protection. Cultural and linguistic plurality of Europe must be taken into account and preserved. Van den Bergh, R. The subsidiarity principle in European Community law: some insights from law and economics, Maastricht Journal of European and Comparative Law, 1/1994, p. Van den Bergh, R. The subsidiarity principle in European Community law: some insights from law and economics, Maastricht Journal of European and Comparative Law, 1/1994, p. 343-4.

The DCFR and Consumer Protection: an Economic Assessment

257

ferent experiences and can help to improve the understanding of the effects that alternative legal solutions have in similar legal problems. This advantage is present both in substantive and procedural rules.97 Diverging preferences is a third factor that could tip the balance in favour of decentralisation. When all parties in one region have identical preferences, cost efficiency considerations might point to harmonising through one single instrument that suits all.98 The pro and contra arguments of harmonisation in consumer law have different implications with regard to default rules and mandatory rules. Further differentiation can be made on the basis of the content of mandatory rules. The specific issues in the DCFR relate to contract law and tort law, which are areas under the almost exclusive competence of the Member States. There is a large diversity of legal solutions in the area of private law. This diversity is a result of fundamental differences in structure, legal philosophy, scientific consideration and language between the contract laws of the Member States. Not only do the solutions differ concerning the same legal problem, but terminology as well. With regard to mandatory rules transaction and information costs savings are more relevant when a centralized uniform law is available. Uniform mandatory rules save transaction costs and create scale economies. However, it cannot serve heterogeneous preferences. It also excludes beneficial learning processes between different legal systems. While centralized mandatory rules might be enacted as a tool of correcting regulatory failures on Member States level, it cannot be guaranteed that similar regulatory failures do not occur at central level. Rent-seeking in the form of lobbying at the central level can be harmful as well. Whether the DCFR as a piece of legislation drafted by academics suffers from such regulatory failures is beyond the scope of this contribution. Harmonization of mandatory rules containing information disclosures has number of advantages. They, first of all, intervene less with free market processes than do mandatory substantive rules do and they actually create optimal conditions for the exercise of freedom of contract. They can save transaction and information costs. Rent-seeking is less of a problem as the content of mandatory rules is merely information. However, regulatory failure might still play a role in the form of adopting inefficient or inconsistent rules and piling up legal instruments. Examples of inconsistency can be found in the diverging notions of consumer, information requirements and withdrawal rights in the EC Directives.99 The arguments that legal diversity raises transaction costs, creates negative welfare effects on trade and competition as well as resulting in legal uncertainty can only be raised with regard to mandatory rules.100 The costs of cross-border legal uncertainty are the costs of collecting information, of legal disputes, setting incentives for pushing through legal claims and other transaction costs. These costs have negative effects on trade and income through 97

98

99

100

Van den Bergh, R. Economic criteria for applying the subsidiarity principle in the European Community: the case of competition policy, International Review of Law and Economics 1996, 16 p. 364. Van den Bergh, R.J. Forced harmonisation of contract law in Europe: not to be continued, in S.Grundmann, J.Stuyck (eds.), An academic Green Paper on European Contract Law, 2002, pp. 249-268; Frey, B.S. A Europe of variety, not harmonizaton in: Josselin, J-M., Marcian, A. (eds.), The economics of harmonizing law in Europe, Cheltenham, Edward Elgar 2002 pp. 209-223. Kerber W., Grundmann S., An optional European contract law code: Advantages and disadvantages, Eur J Law Econ, 2006 21, pp. 215-236. The macroeconomic effects of legal uncertainty have been discussed by Wagner. Wagner H., Economic analysis of cross-border legal uncertainty in: Smits J., The Need for a European Contract Law; Empirical and Legal Perspectives, Europa Law Publishing, 2005, pp. 27-54.

258

Hanneke A. Luth & Katalin Cseres

increasing prices or reducing revenues for enterprises and consumers. Moreover, legal uncertainty delays transactions and reduces cross-border trade. The legal system is one of the key institutions of society. There are different methods for measuring the quality of formal institutions such as law and the effects of cross-border legal uncertainty as a result of institutional differences. The variety of legal rules reflects variety of preferences. Competition between legal systems and regulations can accommodate variety of consumer and enterprise preferences, it can support countries in experimenting and searching for efficient and feasible rules and it can generate a useful learning process. Moreover, such a regulatory competition can reduce rent-seeking and regulatory failures. It is the small entrepreneurs and consumers who face high information costs and costs of organizing their interests and thus they need. The implications of behavioural law and economics for harmonization of contract rules do not, however, differ from the conclusions drawn from traditional law and economics approach. Behavioural law and economics argue for a bottom-up approach of harmonization instead of a top down. The existing diversity of preferences as a result of cultural differences and strengthened by the endowment effect and status quo bias argues against a change imposed from above. Moreover, legislators could also suffer from cognitive distortion and therefore may not be able to filter the distortions of heuristics and biases individual consumers might exhibit. While individual consumers are likely to learn in real market situations about their own cognitive failures and thus learning effects could be substantial, legislators are unlikely to be able to profit from such real world opportunities as market parties. Learning processes among legislators are more likely to emerge when diversity and regulatory competition exists. Some individuals may also need more protection than others. All these arguments point to a preference for diversity instead of uniformity of law.101 Furthermore, coordination problems may arise between contracting parties when they face uncertainty as a result of legal diversity and when they have information problems. Competitive supply of legal rules coupled with institutional meta-rules of mutual recognition and freedom of choice leads to competitive equilibrium. Legal diversity is also an effective constraint on lawmakers’ arbitrariness. Amendment of provisions requires coordination and harmonization with other rules and legal principles and creates chain effects on other provisions and leads to additional legislative fixed costs and adjudication costs.102 Beneficial spill-over effects between competing European legal systems has been registered many times in the development of private law in Europe. Comparative law has proved to be a useful mechanism of regulatory competition. Harmonization would destroy these useful effects. Comparative law is capable of reducing coordination problems. Impulses from foreign systems and from the EC function as “legal irritants” which can induce useful changes in national law.103 Such processes have been started by the Unfair Contract terms Directive on English law. Legal transplants as a tool of comparative law stand for whether legal institutions can be transplanted from one legal system to another.104 Ideas flow through EC legislation as

101

102

103 104

Kerkmeester, H. Uniformity of contract law, in: Smits J., The Need for a European Contract Law; Empirical and Legal Perspectives, Europa Law Publishing, 2005, p. 71. Parisi F, Harmonization of European Private Law: An Economic Analysis, Legal Studies Research Paper Series Research Paper No. 07-4, University of Minnesota Law School. Wilhelmsson, T. Is There a European Consumer Law – and Should There Be One?, 2000. Teubner, G. Legal irritants: good faith in British law or how unifying law ends up in new divergencies, Modern Law Review, Vol. 61, 1998, pp. 11-32; Watson, A. Legal transplants: an approach to comparative law, University of Giorgia Press, 1993, 2000.

The DCFR and Consumer Protection: an Economic Assessment

259

well as directly between Member States. Piecemeal engineering105 can serve as a method of harmonization in the form of small adjustments and re-adjustments offering opportunity for institutional changes with continuous monitoring and improvement.106 The DCFR leaves room for a dynamic competition between alternative legal arrangements. It is in fact one of the alternatives. The future function of the DCFR is at the moment uncertain and therefore it makes little sense to elaborate on its harmonizing effects.

4. The notions of consumer and professional The concepts of “consumer” and “professional” vary among different European directives.107 This inconsistency creates uncertainty and confusion needs to be remedied by adopting one clear definition of both concepts.108 The question remains whether these concepts should be wide or narrow. The narrow notion would define consumers as natural persons acting for purposes that are outside their trade, business or professions, and professionals would be defined as persons (legal or natural) acting for purposes relating to their trade, business and profession. In the DCFR a wider concept of consumer and corresponding notion of business is opted for, namely:109 Consumer A “consumer” means any natural person who is acting primarily for purposes which are not related to his or her trade, business or profession.

105 106

107

108

109

Popper, 1957. Hage, J. Law, economics and uniform contract law: a sceptical view in: Smits J., The Need for a European Contract Law; Empirical and Legal Perspectives, Europa Law Publishing, 2005, pp. 55-70. As indicated in the Green Paper on the Review of the Consumer Acquis, COM (2006) 744 final. Available at: http://ec.europa.eu/consumers/cons_int/safe_shop/acquis/green-paper_cons_acquis_ en.pdf. The Directives covered by the Review are: Council Directive 85/577/EEC on Doorstep Selling; Council Directive 90/314/EEC Package holidays; Council Directive 93/13/EEC Unfair terms in consumer contracts; Directive 94/47/EC Timeshare; Directive 97/7/EC Distance selling; Directive 98/6/EC Price Indications; Directive 98/27/EC Injunctions; Directive 1999/44/EC Consumer Goods. Hondius, E.H. (1983). The legal control of unfair terms in consumer contracts: some comparative observations. In: Th. Bourgoignie, (Ed.), Unfair terms in consumer contracts. Legal treatment, effective implementation and final impact on the consumer, p. 33-76, esp 66-67. Louvain-la-Neuve: Cabay Bruylant. Definitions are given in Annex 1, which is referred to in article I.–1:103 (1) of the draft CFR: The definitions in Annex 1 apply for all the purposes of these rules unless the context otherwise requires; another issue concerning when to consider a party to the contract “consumer”, commented upon in the Green Paper, is whether a consumer dealing via an intermediary should still be considered a consumer. As the draft CFR does not comment upon this situation, this paper will not focus upon this issue here. See for a law and economics view on the topic: Luth, Cseres, An economics and effects based regulatory approach to consumer law: the case of EC consumer policy, draft version.

260

Hanneke A. Luth & Katalin Cseres

Business “Business” means any natural or legal person, irrespective of whether publicly or privately owned, who is acting for purposes relating to the person’s self-employed trade, work or profession, even if the person does not intend to make a profit in the course of the activity.

This paper proposes a new approach: the notions of consumer and producer should correspond to the market failure addressed by consumer policy, namely information asymmetry. As mentioned before, as in transactions between parties one of those parties can be subject to asymmetric information, regulation can correct this market failure by providing protection to that party. It is arguable whether a focus on being a natural or a legal person and a focus on the purpose of the contract are adequate proxies for singling out those situations where consumer protection is warranted. Of course, both elements need to be present in order for a party to be considered a consumer for the application of these regulations. Second, the definition can be confronted with the “image” of the consumer which is adopted in the EC Consumer Policy strategy 2007-2013, stating: “confident, informed and empowered consumers are the motor of economic change as their choices drive innovation and efficiency”.110 The personality of the parties, being natural or legal, is relevant for assessing whether a party should be considered a consumer; only natural persons can be considered consumers. According to the draft CFR, the personality of the party is irrelevant for assessing whether a party can be considered a business party. When assessing this element as a proxy for indicating information asymmetry, the proxy seems quite inaccurate. Legal persons can lack information with respect to the main subject matter of transactions making them eligible for consumer protection. Employing the personality of the parties as a proxy for indicating informational issues might lead to both type 1 and type 2 errors, granting consumer protection where it might not be necessarily to overcome information problems, and withholding consumer protection from parties that do suffer from information asymmetries in contract situations. Of course, the second proxy might corrects this. However, as the legal or natural personality of the parties seems not to correspond to either of them having information with respect to the main subject matter of the contract, this proxy seems irrelevant for indicating parties that should be protected from an economic point of view. A definition that would address the market failure of information asymmetry between the parties to the contract would focus on exactly that issue: whether or not a – thought to be professional – party has or is supposed to have information or knowledge with respect to the main subject matter of the contract, whereas his counterpart – the consumer – to the contract does not. Therefore, whenever a party, natural or legal, is acting outside the field of his professional competences with respect to the subject matter of the contract, whereas his counterpart is not, the first party should be protected by consumer protection regulation. Lacking professional competences, skills and expertise vis-à-vis a contract partner implies being subject to information asymmetry. This reasoning would argue for a wide definition of consumer based on concerns related to information asymmetry. The definition of consumer and professional should therefore be widened even further to contain all situations in which the seller has an information advantage over the buyer.

110

EU Consumer Policy strategy 2007-2013, COM(2007) 99 final.

The DCFR and Consumer Protection: an Economic Assessment

261

Consumers could be defined as: natural or legal persons operating in a field external to their professional competences, skills and knowledge with respect to the subject matter of the contract.111 The definition focuses on the information asymmetry, the market failure underlying efficiency losses in transactions between consumers and professional parties. This definition also incorporates corporate bodies and companies who act outside their field of expertise and therefore possibly lack relevant information in comparison with their contracting partner.112 Information disadvantage is widely acknowledged as being on of the primary reasons for protecting consumers, next to lack of bargaining power. Acting outside business, trade or profession is used as a proxy to capture situations where consumers are at a disadvantage towards professional parties due to lack of information or bargaining power. A proxy that is better able to capture these situations would be more efficient to apply. If the before mentioned definition is considered too vague or too wide to employ within the CFR framework, a wider proxy than merely “acting primarily outside the scope of business, trade or profession” would be more efficient to employ. The Green Paper furthermore mentions the situation of a mixed contract, referred to in the draft CFR in article II.–1:108: Mixed contracts. Within the context of assessing when a party should be considered a consumer or business/professional, a mixed contract would imply two professional parties being involved in a contract, one of whom is acting outside the field of his profession. The concept of a mixed contract and the corresponding example mentioned in the Green Paper will be explored to further explain the merits of employing a wide notion of consumer in the Draft CFR. The example employed in the Green Paper to elaborate on the concept of a mixed contract in this context involves a doctor buying a car to use both professionally, when visiting patients, and privately. This doctor, although clearly a professional, lacks information about the quality of the car and lacks skills to easily assess this quality, as he or she lacks information about other aspects of the contracts about which his counterpart, the professional car salesman, will possess information. The subject matter of this specific transaction is neither familiar to the doctor as he or she is not repeatedly engaging in such a transaction nor does he or she have specific knowledge about it. When the doctor is buying a car, he or she does not differ in his decision-making mechanism from any other consumer. Moreover, the informa-

111

112

See Aquaro, Enhancing the Legal Protection of the European Consumer, 14 European Business Law Review 3, pp. 405-413. But see Heiderhoff and Kenny, Response to the 2007 Green Paper on the Consumer Acquis, accessible at http://ec.europa.eu/consumers/cons_int/safe_shop/acquis/responses_green_paper_acquis_en.htm. Heiderhoff and Kenny argue that small businesses, entrepreneurs or professionals cannot be defined as consumers following Court of Justice decisions (p. 5), most notably Case C-361/89, Criminal Proceedings against Patrice di Pinto [1991] ECR I-1189; Case C-269/95 Francesco Benincasa v Dentalkit Srl [1997] ECR I-3767. Also Beale and Schulte-Nölke comment on the fact that a wide interpretation of the notion of consumer might go against ECJ rulings (Case C-464/01, Gruber/Bay Wa AG [2005] ECR I-00439), and ask to clarify whether the ruling of the ECJ would be remedied by the uniform definition. Beale and Schulte-Nölke, Response submitted on behalf of the Study Group on a European Civil Code and the Research Group on Existing EC Private Law (Acquis Group), accessible at http://ec.europa.eu/consumers/cons_int/safe_shop/acquis/ responses_green_paper_acquis_en.htm.

262

Hanneke A. Luth & Katalin Cseres

tion asymmetry that creates a market failure to the detriment of consumers would continue to exist even if the doctor would buy a car to use it only professionally. Extending the example used in the Green Paper, a doctor could buy a car with the sole purpose of using it to visit his patients. This does not alleviate the information asymmetry the doctor is subject to vis-à-vis a professional car salesman. Consumer protection rules as a means to overcome adverse selection should apply in this situation when taking an economic viewpoint.

5. Standard terms in consumer contracts Not only behavioural economics literature has taken a recent interest in the discussion on desirability of an intervention in the substance of terms in consumer contracts. In policy discussions, a need is expressed for enabling more substantive influence in the terms of consumer contracts. Linked to these discussions is an issue raised in the Draft CFR, until now not yet resolved, namely the suggestion to expand the discipline concercing “unfair terms in consumer contracts” to also cover individually negotiated terms. In article II.–9: 404 on the meaning of “unfair” in contracts between a business and a consumer, the Draft CFR mentions the following: in a contract between a business and a consumer, a term [which has not been individually negotiated] is unfair for the purposes of this Section if it is supplied by the business and if it significantly disadvantages the consumer, contrary to good faith and fair dealing (emphasis added). The same suggestion is made among others in the Green Paper on the Review of the Consumer Acquis. The Green Paper suggests two specific issues in relation to enabling more substantive control over terms in consumer contracts: 113 • a possible extension of the fairness test to cover price and main subject matter of the contract, in addition to only covering standard terms in consumer contracts; • extending the scope of the discipline concerning “unfair terms in consumer contracts” to cover individually negotiated terms. This would basically imply that negotiated terms would be subject to both the fairness test, combined with the (possible) exclusion of terms that feature on the black and/or grey lists. Currently, typical consumer protection regulation with respect to terms in consumer contracts focuses on disclosure duties, readability and intelligibility of terms, availability of standard terms before the contract is signed, and the contra proferentem rule.114 Contra proferentem is a rule of contractual interpretation which provides that an ambiguous term will be construed against the party that imposed its inclusion in the contract. Furthermore,

113

114

Green Paper on the Review of the Consumer Acquis, COM 2006 744 final. However, the preliminary outcome of the consultation, accessible at http://ec.europa.eu/consumers/cons_int/safe_shop/ acquis/index_en.htm, indicates that commentators to the Green Paper would not favour an extension of the scope of the fairness test to cover price, main subject matter and individually negotiated terms. The last issue of more substantive contol over individually negotiated terms is still under discussion, even in the context of the CoPECL network aiming to deliver a proposal for the “Common Frame of Reference” (CFR) for European contract law (www.copecl.org). Johnston, J. S. (2006). The Return of Bargain: an Economic Theory of How Standard-Form Contracts Enable Cooperative Negotiation between Businesses and Consumers. Michigan Law Review, 104(5), p. 860-864.

The DCFR and Consumer Protection: an Economic Assessment

263

onerous terms can be excluded by a fairness test, combined with black lists and/or grey lists.115 Back lists and grey lists serve to create clarity on how the fairness test should be implemented. Blacklisted terms are onerous and grey listed terms are presumed to be onerous, reversing the burden of proof to the party who imposed the term on the other party. Also, contract terms are governed by regulation on misrepresentation, and default rules.116 Recent policy discussions seem to show dissatisfaction with the current ability of regulators and courts to intervene in consumer contracts, concluding that consumers need to be protected from themselves.117 Discussions about consumer credit, lotteries and price festivals, warranties that are different from legal warranties are examples of this sentiment in policy. Information duties do not seem effective in enabling the consumer to include what is considered relevant or even vital information into her decision-making process. With respect to standard terms especially, the availability requirement is more and more criticized as being a proxy, enabling courts to intervene when a term is considered undesirable with regard to substance by the consumer only when a procedural requirement has not been met.118 Standard terms been criticized more or less since the industrial revolution.119 However, whether this calls for more regulation enabling intervention in the terms of consumer contracts which is justifiable from an economic viewpoint, is arguable. Economic theory defends the freedom of contract principle, only to be corrected in case of clear market failures and only then when intervention would be efficient. Behavioural analysis has provided insights indicating that some regulatory intervention going beyond that what is currently suggested in law and economics literature might be welfare-enhancing, but clear policy guidelines can not yet formulated.

5.1. Terms in consumer contracts – a law and economics point of view The desirability of regulating price, main subject matter and negotiated terms, as well as regulation on standard terms in consumer contract will be looked into from an economic viewpoint in the next part of this paper. Special attention will be given to adverse effects of consumer protection regulation with respect to price, main subject matter and negotiated terms.

115

116

117

118

119

The Unfair Terms directive only contains a grey list; other systems like the Dutch regulations of standard terms contain both a black and a grey list. See for a discussion: Howells, G. G. and Weatherill, S. (2005). Consumer Protection Law. (2nd ed.). Aldershot: Ashgate. p. 261-294. This viewpoint in itself is not that recent; it lately has become more widespread. See for instance: Sheldon, James. E, Consumer Protection and Standard Contract: The Swedish Experiment in Administrative Control, 22 Am. J. Comp. L. 17 1974, p. 17. Cserne, P. (2007), Policy considerations in contract interpretation: the contra proferentem rule from a comparative and economic perspective. Kessler, F. (1943). Contracts of Adhesion-Some Thoughts about Freedom of Contract. Columbia Law Review, 43(5), p. 629-642.

264

Hanneke A. Luth & Katalin Cseres

5.1.1. Price, main subject matter and negotiated terms 5.1.1.1. Where is the market failure?

Most national legal systems and the Unfair Terms Directive exclude price, main subject matter and individually negotiated terms from the scope of more substantive regulation, a choice defendable from an efficiency point of view. To apply the insights discussed above, price, main subject matter and individually negotiated terms are by definition salient to the consumer. As such, they are part of the consumer’s decision making process, implying that no information asymmetry exists between the consumer and the seller absent misrepresentation.120 This coincides with the principle of freedom of contract. The consumer is able to base her choice on available information in case misrepresentation is absent and no market failure exists. She has the best information about her own preferences. Some contract terms could very well be knowingly and willingly accepted by the consumer in negotiations even if the terms grant a less favourable position to the consumer than default terms would when this is balanced by a decrease in price. For example, a shorter warranty period could be agreed upon in exchange for a lower price. This does not imply that a less favourable position to the consumer is necessarily welfare reducing; it can be the explicit wish of the consumer to include a less favourable term in the contract to incur a price premium. However, contracts where the price that the consumer has to pay for the product or service and the benefit she expects to gain from the contract depend on her use pattern shed a different light on the clarity of price.121 In for instance consumer credit contracts, fitness club subscriptions and calling plans with respect to mobile phones, consumers have great difficulties in assessing prices and gains before agreeing to the contract, and sometimes even while being in the contract. Biases like overoptimism and overconfidence influence the assessment of prices and gains to contracts, enable sellers to exploit consumer biases. This could be an argument to extend the scope of the fairness test to cover price and main subject matter. However, other policy options might be available to deal with this issue that might be more effective in enabling contracts that correspond to consumer preferences. Extending the fairness test is a very general instrument that will have implications for all consumer contracts, not just that contracts that involve problems with estimating the price and gain of the contract for the consumer. These issues might be better addressed by special regulation applicable to these specific contracts, instead of being applicable to all contracts. Second, the focus of regulation might better focus on correcting the information asymmetry that exists between sellers and consumers. Neither sellers nor consumers have information about the specific price and gains of the contract for each individual consumer. However, the seller has information about average use-patterns of the consumers that he deals with. This information could be made publicly available even by mandatory information obligations assigned to the sellers. In this way, consumers could be debiased, at least to some extent.122

120

121

122

De Geest, Gerrit, The Signing-Without-Reading Problem: An analysis of the European Directive on Unfair Contract Terms, in Festschrift für Claus Ott (2002), pp. 230-231. Bar-Gill, Oren, “Informing Consumers about Themselves” (August 2007). NYU Law and Economics Research Paper. No. 07-44 Available at SSRN: http://ssrn.com/abstract=1056381. Idem.

The DCFR and Consumer Protection: an Economic Assessment

265

5.1.1.2. Encouraging consumer moral hazard

Extending the scope of the discipline of unfair contract terms to cover price, main subject matter and individually negotiated terms could result in consumer moral hazard problems. Responsibility for assessing ex ante the desirability of a particular transaction should lie with all contract partners, the seller and the consumer. The consequences of a transaction should be evaluated before the contract is entered into. This argument is a prerequisite for being able to conclude welfare enhancing contracts. Focusing on both seller and consumer responsibilities corresponds to EC Consumer Policy strategy 2007-2013, which states that: “confident, informed and empowered consumers are the motor of economic change as their choices drive innovation and efficiency”.123 Carelessness on the part of consumers can be induced by allowing consumers the possibility to invoke the fairness test against any contract term.124 This can be seen as an example of consumer moral hazard. Carelessness before entering into contracts would not stimulate welfare enhancing transactions to be concluded. In another scenario, a consumer could agree to take on a certain risk, e.g. a warranty period lower than the standard, for a price premium. Upon occurrence of the event that this risk pertains to, the consumer could seek a remedy against this individually negotiated term, claiming it to be unfair. Consumers are not inhibited from actions like these as professional parties can be.125 As sellers can expect this kind of moral hazard on the part of consumers, they would not be willing to offer these kinds of contracts to consumers, or only be willing to do so after a lengthy negotiation process with several assurances drafted in the contract. Uncertainty, extensive negotiations and extra drafts lead to an increase in transaction costs.

5.1.1.3. Decreasing available options

Extending “unfair terms” regulation to cover individually negotiated terms in addition to standard terms could lead to a decrease in options available for negotiations. Instead of protecting the consumer, an extension of the fairness test to cover negotiated terms could bar some terms from ever being agreed upon. A welfare enhancing term that is subjected to a fairness test might fail that test, if it seems be to the detriment of the consumer vis-à-vis the seller as it allocates more risk to the consumer. The seller could expect moral hazard on the part of the consumer and be unwilling to draft the term requested in the contract. This infringes the freedom of choice, reduces the availability of options and delays transactions to materialize.

123 124

125

EU Consumer Policy strategy 2007-2013, COM(2007) 99 final. See Winick (1992) on how regulatory intervention might prevent individuals from learning from the consequences of their actions, therefore becoming totally dependent on government protection; B.J. Winick, On Autonomy: Legal and Psychological perspectives (1992) 37 Villanova Law Review 1705, 1756. Bebchuk, L. A. and Posner, R. A., “One-Sided Contracts in Competitive Consumer Markets” (November 9, 2005). U Chicago Law & Economics, Olin Working Paper No. 270 Available at SSRN: http://ssrn.com/abstract=845108.

266

Hanneke A. Luth & Katalin Cseres

5.1.1.4. Protecting the active consumers?

When negotiated terms remain excluded from the discipline of unfair terms, less protection seems to be offered to active consumers than to inactive consumers. As active consumers know what they bargain for, welfare decreasing terms will not be agreed upon, save in cases of misrepresentation. Active consumers are well protected by the doctrine of misrepresentation, thus only suffering a decrease in welfare when they are denied the possibility to negotiate the terms they prefer. Consumers might err when negotiating for certain contract term, mistakenly agreeing to a term that decreases their welfare. This issue should be balanced with problems connected to consumer moral hazard, the incentive to thoroughly assess the consequences of one’s actions and decisions to prevent error and the decline in available options to consumers. Consumer error is not a sufficient rationale for mandatory legislation, especially when adverse effects of that legislation are taken into account and protective legislation might even induce consumer error and carelessness. Consumers, and especially the active consumers, might not be so much in need of protection from themselves, but more in need of protection from consumer protection.126 However, behavioural insights could again be applicable to this issue. As behavioural literature details, individuals suffer from several biases and heuristics when assessing risks. Consumers might willingly accept a certain condition for a price premium, whereas the company with her expertise and available data knows that this condition is likely to be less beneficial to the consumer than she expects.127 On the other hand, the question can be raised whether expanding the scope of the fairness test will provide a remedy. Black and grey lists as well as the fairness test can only bar the most onerous terms from being drafted into consumer contracts. A term that is not particularly onerous can still be welfare decreasing for the majority of consumers. Furthermore, not all standard terms that allocate risks are covered by black and grey lists, nor are they easily assessed using the fairness test. Whether cognitive errors in decision making warrant further mandatory provision of terms and at the same time reducing the consumers’ options remains an open question until further evidence is found. Other policy options could be targeted at more efficient ways of dealing with consumer biases.

5.1.1.5. Standard terms

The analysis of standard terms in consumer contracts is more complicated. The discussion below will follow developments in law and economics literature, to give insights to how scholars continue to address the issues of standard terms in consumer contracts. Undoubtedly, the uneasiness policy makers feel due to a sense of not being able to control or address the issue to satisfaction has an influence on the focus in literature.

126

127

Van den Bergh, R. J. (1997). Wer schützt die europäischen Verbraucher vor dem Brüsseler Verbraucherschutz? In: C. Ott, H. B. Schäfer (Eds.), Effiziente Verhaltenssteuerung und Kooperation im Zivilrecht, p. 79-102. Tübingen: Mohr Siebeck. See Bar-Gill, 2007, for a discussion on informing consumers about use-pattern mistakes, an product aspect that sellers are far more likely to have information on than consumers, and where consumers might be highly susceptible to biases. Bar-Gill, Oren, “Informing Consumers about Themselves” (August 2007). NYU Law and Economics Research Paper No. 07-44 Available at SSRN: http:// ssrn.com/abstract=1056381.

The DCFR and Consumer Protection: an Economic Assessment

267

5.1.1.6. First glance – standardization efficient

Standard terms in consumer contracts will be known by the professional party, but are not likely to be known by the consumer. This is a typical case of information asymmetry. The unequal bargaining power, take-it-or-leave-it nature, signing-without-reading and standardization of contracts have given rise to scepticism regarding the fairness of standard terms. However, in the classical law and economics literature, these issues do not pose serious challenges. Competition between companies with respect to their standard terms should ensure a “fair” (welfare-enhancing) contract for all consumers. If consumers disagree with some terms, they could shop for better terms.128 As profit-maximizing sellers do not want to lose consumers, they will adapt their standard forms to correspond to the preferences of the majority of their customers. Furthermore, a “duty to read” clause endorses that parties to a contract will be held responsible for their own decisions, counteracting consumer moral hazard problems.129 Consumers themselves are in the best position to maximize their own welfare. The doctrine promotes stability and reliance on contracts, even if the contract turns out to be less beneficial than was estimated by one of the parties before. “Duty to read’ as well as the general economic approach focuses on the ex ante perspective more than the situation ex post. Standard terms allocate risks between consumers and businesses. The “best” allocations of risks are unlikely to vary between businesses, and the standard forms used by different firms allocating these risks will be comparable.130 Standardization is therefore not a clear sign of unfairness of the standard terms; full information and allocation of risks corresponding to consumer preferences could also result in similar standard terms throughout the business sector.

5.1.1.7. Second glance – market failures: information asymmetry and rational apathy render standard terms possibly inefficient

Since consumers are not able to include non-salient standard terms in the decision-making process, severe doubt can be raised as to whether these standard terms will be welfare enhancing.131 Even when standard terms would be read, too much information precludes assessing all terms, let alone comparing all terms in different contracts. A welfare enhancing contract term would allocate risk with the party that is most able to influence that situation from occurring and place responsibility for certain actions with that party who is most able to induce them at lowest cost. Furthermore, it should correspond to the risk preferences of the parties to the contract. To exemplify, a standard term could specify that in case of damage resulting from delivering a household appliance to the house of a customer the store should be notified in writing within three days. The majority of consumers, upon thinking about it, 128 129

130 131

Overview: George L. Priest, A Theory of Consumer Product Warranty, 90 Yale L.J. (1981). Todd D. Rakoff, Contracts of Adhesion: An Essay in Reconstruction, 96 Harv. L. Rev., at 1174 (1983), at 1187. George L. Priest, A Theory of Consumer Product Warranty, 90 Yale L.J. (1981), at 1300. See both Goldberg (1974) and Korobkin (2003). Goldberg, V.P., Institutional Change and the Quasi-Invisible Hand”, 17 Journal of Law and Economics, 461; Korobkin, Russell B., “Bounded Rationality, Standard Form Contracts, and Unconscionability”. University of Chicago Law Review, Vol. 70, p. 1203, 2003 Available at SSRN: http://ssrn.com/abstract=367172.

268

Hanneke A. Luth & Katalin Cseres

could prefer to have this term extended to seven days, even when that implies incurring a small cost for the extension of the time frame. However, since this term will not be read nor is salient to consumers, the seller is able to draft a term with a short time frame for notification. Prolonging this time frame could result in more notifications of damage made to the seller, and therefore be more costly to the seller. Providing a short time frame for notification is less costly to the seller, enabling the seller to charge a lower price for the good. Competition between sellers for the favour of the consumer only affects salient product attributes. As price is a salient product attribute, competition between sellers induces the price to be set at a competitive level. A standard term regarding notification duties in case of damages caused by the delivery service will just be provided low-cost and low quality, regardless of actual consumer preferences. The latter term will be one-sided, favouring the seller and not the consumer, but not necessarily welfare enhancing. Reputation is questionable as a mechanism to induce less one-sided terms. Sellers can waive a specific form term if a consumer complains.132 This saves reputation, but does not allow all consumers to benefit from the less one-sided ones. Only the consumers that complain accrue the benefits, even though not all consumers would take the trouble of complaining if they expect the firm to invoke the one-sided term against them. Also, consumer standard terms often allocate remote risks. Consumers might not learn of businesses invoking these terms against their costumers, causing little or no effect on the reputation of the firm. Lastly, businesses might also be managed by short-run players who are not concerned with reputation. For all these reasons, the market is likely to produce one-sided, not necessarily welfareenhancing, standard terms.

5.1.1.8. Third glance – Consumer moral hazard and screening: one-sided terms are efficient?

Bebchuk and Posner agree with the viewpoint that standard terms in consumer contracts are one-sided in that they favour the sellers.133 However, they assert that this is efficient due to the fact that the consumer needs to be deterred from behaving opportunistically. Sellers are less inclined to behave opportunistically in a competitive consumer market as they are constrained by reputational considerations. For example, hotels will never ask their clients to pay for an extra night if they check-out ten minutes late (e.g. 11.10 am instead of 11.00 am), even though that is a term of the contract. A term unlikely to be invoked due to reputation constraints enables the hotel to use discretion and only call upon this term when faced with opportunistic consumers. The literature focusing on providing information about standard terms and behavioural theory with respect to standard term implicitly assumes that these terms govern the contract. In effect, these terms might not be enforced regularly by the drafting parties. Analysis of standard terms should be take into account whether the privileges that standard terms give to drafting parties are relied upon, or usually waived unless in circumstances that make it more efficient for the terms to be upheld. Bebchuk and Posner do admit that a different al-

132

133

Ibrahim, D.M, Rewriting the Law of Boilerplate Terms: Fairness, Efficiency, and a Standard Form Principle of Comparative Responsibility (2005). Bebchuk, Lucian Arye and Posner, Richard A., “One-Sided Contracts in Competitive Consumer Markets” (November 9, 2005). U Chicago Law & Economics, Olin Working Paper No. 270 Available at SSRN: http://ssrn.com/abstract=845108.

The DCFR and Consumer Protection: an Economic Assessment

269

location of reputation constraints might result in one-sided terms being invoked inefficiently. Furthermore, social norms might be at play here to constrain consumers. As theoretical considerations based on both behavioural and more traditional economic analysis point to possible efficiency problems, some form of government regulation beyond current regulation could be warranted. However, efficiency problems do not always warrant government intervention; this intervention is only desirable when it increases welfare. Also, government intervention can take other forms than only regulation. Different alternatives in policy instruments should be assessed before more substantive regulation with respect to terms in consumer contract is implemented.

5.2. Looking beyond regulation: other policy options concerning consumer contract terms The current policy initiatives undertaken to enable increased substantive control over consumer contract terms are largely focused on regulation. As argued above, intervention through regulation with respect to price, main subject matter and negotiated terms is likely to have more adverse effects decreasing consumer welfare than it is increasing consumer welfare. To address the growing need for more substantive influence on terms in consumer contracts, alternative policy methods of increasing welfare enhancing terms to be included in consumer contract merit further analysis.

5.2.1. Alternative policy methods concerning standard terms in consumer contracts

A model form of standard terms that enhance welfare for the majority of consumers might be more efficient in providing protection to consumers. An optional model for standard contract terms is currently being suggested in the context of the Network of Excellence on European Private Law (CoPECL). Schulte-Nölke proposed to develop an optional instrument on a European level, calling it a “Blue Button”.134 When concluding a (cross-border) transaction, a consumer would be asked to state his residence. The seller can opt to offer a contract under the law of the residence of the consumer, or offer to conclude the transaction under EU law. The consumer would in the latter case be asked to press a “Blue Button”, hence the name of this proposal. It is not clear yet how the optional model would be constructed and a possible implementation of this instrument is not foreseen before 2015. Alternative policy options could include a certificate for a “fair set” of standard terms or a grading scale for the fairness of standard terms.135

134

135

Beale, Schulte-Nölke, Response submitted on behalf of the Study Group on a European Civil Code and the Research Group on the Existing EC Private Law (Acquis Group) (Response to the Green Paper on the Review of the Consumer Acquis, COM (2006) 744 final, 2007, pp 8, available at: http://ec.europa.eu/consumers/cons_int/safe_shop/acquis/responses/universitat_bielefeld.pdf. See Becher (2007) for a discussion on several options, like a certificate for fair standard terms, Becher, Shmuel I., “A ‘Fair Contracts’ Approval Mechanism: Reconciling Consumer Contracts and Conventional Contract Law” (September 2007). Available at SSRN: http://ssrn.com/abstract=1015736. The certificate for a fair standard terms or a grading scale for the fairness of a set of standard terms will not further be assessed in this paper.

270

Hanneke A. Luth & Katalin Cseres

5.2.1.1. The attractiveness of a model of standard terms

A welfare-enhancing model of standard terms might provide an efficient way of stimulating the inclusion of such terms to be included into consumer contracts. Default rules, and as such model rules, tend to be sticky and have an expressive effect.136 Empirical research indicates that individuals stick to the default rule provided. The default rule could be expressive in that it is interpreted to be the fairest allocation of risk simply because it is drafted in the standard term model form. A default risk allocation in the model standard form could provide the consumer with an anchor or reference point, from which she is more reluctant to deviate to the risk division she would have had in mind absent the default. This is particularly relevant when discussing negotiated standard terms. Policy makers might be more efficient in remedying market failures by resorting to instruments like model standard terms or certificates for standard terms to influence the decision making process of consumers who choose to negotiate standard terms. The availability of options is not diminished, allowing consumers to opt for other terms if they feel confident with deviating from default and thereby increasing their welfare. These options coincide with behavioural insights that government intervention should apply a “lighter hand” and guide consumers to welfare-enhancing options for the majority. Still allowing for deviations when individual consumers feel there welfare would be enhanced by an alternative term, this approach could accrue benefits of government intervention while keeping the costs of this intervention low. Special attention should be given to those terms that allocate risks between parties. Individuals are regarded to be more risk-averse than professionals and companies. Model terms governing the contract should follow these risk preferences allowing a departure from the defaults upon expressed wish of the consumer. Likewise, consumer moral hazard should explicitly be taken into account in order to counteract adverse effects of regulation, such as an increase in transactions costs passed on to consumers. There are several reasons why the alternative policy options might be quite attractive to parties to adopt. The model sets and certificates can be adjusted to the specific features of different business sectors. The drafting of model standard forms might be done by a policy maker, a consumer representative organisation, or it could be the result of negotiations between business and consumer interest groups.137 Organizations who draft model terms, facilitate negotiations or are active in accreditation, could also be an active party when it

136

137

For a discussion of “stickiness” in several contexts, see Jason Scott Johnston, Strategic Bargaining and the Economic Theory of Contract Default Rules, 100 YALE L.J. 615 (1990). In the Netherlands, the Social Economic Council (SER) hosts negotiations between business and consumer interest groups to draft model standard forms for consumer contracts. Viitanen (2007) describes how Nordic Consumer Ombudsmen negotiate sets of standard terms with trade organisations. These Consumer Ombudsmen also serve as the public enforcement agency when the “fairness” of these terms is disputed. As the terms were developed in negotiations with the Consumer Ombudsmen, they are not likely to be set aside by the same institution in a fairness dispute, firms are likely to want to adopt these terms to save on transaction costs arising from potential disputes. Viitanen, Enforcement of consumers’ collective interests by regulatory agencies in the Nordic countries, In: Collective Enforcement of Consumer Law (2007), Europa Law Publishing, eds W. van Boom and M. Loos, p. 85-100.

The DCFR and Consumer Protection: an Economic Assessment

271

comes to the enforcement of the standard terms.138 Terms that have been approved by these organizations are less likely to be successfully questioned in court or other proceedings. When standard term models are the result of negotiations between consumer representative organisations and business organisations, these models are more likely to consist of efficient terms than default rules drafted by policy makers.139 Information about preferences and consequences for parties involved is more likely to be brought up in discussions by the parties whom these preferences and consequences pertain to. Also, as business representative organisations agreed to these terms, companies are more likely to adopt these sets of standard terms than the “set” of standard terms drafted by default rules. Especially medium-to-small enterprises can save on transaction costs by freeriding on the drafting efforts by other parties and are assured that their interests are well represented. Employing a model of standard terms or being accredited a certificate can also be interpreted as a sign of consumer friendliness, a signal for consumers that the terms applied by this company are of higher quality than other companies have drafted.

5.2.1.2. Assessing the options – future research

Further research should be conducted on analyzing the efficiency of alternative instruments to influence the formation of standard terms in consumer contracts. One of the most interesting questions is under which conditions companies would adopt these high-quality-more-expensive terms, as opposed to harsh-but-cheap terms. However, these alternative instruments might be more effective in inducing welfare-enhancing terms in consumer contracts than purely relying on competition between sellers to attract consumers combined with (mandatory) legislation, mostly because options for consumers are not decreased and information enabling the drafting of welfare-enhancing terms can more easily be obtained from the actual parties to the contract.

6. Right of withdrawal The DCFR contains a general right of withdrawal regulated in Articles II.–5:101-106. The period of withdrawal is provided for all kinds of contracts and it is fixed in fourteen calendar days (Article II.–5:103). The exceptions do not deal with narrower or broader periods even though in certain contracts that would be justified as consumers might need longer or shorter periods to reflect on their decision. Cooling-off periods are legislated to cure consumers’ irrational decision-making, situational monopolies and information asymmetries in an environment of incomplete information and high pressure marketing. Consumers can make sub-optimal decisions when they face temporary market power. Cooling-off periods provide consumers time for reflection, to process all the relevant information, search for additional information or advice and establish whether the agreement indeed reflects their individual preferences. 138

139

As is the case in both the Nordic countries and Israel, see Becher (2007) and Viitanen (2007). In Israel, a special tribunal for standard terms deals with accreditation and enforcement of standard terms in consumer contracts. Bernstein, L. (1992). Opting out of the Legal System: Extralegal Contractual Relations in the Diamond Industry. Journal of Legal Studies, 21(1), p. 115-158.

272

Hanneke A. Luth & Katalin Cseres

Nevertheless, consumers may as well engage in opportunistic behaviour by using the product during the cooling-off period and then returning it back to the seller claiming bad quality. Cooling-off periods delay transactions, create uncertainty and thus increase the transaction costs of parties. Mandatory provisions on withdrawal will raise prices of the products and services and such costs will be passed on to consumers. Thus, the optimal period of withdrawal strikes a balance between the information asymmetry the consumer faces and the opportunity for moral hazard. Moreover, granting withdrawal rights reduces the need for information disclosure requirements. When consumers as the cheapest information costs providers can obtain the relevant information themselves during the cooling-off period sellers should not be required to provide the same information. These extra information provisions only increase the costs of the transaction and thus the costs for consumers. This is, for example, the case in the Directive on doorstep selling where sellers have to provide the information prior and after the conclusion of the contract. In the EC directives the duration of cooling-off periods differ from seven days in the Directive on doorstep selling to thirty days in insurance contracts. In certain cases the prescribed period should be longer where the economic distortion cannot be remedied within that period in case of timesharing for example. In order to avoid consumers abusing the right of withdrawal monetary compensation, a rental payment could also be introduced.140 The DCFR provides a right to withdraw in consumer contracts where the consumer’s offer or acceptance was expressed away from business premises (Article II. 5:201). The number of contracts negotiated away from business premises is growing and thus this right might become a standard right. Moreover, it is not clear in the DCFR how the various information remedies such as cooling-off periods and requirements of information disclosure are coordinated. Are consumers entitled to both rights? That would result in piling up remedies that target the same information problem and thus impose unnecessary costs on business.

Conclusions This paper reviewed the consumer protection rules of the DCFR. It sketched a general picture on the DCFR’s model of consumer protection and contrasted it with the economic function of mandatory rules. Three conclusions can be made. First, while the DCFR firmly stands on the basis of freedom of contract as one of its core principles and accommodates distributive justice goals merely to the extent that it is absolutely necessary and this approach has not been implemented in the specific rules of the DCFR. In fact there is a striking contrast between the fundamental principles and the picture that emerges when one examines the specific consumer rules. The rules on information duties, right of withdrawal, standard contract terms, notion of consumer and the way the DCFR envisages the consumer point to a paternalistic market philosophy which regards consumers as the typically weak parties and provide them broad protective rights. Second, the DCFR lacks a clear model of consumer behaviour. It fails to explain what is the starting point of its legal rules, what is the expected consumer behaviour they aim to address. This would be essential in order to underlie the regulatory approach the DCFR has chosen and the balance it strikes between default and mandatory rules. Furthermore, the 140

Rekaiti, p. , Van den Bergh, R.J. Cooling-off periods in the consumers laws of the EC Member States. A comparative law and economics approach, Journal of Consumer Policy, 23 (4) 2000.

The DCFR and Consumer Protection: an Economic Assessment

273

DCFR could make use of both neo-classical economics on the rational choice theory and behavioural economics on bounded rationality. The main message of this latter stream of research is that the assistance of (vulnerable) consumers can be provided by determining how information disclosure takes place. Third, the DCFR can profit from the following economic arguments. In the first place, it should consider the distributional effects mandatory rules have on different groups of market actors. Not only the costs of intervention, but also the costs imposed on business and the different groups of consumers (informed, sophisticated, uninformed and vulnerable) have to be considered. Consumer protection has its price and only the analysis of the distributional effects of the envisaged legal rules can tell whether intervention is cost-effective or whether consumers or sellers pay the price of protective rights. This cost conscious economic approach is missing in the DCFR. Second, highly complex systems of information disclosure originally aiming at lowering information costs will obviously restrict competition and will have counterproductive effects. Consumers unable to make informed choices are forced to employ expensive intermediaries and business has to bear the costs of the ineffective disclosure. Some consumer protection measures create barriers to entry that limit the freedom of sellers and might eventually lead to higher prices for consumers. Interventions therefore should be evidence-based and evaluated to ensure that it is not unnecessarily applied. It should be examined why the market-based solution does not work or why that solution might be socially sub-optimal.141 The DCFR is addressing information failures in order to prevent inefficient transactions, it could, however, establish a more out-spoken guidance taking account of the above discussed insights and evidence from neoclassical economics and recent empirical research. It could set up a checklist and toolkit for consumer protection issues, similar to what is being worked on presently within the OECD.142 141

142

These are self-correcting mechanisms that are based on private law norms of tort, contract and property rights that they are the result of government action. Hadfield, Howse, Trebilcock (1998, p. 155). It also has to be demonstrated why government regulation is going to be better than markets in providing low-cost information. Even where a relevant market failure has been identified, government should only act when this is feasible and it is cost-effective to do so. The costs and benefits of particular forms of intervention and alternatives thereto should be examined and represented. OECD, Roundtable on economics for consumer policy, Summary Report, DSTI/CP(2007)1/FINAL, p. 37.

6. Non-contractual Liability

The Boundary between Torts and Contracts: A Law and Economics Perspective* Assunção Cristas** & Nuno Garoupa***

Abstract The boundary between tort law and contract law (i. e., the boundary between contractual and extracontractual liability) is relevant in any legal system. However, the extension and the practical importance of this boundary is dependent on the specificities of the legal system. Our paper approaches the boundary between torts and contracts from a law and economics perspective. The interaction between a contractual relationship and the possibility of a tort claim generates interesting legal issues that can be analyzed from an economic perspective. The second contribution of our paper is the use of the DCFR (the Draft Common Frame of Reference) as the case-study for discussion. We identify and discuss four classes of situations that are relevant from this point of view, namely (i) third-party effects (contracts that cause a potential wrongdoing); (ii) breach of contract induced or caused by a third party; (iii) quasi-contracts; (iv) breach of contract or other forms of contractual under-performance that cause a potential wrongdoing.

Introduction The boundary between tort law and contract law, that is, between contractual and extracontractual liability is relevant in any legal system. However, the extension and the practical importance of this boundary is dependent on the specificities of the legal system. In common law, for example, torts and contracts are perceived to be two separate and well-defined bodies of law. The intersection between both is quite limited, although, recently in some, this distinction has been somewhat blurred. Despite relevant distinctions, in the Continental legal systems, tort and contract liability are encompassed by the law of obligations codified

*

** ***

The authors are grateful to John Colombo, Anthony Ogus and the participants at the Barcelona meeting of COPECL for comments and suggestions. Two anonymous referees provided stimulating comments. Yeny C. Estrada has provided superb research assistantship. Garoupa acknowledges financial support by FCT, PPCDT/JUR/55752/2006, Portugal. Cristas acknowledges financial support by FCT, PPCDT/JUR/65141/2006, Portugal. The usual disclaimers apply. FD, Universidade Nova de Lisboa. University of Illinois College of Law.

278

Assunção Cristas & Nuno Garoupa

more than 200 years ago.1 The law of obligations effectively makes the line between both less distinct.2 In common law, tort law was developed from a large body of unrelated doctrines and case law; including trespass, nuisance, negligence, and deceit. Presumably torts have become more relevant in the last hundred years due to technology developments, new hazardous products and more dangerous activities. Therefore, the boundary between torts and contracts has become an important issue. The lack of an adequate systematization in common law for historical and procedural reasons, left the issues of the boundary between torts and contract without a clear treatment or logic.3 For example, in Tai Hing Cotton Mill Ltd. v Liu Chong Hing Bank [1986], the Privy Council rejected that tort liability could arise independently of a contract. However, the House of Lords in Henderson v Merrett Syndicates [1995] came to the opposite conclusion. Apparently in English common law, under certain conditions, the plaintiffs can seek compensation under the most favorable regime (for example, if the contract does not specify otherwise). Similar developments took place in the United States (e. g., in the area of defense contracts).4 It is fairly exceptional to have restitution in damages for breach of contract cases (but not impossible5). Still, there is a sharp distinction between contractual and tort damages in common law.6

1

2

3 4

5

While French Civil Code has one general tort clause (article 1382), BGB opted for considering three typical general situations (stated in §§823 and 826). See K. Zweigert and H. Kötz, An Introduction to Comparative Law (translation T.Weir), 3rd ed. (1998), respectively, pp. 615-621 and 598-605. For a comparative analysis of legal regimes of Member States in Europe at fifteen see C. von Bar and U. Drobnig, The Interaction of Contract Law and Tort and Property Law in Europe. A Comparative Study (2004). But does not necessarily mean that material solutions differ considerably, tough with different starting point and procedural options. See K. Zweigert and H. Kötz, supra 1, pp.621-628 and C. von Bar and U. Drobnig, supra 1, p. 15. See Anthony Ogus, Costs and Cautionary Tales: Economic Insights for the Law (2006), pp. 67-68. It is common to have overlap between contracts and torts in the area of contract defenses. Fraud and misrepresentation are torts, but they are also defenses to contractual liability and are perfect examples of our short discussion later about “picking a theory” depending on the remedy. In general, alleging tortious fraud would allow one to recover damages for pain/suffering and perhaps even punitive damages, whereas alleging fraud as a contractual defense probably only nets the recession of the contract and the return of any consideration already paid. It is fairly routine in the US for someone who breaches a contract to bring an action for restitution to recover some value of their performance. So, for example, if a contractor builds a house for a second party and breaches the contract in a way that permits the second party to discharge the contractor (a “total” breach that permits the owner to declare the contract terminated), it is not at all unusual for the contractor to bring a claim against the second party for restitution. This claim would not permit the contractor to recover his profit on the contract (e. g., expectation damages in contract), but might well get him some compensation for materials used in the contract and for labor. It is true that the basis of this action is not “contract” but rather restitution. Restitution is not generally a remedy that is sought (or awarded) to the non-breaching party in a contracts case being tried on a contracts theory, but a breaching party in fact typically tries to recover something in restitution, if only as an offset to whatever contractual damages are available to the non-breaching party. For discussion, see S. Levmore, Explaining Restitution, 71 Virginia Law Review 65 (1985).

The Boundary between Torts and Contracts:

279

Clearly, a claim in tort cannot generate contractual liability. The fuzzy line exists because the presence of a tort might constitute a breach of contract even in cases when the tort is directed against a third party7 or because contract itself can potentially generate a tort claim8. Because codification in civil law leaves many of the statements concerning the boundary between torts and contracts open to judicial interpretation, it is of no surprise that judgemade law has been quite influential in this specific area.9 However the French principle of non-cumul contrasts with English or German law; the victim of breach of contract cannot sue in tort when there is contractual liability.10 Our paper approaches the boundary between torts and contracts from a law and economics perspective. The interaction between a contractual relationship and the possibility of a tort claim generates interesting legal issues that can be analyzed from an economic perspective.11 The reason why this blurred line has not deserved so much attention in the classical law and economics literature could well be because it is of less pertinence in common law jurisdictions than in civil law jurisdictions where contracts and torts belong to the law of obligations.12 From that perspective, our paper fills an existing gap in the literature. We propose a comprehensive analysis of the boundary between torts and contracts. In relation to some aspects, we put together different parts of the existing economic literature. On the less underdeveloped aspects, we advance the discussion by proposing new economic explanations for certain rules. The second contribution of our paper is the use of the DCFR (the Draft Common Frame of Reference)13 as the case-study for discussion. The importance of the DCFR for coding the principles and model rules of European private law is currently well established. Presumably, at a minimum, the DCFR will play an influential in the development of private law in Europe in the nearby future. For the more optimists, the DCFR will effectively harmonize private law across Europe. Our paper does not enter into the discussion of the benefits and costs

6

7 8

9 10

11 12

13

In certain cases, breach of contract can result in tort-like damages. These cases involve requests for damages for emotional distress (typically only awarded in tort), and traditionally have been limited to circumstances in which the subject matter of the contract involved a peculiarly personal or emotional subject matter – for example, the funeral home that failed to properly handle a body may be liable for emotional distress damages even though the failure is a contractual breach, not a tort. One might say that these cases were ones in which the contract itself had a predominant purpose to protect the emotional well-being of the contracting party, and that the breach by its nature injured that emotional well being. It is true, however, that these cases are rare. See C. von Bar and U. Drobnig, supra 1, p. 15. As C. von Bar and U. Drobnig, supra 1, p. 189, write, “tort and contract law are only practically completely separate from one another when accidents take place between «strangers», that is to say between persons who have no other legal connection other than the damaging event”. See H. B. Schaefer, Tort Law: General, Encyclopedia of Law and Economics (1998), 569-596. See discussion by Ogus, supra 4, at 86-92 and A. Ogus and M. Faure, Économie du Droit: Le Cas Français (2002), pp. 103-104. For a comparative view, see T. Weir, International Encyclopedia of Comparative Law (Torts), Vol. XI, (Complex Liabilities) Ch.12, pp.24-39 (1976) and C. von Bar and U. Drobnig, supra, pp.189 ff. See the importance of this perspective stressed in C. von Bar and U. Drobnig, supra, p. 17. See Ogus, supra 3, and W. Bishop, The Contract-Tort Boundary and the Economics of Insurance, 12 Journal of Legal Studies 241 (1983). We use the March 2008 version of the DCFR (available both on line and in paper, by Sellier).

280

Assunção Cristas & Nuno Garoupa

of harmonization, for which by now there is a large economic literature,14 but rather how the DCFR treats the boundary between contracts and torts from an economic perspective. Harmonization, under the DCFR, means that certain principles prevail over others. We provide an economic analysis to establish the economic superiority of one approach over the others. Summing-up, our paper makes a methodological contribution by advancing the economic analysis of contract and tort law and a legal policy contribution by applying the economic analysis to the DCFR. Law and economics tends to look at contracts and torts by making use of different models that make the analysis of the boundary between apparently methodologically difficult. On one hand, the economic theory of torts15 looks at two important issues: • Deterrence or avoidance of accidents: from the least cost avoider principle to the different liability rules, the objective is to achieve efficient levels of precaution ex ante facto in order to internalize the negative externality caused by the accident; the costs of precaution must be balanced against the costs of the accident. • Compensation of victims: having determined the efficient compensation, the objective is to develop the best mechanism to make compensation for victims effective where usually private insurance markets are paramount. On the other hand, the economic theory of contract law16 studies the efficient allocation of rights and duties between different parties. Two moments are of critical importance, contract formation and contract breach. Default rules, enforcement and damages are studied from the perspective of efficient formation (only contracts that generate an expected surplus should be formed) and efficient breach (only contracts that ex post facto generate a surplus should be performed). Furthermore, economic analysis of contracts is based on the existence of a market relationship that is repeated in nature (i. e., a consensual reallocation of resources, rights and obligations) whereas the economic analysis of torts looks at the interaction of players who lack a market relationship (i. e., a nonconsensual reallocation of resources, rights and obligations usually between strangers). The boundary between torts and contracts necessarily mixes traits of these two models. These two fields of law and economics seem to be totally unrelated but provide important insights when the intersection is considered. In other words, an economic analysis of the boundary between torts and contracts can use both models as the main source of reasoning, but needs to identify the relevant aspects to achieve the appropriate balance. Without loss of generality, we consider four significant areas of interaction: (i) third-party effects (contracts that cause a potential wrongdoing); (ii) breach of contract induced or caused by a third party; (iii) quasi-contracts; (iv) breach of contract or other forms of contractual under-performance that cause a potential wrongdoing (including the possibility of a “picking a theory”). Notwithstanding, differences between contractual liability and torts should be taken into account in order to understand and measure the consequences of assigning each of these 14

15

16

See, among others, N. Garoupa and A. Ogus, A Strategic Interpretation of Legal Transplants, 35 Journal of Legal Studies 339 (2006). From a legal perspective see C. von Bar and U. Drobnig, supra 1, p. 19. See Schaefer, supra 9. See also Robert Cooter and Thomas Ulen, Law and Economics (2007), ch. 8 or A. M. Polinsky, An Introduction to Law and Economics (2003), ch. 6. See R. Craswell, Contract Law: General, Encyclopedia of Law and Economics (1998), 1-24. See also Cooter and Ulen, supra 15, ch. 6 or Polinsky, supra 15, ch. 5.

The Boundary between Torts and Contracts:

281

four particular areas to one type of liability or another. The main differences lay typically in two central areas17: measurement or calculation of damages18 (and standard related problems such as “forseeability” for contract liability and causation and prevention for tort law), and exclusion and limitation of liability (i. e., issues related to standing broadly defined). In what concerns the latter, significant differences usually occur in contractual liability as compared to extracontractual liability (because typically there is not a relation between the involved parties previous to damage occurrence in torts).19 Taking into account these aspects is important to understand the behavioral incentives and outcome consequences of the different options we consider for regulating the boundary of contracts and torts. The DCFR expressly admits cases of exclusion and limitation of extracontractual liability [Article VI.–5:401] not too far from those admitted in contractual liability (Article III.–3:105(2) permits, in case of negligence, exclusion or restriction of remedies for nonperformance by a term of the contract limited to good faith and fair dealing). It seems that a relevant legal regime difference has been blurred. A difference might still exist in the possibility of previously determining the payment for non-performance (Article III.–3:510). The point of our paper is not to explain why from an economic perspective tort and contracts are subject to different legal regimes, but rather considering the boundary. Nevertheless, given the current state of the art in the economics of private law, the blurring of the legal regimes does not seem to be a good direction to take.20Another important and relevant area in national laws, which in practice might explain differences in many court decisions across Europe, is the statutes of limitation.21 Normally and traditionally the period of limitation is much longer in contractual liability. However, following the PECL (Principles of European Contract Law), the DCFR adopts a unified regime of limitation. The same option has been made by the German Schuldrechtmodernisierung,22 which might be seen as part of an international trend. Therefore, limitation will be no longer an argument for deciding whether opting for a particular situation should be treated as contractual or extracontractual liability. However, from an economic perspective, there are good reasons to think that periods of

17

18

19

20 21

22

For more detailed information see C. von Bar and U. Drobnig, supra 1, p. 43, who point out seven domains of important differences between contractual and tortious liability: role of fault, burden of proof, rules on the type and extend of compensation to be made, recoverability of “pure economic loss”, liability for assistants, contractual freedom to restrict or exclude liability, and time-barred limitation of actions. An economic analysis of “pure economic losses” can be found in a special issue of the International Review of Law and Economics (2007) 27, 1-109. See a very concise summary in U. Magnus, Comparative Report on the Law of Damages, Unification of Tort Law: Damages (2001), 185, 192. We do not discuss here why different rules apply to standing (remoteness) and limitation; we just observe that they are different and therefore should matter for the discussion about the boundaries between torts and contracts. An economic theory for these observed differences is explained by Ogus, supra 3, at 88-92. The main reasoning is that remotness of harm and limitation periods should differ when consent for wealth transfer is explicit (contracts) from when it is implicit (torts). See discussion in Ogus, supra 3, at 88-92. For example, with a very critical view, M.Yzquierdo Tolsada, Sistema de Responsabilidade Civil, Contractual y Extracontractual (2001), 85, 105. For a comparative analysis, see R. Zimmermann, The New German Law of Obligations. Historical and Comparative Perspectives (2005), 122 ff.

282

Assunção Cristas & Nuno Garoupa

limitation should be different given the mutual consent and repeated nature of contracts, which is not present in torts.23 In conclusion, the DCFR seems to reduce the procedural differences between contracts and torts. Obviously, this makes the economic and legal treatment of the boundary less cumbersome. However, from an economic perspective, there are good reasons provided by the literature to suspect that such a path is ill-advised.

1. Third-party Effects A wrongdoing committed by contractual parties where the injured party is a third-party to the contract, can be studied in law and economics from the perspective of vicarious liability. There is a vast literature on this topic, namely how to assign liability between the two contractual parties (e. g., employee and employer in an employment contract, or owner of car and driver in a car accident, or insured party and an insurance company in an insurance contract).24 The standard concerns include, among others, the structure of incentives for the contractual relationship, insolvency of parties, risk allocation, transaction costs and other potential distortions. Let us take three examples in order to illustrate how economic principles can be discussed in the context of the solutions proposed by the DCFR: Example 1: A works for B, a company that sells and installs windows. A went to C’s apartment to install a new window in the living room. He destroyed a pricy carpet while trying to pass with the window. From an economic perspective, as long as A and B can negotiate their contract freely, any solution is efficient. Shifting liability from A to B will be internalized ex ante facto by a renegotiation of the employment contract by which the salary will be reduced in the amount of expected damages. However, with transaction costs that preclude the internalization ex ante facto, different assignments of liability between A and B could lead to different outcomes. It is plausible that the best solution is for C to sue A since any other solution could create serious moral hazard problems (basically A would get insurance from B without actually paying for it). Though, the potential problem of judgment-proof would not make assignment of liability to B emerge as the obvious solution. Although, it could be that judgment-proof has been the driving force behind vicarious liability regimes, in law and economics, however, liability should concentrate on incentives because insurance mechanisms are usually more adequate for judgment-proof problem. In order to justify economically that B should be liable, we would need to have a clear mechanism by which the likelihood of accident caused by A is significantly affected by B’s behaviour. That could be, for example, because B imposes a level of activity or performance goals upon A that increase the likelihood of accident. Shifting liability from A to B in the absence of an activity level issue or another mechanism to justify it 23

24

Notice that, following Ogus, supra 3, at 88-92, a unified regime for limitation is clearly problematic from an economic perspective given the different characteristics underlying issues in contract and tort disputes, See R. H. Kraakman, Vicarious and Corporate Civil Liability, Encyclopedia of Law and Economics (1998), 669-681. See also Polinsky, supra 15, ch. 9.

The Boundary between Torts and Contracts:

283

will, in a first instance, cause moral hazard that generates more accidents (more carpets being destroyed) and, in a second instance, will drive selling and installing windows companies out of the market because the marginal cost of the activity has been effectively increased. According to the DCFR [Article VI.–6:201], C is entitled to sue B, as long as A is himself accountable for that damage. Considering his action as an offence to a property right, A has caused a legal relevant damage and will be liable under Article VI.–2:101 (1)(b). C might opt to sue B, A or both. Obviously, in most cases, the most attractive way is to sue B (even if A has to be part of the action). In what concerns internal relations the DCRF has no answer, as clearly excludes its application to liability of employees whether to co-employees, employers or third parties and of employers to employees arising in the course of employment [Article VI.–7:104]. If there is a notion of “course of employment” used to exclude cases where the employee was at fault in a situation which would be outside of the “sphere risk” created by employment, and hence, the employee alone is liable, the point about moral hazard is minimized.25 Example 2: A owns an apartment and has an insurance contract with B to cover all risks concerning fire. A forgets the heater on and provokes a fire. His apartment and his neighbour C’s apartment were totally destroyed by the fire. C wants to be paid for his losses. C sues B and A. A refuses payment because sustains that B is obliged to pay under the insurance contract. Apart from potentially relevant issues such as the risks covered by the contract, and thus assuming that B is liable towards A, the question is the extent to which a legal regime should permit C to choose who to sue, either alternatively or concurrently? The answer is easy if the insurance contract between A and B explicitly provides a claim for C. So, let us assume no such claim exists. Then, the alternative is between C can sue B directly and independently or C must sue A, and then A eventually turns to B as a guarantor. Under the current DCFR, A would be liable for causing a relevant legal damage under non-contractual liability [Article VI.–2:101 (1) (b). Taking into account that under the insurance contract B was liable as well and C was entitled to reparation, the DCFR does not provide a direct answer. National laws have the last word in determining whether or not liability under DCFR Book VI is limited or excluded because of the existence of another source of reparation [Article VI.–7:105]. Under national laws, C might have a full option or might be limited to sue either A or B. From an economic perspective, the efficient solution would be for C to sue A only if, as likely, A is the least-cost avoider unless other transaction costs play a significant role here (for example, some complex risk management). The relationship between A and B should be determined by standard economics of insurance contracts. Example 3: A contracts with B the total renewal of his apartment. After all the works have finished, A sells the apartment to C. The construction was poor, and humidity immediately appeared in the bathrooms and closets. C has no contract with B, but wanted to sue him for damages. Is it possible? 25

This example could also be justified under the different economic rationales for strict liability, in particular the control of the activity level by the employer in order to keep the activities of the company at the efficient level. This discussion, however, would be on the efficient design of tort law, not how liability should be allocated, and thus shape the boundary of contract and torts.

284

Assunção Cristas & Nuno Garoupa

From a legal perspective, C might be able to sue A for lack of conformity to the contract and obtain damages under contractual liability. But it could be more interesting for him to sue directly B (suppose, for example, A moved to a different jurisdiction). Except when it comes to product liability, that is, accountability of the manufacturer for damage caused by defective products [Article VI.–3:204, where included property damage in relation to consumers], the DCFR does not answer this question. From an economical perspective, the best solution is quite complex. As before, in absence of transaction costs, C suing A or B would be pretty much indifferent. However, in this example, there are potential serious transaction costs, namely information asymmetries between the different agents . Suppose the costs of A finding the quality of the repair or the renewal of the house prior to the selling are too high. Then the efficient solution is for C to sue B, and not A. By contrast, suppose A monitoring B could have been done at very low cost. Then the efficient solution could be for C to sue A or concurrently with B. In summary, tort wrongdoings, on third-parties arising within a contractual relationship, have been extensively considered by the law and economics literature. The three economic arguments to keep in mind when assessing third-party effects are: (1) in absence of transaction costs, it is irrelevant which contractual party is sued, alone or concurrently (except for differences in legal costs); (2) transaction costs should determine who is the least-cost avoider to be sued by the third party; (3) judgment-proof arguments are not overwhelming since insurance contracts should be a primarily source of solution. The DCFR presents a scratched response to the examples we have considered, not always consistent with the economic analysis.

2. Breach of Contract induced by a Third Party This is a particular case where a contractual obligation is breached by a third-party who is not part of the contract, and therefore, can only be sued in tort.26 Here there are two substantive questions, namely the extent to which: (i) efficient breach can be induced by a third-party and (ii) an inefficient breach induced by a third party should be punished by tort law. As to the first question, law and economics takes the view that, for example, a better offer by a third party who causes breach should be allowed in certain circumstances. Generally speaking, imposing doctrinal principles that penalize better offers creates inefficiencies, forces performance of suboptimal contracts and reduces efficient contract formation.27 However, one should be careful in jumping to the conclusion that a third-party induced breach of contract should always be allowed. Economists agree that it should be only allowed when the induced breach is efficient. The problem is of course that quite frequently there is asymmetric information, transaction and administrative costs, and therefore, it could be difficult for a court to determine the extent to which a specific breach induced by a third-party is actually efficient. This is quite an important relevant shortcoming. An inefficient breach induced by a third party should be deterred. The essential question is how to allocate punishment between the breaching party (contractual liability) and the third party (tort liability). Incentives have to be balanced against administrative and litiga-

26 27

For a legal comparative perspective, see C. von Bar and U. Drobnig, supra 1, p. 211 ff. See A. N. Hatzis, Rights and Obligations of Third Parties, Encyclopedia of Law and Economics (1998), 200-222.

The Boundary between Torts and Contracts:

285

tion costs. Also, because of asymmetries of information and court mistakes, one should be careful not to create over-deterrence (i. e., dissuade efficient breach). Generally, tort interference is mistrusted by legal economists. Obviously, unrestricted tort interference with contracts is inefficient since it will deter efficient breach. However, restricted tort interference raises substantive issues (for example, the extent to which the presence of a third party generates an exchange value for the underlying contract, that is, different from the use value, and therefore, tort liability is required because contractual liability is insufficient to tackle such difference) as well as pragmatic concerns (asymmetries of information between parties and court mistakes concerning adjudication are important once one wants to limit the use of tort liability). Some legal economists conclude that due to all these complications, breach of contract induced by a third party should never be allowed.28 It is important to emphasize that the efficiency of breach includes not only the arrival of a better proposal that increases the surplus but also the effect on contract formation. A completely unrestricted tort interference deters breach which reduces the ex ante facto surplus of the contract, thus undermining the general willingness to engage in contractual formation. The exclusion of tort interference fosters opportunistic breach, which also reduces the ex ante facto surplus. Therefore, in theory, the argument goes for a moderate or restricted tort interference if we explicitly consider contract formation. The DCFR provides for the liability of a person that induces a third person to breach an obligation, considering that the obligation was owed to the person sustaining the loss and the person inducing the breach intended the third person to breach the obligation and did not act in legitimate protection of their own interest [Article VI.–2:211]. Considering this provision, it is clear that under the DCFR, the breach of an obligation (either contractual or not) by a third party is to be considered as a violation of a right conferred by the law, which, if origins a loss or injury, is a “legally relevant damage” [Article VI.–2:101]. The view taken by the DCFR seems to be that the criteria to promote or deter a breach induced by a third party, or even caused by a third person who induced another to breach the contract, should not vary. From the aggrieved party point of view, the fact that others besides the party to that contract are liable represents an option concerning whom to sue. Considering that losses are the same, the aggrieved party only has to decide whether to sue the party, the third party or even the person who induced the third party all together, concurrently, or separately. This multiple choice is particularly relevant when one of them is insolvent (notwithstanding the role of insurance as discussed in the previous section). Normally, it is expected that the injured party will follow the easiest way, being this to sue the other party to the contract. However, as the view take by the DCFR is that the possibility of suing the third party cannot be excluded. From an economic perspective, the solution found by the DCFR can be seen as too favorable to tort interference. However, the way the solution is formulated leaves a reasonable degree of action to the courts; a more restrictive interpretation by the courts could produce a result closer to the economic reasoning we have discussed before.

28

Id.

286

Assunção Cristas & Nuno Garoupa

3. Quasi-contracts Another class of issues concerns semi-contractual relationships that are not embodied by a formal contract. Some examples include hypothetical contracts imposed by duties (i. e., duty of rescue) that are quite important in certain contexts (i. e., the medical profession), performance to another, unauthorized intervener, benevolent intervention in another’s affairs, or pre-contractual rights and obligations (when parties have manifested an intention to form a contract).29 Quasi-contracts are not contracts in the sense that there is not an explicit consent (as required by the rules of contract formation). Quasi-contracts presuppose implicit consent and can be rationalized as situations where requirement of explicit consent would be very costly for the parties to obtain (hence, quasi-contracts are appropriate when they lower transaction costs). Therefore, quasi-contracts should only be enforced under two conditions: (i) transaction costs deter a contract and (ii) it is clear that the wealth transfer is positive.30 The economic literature is divided by those who view quasi-contracts as a particular form of torts and those who argue that quasi-contracts should be the object of special and more appropriate liability rules. The most controversial issue is how different liability regimes affect the structure of incentives to obtain explicit consent and therefore shape contract formation.31 The economic approach to contract formation provides some guidance to the extent we should accept quasi-contracts claims and how to treat them from a contract vs. tort perspective. For example, the treatment of pre-contractual liability has been based on the observation that when parties manifest an intention to form a contract, some pre-contract investment might be necessary (for example, searching for contractual partners, acquiring and disclosing information about the value of the contract; it could also involve production activities that potentially generate gains if the contract is successful). Naturally, there is a risk that the contract is not actually formed, and therefore such pre-contract investment is risky (furthermore, in many instances, such pre-contract investment is contract-specific and is valued zero if the contract does not take place). A rule recognizing pre-contractual liability shifts the risk to the party undermining contractual formation. However, that raises the potential costs of entering in pre-contract manifestation and, in the limit, transforms

29

30

31

The designation of “hypothetical contracts” comes from R. A. Long, A Theory of Hypothetical Contracts, 94 Yale Law Journal 415 (1984). For a general overview of pre-contractual liability, see Precontractual liability in European Private Law, John Cartwright and Martijn Hesselink, eds., Cambridge University Press, forthcoming (2009). See discussion by G. de Geest, Comparative Law and Economics and the Design of Optimal Legal Doctrines, in Law and Economics in Civil Law Countries (ed. by B. Deffains and T. Kirat, 2001). He provides four sources of transaction costs (time, incapacity, errors and serious bilateral monopolies) and seven situations (duty of rescue, gestion d’affairs, compulsory licensing, obligation to sell, the right to make a wall common property, the right to take a way out in a parcel of land, and the use of a fence to an owner). See also Wouter P. J. Wils, Who Should Bear the Costs of Failed Negotiations? A Functional Inquiry into Precontractual Liability, 4 Journal des Economistes et des Etudes Humaines 93 (1993). See C. T. Wonnell, Unjust Enrichment and Quasi-Contracts, Encyclopedia of Law and Economics (1998), 795-807. See also Ogus, supra 3, at 168-173.

The Boundary between Torts and Contracts:

287

pre-contract bargaining into actual contractual formation without demanding formal explicit consent (with courts establishing contract liability in early stages of the negotiation).32 Some degree of pre-contractual liability seems optimal, not full since it would provide excessive reliance (moral hazard), not zero since it could provide for less than optimal precontract investment. The controversial question is the extent to which pre-contract investments are productive (hence they need to be incentivized by pre-contractual liability) or are merely private redistribution between parties (that is, they might be excessive even in the absence of moral hazard, therefore, a pre-contractual liability rule would be inefficient). Generally, the question of quasi-contract liability should be addressed only if, in the absence of a liability rule, quasi-contracts underperform. It makes little sense to engage in an economic analysis of the appropriate liability regime for quasi-contracts if they perform reasonably well in the absence of such regime. This would lead to waste and important distortions.33 Thus, it is possible that some quasi-contracts should be subject to some liability rule in order to be efficiently incentivized, but not all. For example, if market forces are enough to guarantee appropriate pre-contract investment, then this type of quasi-contract should not be the object of liability.34 Establishing the need for a liability regime for quasi-contracts is not easy, however, it is certainly not the most difficult issue. The nature of such liability regime is less obvious, and therefore, more complicated. Whereas contracts require explicit consent (which is part of efficient contract formation) and torts presuppose a nonconsensual transfer (hence the emphasis on deterrence), quasicontracts are hybrid because the line between implicit consent and non-consent is not easily observed and hard to verify in many contexts. A pure contractual liability approach could over-deter quasi-contracts (by effectively transforming implicit consent into non-consent) and would raise serious questions concerning the determination of damages because observing the value of the unformed contract might not be feasible.35 A liability regime that dilutes explicit consent to include contracts and quasi-contracts could generate problems of underdeterrence (thus blurring contractual formation and generating too much opportunism).36 Tort liability is also a costly alternative. Even if needed to incentivize pre-contract investment, for example, it needs to be developed in a creative way to avoid over-deterring parties from entering into pre-contractual bargaining (which in turn would reduce effective contract formation). Presumably, tort liability is more effective than contracts to guarantee restitu32

33

34

35

36

See B. Hermalin, A. Katz and R. Craswell, The Economics of Contract Law, in the Handbook of Law and Economics (ed. by S. Shavell and A. M. Polinsky, 2007), volume 1, section 3. See A. Katz, When Should an Offer Stick? The Economics of Promissory Estoppel in Preliminary Negotiations, 105 Yale Law Journal 1249 (1996). He takes the view that liability for early reliance would probably lead to significant waste. See also Juliet P. Kostritsky, Bargaining with Uncertainty, Moral Hazard and Sunk Costs: A Default Rule for Precontractual Negotiations, 44 Hastings Law Journal 621 (1993). See discussion by O. Grosskopf and B. Medina, Regulating Contract Formation: Precontractual Reliance, Sunk costs and Market Structure, 39 Connecticut Law Review 1977 (2007) See Grosskopf and Medina, supra 34, making the point that expectation damages in this context are difficult to assess since the value of the contract not formed is difficult to calculate. The authors argue that presumably it could be approximated by the investments incurred by whole parties, but then the market should have been the best mechanism in the first-place. However consider O. Ben-Shahar, Contracts Without Consent: Exploring a New Basis for Contractual Liability, 152 University of Pennsylvania Law Reveiew 1829 (2004) proposing contractual liability without consent.

288

Assunção Cristas & Nuno Garoupa

tion, but the incentives in the contractual bargaining are also determined by the relevant expectations about the future contract. The debate shifts into the extent it is conceivable that courts can apply tort liability to quasi-contracts without damaging the incentives to engage in those same quasi-contracts in the first place.37 In that respect, the economic efficient solution cannot be provided without a serious empirical effort. In common law, actions that seek damages in a quasi-contractual nature are regulated by the law of restitution. Many of these actions are anomalous because liability is benefit-based rather than fault- or harm-based. In continental law, most cases are dealt in the general frame of the unjust enrichment which determines an obligation to restitute. But when a right to damages emerges it is controversial whether the regime should follow contract or tort law. The particular case of pre-contractual liability has raised substantive comparative work.38 The trend from traditional (maximal) contractual formalization to the modern informality and flexibility has apparently narrowed down pre-contractual liability and expanded contractual liability. The efficiency of such trend is debatable since implicit consent is potentially unverifiable. In fact, little economic difference can be made of verifiable implicit consent and explicit consent (in that respect, the trend is not problematic). The less obvious question is the strategic or opportunistic behavior that might emerge in terms of unverifiable implicit consent. Apparently this trend is more of a consequence of the changes in the technology of contracting and judge-lawmaking, and less due to reforms of private law. For example, culpa in contrahendo was fundamentally case law (derived from business litigation in court) and only later come into statute law in some jurisdictions. Comparativists identify pre-contratual liability as (i) quasi-contractual liability (German & English laws), (ii) tort liability (French law), (iii) liability for breach of a promissory estoppel (in common law, namely the US). The view seems to be that the US solution is less demanding for the breaching party (doctrine of estoppel) than in German law (culpa in contrahendo), but these solutions have converged somehow.39 The French option for torts can easily fit our economic model, where it is clear if the solution incentivizes pre-investment appropriately.40 The doctrine of estoppel is less straight37

38

39

40

The leading article by L. Bebckuk and O. Ben-Shahar, Precontractual Reliance, 30 Journal of Legal Studies (2001) makes the point that the efficient solution is an ingenious form of liability rules to protect reasonable pre-contract reliance without incurring in superfluous investments plus promote efficient formation of contracts. They argue that tort liability could have such role. In their model, the investments in reliance are suboptimal in the absence of liability but excessive if strict liability is imposed. They also argue that pre-contractual liability does not necessarily deter parties from entering the bargaining. See also R. Craswell, Offer, Acceptance and Efficient Reliance, 48 Stanford Law Review 481 (1996). See Paula Giliker, Precontractual Liability in English and French Law, Kluwer Law International (2002) and Joachim Dietrich, Classifying Precontractual Liability: A Comparative Analysis, 21 Legal Studies 153 (2001). See discussion by Alyona N. Kucher, Precontractual Liability: Protecting the Rights of the Parties Engaged in Negotiations, working-paper (2008). She also argues that the use of ethical rules could be more appropriate for precontratual stage. For a discussion of gestion d’affaires (when one person voluntary or unintentionally does something that benefits another person thus creating a non-contractual obligation) in French law, see John Bell, Sophie Byron and Simon Whittaker, Principles of French Law, Oxford University Press (1998) and gestione di affair (assuming responsibility for actions that benefit another without a contractual

The Boundary between Torts and Contracts:

289

forward because some authors see it as effectively closer to a “no liability” situation (that is, a solution abstaining from extending a good faith or a duty of care principle to pre-contractual stage) whereas others argue that is closer to contracts since expectations matter (for example, reasonable rendered services).41 The German culpa in contrahendo, being an intermediary solution, has raised more debate, namely in the nature and basis for liability. The basis for pre-contractual liability seems to be closer to contractual than tort.42 As to the nature, it apparently draws from both since failed contractual negotiations do not imply a negligent or intentional behavior to create and defraud expectations. Presumably, the German solution is the best candidate to fit with the complicated liability regime advocated by legal economists, but it becomes an empirical question the extent to which the calibration between mere sunk costs of bargaining and foregone opportunities due to opportunistic behavior by the other party is achieved efficiently by courts.43 Apart from relevant considerations about pre-contractual liability as contractual or extracontractual liability, some thoughts about measurement of damages are also important as they might differ from one sort of cases to another. Three different typical cases should be analyzed: contracts void or avoidable that have been avoided; contracts avoidable that have not been avoided; contracts that have not been concluded due to a break of negotiations. All these cases are addressed by the DCFR. Article II. – 3:301 considers liability for negotiations contrary to good faith. That includes entering into a negotiation knowing that they will never result in a contract or to continue negotiations after deciding not to conclude the contract, breaking off negotiations contrary to good faith or fair dealing. As explicitly pointed in the PECL comments, it is possible that this same situation consists of misrepresentation. In that case, non-contractual liability (presumably of tortious nature) may be called as well. This article will constitute, therefore, an alternative basis for the claim. In fact, Article VI.–2:210 considers legal relevant damage any loss caused by a person as a result of another’s fraudulent misrepresentation. In comment G to PECL Article II.–2:201 (now Article II.–3:301) can be read that losses include expenses, work done, loss on transactions made in reliance of the expected contract and in some cases loss of opportunities. But the aggrieved party cannot claim to be put into the position in which that party would have been had the contract been duly concluded and performed. Accordingly, Article II.–7:214 (similarly to Article II.–7:304 applicable to contracts void or avoidable due to infringement of fundamental principles or mandatory rules) determines that a party who has the right to avoid a contract is entitled to damages from the other party for any loss suffered as a result of the mistake, fraud, coercion, threats or unfair exploitation,

41 42

43

obligation) in Italian law, see Thomas Glyn Watkin, The Italian Legal Tradition, Dartmouth Publishing Company (1997). See Dietrich, supra 38. According to Dietrich, supra 38, it includes failed contractual negotiations, plaintiff believes a contract was to exist, a plaintiff has relied detrimentally on that belief or expectation of that contract, the defendant’s conduct was obviously opportunistic. See also Peer Zumbansen, The Law of Contracts, in Introduction to German Law (ed. J. Zekoll and M. Reimann, 2nd edition, 2005). According to him, culpa in contrahendo is a notable expansion of pre-contractual duties promoted by the courts and codified only recently (reform of BGB 2002). It is independent of whether or not the contract is actually concluded. Although designed to compensate for the exposure of a potential contractual partner to risk in pre-contractual settings, a mere social interaction or contact is explicitly excluded.

290

Assunção Cristas & Nuno Garoupa

provided that the other party knew or could reasonably be expected to have known of the ground for avoidance. This last requisite is probably met when we talk about negotiations without good faith. Damages are to place the aggrieved party in the position it would have been if the contract has not been concluded, whether or not the contract has been avoided. If the contract has not been avoided, but maintained instead, damages will not exceed the loss caused [Article II.–7:214 (2)]. That means that a decline in the value of property will not be considered (see Comment C). For other respects, rules on damages for non-performance of contractual obligations are applicable with appropriate adaptation. This makes it clear that, when concerning measure of damages, the rule differs from the expectation damages contractual liability rule. However, considering the case of fraudulent misrepresentation, Article VI.–6:101 (1) will be applicable as well. The rule here is to reinstate the person suffering the legally relevant damage in the position that person would have been in had the legally relevant damage not occurred. If the contract would have not been concluded, the solution matches with the solution referred above. But considering that the contract would have been concluded but with different content, is the party entitled to expectation damages or only to reliance damages? If we answer positively to the first part, then we will come to the conclusion that the remedy is contrary to the solution of the DCFR in Book II section 2. Where fraudulent misrepresentation has to do with inobservance of information duties, the DCFR previews as well entitlement to damages when a loyalty duty has been violated. That happens not only when negotiations are not concluded contrary to good faith [Article II.–3:301 (2) and (3)] but also when breach of confidence occurs [Article II.–3:302, but also VI.–2:205]. Breach of confidence might occur in a pre-contractual stage, during the contract or even after extinction of the contractual relation (post-contractual liability). In what concerns pre-contractual liability, it is not totally clear if the measure of damages might differ depending on the basis of the claim being articles in Book II or in Book VI. Under Article II.–3:302 (4), the party in breach of the confidentiality duty might be ordered to pay any benefit obtained by the breach. If we do not take into account rules on Book VI, the DCFR gives a similar answer to the three cases referred above: reliance damages. From an economic perspective, although reliance damages are usually not perceived as efficient for contractual breach, they are presumably the second-best solution for pre-contractual liability as earlier discussed (since the firstbest solution is presumably not feasible due to the nature of unformed contracts). Generous expectation damages would over-compensate the victim of breach of a quasi-contract (hence, reducing formation of quasi-contracts) whereas restitution damages would seem to approximate the opportunity costs of the failed negotiations.44

4. Wrongdoings caused by Breach or Non-performance of Contract The last class of issues is vast and includes wrongdoings caused by breach of contract. We look at two quite distinct situations. One refers to the important distinction in law and economics between torts among strangers without any market relation (pedestrians and drivers) and those with a market relation underlying the wrongdoing (such as product liability). Wrongdoings caused in the market setup are analogous to a breach of contract in the sense that the market relationship presupposes a contract (in the economic sense, not 44

See Craswell, supra 16, and Ogus, supra 3, at 123-125.

The Boundary between Torts and Contracts:

291

necessarily in the legal sense). They are different from other torts because one party (the seller) can use the price of the good or service to make the other party pay ex ante facto for the wrongdoing (the buyer). The extent to which this price effect is detrimental to liability or should be taken into consideration when determining damages is a matter of debate.45 The ability of manufacturers to make consumers pay ex ante facto for liability depends crucially on the market structure, and therefore the price effect becomes an important aspect. Also, due to the nature of a market relationship, it could be that market incentives are enough to provide for efficient care. Firm reputation and branding are good candidates. The second situation considers those few cases where the injured party can strategically choose to pursue breach of contract under contractual liability or tort liability (for example, in restitution).46 Generally, the legal provisions are not so generous that a contract can always generate a tort claim, but there are particular situations when the injured party can strategically choose between pursuing compensation by contractual liability or by tort liability, a “picking the theory” choice. This strategic decision is driven by the determinants of litigation, including the extent to which procedure, discovery, evidence rules, determination of damages can be different and therefore could generate a more pro-plaintiff setup under tort liability than under contractual liability or vice-versa. The general view seems to be that it is not the case under the law of restitution in American courts, where wrongdoings caused by breach of contract are occasional.47 Apparently the English regime is more flexible in that respect48 and definitely is the German49 or the Italian50 solutions. On the contrary, under the French principle of non-cumul, a victim of breach of contract cannot pursuit a tort claim concurrently, that is, when an obligation exists by virtue of a contract, it cannot also exist in tort.51 Notice that this analysis only makes sense if there are differences across contracts and torts. A legal regime that effectively eliminates these differences will also reduce the “picking the theory” choice. The option for a principle of non-cumul seems wise from an economic point of view. First, because obligations freely negotiated should supersede potential tortious wrongdoings. 45 46

47 48

49

50

51

See Schaefer, supra 9 and Polinsky, supra 15, ch. 14. See S. A. Smith, Concurrent Liability and Unjust Enrichment: The Fundamental Breach Requirement, Law Quarterly Review 245 (1999). In particular, if the facts satisfy the elements of two causes of action, a breach of contract can also support an act in tort, or vice versa. See Wonnell, supra 31. See Ogus, supra 3, at 88 and Raymond Youngs, English, French and German Comparative Law, 2nd edition, Routledge (2007). In particular, after Henderson v Merrett Syndicates (1994). See also Jack Beatson, Restitution and Contract: Non-Cumul?, 1 Theoretical Inquiries in Law (2000). Which allows concurrent claims in contract and tort. That has to do with several reasons, namely because damages for pain and suffering are not available in a contract suit. See T. Weir, supra 10, at 32. However, different limitations rules and a common reason for concurrent claims no longer hold after the 2002 changes. In Italian law, a wide range of (contractual) damages can be recovered in torts (non-contractual liability); although the defendant is not liable for all every single consequence of breach, damages are not limited to foreseeable losses. Indirect damages are recoverable when they can be attributed to breach through standard principles of causation. Even future losses can be recovered, if they are based on an inevitable situation. See Guido Alpa and Vincenzo Zeno-Zencovich, Italian Private Law, Routledge (2007), in chapter seven. See, for example, M.Fabre-Magnan, Les Obligations (2004), pp.641-643, for a short and clear explanation of the principle of non-cumul.

292

Assunção Cristas & Nuno Garoupa

Second, the possibility that breach of contract could generate a tort claim undermines efficient breach (unless contractual damages are too low and therefore tort damages operate as a mechanism to achieve efficient breach, an argument that seems quite difficult to make in general terms since expectation damages tend to prevail). Third, ex ante facto, a potential tort claim could deter formation of contracts or increase negotiation costs to overcome potential future tort claims.52 As a consequence, allowing tort claims concurrently to breach of contract can only be efficient in very exceptional conditions, for example, when contractual damages are unable to internalize the losses of non-performance due to externalities or the existence of serious asymmetries of information that undermine the optimality of contractual rules.53 Furthermore, given the different nature of contracts and torts (consensual and nonconsensual transfers), it seems pernicious to allow tort liability as a default within contracts. Unless hybrid situations arise, contracts and torts should not overlap due to the very different issues that each generates as earlier explained in the introduction (namely, efficient deterrence of torts versus efficient formation and breach of contracts). There is also a dynamic efficiency argument for the French solution. In French law, because non-cumul is the rule, over the years, a large body of law has evolved under contract law to extended responsabilité contractuelle to include actions that are very close to tort law in substance. Due to the non-cumul, such rules have to be housed within contract law. In other words, either the legal system sticks to the non-cumul as under French law and accepts the growth of responsabilité contractuelle or the system decides that these cases are actually more like tort law cases and allows them to be dealt with as such, despite the presence of a contractual relationship. This raises the economic question if such actions are better treated under tort law (concurrently with contracts) than under an extended contractual responsibility. The nature of mutual consent in contracts seems to indicate that the French approach is dynamically more efficient too. The only exceptions should be damage situations that require high transaction costs to achieve mutual consent ex ante facto. Inserting these cases into an extended contractual responsibility would raise the problems we have discussed with quasi-contracts, either by diluting the definition of mutual consent or by increasing the costs of contractual formation (since those transaction costs become part of the costs of contract formation). In fact, example 1 given above can be used here as well. Besides the extracontractual liability, it could be asked if A’s behavior might be qualified as well as defective performance of the contract and therefore generate contractual liability. C would be in advantage as B would be liable unless proved that non-performance was excused [Article III.–3:701]. Should this obligation to act carefully when performing a contractual obligation be treated among contractual liability as well? Protection duties are normally invoked related to physical integrity of the other party: when entering into a contract and also during the contract parties should act in order to assure protection to the other party (especially personal rights). This duties or obligations might in fact function independently from the main contractual obligation. Another impressive example might be that one of the defective washing machine that totally ruins a great number of clothes. In what concerns the defective machine itself, a wide range of remedies is applicable and damages might be awarded in the context of contractual 52 53

See Ogus, supra 3, at 86-87 for further discussion. French law allows for derogation of the principle of non-cumul for reasons of public interest which could be interpreted as serious negative externalities. See Ogus and Faure, supra 10, at 104.

The Boundary between Torts and Contracts:

293

liability. But damage caused to the particular clothes is to be included in the contractual liability or should be considered as extracontractual liability due to an offence to property rights? When selling the machine is it admissible to limit liability for any damage caused by the machine to a certain amount of money? And if that is the case, calling extracontractual liability is precluded? An economic answer would have to balance out the relevant transaction costs (including risk sharing) as we have argued before. If a great number of clothes are ruined solely caused by a defective washing machine, we could be tempted to endorse a regime of strict liability in torts. The well-known problem with such solution is that the producer will use the price of the washing machine to internalize such liability. If the producer has a dominant position in the market for washing machines, this will essentially make the consumers to pay for potential liability. Obviously there are still arguments that could go one way or the other (different risk preferences, different time preferences), but the main result is that in expected value, the consumer would be pretty much indifferent between strictly liability imposed on the producer, any form of contractual liability, or no liability. However, if the producer has no dominant position in the market for washing machines (that is, the market for washing machines is very competitive), then the producer cannot use the price to internalize the externality. Presumably these complex balances should be left to the courts in a case-to-case approach rather than to the statute. Should contractual liability be broader in order to take into account accessory obligations – e. g. emerging from good faith and fair dealing principles – and not only principal and secondary contractual obligations54? The so-called security or protective duties when performing a contractual obligation should be considered among torts or contractual liability? The answer to these questions requires a careful balance of the ex ante facto effects on avoidance (in conformity with the principle of least cost avoidance) and ex post facto administration of appropriate adjudication.

Concluding Remarks The economic analysis of law has not given a lot of attention to the boundary line between contracts and torts. We have identified four classes of situations that are relevant, namely (i) third-party effects (contracts that cause a potential wrongdoing); (ii) breach of contract induced or caused by a third party; (iii) quasi-contracts; (iv) breach of contract or other forms of contractual under-performance that cause a potential wrongdoing. Whereas the current economic literature offers good guidance for the first two classes, we have seen that quasi-contracts and breach of contract that causes tortious claims are less established. This paper helps filling that gap. As to quasi-contracts, we have argued that approximating their regime to a contractual setup would raise the costs of forming quasicontracts, and therefore, inefficiently deter them. However, torts of omissions are delicate and should be treated with care because they are subject to serious information problems that could undermine the achievement of efficient prevention. As to contractual and tort liability operating concurrently, we have argued that unless there are serious shortcomings with contractual damages (that is, when contractual damages are unable to internalize the losses of non-performance due to serious externalities for

54

Strongly defending a broader application of contractual liability, criticising restrictive court decisions, M. Yzquierdo Tolsada, supra 21, at 96-99.

294

Assunção Cristas & Nuno Garoupa

example), a general principle of noncumulative (presumably with some possible but clearly exceptional derogations) seems as a more adequate legal framework. We have used the DCFR as a case-study for our economic approach. Although, the DCFR does not tackle the theme of our paper systematically, it definitely promotes an important set of principles that should apply to the boundary between contracts and torts in Europe. Quite surprisingly, on the main point we have discussed in the paper (contractual and tort liability operating concurrently), the DCFR does not provide a clear statement, only a general impression that the contractual and tort routes for litigation are both open concurrently. In that sense, our paper provides a skeptical view, and might be a good starting point to rethink such an impression.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR Pierre Larouche*

Most of the draft CFR (DCFR) concerns contract law, yet in Book VI the DCFR deals with another major component of private law which could not be ignored in the work of the Economic Impact Group (EIG). What the DCFR accurately if dryly defines as “noncontractual liability arising out of damage caused to another” roughly corresponds to tort law as it is known in common law systems, or the law of delict (responsabilité civile délictuelle, Haftungsrecht), a sub-part of the law of obligations in civil law systems. This contribution is the only one to deal solely with Book VI DCFR,1 and accordingly it cannot cover every interesting feature of that book. Rather, this contribution focuses on one central aspect of the law of non-contractual liability, namely the general limitations on the scope of non-contractual liability.2 After a preliminary discussion on the idea of limiting liability (1), this contribution then answers the two issues which are central to the work of the EIG, namely whether there is a need for unification or harmonization of the law on this issue (2) and whether the substantive solution retained in the DCFR is optimal from a law and economics perspective (3).

1. Preliminary matters – Limiting liability As a preliminary matter, it is useful to explain at greater length what is meant by limiting liability.

*

1

2

Professor of Competition Law and Co-Director, Tilburg Law and Economics Centre (TILEC), Tilburg University. The author wishes to express his thanks to the members of the Economic Impact Group (EIG) created as part of the CoPECL network of excellence funded via the 6th EC Framework Programme for Research and Development for their comments and suggestions, and in particular to Michael Faure and Filomena Chirico. Comments from two anonymous referees as well as from the participants in a Workshop on the DCFR at the Universität Bonn in November 2008, in a TILEC seminar in December 2008 and in a TICOM talk in February 2009, are gratefully acknowledged. Responsibility for the content of this paper remains the author’s alone. In their contribution, Cristas and Garoupa touch upon non-contractual liability as well, but only in its relationship with contractual liability and in particular the cases where the two areas (contract law and non-contractual liability) overlap. For more legal background on this issue, see W. van Gerven, J. Lever and P. Larouche, Tort Law (Oxford: Hart Publishing, 2000) [hereinafter Tort Law] at Chapter 2 (scope of protection), as well as J. Spier (ed.), The Limits of Liability – Keeping the Floodgates Shut (The Hague: Kluwer Law International, 1996).

296

Pierre Larouche

1.1. Basic conceptual architecture of non-contractual liability At the root, non-contractual liability is about shifting losses from the victim (usually the plaintiff) to the defendant. The basic conceptual architecture is very simple, turning as it does around three concepts: a loss or injury suffered by the victim (usually designated as “damage”3), an event attributable to the defendant and a causal link between the two. Nowhere is this basic architecture better expressed than in the main provision of the French Civil Code concerning non-contractual liability, Article 1382: Tout fait quelconque de l’homme, qui cause à autrui un dommage, oblige celui par la faute duquel il est arrivé, à le réparer. Anyone who, through his act, causes damage to another by his fault shall be obliged to compensate the damage.

The basic rule of Article VI.–1:101(1) DCFR echoes this conceptual architecture, 200 years later, the bold spirit of the Enlightenment leaving way to decidedly more technocratic drafting.

1.2. Accountability as a limit on liability Yet, in all legal systems, limitations are put on this simple conceptual architecture: not every event attributable to the defendant and which causes damage to the plaintiff will lead to liability. The issue is where to attach these limitations on the event-causation-damage architecture. In every system, limitations are attached in relation to the event: not every course of action (or omission) of the defendant can lead to liability. Very broadly, the most common limitation, found in every system (including in Article 1382 C.civ., quoted above), involves the notion of fault:4 the course of action must somehow be at variance with the care to be expected from the defendant if liability is to follow. Alternatives to fault comprise notions such as the creation of risk or the control over other people or things, which are often linked with liability not based on wrongful conduct (or ‘strict liability’ to use the most widespread term5). The DCFR regroups these limits under the notion of “accountability”: somehow the defendant must be answerable for the event. Law and economics scholars have already dedicated considerable resources to investigating these types of limitations attached to the event. Indeed, since these limitations are attached to the concept closest to the defendant, it is possible to analyse them in a consequentialist fashion, with a view to the deterrence or compensation functions attributed to non-contractual liability in the literature.

3

4

5

Not to be confused with damages, which is the remedy whereby the defendant is ordered to pay a sum of money to the plaintiff by way of reparation. We leave aside here the complex distinctions between fault and wrongfulness and assume for the sake of discussion that the two are subsumed under “fault”. See Tort Law, supra, note 2 at Chapter 6 (Liability not Based on Conduct).

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

297

1.3. A priori limits on liability Beyond accountability, legal systems will often add extra limits to non-contractual liability. These limits are not linked to the defendant as such: they can be attached to the event, the causal link or the damage. In a sense, they are a priori limits, which seek to keep certain cases outside of the law of liability altogether, before the actual facts of any given case are examined. By way of a summary and non-exhaustive list: • the very essence of English tort law is that not every type of conduct gives rise to liability: rather, the conduct must fall within the conditions of a specific tort. Amongst the various torts, the tort of negligence6 is the largest, but others are also significant: nuisance, trespass, negligent misrepresentation, etc.; • the largest tort under English law, negligence, requires that the plaintiff establish that he or she was owed a duty of care by the defendant, i. e. that the defendant actually needed to exert care towards that specific plaintiff. How the duty of care is established has been a controversial issue in English law for the past 100 years;7 • under German law, the plaintiff must fit the claim under one of three main provisions. Under § 823(1) BGB, liability arises only if the damage concerned one of a list of protected interests. Under § 823(2) BGB, liability arises as a consequence of a breach of another legal provision,8 but then only if the plaintiff and the damage suffered by the plaintiff fell within the protective scope of the provision in question (Schutzzweck der Norm).9 The last main heading of liability, § 826 BGB, does not offer a limited scope of protection, but it applies only to conduct which is contra bonos mores.

1.4. A priori limits under the DCFR: the notion of ‘legally relevant damage’ The conceptual architecture of non-contractual liability under the DCFR is set out in the basic rule of Article VI.–1:101: A person who suffers legally relevant damage has a right to reparation from a person who caused the damage intentionally or negligently or is otherwise accountable for the causation of the damage [emphasis added].

The three basic elements described earlier are present: damage is suffered by the plaintiff/ victim, there is an event for which the defendant is accountable (because of intention, negligence or otherwise) and the damage and event are causally linked. Yet there is an additional element, extraneous to these three: the damage must be “legally relevant”. Via this concept of “legally relevant damage”, the DCFR introduces a priori limits to liability of the kind alluded to at the outset.

6 7 8 9

Not to be confused with negligence as a standard for assessing the conduct of the defendant. Tort Law, supra, note 2 at 50-57. Provided that the defendant was at fault. Tort Law, supra, note 2 at 227-231. The Schutzzweck der Norm has been elevated to the rank of a general and normative causation theory in German literature (ibid. at 396-398, 403-405), but its main use is under § 823(2), § 839 BGB and for the various risk-based liability (Gefährdungshaftung) regimes.

Pierre Larouche

298

Beyond this seemingly clean concept lurks a complex set of provisions. At Art. VI.–2:101 DCFR, “legally relevant damage” is defined as loss10 or injury either (i) to a right conferred by law outside of the DCFR or an interest worthy of protection, but then only if certain conditions are fulfilled or (ii) falling within one of the cases explicitly recognized in Art. VI.–2:201 and ff. The latter provisions are a mixed bag, comprising both protected rights and interests as well as specific courses of conduct. The following illustration gives an overview of the definition of “legally relevant damage”. Legally relevant damage (VI.-1:101) Right conferred by law (VI.-2:101(1)(b)) Interest worthy of legal protection (VI.-2:101(1)(c))

OR

OR

ONLY IF fair and reasonable (VI.-2.101(2) and (3)): Ground of accountability Nature/proximity of damage Reasonable expectation of victim Public policy

Protected interest explicitly listed in DCFR itself: Personal injury (VI.-2:201, 202) Dignity, liberty and privacy (reputation) (VI.-2:203) Property or lawful possession (VI.-2:206) Unlawful impairment of business (VI.-2:208) Environment (for State) (VI.-2:209) Specific course of conduct Communication of incorrect information (VI.-2:204) Breach of confidence (VI.-2:205) Reliance on incorrect advice (VI.-2:207) Fraudulent misrepresentation (VI.-2:210) Inducement of contractual breach (VI.-2:211)

Table 1 – ‘Legally relevant damage’ in Book VI DCFR Without engaging into an in-depth comparative discussion, it can be seen that: • the conceptualization of damage as a violation of a right or interest (as opposed to the actual loss or injury) is borrowed from German law and similar systems such as Italian law (danno ingiusto); • the condition that the damage be proximate and that it be fair and reasonable for noncontractual liability to extend to the violation of the right or interest in question resembles the requirement of duty of care under the tort of negligence, the most important tort under English tort law. It also echoes the notion of Schutzzweck der Norm as it is embedded under German liability law. 10

Art. VI.–2:101(1) states that “loss” includes economic or non-economic loss, so that as long as the damage was legally relevant, the non-economic loss (pain and suffering, loss of amenity, etc.) resulting from such damage will also be compensated. Among the particular instances of legally relevant damage defined at Art. VI.–2:201 and ff., only Art. VI.–2:202 (loss suffered by third persons as a result of another’s personal injury or death) distinguishes between economic and non-economic loss.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR •

•

299

the limitation of liability in line with a list of protected rights and interests comes directly from German liability law, originally at § 823(1) BGB but also reflected in regimes of risk-based liability (Gefährdungshaftung) as well.11 The rights and interests listed in the DCFR are also broadly similar to those found in German liability law; the limitation of liability in line with a list of specific courses or types of conduct reflects the basic structure of English tort law, where the main tort, negligence, is said to evolve in step with the “categorization of distinct and recognizable situations as guides to the existence, the scope and the limits of the varied duties of care which the law imposes”12. In addition to the broad tort of negligence, the law comprises a series of more specifically delineated torts.13 More specifically, the types of conduct enumerated in the DCFR as giving rise to legally relevant damage tend to match well-known categories under the law of negligence or specific torts under English law.14

In practice, in many cases, the damage will be legally relevant on account of two or more aspects. For instance, let us assume that as a result of the negligent disclosure and use of confidential information (say, medical information) about a person in the course of a defamatory attack, that person suffers a serious psychological problem. Here the injury will be legally relevant because (i) it is physical injury (Article VI.–2:201 DCFR), (ii) reputation and privacy are also protected (Article VI.–2:203 DCFR) (iii) it results from a breach of confidence (Article VI.–2:205 DCFR) and (iv) in all likelihood privacy qualifies as a right conferred by the law within the meaning of Article VI.–2:101(1)(b) DCFR. Yet despite all this redundancy, some damage could still not qualify. Indeed the impact of this complex definition can be best understood by ascertaining what is left outside, i. e. what will not qualify as “legally relevant damage”. At first glance, the main item of damage which would not qualify is pure economic loss, i. e. the financial loss (whether loss of profit or sunk costs) which is not linked with any physical loss (injury to the person, damage to property).15 That type of loss is not featured in the list of protected interests in the DCFR. Yet it must be noted at the outset that the DCFR does not completely exclude pure economic loss. First of all, such loss can qualify as legally relevant damage if it meets the criteria of the general definition of Article VI.–2:101 (b) or (c) (violation of a right conferred by law or an interest worthy of legal protection). Conceivably, if pure economic loss in general is conspicuously absent from the particular instances of legally relevant damage at Article VI.–2:201 and ff. (contrary to other protected interests), it probably reflects the opinion of the drafters of the DCFR that it should not qualify under the general definition of Article VI.–2:101, but courts may differ. Secondly, those particular instances do allow for the compensation of pure economic loss in a number of well-circumscribed circumstances. Pure economic loss becomes legally relevant damage if it either so extensive that it qualifies as 11 12

13 14

15

Tort Law, supra, note 2 at 63-68, 546. House of Lords, Caparo Industries plc v. Dickman [1990] 2 AC 605 at 618, referring to T. Weir, A Casebook on Tort (London: Sweet & Maxwell). Tort Law, supra, note 2 at 44-47. With the exception perhaps of the unlawful impairment of business at Art. VI.–2:208 DCFR, which seems to descend from the Recht am eingerichteten und ausgeübten Gewerbebetrieb as it grew out of the case-law under § 823(1) BGB. By and large, injury to contractual relationships falls within the same category, since in general it results in pure economic loss. It could be argued that it is in fact just a more conceptual way to describe pure economic loss.

300

Pierre Larouche

unlawful impairment of business within the meaning of Article VI.–2:20816 or when it results from one of the specific courses of conduct listed in the DCFR. Indeed, the inclusion of these specific courses of conduct only make sense because of the exclusion of pure economic loss, since any other type of damage will already be covered in the list of protected interests. Beyond pure economic loss, the other cases of damage which would not be legally relevant are those where a person suffers no measurable or even perceptible loss (even if nonmaterial), but is rather hurt in his, her or its beliefs and convictions or in his or her position as citizen. For instance, a card-carrying member of Amnesty International is deeply offended by the police methods used by the State or the Greenpeace organisation itself is concerned about the mismanagement of waste. In principle, injury to such collective interests does not suffice as legally relevant damage for the non-contractual liability. However, the DCFR does allow the State to claim for environmental impairment at Article VI.–2:209. While the second cases are interesting, they are less frequent than the first ones, and furthermore it seems from the DCFR that the main purpose of the very concept of ‘legally relevant damage’ is to enable pure economic loss to be excluded except in limited cases. Accordingly, the rest of this article will focus on the first set of cases.

2. Desirability of unification The first issue is whether it is desirable at all to unify the law of non-contractual liability, at least as far as those a priori limits on liability are concerned.

2.1. Difference between contract law and non-contractual liability On that point, non-contractual liability generally differs from contract law. Indeed for the latter, it is possible to find an elegant way out of the desirability issue by framing the DCFR as an optional regime which could be made available to parties and serve as guidance to law reformers.17 One can argue whether it is useful to come up with yet another set of contract law, but at least the overall impact on welfare remains limited. Yet that line of argument ignores that certain parts of contract law cannot readily be at the disposal of the parties, for instance consumer law. When it comes to non-contractual liability, the problem of non-disponibility cannot be ignored. Because by definition the victim/plaintiff and the defendant cannot bargain ahead of the occurence of damage since they were not in contact,18 the unified law of non-contractual liability as proposed in the DCFR cannot be offered to the choice of the parties. Either it is meant as a candidate for enactment as a unified (or at least harmonized) European law of 16

17

18

This provision appears closely modelled on the case-law concerning the Recht am Gewerbebetrieb in German law, which evolved to remedy the outright exclusion of pure economic loss (reiner Vermögensschaden) from § 823(1) BGB: Tort Law, supra, note 2 at 183 and ff. Even though the original intent of many of the participants in the drafting groups of the DCFR was clearly to develop a European Civil Code – certainly for many in the Study Group on a European Civil Code, as the name indicates. Except perhaps where non-contractual liability applies as between contractual parties, for instance in the case of product liability (when the purchaser bought directly from the producer) or professional malpractice.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

301

non-contractual liability or it is an academic exercise offered as a model for law reform at national level (and as a frame of reference at European level). A serious discussion of whether it is desirable to unify non-contractual liability law at European level cannot be escaped. It is very difficult to conduct this discussion for a specific part of the law – such as here the a priori limits to liability – without first examining desirability in general and then turning on whether parts of the law can be unified without touching others. That examination will be done against the background of the analytical framework outlined in an earlier piece.19

2.2. The prospects for harmonization or unification As a starting point, it is useful to briefly assess how much and why the law of non-contractual liability may differ from one European State to the other. The law of non-contractual liability has grown into a fairly large body of law (if only case-law) in all jurisdictions – even in French law and similar systems where the legislation has been kept short. The various jurisdictions also seem to have followed different paths in the development of their respective law of non-contractual liability, even though they often evidence a fair amount of convergence in the outcomes. The law of non-contractual liability has greatly evolved in all jurisdictions as a result of the rise in the number of accidents in step with industrialization in the 19th and 20th centuries. Since this was also the golden age of the nation State, it is not surprising that national approaches would prevail. Divergence between the various national law of non-contractual liability can be explained along three lines: (i) rational but not deliberate (network effects), (ii) rational, deliberate and benign (local preferences) and (iii) rational, deliberate and not benign (public choice).20 Overall, the main factor of divergence is probably local preferences (explanation (ii)), i. e. different policy choices as to how non-contractual liability law should be designed, including for instance the balance between contract and non-contractual liability,21 the balance between industrial and consumer interests, etc. Indeed, of the main areas of private law, non-contractual liability is probably the one where policy choices have the greatest impact.22 It cannot be excluded that the policy-making process has also been captured by some interests (explanation (iii)), so that local preferences would not adequately be reflected (which does not take the divergence away). Beyond that, there is also a fair amount of conceptual “noise” induced by respective national legal communities working essentially in isolation

19

20 21

22

F. Chirico and P. Larouche, “Conceptual divergence, functionalism, and the economics of convergence” in S. Prechal & B. van Roermund (Eds.), The Coherence of EU Law. The Search for Unity in Divergent Concepts (Oxford: OUP, 2008) 463. Ibid. The contract-tort boundary is drawn differently from one legal system to the other, and to some extent systems which give contract law a broad reach (such as German law) thereby compensate for perceived weaknesses of tort law. This could perhaps be linked to the fact that non-contractual liability is also the area of private law where the will of the parties plays the least role. Because the State is the prime influence in designing these non-contractual relationships, policy – as opposed to commercial practice, usage, etc – drives lawmaking.

302

Pierre Larouche

from one another.23 Local preferences are thereby amplified and cemented in the law (explanation (i) above). Where the different jurisdictions follow similar policies,24 typically it will be observed that the laws converge once the conceptual noise has been dampened though proper methodology.25 But as long as these different local preferences are not acknowledged and therefore no meaningful discussion of the policy choices underlying non-contractual liability law has taken place,26 the prospects for harmonization or unification are dim. The combination of different policy choices exacerbated by conceptual noise also creates inertia in each national legal system – sometimes referred to as “legal culture”. Such inertia creates a significant hurdle for any attempt at harmonization or unification – whether complete or a fortiori if partial. It could also perversely undermine such attempts by replacing open and explicit divergence with hidden conceptual divergence,27 where what appears to be a single legal framework is in fact interpreted differently across jurisdictions.

2.3. Unification or harmonization of non-contractual liability in general The arguments in favour of and against the unification – or even just the harmonization – of non-contractual liability law in Europe have already been well rehearsed in a number of publications.28 The proponents of unification put forward in essence two lines of argument. The first one centres on the internal market: mobility of factors across borders is hampered if the rules of non-contractual liability vary from one Member State to the other, all the more since the lex loci delicti tends to apply under the private international law of most Member States. Firms and individuals therefore face a patchwork of differing rules, which creates transaction costs and acts as a deterrent to cross-border activity. A variant of that line of argument would also suggest that differing laws create an uneven playing field, i. e. distort the conditions of competition across the EU. Yet even more informed variants pick on more sophisticated economic reasoning, namely the risk of cross-border externalities and the danger of a race to the bottom between the Member States. 23

24

25

26

27 28

Such conceptual confusion is often at the root of the current attempts to put forward a unified European tort law. This holds not just for the DCFR, but also for the European Group on Tort Law, which produced the Principles of European Tort Law (see European Group on Tort Law, Principles of European Tort Law. Text and Commentary (Wien: Springer, 2005). See N. Jansen, “Principles of European Tort Law? Grundwertungen und Systembildung im europäischen Haftungsrecht” (2006) 70 RabelsZ 732. Whether they result from genuine local preferences or captured preferences matters less, as long as these expressed policy choices are convergent. See the conclusions reached in Tort Law, supra, note 2 as regards causation (452 and ff.) and liability for the conduct of others (522 and ff.). In the course of discussing these different policy choices, the distinction between genuine local preferences (explanation (ii) above) and captured preferences (explanation (iii) above) can play a key role: in principle, captured preferences are more easily questioned. See Chirico and Larouche, supra, note 18 at 487. See among others U. Magnus, “Towards European Civil Liability” and the reply of T. Hartlief, “Harmonization of European Tort Law. Some Critical Remarks”, in M. Faure et al., eds., Towards a European Ius Commune in Legal Education and Research (Antwerp: Intersentia, 1992) 205 and 225, as well as R. van den Bergh and L. Visscher, “The Principles of European Tort Law: The Right Path to Harmonization?” (2006) 4 Eur Rev Priv L 511.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

303

The second one is rights-based: all EU citizens are entitled as a matter of right to equal protection from the law of non-contractual liability. Neither of these lines of argument is convincing. 2.3.1. The internal market argument

The internal market argument must be put in the balance against the standard public interest justification for divergence between legal systems, namely the rational and deliberate reflection of differing local preferences in each legal system.29 A trade-off must be made between the welfare gains arising from unification or harmonization and those following from a greater alignment with local preferences. Typically, if the former do not go beyond a reduction in transaction costs in cross-border situations, it is unlikely that the trade-off will favour unification or harmonization, all the more since the market often offers ways to minimize or even overcome the transaction costs. For instance, as far as liability for road traffic accidents is concerned, Member States such as France, Germany and the UK have made different policy choices – the former two introduced risk-based liability regimes, while the latter left these accidents to be dealt with under general tort law, i. e. usually under negligence.30 It can be thought that this reflects different sets of local preferences: all three countries have comparable levels of development and presumably comparable rates of road traffic accidents and resulting damage.31 Of course, the different liability regimes impose costs on individuals and firms from one Member State as they conduct activities in another State. At the same time, it is relatively easy to work through insurance to equalize these differences and provide policyholders with a uniform level of coverage against first- or third-party losses.32 To the extent that there might be an externality on third-party insurance, the EU indeed intervened to compel third-party liability insurance such as would cover the liability arising under the law of any Member State where damage is caused.33 Arguments based on the internal market are more convincing when they invoke not merely transaction costs, but rather externalities or a race-to-the-bottom argument. Externalities arise in specific cases which have a cross-border dimension, the most famous being liability for environmental damage. In such cases, unification or harmonization is warranted in order to remove the externality and ensure an efficient level of care.34 29

30 31

32

33

34

Divergence can also be explained by rational but not deliberate phenomena (network effects, informational imperfections) or by rational, deliberate yet not benign phenomena (capture of local legal system by certain interest groups). See Chirico and Larouche, supra, note 18 at 466-471. Tort Law, supra, note 2 at 583 and ff. For information on road traffic accident statistics, see the compilations made by the UN Economic Commission for Europe at www.unece.org. France seems to have a lower rate of road traffic accidents involving physical injury or death than Germany or the UK. Transaction costs within the insurance industry could be reduced through unification or harmonization of tort law, however, subject of course to the remarks made above concerning the inertia of national legal systems and the scope for hidden divergence to persist. Directive 72/166 of 24 April 1972 on insurance against civil liability in respect of the use of motor vehicles, and to the enforcement of the obligation to insure against such liability [1972] OJ L 103/2, as amended. See for instance the efforts made over the years at EU level, so far leading to Directive 2004/35 of 21 April 2004 on environmental liability with regard to the prevention and remedying of environmental damage [2004] OJ L 143/56.

304

Pierre Larouche

The race-to-the-bottom argument would imply that Member States would compete to attract factors of production via the law of non-contractual liability: firms would choose to settle in Member States which impose a lesser liability burden on them. This ties in with the level-playing field argument: firms operating under the laws of Member States with less burdensome liability regimes would benefit from an unfair competitive advantage. While appealing at first sight, these arguments are also flawed. First of all, they assume that the state of non-contractual liability law plays a central role in the location decisions of firms, more so than labour or material costs, for instance. Secondly, they ignore the effect of private international law (choice of law) rules. Since as a general rule the law of the jurisdiction where the damage occurred will apply,35 all firms who are active in a given jurisdiction are on the same competitive footing as regards exposure to non-contractual liability arising out of damage occurring in that jurisdiction. The specific content of non-contractual liability law in that jurisdiction does not therefore affect the conditions of competition, at least as far as the downstream activities36 of the firm are concerned. Location decisions can affect non-contractual liability only with respect to liability arising from the upstream activities37 of the firm, including in particular environmental liability for damage caused at production sites. Here there might indeed be a risk that jurisdictions compete with each other to offer favourable environmental liability regimes. In addition to the direct externalities linked to cross-border damage, as outlined above, harmonization of environmental liability might also be warranted in order to avoid a race to the bottom. Yet thirdly, even in the case of environmental liability, we come back to local preferences. If a given jurisdiction wants to use its law of non-contractual liability to create a competitive advantage for itself, the fact remains that damage does not disappear. It is bound to occur in that jurisdiction and somehow must be dealt with.38 If injury and loss are not directly shifted to a defendant via liability law, they will be absorbed through other mechanisms such as social security (with an impact on taxation) or insurance (with an impact on spending). To some extent, non-contractual liability can contribute to shifting the burden of damage – including environmental damage – from the taxpayers and consumers of the jurisdiction to all the customers of the defendant firm (who might be located within or outside the jurisdiction), thereby creating a positive externality for the jurisdiction in question. Accordingly, the incentive to engage into a race to the bottom might be counter-balanced by an incentive

35

36

37

38

Regulation 864/2007 of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) [2007] OJ L 199/40, Art. 4. I. e. those taking place closer to the end-user, for instance distribution and retail. For product liability, in addition, specific rules are set out at Art. 5 of Regulation 864/2007, ibid., whereby the applicable law is also closely linked to the plaintiff and not to any location decision made by the defendant. The specific rules for liability arising out of unfair competition or restrictions of competition, at Art. 6, are also likely to point to the law of the jurisdiction where the end-user is located. I. e. those taking place closer to the production location. In addition to environmental liability (dealt with at Art. 7 of Regulation 864/2007, ibid.), this might also cover liability for damage caused to third parties by employees in the course of production, for instance. The law of the jurisdiction where production is located is likely to apply since the damage is likely to have been incurred there. In fact, if the law of non-contractual liability is producer-friendly and results in the transfer of factors of production to the jurisdiction in question, the risk of damage resulting from the activities of these factors will increase because of the higher level of activity.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

305

to “race to the top”, so the state of the law of non-contractual liability in a given jurisdiction should reflect a careful trade-off which should be respected.

2.3.2. The rights-based argument

In light of the above, it should also be apparent that the rights-based argument is in fact circular from a law and economics perspective. Since the rights enjoyed under non-contractual liability are a function of the local preferences, using these rights as an argument to unify or harmonize the law amounts to argue that the local preferences should be disregarded and replaced with a more global trade-off at European level, which is precisely the issue to be discussed. The validity of the rights-based argument therefore depends on specific political decisions whereby local preferences would be set aside. So the case for harmonization – let alone unification – of tort law is weak, except in a limited number of areas, first and foremost environmental liability because of externalities caused by cross-border damage.

2.4. Partial unification or harmonization The next question is then whether it is feasible and desirable to harmonize only part of the law of non-contractual liability, namely the issue of a priori limits to liability. As mentioned earlier, the inertia of national legal systems creates an obstacle for unification or harmonization, all the more so when it is only partial. In the face of an attempt at harmonizing only part of the law of non-contractual liability, national legal systems are likely to react to the introduction of a “foreign element” within the law of non-contractual liability by re-asserting their – self-perceived – specificity. The scope of the harmonized part could be construed restrictively, or the harmonized part could be interpreted centrifugally, to improve its fit within the national legal system at the expense of harmonization. For instance, Directive 85/374 on product liability – a well-focused intervention, centred on a reasonably well-circumscribed area of non-contractual liability law – still did not entirely achieve its objectives.39 Available studies suggest that it remains subject to diverging interpretations from one Member State to the other.40 In addition, at Article 18, that Directive left national 39

40

Directive 85/374 of 25 July 1985 on liability for defective products [1985] OJ L 210/29. It is interesting to note that, even if Directive 85/374 is based on Article 94 EC (since Article 95 EC did not exist yet at the time of enactment), the case for the harmonization of product liability cannot rest on the internal market (at least not for a regime along the lines of Directive 85/374). Product liability is one area where producers can relatively easily overcome transaction costs arising from different liability regimes through insurance and ultimately through refraining from marketing their products in certain jurisdictions. If anything, the case for the Directive is much stronger from a consumer rights perspective, which would however warrant minimum harmonization. In that case, the internal market will not be achieved since local preferences will trump the Directive when they are more consumer-friendly. In practice, because of a number of specific features of Directive 85/374 (Article 13 and the optional elements at Article 15), Directive 85/374 has had that effect. See the most recent study done by Lovells, Product liability in the European Union – A report for the European Commission (February 2003), available at http://ec.europa.eu/enterprise/regulation/goods/ docs/liability/studies/lovells-study_en.pdf.

306

Pierre Larouche

law to apply in parallel to the harmonized product liability regime,41 and predictably in some jurisdictions national law remained the vehicle of choice for product liability claims.42 By the same token, attempts to introduce a harmonized private right of action for damages for breaches of EC competition law in the non-contractual liability laws of the Member States are bound to run against that inertia.43 In the light of the above, it would appear that the unification or harmonization of such a central element of non-contractual liability law as the a priori limits to liability cannot be conceived on a stand-alone basis. If anything, it would have to take place within a broader effort at unifying or harmonizing the whole of the law of non-contractual liability. Even then, the effectiveness of a broad unification or harmonization of tort law would be limited by factors going beyond tort law, such as the border between tort law and contract law, rules of evidence or even the absence of a common judiciary delivering binding interpretations. As seen above, there is no pressing case for the unification or harmonization of the law of non-contractual liability across the EU. If nonetheless such unification or harmonization is undertaken, its success will hinge on the quality of the underlying comparative law work. The conceptual noise must be removed, but more importantly underlying policy differences must be identified and bridged.

3. Assessment of the solution chosen in the DCFR For the remainder of this paper, it will be assumed for the sake of argument that unification or harmonization of the law of non-contractual liability is desirable – at least as regards the a priori limits to liability. The second issue is then whether the concept of ‘legally relevant damage’ in the DCFR represents the best outcome from a law and economics perspective. After some critical remarks on the underlying comparative method (3.1.), the solution chosen in the DCFR will be assessed from a law and economics perspective (3.2.).

3.1. The comparative method used in Book VI DCFR When putting together all the elements of the concept of “legally relevant damage”, one cannot escape the impression that the drafters of the DCFR followed a superimposition 41

42

43

See on the scope of Article 13 of Directive 85/374: ECJ, 25 April 2002, Case C-183/00, González Sánchez [2002] ECR I-3901. Wherever for one reason or another the features of national law were thought preferrable to those of the regime of the Directive: see Tort Law, supra, note 2 at 674-683 and the Lovells Report, supra, note 36. See Commission, Green Paper on Damages actions for breach of the EC antitrust rules COM(2005) 672 final (19 December 2005). In more recent policy documents, however, the Commission appears to move away from the introduction of a complete liability regime, towards an advocacy programme coupled with specific and discrete harmonization measures: Commission, White Paper on Damages actions for breach of the EC antitrust rules COM(2008) 165 final (2 April 2008). It is not clear if Article VI.–2:208(2) DCFR is meant to deal with these cases when it refers to “unfair competition”. If so, then it is much more limitative than what the Commission is proposing, since it appears restricted to “consumers” as opposed to any downstream customer of the firm(s) engaging into a violation of EC competition law.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

307

technique. The a priori limits to liability found in German and English law (and in other similar systems) are added to one another, and the resulting configuration is taken up in the DCFR. It was explained earlier above why this method is deficient: by failing to take into account divergences arising from different policy choices reflecting different local preferences, such superimposition creates the illusion that the DCFR represents the common law of Europe. In fact, it does not even represent the “common denominator” to the laws of the Member States, or in terms of set theory, the intersection between these laws. If anything, “legally relevant damage” comes closer to the union than the intersection. The protected interest and specific torts approaches of German and English law, respectively, are brought together without any apparent assessment of whether they overlap with each other or contradict each other. Even then, the DCFR misses one key element of the English approach to a priori limits through the use of specific torts and the prudent development of the law of negligence, namely that the categories are open-ended.44 Methodologically, the English law of noncontractual liability evolves constantly, in an inductive (bottom-up) fashion. In their wisdom, common law courts can always choose to extent a duty of care under negligence to new categories, or to recognize a new specific tort. Indeed, liability for loss made by reliance on incorrect advice or information (Article VI.–2:207 DCFR) was still being toyed with by the House of Lords a mere decade ago.45 The DCFR may incorporate the English approach in substance, but it freezes the law in time, in disregard of the open-ended method of English law in this area. What is more, French law and similar systems evidence no a priori limits on liability using protected interests or specific conduct. This does not mean that somehow that issue would have been overlooked under French law. Rather, different policy choices were made: either a choice was made not to exclude compensation for pure economic loss a priori, or perhaps more accurately, other limitation devices (refinements on causation, restriction on the quantum of damages awarded) were preferred the use of a priori limits.46 The superimposition method followed in the DCFR seems to ignore French law and its choices altogether. The methodology used in Book VI DCFR is deficient not only in its apparent ignorance of French law and related systems, but also in that, in keeping with a blind spot in much comparative law scholarship, it fails to factor in developments at European level. That failure is especially damning in the context of the DCFR, which was drafted at the behest of the European Commission, in order to guide law reform in the EU. In its case law over the years, the European Court of Justice (ECJ) identified two liability regimes for breaches of EC law, both of which are directly based on the EC Treaty, and in particular on the principle that the Treaty provisions must be made effective (effet utile). Firstly, Member States are liable to individuals for damage arising from breaches of EC law.47 Secondly, individuals are liable to 44

45

46 47

It could be argued that the general definition of legally relevant damage at Art. VI.–2:101(1) DCFR provides an element of open-endedness, but in such a case the efficiency of adding a long list of particular instances at Art. VI.–2:201 and ff. must be questioned. See infra, heading 3.2.3. See Caparo Industries v. Dickman, supra, note 11 and House of Lords, White v. Jones [1995] 2 AC 207, as discussed in Tort Law, supra, note 2 at 215-224. Tort Law, ibid. at 197-200. As established by ECJ, 19 November 1991, Joined Cases C-6/90 and C-9/90, Francovich [1991] ECR I-5357 and confirmed on a broader basis in ECJ, 5 March 1996, Joined Cases C-46/93 and C-48/93, Brasserie du Pêcheur [1996] ECR I-1029.

308

Pierre Larouche

other individuals for damage arising from breaches of EC competition law.48 Both regimes are based on a core of rules set out in the case law of the ECJ and completed with rules taken from national law, in accordance with the principle of national procedural autonomy, which prevails for remedies under EC law. National procedural autonomy is subject to two reservations, however: the national rules must treat claims made on the basis of EC law on the same footing as claims based under national law (equivalence) and they cannot work so as to render remedies under EC law ineffective (effectiveness). Under both regimes, the loss suffered by the plaintiff will most likely consist in pure economic loss. In preliminary references, the ECJ was asked whether pure economic loss could be excluded under those EC liability regimes when the relevant national law contained an a priori bar on pure economic loss. As regards Member State liability for breaches of EC law, the ECJ answered the issue unequivocally in Brasserie du Pêcheur:49 The Bundesgerichtshof asks whether national legislation may generally limit the obligation to make reparation to damage done to certain, specifically protected individual interests, for example property, or whether it should also cover loss of profit by the claimants … Total exclusion of loss of profit as a head of damage for which reparation may be awarded in the case of a breach of Community law cannot be accepted. Especially in the context of economic or commercial litigation, such a total exclusion of loss of profit would be such as to make reparation of damage practically impossible. As regards the liability of individuals for breaches of EC competition law, the ECJ answered in the same terms in Manfredi.50 So, to the extent that non-contractual liability is based on either of these regimes, an a priori limit on liability which works so as to exclude pure economic loss would deprive the liability regime of its effectiveness. More generally, given the need to ensure that remedies under EC law are effective, it is not surprising that the ECJ rejects a priori limits on liability in favour of case-by-case determinations on the basis of general conditions for liability. A priori limits are rough and can result in the exclusion of meritorious cases.

3.2. Analysis of a priori limits and of the concept of ‘legally relevant damage’ The inclusion of a priori limits to liability in Book VI DCFR via the concept of “legally relevant damage” is methodologically flawed. In addition, from a law and economics perspective, it is doubtful whether that construction is the best choice in substance. The common justification for a priori limits to liability is a concern that the defendant would be overburdened with liability, and accessorily that the judicial system would be flooded with liability claims (the famous “floodgates” argument). In principle, this overburdening concern would appear to be based on economic considerations. After all, if non-contractual liability is too burdensome, the risk of liability will act – 48 49

50

As established by ECJ, 20 September 2001, Case C-453/99, Courage [2001] ECR I-6297. Supra, note 42 at Rec. 86-87. That point was confirmed in ECJ, 8 March 2001, Joined Cases C-397/98 and C-410/98, Metallgesellschaft [2001] ECR I-1727 and 17 April 2007, Case C-470/03, A. G.M.-COS.MET Srl [2007] ECR I-2749. ECJ, 13 July 2006, Joined Cases C-295/04 to C-298/04, Manfredi [2006] ECR I-6619 at Rec. 95-96.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

309

for the defendant and for any firm which could potentially find itself defending a liability claim – as a disincentive to economic activity and a brake on entrepreneurship. But the overburdening concern rests on a myopic perspective. On a proper social welfare analysis, the situation is more complex, as the law and economics scholarship on noncontractual liability illustrates very well. We will look at a priori limits against the three main functions of non-contractual liability: 51 (i) deterrence and the reduction of the social cost of accidents, (ii) compensation and loss-spreading and (iii) reduction of transaction and administrative costs.

3.2.1. Reducing the social cost of accidents – deterrence

In a first stage, we examine whether the introduction of a priori limits to liability improves the ability of non-contractual liability to reduce the social cost of accidents by deterring potential injurers from entering into the course of conduct which could lead to loss or injury. The overburdening concern assumes that non-contractual liability somehow generates additional cost out of nothing, thereby putting an extra burden on the defendant. This ignores the starting point, namely that some loss (damage or injury) has occurred and been suffered by the plaintiff (victim). As will be seen further below,52 that loss is very often a social loss: production or consumption is negatively affected, without any corresponding increase elsewhere that would offset the loss. That social loss can be avoided by deterring the defendant from entering into the course of action which causes that loss or injury, at least to the extent that it would be more efficient for the defendant to refrain from that course of action. Non-contractual liability is one means of deterrence. This deterrence function of tort law is best encapsulated in the basic model expounded by Shavell.53 As a starting point, the social cost function is:54 C = p(x, y)L + P(x) + D(y) (1) where C is the sum of expected accident costs and the costs of care P is the cost of care for the plaintiff (victim) D is the cost of care for the defendant x is the level of care of the plaintiff y is the level of care of the defendant p is the probability that the loss will occur given the levels of care x and y on the part of the plaintiff and defendant L is the magnitude of the loss.

51

52 53

54

These three functions have been standard ever since the work of G. Calabresi, The Costs of Accidents: A Legal and Economic Analysis (New Haven: Yale University Press, 1970) at 26-31. Infra, heading 3.2.2. See S. Shavell, “Strict Liability versus Negligence” [1980] 9 Journal of Legal Studies 1, as well as S. Shavell, Economic Analysis of Accident Law (Cambridge: Harvard University Press, 1987) at 5 and ff. This is the version of the Shavell model used in M. Faure, “Economic Analysis of Fault”, in P. Widmer, ed., Unification of Tort Law: Fault (The Hague: Kluwer, 2005) 311 at 312.

Pierre Larouche

310

The cost function of the defendant is D(y), so that a rational defendant would take no precaution since it only leads to additional costs for the defendant. It is clear that for society, setting y at 0 (and thereby also increasing p) can lead to externalities, given that social cost will ensue from the actions of the defendant. Shifting L onto the defendant – for instance through the law of non-contractual liability – can help to minimize C by influencing y.55 The defendant’s cost function then becomes D(y) + p(x, y)L

(2)

where p is inversely linked to y. The defendant will bring its level of care y up to the point (y*) where the marginal cost of increasing care further ΔD(y) is more than the decrease in the expected cost of loss Δp(x,y)L which the law of liability shifts onto the defendant. By the same token, C will be minimized. In the previous paragraph, it is assumed that the law of liability simply shifts L onto the defendant. However, as was seen above, non-contractual liability generally incorporates limits on liability, typically relating to the accountability of the defendant. The most common is the requirement that the defendant be at fault.56 It modifies the model above by introducing a normative standard as to the level of care ynorm and turning the defendant’s cost function into D(y) + p(x, y)L if y , ynorm D(y) if y $ynorm

(3)

The effects of introducing a fault requirement to limit liability are manifold. First, since the cost to the defendant is significantly decreased to the extent that it meets ynorm, the defendant receives a stronger incentive to set its level of care at ynorm.57 Hence it is crucial that the law somehow ensures that ynorm = y*, the optimal level of care from a social welfare standpoint. Secondly, a fault requirement induces the defendant to reveal the techniques it used to try to prevent or reduce third-party damage; such information is useful to the general public. Thirdly and more importantly, a fault requirement can induce the defendant to engage into a potentially injurious or damaging activity to a greater extent than is socially desirable. Indeed so far the discussion has assumed that the loss was caused through private conduct with no or limited social benefit, for instance carelessly running into someone while walking outside to enjoy the nice weather. Of course, this assumption is unrealistic, since most 55

56

57

We assume a constant level of care x by the plaintiff for the sake of discussion. See Shavell, supra, note 48 for the more elaborate discussion where x is also variable, i. e. where contributory negligence is brought into the analysis. Besides fault, other limits related to accountability are sometimes imposed in a context of strict liability. They have a different impact on the model. In a nutshell, these accountability limits are designed not so much by reference to the level of care, such as to affect the incentives of the defendant to reach y*, but rather by reference to the activity engaged into by the defendant (creation of risk, control over another). Their aim is then to ensure that the shifting of liability to the defendant is limited to those cases which come within the sphere of influence of the defendant, in order to avoid overcorrecting the externalities mentioned in the main text and negatively influencing the level of activity of the defendant. Perhaps this is also useful for behavioural reasons, see M. Faure, “Calabresi and Behavioural Tort Law and Economics” (2008) 1:4 Erasmus L Rev. 75.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

311

of the time the defendant will inflict damage or injury in the course of an activity which is itself socially beneficial. When looking at social welfare, the benefits from that activity must also be taken into account, so that W = B(α) + C(α) (4) where W is the overall welfare α is the level of activity B is the social benefit arising from the activity of the defendant C is the social cost of such activity, as defined under (1) above The literature shows that the basic liability rule set out above under (2) also leads to an efficient level of activity.58 However, introducing a fault requirement as with (3) above can lead the defendant, once it meets the required level of care ynorm, to engage into the activity in question to a greater level than is socially optimal, given that it does not have to bear the cost of the loss inflicted on the plaintiff/victim (while still deriving some private benefit from the activity, otherwise no activity would take place).59 This brief analytical foray indicates that at first sight the law of non-contractual liability is not likely to overdeter the defendant, i. e. to cause the defendant to refrain from a course of action that would otherwise have been efficient from a social welfare perspective. The overburdening concern appears exaggerated.60 What is more, the analysis made above for the fault requirement can be applied to the a priori limits to liability described above. Since these limits are unrelated to the defendant, the first two effects – heightening deterrence by increasing incentives to take the socially desirable level of care and promoting the disclosure of damage prevention techniques – are absent. However, these a priori limits do relieve the defendant from liability for a range of conduct. It would follow that they also produce the second effect, i. e. inducing the defendant to engage into a level of activity which is higher than socially optimal. Furthermore, even if a priori limits on liability could send correct incentives to guide the behaviour of market parties, the specific concept of ‘legally relevant damage’ as set out in Book VI CFR cannot provide any significant incentive. As Table 1 shows, ‘legally relevant damage’ is very complex, with a definition spanning no less than twelve articles. That definition includes general standards at Article VI.–2:101, and a set of more specific instances articulated along either protected interests or specific courses of conduct. What is more, those specific courses of conduct are defined in a way which includes elements of accountability (negligence, knowledge, reasonable expectations). Accordingly, it is unlikely that such a circonvoluted concept can produce any noticeable incentive effect which would bring parties to adopt a socially optimal level of care and of activity.

58 59 60

Shavell, Economic Analysis of Accident Law, supra, note 48 at 8-9. Ibid. at 23-26. According to Shavell, ibid. at 29-32, the main situation in which overdeterrence could occur is when the law of non-contractual liability would not include a contributory negligence defence. That defence allows the defendant to invoke the failure of the plaintiff to meet the required standard of care xnorm in order to extinguish liability (contributory negligence in the classical sense) or to reduce it proportionately (comparative negligence).

312

Pierre Larouche

3.2.2. More efficient loss-spreading

Once the loss or injury has occured, its impact on social welfare can perhaps be mitigated by limiting its consequences or spreading it over many so that it is absorbed. In practice, many mechanisms intervene on each side to spread the loss (if it stays with the plaintiff) or disperse the burden of liability (if the defendant is found liable). For the plaintiff, these include first-party (damage) insurance, insurance coverage from a third party (workers compensation insurance, etc.) and social security (whereby the loss is spread over the collectivity). For the defendant, these include third-party (liability) insurance, industry-wide insurance funds, etc. In addition, if the plaintiff or the defendant is a firm, the loss or the burden of liability, respectively, can be passed on to others via price increases.61 Non-contractual liability is therefore only one among many mechanisms for spreading liability, and its loss-spreading effects can only be understood against the background of these other mechanisms. As set out above, a priori limits to liability are usually justified by a concern for overburdening the defendant. This overburdening concern reduces non-contractual liability to litigation between an individual plaintiff and an individual defendant, where the loss must either be left to lie with the plaintiff or be shifted to burden the defendant. In reality, as indicated in the previous paragraph, both plaintiff and defendant are likely to have loss-spreading mechanisms at their disposal. While non-contractual liability may superficially appear to shift loss entirely from one side to the other, in such cases in fact it is simply transferring the loss from set of loss-spreading mechanisms (the plaintiff’s) to another (the defendant’s). From a social welfare perspective, the issue is then whether the loss is more efficiently spread through the mechanisms available to the plaintiff or through those available to the defendant. This in turn depends on a number of factors, for instance the ability to identify, isolate and quantify the risk of loss so as to obtain insurance cover efficiently: the risk that a driver causes accidents is easier to insure efficiently than the general risk that someone suffers injury. This explains why specific liability regimes – often strict (not based on wrongful conduct) – have been put in place for identifiable risk-creating activities such as driving a motor vehicle, placing manufactured products on the market, etc. In any event, all other things being equal, if no or few means of spreading losses are available to the plaintiff, it is sensible to shift the loss when the defendant is well-positioned to spread the loss. It was seen earlier62 that the outcome of “legally relevant damage” is primarily to exclude recovery of pure economic loss except if such loss meets the general criteria of Article VI.–2:101 or falls within one of the particular instances listed at Article VI.–2:201 and ff. So if that exclusion is welfare-neutral or even positive because pure economic loss should not generally be shifted from the victim to the defendant, that might offset the negative effect of a priori limits to liability on the deterrence function of tort law. The policy arguments for not extending liability to pure economic losses are well articulated by Lord Denning MR is his famous opinion in Spartan Steel & Alloys Ltd. v. Martin & Co. Ltd.:63 61

62 63

To the extent of course that this will not affect the market position of the firm in question too adversely. Supra, heading 1.4. [1973] QB 27 at 34-39. This is a so-called “cable case” where a road building company negligently severs a power cable and prevents a neighbouring plant from operating. Policy arguments which are specific to the case are omitted.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

313

[Nature of the hazard] This is a hazard which we all run … And when it does happen, it affects a multitude of persons: not as a rule by way of physical damage to them or their property, but by putting them to inconvenience, and sometimes to economic loss … [T]he economic loss is not very large. Such a hazard is regarded by most people as a thing they must put up with – without seeking compensation from anyone … [T]hey do not go running round to their solicitor. They do not try to find out whether it was anyone’s fault. They just put up with it. They try to make up the economic loss by doing more work next day. This is a healthy attitude which the law should encourage. [Floodgates] If claims for economic loss were permitted for this particular hazard, there would be no end of claims … It would be well-nigh impossible to check the claims. If there was economic loss on one day, did the claimant do his best to mitigate it by working harder next day? And so forth. Rather than expose claimants to such temptation and defendants to such hard labour – on comparatively small claims – it is better to disallow economic loss altogether, at any rate when it stands alone, independent of any physical damage. [Best loss-avoider] In such a hazard as this, the risk of economic loss should be suffered by the whole community who suffer the losses – usually many but comparatively small losses – rather than on the one pair of shoulders, that is on the contractor on whom the total of them, all added together, might be very heavy. The floodgates concern pertains more to the next function of non-contractual liability, but the first and third concerns are relevant here. According to Lord Denning, pure economic loss should be considered as part of the normal risk of conducting economic activities. Instead of channelling it to one party, it is best left to be shouldered by everyone through an extra effort (or other means such as first-party insurance), since it tends to be widespread and relatively limited. These concerns were placed in a law and economics analytical framework by W. Bishop.64 He sought to develop a comprehensive theory to explain the state of the common law as regards pure economic loss. The starting point is that some losses are private losses only; they imply only a transfer of wealth between private actors without any social cost. These private losses should not be covered by non-contractual liability and are best handled via other loss-spreading mechanisms at the disposal of the parties. In other words, the loss L in equation (1) above must represent a social cost, since equation (1) is about social costs. As far as deterrence is concerned, making the defendant liable for a private loss without social cost, i. e. including private losses in L, would distort the incentives given to the defendant through the liability system, as resulting from equation (2): because L is inflated beyond social costs, the defendant will be led to adopt a level of care y which is above the social optimum y*. From a compensation perspective, in the absence of social cost, it can be argued that the spreading of private losses is best left to the parties themselves. Bishop considers that pure economic loss is often a mere private loss. By way of base scenario, if competitors of the firm suffering pure economic loss (loss of profit because of inability to supply, in the quintessential cable case) have spare capacity and pick up the demand, then profit is transferred from the firm to its competitors, without social loss. In such a case, it is not efficient to grant recovery. But, as Bishop notes, spare capacity is not always available. In a number of scenarii, many of them more realistic than the base scenario, resources must be diverted and social 64

W. Bishop, “Economic Loss in Tort” (1982) 2 Ox. J. Leg St. 1.

314

Pierre Larouche

cost ensues. Liability could then be envisaged. For instance, competitors may have to divert resources to satisfy the demand left unmet when a firm is prevented from supplying. Or the injured firm is perhaps risk-averse and it used some of its resources to take insurance against the occurrence of pure economic loss. Another possibility is that, in the base scenario, the potential victim could avoid a private loss at a cost which is inferior to the expected loss, but superior to the avoidance cost of the defendant. In the absence of social cost it would be socially optimal not to invest anything to avoid the private loss. Yet here the victim has an incentive to invest in avoidance, even though as a second best solution it would be more efficient for the defendant to do so since it has lower avoidance costs. For all these cases, however, Bishop argues that the courts are ill-equipped to isolate the social cost, and that in any event it is likely to be small in comparison to the private loss. Nevertheless, the basic argument made by Bishop about the lack of social cost does not appear to be valid outside of fairly narrow assumptions. In a more recent piece, Dari-Mattiacci and Schäfer critically revisit Bishop’s arguments in the light of subsequent developments in the literature and in the law.65 They argue that, even in the base scenario set out by Bishop, there is a social cost, in that the spare capacity available to offset the impairment of the victim’s ability to supply and make profits is endogenous and not exogenous as Bishop assumes. In other words, firms voluntarily invest resources in creating excess capacity in order to be able to cater to variations in demand and supply due to accidents, among others. There is thus a market for “third-party mitigation of accidental loss” for which firms decide to commit resources, hence there is a social cost to accidents even if the loss is purely economic. Going further, Dari-Mattiacci and Schäfer argue that compensation should be paid in some cases even in the absence of social cost. They present the case where an expert paid by A incorrectly overvalues a house which A thought of buying. A loses interest in the house but his friend B likes the house. A transmits the expert report to B who buys the house at too high a price. There is no social cost, only a private transfer from B to the seller. Nevertheless, the expert should be liable to B because the expert has internalized the liability risk in the price of the expertise, and it matters not that B used the expertise instead of A. The situation would be different if the liability costs would not be internalized, for instance if a financial statement made for internal auditing purposes is then used as a basis for an acquisition by a third party. In light of their analysis, the authors conclude that denying compensation for pure economic loss a priori is not justifiable from an economic point of view, and they offer alternative designs for non-contractual liability. Dari-Mattiacci and Schäfer’s analysis appears more robust than Bishop’s. In the end, there is no support for the exclusion of pure economic losses priori from liability. In any event, even Bishop had to accomodate within his theory the various cases where recovery of pure economic loss has been allowed under the common law (typically the cases listed in Table 1 under “specific course of conduct”). Indeed the state of the law is that the a priori exclusion of liability for pure economic losses is moderated by a number of exceptions,66 as the DCFR itself acknowledges when it opens the door to liability for unlawful impairment of business (Article VI.–2:208) and for the various specific courses of conduct just mentioned. In the end, therefore, there is no clear increase in welfare from excluding a priori pure economic loss from loss-spreading through the operation of non-contractual liability, all the 65 66

C. Dari-Mattiacci and H.B. Schäfer, “The core of pure economic loss” (2007) 27 Int. Rev. L. Econ. 8. F. Parisi, V. V. Palmer and M. Bussani, “The comparative law and economics of pure economic loss” (2007) 27 Int. Rev. L. Econ. 29 review these from a law and economics perspective.

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

315

more if – as in the case of the DCFR – a substantial number of exceptions are made to this a priori exclusion. In a large number of pure economic loss cases, there is indeed a social cost, which warrants the use of non-contractual liability, in order to channel the cost towards the least-cost avoider. By the same token, private actors are also given proper incentives to take the requisite care to minimize the social cost of accidents.

3.2.3. Reduction in transaction costs

Non-contractual liability is typically distinguished from contract law on the grounds that the parties to the former (victim/plaintiff and defendant) did not choose to be in contact with each other and could not possibly have negotiated a contract which would have covered the instant case.67 Because of that, it is thought preferable to abandon the idea that the law should enforce the bargain that the parties would have made had they been able to draft a perfect contract; instead, the law of non-contractual liability would be framed in line with public policy objectives. At the same time, from a welfare point of view, the distance between contract and non-contractual liability might not be so large: even if the victim/plaintiff and defendant were not in direct contact with one another prior to the occurrence of the damage to the victim, in reality they are both part of large web of State (social security, etc.) and private (market, insurance) institutions designed to spread losses in an efficient way. On the assumption that individuals and firms behave rationally in the face of the risk of loss,68 they are likely to arrange their affairs (level of activity, insurance coverage, prices, etc.) in such a way as to achieve an efficient level of protection against loss or injury. Accordingly, one could reason along the lines of the Coase theorem for non-contractual liability law as well: the role of the law would be to overcome transaction cost problems and ensure that such an efficient level of protection against loss or injury is indeed achieved. In so doing, non-contractual liability should itself seek to minimize the transaction or administration costs which it generates. This is the third, subsidiary function of tort law.69 A priori limits to liability might play a role here, to the extent that they could simplify the administration of the law by enabling a quick resolution of certain classes of cases (much like per se rules in competition law). Having concluded above that the concept of “legally relevant damage” in the DCFR does not increase welfare either on the deterrence or the loss-spreading functions of noncontractual liability, the one remaining welfare gain would be in reducing the administration costs of the law. The main type of damage which is affected through the operation of these a priori limits – pure economic loss – is in any event not entirely excluded, as was seen in the previous heading. Contrary to other types of damage which are expressly listed in the list of protected interests set out above in Table 1, the DCFR regulates the recovery of pure economic loss through a rule and exception mechanism. Pure economic loss falls outside of the realm of liability unless it is covered by the general clause of Article VI.–2:101 DCFR or it arose out of one of the specific courses of conduct listed in Table 1.

67 68

69

See the basic assumptions set out in Shavell, Economic Analysis of Accident Law, supra, note 48 at 5. An assumption which tends to be questioned by behavioural economics: Faure, supra, note 52 at 87-88. Calabresi, supra, note 46.

316

Pierre Larouche

The issue for analysis is therefore whether the use of a priori limits to liability via the complex conceptual construct of “legally relevant damage” is an efficient legislative technique, compared to alternatives. The main alternative to a priori limits such as “legally relevant damage” is simply to weed out unmeritorious cases ex post through the application of the general conditions for liability which must be assessed in each case, i. e. fault of other accountability device, causation and damage. The choice between a priori limits and control through the general conditions of liability also has an institutional dimension. In the latter case, the legislature – here the drafters of the DCFR – is essentially leaving it to courts to keep the law of non-contractual liability within bounds, through case-law. In all likelihood, courts will pay attention to the categories which are used to define “legally relevant damage” in the DCFR, such as factors listed at Article VI.–2:101(2) and (3), the type of interest which has been injured, the course of conduct, etc. However, courts are likely to be much more nuanced in their approach, wary as they are of not binding themselves too tightly in precedent. With a priori limits, however, the legislature plays a more active role in mapping out the boundaries of liability, effectively depriving the courts of discretion and already indicating to private actors that in certain circumstances they need not fear liability. In practice, beyond these a priori limits, non-contractual liability will not exert any deterrent effect on defendants, nor will it affect the other mechanisms which the parties might use to spread their loss. The added value of “legally relevant damage” resides in its alleged ability to simplify the law of non-contractual liability by introducing an a priori limit to liability. The number of cases is thereby reduced. Of course there is a trade-off: meritorious cases are perhaps kept outside, which affects deterrence and efficient compensation, as explained before. In this respect, there is cause for concern. First of all, just as “legally relevant damage” cannot produce much of an incentive effect because of its complex and methodologically deficient definition, it is unlikely to result in significant efficiency gains in the administration of the law. Quite to the contrary, it could become an additional source of litigation, given that it adds a number of provisions which are themselves subject to interpretation. Secondly, the legislature would be playing sorcerer’s apprentice by trying to predict in advance which cases should be excluded a priori from non-contractual liability and then enshrining that prediction in a legislative enactment. What is more, that enactment would be a codification, which always ranks as a paragon of stability, not to be toyed with frequently and repeatedly. Of the three major legal systems which have inspired most of the non-contractual liability laws in Europe, two, French and English law, share a basic openendedness. In the case of French law, the applicable legislative provisions are very general and leave considerable room to courts. The open-endedness of English tort law is less apparent to the eye, but new torts can be, and are frequently, added to tort law; furthermore, the main tort, negligence, constantly evolves. German law takes a more prescriptive approach, but there as well courts have devoted considerable time to trying to overcome the perceived limitations of some of the more specific provisions of the BGB and opening them to new circumstances.70 70

See Tort Law, supra, note 2 at 142 and ff. (general Persönlichkeitsrecht to overcome limitative list of protected interests at § 823 (1) BGB), 183 and ff. (Recht am Gewerbebetrieb to overcome limitative list of protected interests at § 823 (1) BGB) and 480 and ff. (devices to circumvent the limitations of § 831 BGB).

‘Legally Relevant Damage’ and A Priori Limits to Non-Contractual Liability in the DCFR

317

In the end, the concept of ‘legally relevant damage’ in the DCFR offers no prospect of any notable decrease in administration costs, in part because it is so complex. Against that, by introducing an a priori limit to liability, it deprives non-contractual liability law of some of the flexibility and adaptability which has been its hallmark in many European jurisdictions, and brings with it the risk that meritorious cases are excluded a priori, thereby undermining the deterrence and compensation functions of non-contractual liability. In addition, the practical impact of ‘legally relevant damage’ is bound to be limited. Indeed of the main sources of accidents, many have been taken away from the general law of non-contractual liability to be included in specific liability regimes. Such is the case for product liability,71 environmental liability72 and in most Member States for motor vehicle liability as well.73 The main areas left to the general regime are the liability of the providers of professional services (physicians, lawyers, etc.) and the catch-all category of ‘real’ accidents as between individuals. It is doubtful whether the concept of ‘legally relevant damage’ is worth introducing for these classes of cases only, given the lack of a clear positive welfare effect.

Conclusion This paper aimed first to assess whether the law of non-contractual liability – and in particular the a priori limits to liability – was a good candidate for harmonization or unification. In line with most of the law and economics literature, the answer should be no, both for the law of liability in general and for the specific case of a priori limits. The internal market and rights-based arguments advanced to support harmonization are not convincing. Assuming for the sake of argument that the law would nevertheless be harmonized, the second issue was whether the a priori limits found in Book VI DCFR, as embodied in the concept of ‘legally relevant damage’, represented the best choice for the harmonized law from a law and economics perspective. As a preliminary matter, the comparative method apparently used in Book VI DCFR, and in particular for the elaboration of ‘legally relevant damage’, is deficient. It is a mere superimposition of the very different devices used in German and English law to limit liability for pure economic losses, without regard for the underlying policy choices. The open-endedness of English tort law is jettisoned, and French law seems altogether ignored. What is more, developments in European level seem to have played no role in what is a European project: the ECJ has expressly and repeatedly stated that any a priori exclusion of pure economic loss from liability for breaches of EC law would violate EC law. The DCFR is thus at odds with EC law. In substance, the use of a priori limitations on liability through the concept of ‘legally relevant damage’ has no significant effect on welfare; if it had any effect, it would rather be negative. As regards the first objective of liability law (deterrence of harmful conduct up to the point where the social cost of accidents is minimized), the concept of ‘legally rel71

72

73

See at EC level, Directive 85/374, supra, note 35. Article VI.–3:204 DCFR also creates a special regime for product liability roughly along the lines of Directive 85/374. See at EC level, Directive 2004/35 of 21 April 2004 on environmental liability with regard to the prevention and remedying of environmental damage [2004] OJ L 143/56. See also Article VI.–3:206 DCFR. The EC directives on motor vehicle liability insurance, supra, note 29 do not affect the liability regime in place in the Member States. Article VI.–3:205 DCFR, however, sets out a specific liability regime for damage caused by motor vehicles.

318

Pierre Larouche

evant damage’ under the DCFR is so complex that it cannot produce any strong incentive. Furthermore, a priori limits on liability will tend to induce a greater level of activity that might be socially desirable. Still the a priori exclusion could also be efficient if it led to more efficient loss-spreading (second objective of liability law). Indeed the traditional analysis (Bishop) states that the main type of loss not generally falling under ‘legally relevant damage’ – pure economic loss – is mostly a private loss with no or limited social costs. Yet that analysis is rife with exceptions (as is the law). More recent writings put forward a more robust theory whereby pure economic loss usually includes at least some social cost, and whereby even in the absence of social cost it would be justified to extend liability to pure economic losses in certain cases. In the end, it is difficult to justify an a priori exclusion of pure economic loss. Perhaps the use of an a priori limit could be efficient with respect to the tertiary objective of liability law, namely the reduction of its own administrative costs. Such a limit would then allow unmeritorious cases to be weeded out early and save unnecessary use of resources. However, the concept of ‘legally relevant damage’ in the DCFR is too elaborate to allow expeditious administration. Furthermore, any gain is probably cancelled out by the loss of flexibility and accuracy which ensues when the legislature tries to map out the scope of liability law ex ante in detail. In the end, no efficiency or welfare gain follows from the introduction of an a priori limit such as the concept of ‘legally relevant damage’ at the DCFR, and it should be taken out. This is not to say that no control should be exerted over the type of damage suffered. Rather, there is no reason to believe that the use of a priori limits adds value beyond the general conditions of liability. It must be noted in this respect that, for all the attention which went into crafting the concept of ‘legally relevant damage’, the DCFR is on the other hand stunningly brief on causation at Article VI.–4:101. In all legal systems, causation is used as a filter to limit the scope of liability, using devices such as foreseeability, adequacy or scope of risk.74 Indeed it seems more natural to conduct this examination together with the causation inquiry, since this is where the event for which the defendant is accountable is put in relation with the damage which the plaintiff/victim suffered. It is beyond the scope of this paper to ascertain how a proper concept of causation could be formulated. The drafters of the DCFR chose to include some of the elements to be considered in the general definition of ‘legally relevant damage’ at Article VI.–2:101(2) and (3) DCFR. This could perhaps be envisaged as a second best solution, albeit an imperfect one. In such a case, however, Article VI.–2:101(1)(a) as well as Articles VI.–2:203 to VI.–2:211 should still be removed from the DCFR. As it stands now, the concept of ‘legally relevant damage’ would complicate the law in almost every Member State, as compared to the current state of the law, if the DCFR were to replace national law. It is no coincidence that that concept does not fare well under a law and economics analysis. Accordingly, it should be re-assessed urgently. 74

See Tort Law, supra, note 2 at 395 and ff.

Conclusion Filomena Chirico*, Eric van Damme** & Pierre Larouche*** The fifteen contributions from the members of the Economic Impact Group (EIG) covered many of the main themes of the academic DCFR, certainly as regards contract law. The first part of this conclusion revisits basic principles which came through the various contributions and underpinned the work of the EIG. The second part briefly reviews the outcomes reached in each of the contributions. The third part draws more general conclusions arising from the contributions seen together.

1. Basic principles 1.1. Function of contract law from an economic perspective Well-functioning legal institutions are needed for a free market economy to operate properly. In order for individuals to benefit from the gains of trade, there need to be well-defined property rights that are also enforced, for example. Other aspects of the legal system need to be developed as well: liability law (for protecting individuals and ownership against intrusion), competition law (to avoid the deadweight losses associated with monopolies), bankruptcy laws, etc. Within this general system, contract law determines the rules for how sets of claims can be traded against each other. From an economic perspective, the basic benefit of contracts is that they allow individuals and market parties to make binding commitments. Within game theory, a basic distinction is made between cooperative and non-cooperative games. In a cooperative game, individual players can make binding commitments and coalitions of players can enter into binding contracts. In a non-cooperative game, neither is possible. As the well-known prisoners’ dilemma shows, in a non-cooperative game, players may end up in an outcome that is inefficient. With contracts, an efficient outcome becomes feasible and stable. Contracts thus allow players to reach more efficient outcomes. The rules of contract law govern who can sign contracts, what type of contracts can be signed, under what conditions a contract is established, which contracts will be externally enforced, and what can be done, or what will happen if one of the parties, or both, violate the contract. *

**

***

Official at the Directorate General for Competition of the European Commission – At the time she contributed to these conclusions, she was Assistant Professor and Senior Member, Tilburg Law and Economics Center (TILEC), Tilburg University, and coordinator of the CoPECL Economic Impact Group. The opinions expressed in this article represent only the author’s personal views and not necessarily those of the European Commission. Professor , Department of Economics and CentER, School of Economics and Business and CoDirector, TILEC, Tilburg University. Professor of Competition Law, Tilburg Law School, and Co-Director, TILEC, Tilburg University.

320

Filomena Chirico, Eric van Damme & Pierre Larouche

Contracts may be especially useful when the exchange involves a time element. In such situations, for an efficient outcome to be reached, it may be necessary that one of the parties makes an investment first, to be followed by an investment of the second party. If the first investor cannot trust the second one to make the investment, he or she might be inclined not to make the first one, and an efficient outcome will not be reached. Again, a contract may change the incentives in such a way that it becomes in the second player’s best interest to co-invest after an investment of the first. In this case, the investment of the first player is “protected”, and an efficient outcome can result. A similar argument applies to insurance contracts. A contract in which a risk-neutral party insures a risk-averse player against the risk of an accident can benefit both players: the insured pays a premium in return for compensation from the insurance company when the accident occurs. Since ex post, the insurance company may have an incentive not to pay compensation, a contract that commits the insurance company to do so may be needed in order to realize the mutual gains from trade. In short, access to contracts induces a different game. Contract law allows the players to change the rules of the game, thereby inducing equilibrium outcomes that are to their mutual advantage, that is, are more efficient. Contract law determines in what way the game may be changed. As such, contract law influences what can be achieved and what not. Specific rules may be needed in order to reach efficient outcomes. In this respect, certain systems of contract law may perform better than other. As one system may function better than another, the inquiry then turns to finding the best contract law system, that is, that system that guarantees efficient outcomes in most of the cases, or at least in those cases that are encountered in practice. Here, it is important to mention three considerations. First of all, unlimited “freedom of contract” (allowing parties to conclude any contract that they would want) may not be the best system. The reason is that a contract between two parties may impose negative externalities on third parties, and, as such, while being beneficial for the two parties concerned, may influence the third party negatively and reduce total efficiency instead of increasing it. Secondly, as stated at the outset, contract law should be viewed in the context of the entire legal and institutional system that supports trade and other forms of social interaction. The various parts of the system are interdependent: which contract law is optimal may depend on the state of liability law, property law, etc. Thirdly, a similar remark applies to other characteristics of the environment, such as personal preferences and endowments. Contract law responds to the trading opportunities (and associated potential problems) that arise; different opportunity sets may induce different systems of contract law that are “optimal”.

1.2. Methodology and the efficiency standard Most of economics is based on “methodological individualism”: the analysis starts from individual agents and their personal interests (and resulting incentives), while outcomes resulting from the interaction between these individuals are judged according to how these individuals value them. The individuals each have preferences over outcomes, and an outcome is said to be Pareto efficient if there is no other outcome that is preferred by all individuals. In a context of “general equilibrium”, economists typically work with this Pareto efficiency criterion. It should be noted that Pareto efficiency leaves distributional issues out of consideration. Alternatively put, there are typically many Pareto efficient outcomes. For example, assume that all individuals, in addition to all kinds of other things, also value money, and that they

Conclusion

321

prefer more to less. In this case, if X is a Pareto efficient outcome and Y differs from X only in that individual 1 gives € 1 to individual 2, then also Y will typically be Pareto efficient. The two outcomes X and Y cannot be compared under the Pareto criterion: individual 1 prefers X, while individual 2 prefers Y. In cases of “partial equilibrium” (when the emphasis is on individual transactions or on single markets), it is frequently assumed that preferences are “quasi-linear”, that is, that all individuals value money in exactly the same way, hence, that money can be used to transfer utility from one individual to another. In such a case, the utility ui(X) that individual i attaches to the outcome X can be viewed as equivalent to the monetary value that i attaches to X and an outcome X is Pareto efficient if and only if there is no other outcome Y such that ∑in=1ui(Y) . ∑in=1ui(X). In other words, for quasi-linear preferences, Pareto efficiency corresponds to maximizing the utilitarian social welfare function. This assumption of utilitarianism is maintained in most of the contributions in this volume. A final remark is in place on whether we should look at total welfare or consumer welfare. To a certain extent, this question is misleading. In the above discussion, we looked at individuals, and in the social welfare function, we looked at all individuals. Hence, we looked at total welfare. The distinction between total welfare and consumer welfare appears when, in partial equilibrium, the attention turns to actual markets; it plays a specific role in discussions in competition law. In such specific cases, one distinguishes between producers and consumers, and when one looks at consumer welfare, one looks at the sum of the utilities of the latter group only. One should, however, realize that profits are paid out as dividends to shareholders as well. What would be the justification of leaving out the well-being of the shareholders? The general perspective that we have taken recognizes that consumption is the ultimate aim of all production, and that shareholders ultimately are consuming as well. Total welfare is the appropriate welfare criterion, but in a well-specified model, there is no conflict between total welfare and consumer welfare.

2. Specific conclusions 2.1. General aspects of contract law In her contribution on the function of European Contract law, Filomena Chirico shows that some basic issues concerning the DCFR – and in particular its contract law core – remain unanswered. While considerable attention has been devoted to the functions of a European contract law in the course of the debate on whether it is desirable to have such a law, surprisingly little was written or said on the function of contract law in general. Yet without a shared understanding of the function of contract law, it will be difficult to reach any conclusion on the desirability of a project such as the DCFR. As it went through various iterations, the DCFR did not become clearer as to the functions of contract law.1 Law and economics literature could have helped in that respect, as set out above. As to the higher-level question 1

In the Christian von Bar et al., eds., Principles, Definitions and Model Rules of European Private Law – Draft Common Frame of Reference (DCFR), Outline edition (Munich: Sellier, 2009), published after the contribution of Filomena Chirico was completed, the ‘functions’ and ‘aims’ have been replaced by a set of ‘principles’, comprising ‘underlying’ principles (freedom, security, justice and efficiency) as well as ‘overriding’ principles (protection of human rights, promotion of solidarity and social responsibility, preservation of cultural and linguistic diversity, protection and promotion of welfare

322

Filomena Chirico, Eric van Damme & Pierre Larouche

regarding the function of European contract law, law and economics literature also offers a methodology to assess whether and how the DCFR would indeed fulfil the functions which it is meant to fulfil, namely the improvement of the internal market. The mere fact that national contract laws differ is no sufficient justification; here as well, there are some tradeoffs, and attention must be paid to externalities, transaction costs, economies of scale, local preferences and the costs generated by harmonization or unification. In her second contribution, Filomena Chirico concentrates on the principle of good faith in the DCFR. While good faith is known in civil law systems and some functional equivalents can be found in common law systems, the notion remains vague and fragmented, especially as regards good faith outside of contractual relationships (pre-contractual duties). It is not the best candidate for harmonization at European level. In the DCFR, good faith is defined at Article I.–1:103, but there is no overarching good faith principle. Rather, the DCFR refers to good faith in various specific contexts throughout. In economic terms, good faith cannot be a mandatory rule, but rather a gap-filling tool that parties are free to contract around. In fact, this is implicitly recognised in other rules of the DCFR (Art. II.–9:101 for example), as well as explicitly stated in the Comments. The formulation should be made clearer and more explicit. A mandatory rule is justified only in specific cases where the risk of opportunism is structural (possible both in the pre-contractual phase and in the course of performance). However, such concerns are better addressed by more specific rules rather than a general principle. In her contribution, Ann-Sophie Vandenberghe found that the principle of non-discrimination as stated in the DCFR seems over-inclusive, failing as it does to distinguish between taste-based and statistical discrimination. In competitive markets, taste-based discrimination could be eliminated (or at least minimised) by market forces; acting against such discrimination can be justified under certain circumstances, but the intervention could seek to educate rather than prohibit. In contrast, statistical discrimination is generally efficient. The use of statistical proxies to guide decision-making reflects a trade-off between accuracy and information costs. A outright prohibition on both types of discrimination is likely to reduce the signalling function of certain – statistical – discriminatory forms of behaviour which are suitable to minimise costs for society (although this is an empirical question to be investigated – insurance, hairdressers, car dealers etc.). Against that background, the DCFR at Article II.–2:101 and ff. does not necessarily focus on the most injurious grounds of discrimination, and furthermore efficient statistical discrimination could still fall outside of the justification clause at Article II.–2:103.

2.2. Formation and interpretation of contracts According to Mitja Kovač and Gerrit de Geest in their joint contribution, the rules of the DCFR on the duty to disclose information (Articles II.–3:101 and ff.) and on contract formation as it relates to informational defects (Articles II.–7:201 and ff.) are generally consistent with economic analysis. That conclusion is in part based on a reading of the provisions together with the accompanying comments; a number of key provisions (Articles II.–3:101 on the duty to inform and II.–7:201 on mistake) would gain from being worded more precisely, in line with those comments. Furthermore, a number of provisions require a showing of maand promotion of the internal market): see para. 15-22, pp. 13-17 and the discussion of the new ‘Principles’ section infra.

Conclusion

323

teriality, i. e. that a party would not otherwise have contracted (if she had better information, at Article II.–7:204, if there had been no mistake, at Art. II.–7:201). The filtering function of such a requirement would be more efficiently achieved with other means: procedural rules (to discourage nuisance litigation) or fine-tuning the remedies (for example allowing only damages and not the possibility to avoid the contract in minor cases). Finally, Kovač and de Geest conclude that there is no need for separate doctrines on mistake, fraud, misrepresentation or latent defect – all that is needed is a duty-to-inform doctrine. As regards the interpretation of contracts, Geerte Hesen and Robert Hardy set out how, in contract theory, transaction and enforcement costs and information asymmetries can lead parties not to conclude complete contracts, which would provide for every possible future state of the world. Incomplete contracts imply that interpretation is needed to fill in the gaps in the contract. Law and economics literature puts forward a number of different interpretation rules. If the starting assumption is that parties want to make complete contracts, a textual or objective rule (“plain meaning”, “four corners” of the agreement) would be more efficient. If on the other hand it is assumed that parties make incomplete contracts, a contextual or subjective inquiry into the intent of the parties could be preferable. In addition, the choice of default rule (majoritarian, hypothetical standard, penalty) affects the outcome considerably. The DCFR, at Articles II.–8:101 and 8:102, retains a subjective interpretation rule which is at variance with the practice of international transactions, where the textual or objective rule usually prevails. Other interpretation rules at Art. II.–8:104 and 8:107 are more in line with what law and economics literature would predict. Information asymmetries are also central to the regulation of standard contract terms, according to Hans-Bernd Schäfer and Patrick Leyens. Indeed, contrary to what is often assumed, market power does not provide an economic justification for judicial control of standard contract terms. Rather, the party which receives standard terms will typically not be willing to incur the information cost of reviewing them, in view of the limited gain expected from that operation. Competition in the formulation of standard terms will accordingly not work, with the risk of a race to the bottom. Intervention would be justified in the presence of such a market failure. Against that background, the judicial control of standard contract terms should not extend to terms which the parties have negotiated, and Article II.–9:403 DCFR should be formulated accordingly. The exclusions at Art. II.–9:406 (compulsory terms, subject-matter of contract, price) are also in line with economic analysis. However, the differentiated tests applicable to consumer and business contracts at Art. II.–9:403 and 9:405 respectively are difficult to justify: the information asymmetry rationale set out above applies to both with equal force. It would have been preferable simply to put a cap on the contractual value of business contracts for which standard terms are reviewable, on the ground that above a certain contractual value it is worth it for a business to incur the information costs linked to assessing the standard terms. In order not to give adverse incentives to the party formulating standard terms, judicial control should lead to the nullity of the term as against the receiving party, as is provided at Art. II.–9:408. Judicial adjustment of terms so as to reduce them to the greatest burden still acceptable to the receiving party (salvatory reduction) is efficient only in the limited areas where the legality of the standard term is not clear to the party stipulating it.

324

Filomena Chirico, Eric van Damme & Pierre Larouche

2.3. Performance of contracts Among the main issues of contract law, performance of contracts has received much attention lately, with the adoption of EC-level legislation touching upon key elements of contract performance – at least as regards sales – and reforms in some major legal systems (Germany). Law and economics literature on the topic is also quite diverse. The contributions of Gerrit de Geest, Anthony Ogus and Urs Schweizer all touched upon different aspects of contract performance. At the outset, as Anthony Ogus points out in his contribution, the case for harmonising the principles of contractual damages as remedies is weak. There are clear differences in local preferences, and they affect distributional issues, which are best left at national level (similarly to taxation). At the same time externalities and cross-border transaction costs do not seem to play a big role. There are, however, economic arguments in favour of providing a set of default principles that parties could choose, or else guidelines to which courts may refer in interpreting national laws. As a general starting point, it is clear from all contributions that from an economic perspective, remedies for breach are not important because of the necessity to compensate or to restore rights, but in so far as they give ex ante incentives to enter into efficient contracts and to perform efficiently, i. e. in a way that increases the welfare of the parties and of society. In particular, law and economics has developed the notion of ‘efficient breach’: if circumstances change such that the parties no longer attach the same value to the performance of a contract, then at a certain point the performance of the contract is no longer welfare-enhancing (value is destroyed overall). From a welfare perspective, it is then more efficient that the contract is not performed. According to Gerrit de Geest in his contribution, in the ‘efficient breach’ zone, in principle no remedy should be awarded for non-performance, save in limited circumstances. In contrast, in the ‘efficient performance’ zone, three remedies could be available in order to give sufficient ex ante incentives not to breach: specific performance, substitute performance and expectation damages. Specific performance is applicable only in very few cases, and in practice substitute performance is the superior remedy, with expectation damages for the remaining cases. The DCFR seems to follow the law and economics literature with a provision which comes close to the efficient breach rule at Article III.–3:302(3) and III.–3:104 (in cases where production costs have increased markedly) and III.–3:301 (where the utility to the creditor has decreased). In cases where a breach is efficient, the rules found at Articles III.–3:701 and III.–3:104 are also in line with law and economics analysis. Similarly, in cases where performance is efficient, the solutions found at Articles III.–3:302 and III.–1:110 are adequate. However, Article III.–1:110 (3)(d), which imposes upon contract parties a duty to renegotiate in good faith before one of them can proceed to terminate a contract, is superfluous (parties can always negotiate if this is more efficient) and at worst harmful (parties have an incentive to play a game of chicken and waste time). In his contribution, Anthony Ogus looks at which standard should be used to measure contractual damage awards: expectation, reliance or opportunity cost. In line with the ‘efficient breach’ theory, Article III.–3:702 DCFR provides for expectation damages as a default rule, i. e. the creditor is put in the position where it would have been if the contract had been performed. However, law and economics literature outlines circumstances where expectation damages might be inadequate. For instance, the creditor might have difficulties in monitoring performance, in which case punitive damages, going beyond the expectation standard, are necessary to cure incentive problems which could arise on the side of the debtor. Except for

Conclusion

325

the allowance for penalty clauses and liquidated damages at Article III.–3:710, the DCFR does not acknowledge that issue. Conversely, under certain circumstances, it is more efficient to depart from the expectation damage standard and award less to the creditor. Firstly, the creditor is meant to mitigate losses after the breach (the loss becomes then the opportunity cost to the creditor); Articles III.–3:705-707 are in line with economic analysis in this regard, Secondly, the creditor should take precautions to ward off the risk of breach, while not going beyond what it reasonable: the rule of Article III.–3:704 deals with the case of failure to take preventive steps. Article III.–3:703 DCFR, by limiting recovery to foreseeable damages, gives an incentive for the disclosure of information and for a limitation of reliance expenditure by the creditor. The treatment of penalty clauses in the DCFR is ambivalent: on the one hand, Article III.–3:710(1) allows penalty clauses, recognizing that they make economic sense, as a signalling device (as to the reliability of the promisor) and as an insurance (for the non-pecuniary value of performance, for example). Yet paragraph 2 of the same rule is problematic because, by enabling courts to reduce “grossly excessive” damages, it takes away such signalling function without qualification and without putting it in the correct context (of market failures in negotiating the clause or contractual externalities, for example). In his contribution, Urs Schweizer analyses these provisions with the help of a gametheoretical model, where the debtor has to decide on whether and how to perform a contract, and where the decision on performance has an impact on the payoff for the debtor, but also for the creditor. The sum of both payoffs is the social surplus, which is maximized for a level of performance which may or may not coincide with what is stipulated in the contract. By affecting the payoffs, the law can give an incentive to the debtor as to whether and how to perform. Using that model, it can be shown that the expectation standard for damages at Article III.–3:702 DCFR is efficient, albeit that practical difficulties are likely to result in a downward bias in damages awards, which distorts the incentives of the debtor. Article III.–3:601 offers an alternative, in that the creditor may opt to reduce the price proportionately to the decrease in value, in view of inadequate performance. ‘Decrease in value’ is left indeterminate: is it the subjective willingness to pay of the creditor (which is hard to assess) or something more objective? That remedy is superfluous if it exists in addition to damages; in the absence of damages, it may lead to distortions. Instead of the decrease in value, it would have been more efficient to refer to a hypothetical bargaining (as in Articles III.–3:513 and VII.–5:103) as the standard for the price reduction. On the creditor’s side, the requirement to mitigate losses (Article III.–3:705) is not necessary to achieve efficient incentives for the creditor, but it is not harmful. The ability to obtain damages may generate excessive reliance incentives; accordingly, it is efficient to allow the debtor to excuse non-performance at Article III.–3:104. However, that provision refers to a vague ‘impediment beyond the debtor’s control’ instead of simply excusing non-performance whenever it is not profitable, which would have given efficient incentives to the creditor as well. As regards termination of contracts, the finely-tuned provisions of Articles III.–3:502 and 3:503 – with the distinction between fundamental and non-fundamental breaches (and the additional notice period in the case of the latter) – are hard to justify from an economic perspective; a simple right to terminate for non-performance would have been equally efficient. Finally, with the help of his model, Schweizer also shows that the law should allow for compensation (if only partial, based on statistics) in cases of uncertain causation (including loss of a chance); the DCFR is silent on this point.

326

Filomena Chirico, Eric van Damme & Pierre Larouche

2.4. Termination of contracts In his contribution, Mitja Kovač examines the general rules on the termination of contracts – by agreement or by unilateral notice – found at Articles III.–1:108 and III.–1:109 DCFR. As a starting point, from an economic perspective, it is sensible not to allow parties to terminate contracts unilaterally, but rather to use the principles developed concerning contractual performance (the concept of “efficient breach” set out above). Indeed, allowing for termination risks increasing transaction costs, encouraging opportunistic behaviour, destroy incentives to contract and generally cause inefficiencies; these conclusions hold in particular for fixed-term contracts. Termination as a default rule makes economic sense in two cases only: for long-term relational contracts (more below in the contribution of Gomez) and if parties themselves provide for termination in their contract. Accordingly, Article III.–1:108 is in line with law and economics literature by allowing parties to agree on termination. Article III.–1:109, which governs termination by notice (unilateral), suffers from two shortcomings, however: it contains no requirement of good faith or other requirement which would act as a brake on opportunism and cheating, and furthermore the criteria to assess the length of the notice period could be clarified by referring to the time required to recoup specific investments, cease to incur reliance expenses, etc. One of the most innovative elements of the DCFR remains its treatment of service and commercial contracts. Fernando Gomez assessed the provisions concerning commercial agency, franchising and distributorship at Articles IV. E.–1:101 and ff., and in particular the rules on termination and eventual compensation, usually the most vexed legal issues in such long-term relationships. Economics analyses these relationships as relational contracts, where the open-endedness of the relationship changes the incentives of the parties (it becomes an infinitely repeated game). In such relationships, termination (including the threat thereof) is both a powerful disciplining tool and a potential expropriation mechanism. Indeed, empirical evidence shows that when legislation imposes restrictions on termination, incentives are distorted: one observes more terminations, and for more trivial reasons. In particular, restrictions on termination for non-verifiable instances of breach are objectionable. In addition, when it comes to compensation upon termination, Article IV. E.–2:303 provides for damages (expectation interest) if no reasonable notice is given. In practice, if compensation is not linked to some behaviour of the terminating party which amounts to a breach (typically, opportunistic or hold-up behaviour), then the availability of compensation distorts the incentives of parties during the life of the contract. Article IV. E.–2:303 seems to aim at protecting “general investment” (not specific to the relationship), as opposed to specific investment (which is exposed to the risk of opportunistic behaviour and hold-up). However, economic analysis predicts that it is unnecessary and even detrimental to protect general investment, because parties will likely deal with it (absent transaction costs, as it seems the case) and the will of the parties may be different from what the law imposes (hence a mandatory rule will reduce efficiency – except if bounded rationality is present). Protecting specific investment, especially if it is cooperative (carried out for the benefit of both parties), is however efficient, and there expectation or reliance damages perform better than gainbased liability as set out at Article IV. E.–2:305 DCFR. The DCFR represents a welcomed systematisation, as compared to the Commercial Agents Directive: for example, the DCFR eliminates the reference to franchising, for which empirical data contradicts the approach of the Directive. However, reliance on the acquis does limit the added value of the DCFR, which could otherwise have offered an opportunity to improve the choices made in the Directive (unsatisfactory outcome of a compromise between national systems). The Directive seeks

Conclusion

327

to protect commercial agents, a policy objective which is not necessarily appropriate in the context of franchising or distributorship. It is acknowledged, however, that full systematisation is a very difficult task, given the state of the research.

2.5. Specific areas of contract law Although a project group was in charge of developing Principles of European Insurance Contract Law (PEICL), these principles were ultimately not included in the DCFR. In his contribution, Tomas Kontautas assesses the PEICL. As a starting point, the failure to include insurance contracts in the DCFR is likely to increase transaction costs because cross-border insurance contracts do not take place in a vacuum but in relation to other cross-border contractual relationships which may, ideally, refer to the DCFR. In general, the PEICL seem to ignore that insurance contracts fulfil an essential economic function. The PEICL approach seems to be too legalistic, often aiming just at the protection of one party without due regard to the consequences of the legal rules on behaviour of other parties and, hence on the effectiveness of the rules themselves as well as on the efficiency of the final outcome. For instance, for large-risk insurance contracts, mandatory limitations (inspired by consumer protection) are not justified, and the PEICL are then better used as a default set. Furthermore, the economic function of insurance contracts includes risk allocation. This should be stated in the definition of insurance contracts, in order to make clear to judges that interpreting the contract in favour of the insured has effects not only on the insurer, but also on society. Finally, the disclosure duties of the policyholder are fundamental and should be strengthened in the draft, including all intentionally withheld information, even beyond the questions explicitly asked by the insurer. The PEICL also introduces duties on the part of the insurer (as to the scope of coverage). The strategic interactions between parties as to reciprocal disclosure of information may give rise to opportunistic behaviour. In their contribution, Kati Cseres and Hanneke Luth examine how the DCFR fares as regards consumer protection. After all, the set of EC directives which inspired the work of the Acquis Group (which then fed into the DCFR) are mostly concerned with consumer law. It is interesting to assess whether and how the consumer protection objectives of the original directives survive in the more general DCFR. Cseres and Luth first note the discrepancy between the principles which are supposed to underpin the DCFR (contractual freedom) and the paternalistic approach behind the provisions concerning information duties, right of withdrawal and standard contract terms. There is no clear model of consumer behaviour which informs these consumer-related provisions: are consumers to be protected because they are weak or because they suffer from information asymmetries? Are consumers fully rational (as neo-classical economics would assume) or is their rationality bounded (as newer strands of research in behavioural economics would hold)? For instance, Article II.–3:102 DCFR refers to the disclosure of information “as the average consumer needs in the given context to take an informed decision on whether to conclude a contract”, without further explanation. As a consequence, the rules actually included in the DCFR do not seem to be based on the analysis of the relevant trade-offs, including the identification of market failures and the costs of consumer protection. For one, the definition of ‘consumer’ at Article I.–1:105 DCFR avoids the trap of excluding businesses acting outside of their trade, but it could be more focused on the key issue, namely information asymmetries. Similarly, the right of withdrawal is generalized at Article II.–5:101 and ff., but it is not clear that a generally defined right is adequate in all situations (the market failure might vary from one situation to the other),

328

Filomena Chirico, Eric van Damme & Pierre Larouche

raising the possibility that consumer remedies might be too generous in certain cases, thereby opening the door to opportunistic behaviour from consumers.

2.6. Non-contractual liability In their contribution, Assunção Cristas and Nuno Garoupa venture into the boundary between contract law and tort law, an issue where legal systems are known to diverge. In law and economics literature as well, the models used to analyse contract and tort go out from different assumptions. Cristas and Garoupa look at a number of flashpoints. First of all, contractual breaches can result in injury to third parties. In a case where repair work is badly executed but the prejudice becomes manifest only when the good is in the hands of a subsequent buyer, the DCFR would direct the subsequent buyer to sue the seller. Economic analysis would rather leave the door open to a suit against the repairperson as well, depending on transaction and monitoring costs for the quality of the repair. Secondly, if a third party induces a breach of contract, the concept of ‘efficient breach’ discussed earlier would be decisive: if the induced breach is efficient, the law should not punish it. Only inefficient breaches should be deterred. At Article VI.–2:211, the DCFR does not make that distinction, and allows liability for any inducement to breach a contract. Thirdly, a number of situations are quasi-contractual in nature (including pre-contractual dealings). In these situations, the DCFR (see Articles II.–3:301, II.–7:214, VI.–2:210 read with VI.–7:214) provides for reliance damages, which are the second-best solution (the first-best being impracticable). Fourthly, a contractual breach can also qualify as a tort as between the two contractual parties: this is the famous cumul issue. Here economic analysis would advocate caution in using tort liability to upset the agreement of the parties, considering that the price internalized the possibility of loss. Any ex post benefit in allowing cumul would have to be offset against these ex ante effects. The last contribution, by Pierre Larouche, assesses first whether the law of non-contractual liability is a good candidate for harmonization or unification. In line with most of the law and economics literature, the answer should be no. The internal market and rightsbased arguments advanced to support harmonization are not convincing. Subsequently, the contribution looks more closely at a central element of the system of tort law of Book VI DCFR, namely the use of the concept of ‘legally relevant damage’ to place a priori limits on non-contractual liability (and in particular, to exclude liability for pure economic loss). As a preliminary matter, the comparative method apparently used in Book VI DCFR, and in particular for the elaboration of ‘legally relevant damage’, is deficient. It is a mere superimposition of the very different devices used in German and English law, without regard for the underlying policy choices. The open-endedness of English tort law is jettisoned, and French law seems altogether ignored. What is more, developments in European level seem to have played no role in what is a European project: the ECJ has expressly and repeatedly stated that any a priori exclusion of pure economic loss from liability for breaches of EC law would violate EC law. The DCFR is thus at odds with EC law. In substance, the use of a priori limitations on liability through the concept of ‘legally relevant damage’ has no significant effect on welfare. It is so complex that it cannot produce any strong incentive to reach the appropriate level of care. It does not lead to more efficient loss-spreading either; more recent economic analysis no longer considers that pure economic loss is essentially private and not social loss. It does not either reduce the cost of administering tort law, since it is too elaborate to allow expeditious administration. Furthermore, any gain is probably cancelled out by the loss of flexibility

Conclusion

329

and accuracy which ensues when the legislature tries to map out the scope of liability law ex ante in detail. In the end, no efficiency or welfare gain follows from the introduction of an a priori limit such as the concept of ‘legally relevant damage’ at the DCFR; at the very least, the specific rules of Article VI.–2:201 to 2:211 should be taken out, and unmeritorious cases should be filtered with the usual concepts of accountability and causation instead.

3. General conclusions The aim of the Economic Impact Group was to investigate specific parts of the DCFR, as opposed to producing an additional collection of essays on the pros and cons of a harmonized or unified European private law. At the end of the exercise, a number of trends emerge, from which useful general conclusions can also be drawn. The analysis conducted within the Economic Impact Group played out at two levels. The first one is the choice of the optimal regulatory level (whether to harmonise certain elements of the law or not); the other one is the choice of the optimal design of the rules. Economic analysis can contribute to both levels. Starting with the first level (appropriateness of harmonization or unification at European level), contributors were generally reserved about top-down approaches (not to mention European-level codification). In this respect, the oft-cited transaction costs arising from different national legal systems in cross-border transactions are important but at the same time they are only one of the relevant concepts used in economic analysis.2 A full analysis takes into account the costs and benefits, both of the current situation and of any harmonization or codification. The current state may impose costs by way of cross-border externalities and transaction costs due to divergence, but at the same time it may provide benefits in respecting local preferences3 or allowing for dynamism and experimentation in the development of the law (which benefits society as it searches for the optimal law). Similarly, while harmonization or unification may bring benefits by removing the costs of divergence or reaching economies of scale in the production of law, it can also generate costs to adapt local systems, induce distortions in the coherence of local systems, produce hidden divergence despite superficial harmonization and ultimately fail to attain its objectives. In the area of contract law, much depends on the legal status of the DCFR. Of course, the DCFR can be a reference tool, a restatement without binding force, in which case the discussion is moot. If the DCFR is an optional instrument – a 28th legal system in the EU – the downsides of harmonization or unification, as described above, are less likely to arise. At the same time, the benefits will probably also be smaller. In a dynamic perspective, however, an optional DCFR might contribute to creating the momentum necessary for bottom-up convergence to occur, through regulatory competition or other mechanisms whereby legal systems are subject to pressure to change and improve (impact assessments, law reform exercises, proportionality test in constitutional or EC review, etc.). 2

3

In its Communication on European Contract Law and the revision of the acquis: the way forward, COM(2004)651 final (11 October 2004) at 11, the Commission concludes from available studies that “there are no appreciable problems arising from differences in the interaction between contract law and tort law in the different Member States”. These local preferences must be ‘genuine’, in the sense that the preferences expressed by local decision-makers might be a result of path dependency or government failure (rent-seeking) at local level instead of reflecting the actual preferences of the local population.

330

Filomena Chirico, Eric van Damme & Pierre Larouche

In other areas, such as non-contractual liability (Book VI), but also other areas of noncontractual obligations (Books V and VII) or in property matters (Books VIII to X), the choices are starker. These areas of law do not lend themselves so easily to optional regimes, given that they are concerned with situations where the actions of individuals impose costs or provide benefits on third parties without the latter having consented. The DCFR in these areas can then either remain an invaluable reference resource, or be enacted as mandatory law. In the latter case, as outlined above, it is open to question whether harmonization or unification will truly be beneficial at this stage in the evolution of European private law. Moving now to the substance of the DCFR, the contributions show that for a significant number of the rules and principles under study, the DCFR is in line with law and economics analysis. For instance, the rules on formation and interpretation of contracts, on performance, on termination are by and large in line with economic theory, when properly interpreted. However, the formulation of these rules may sometimes support a different interpretation, which may result in inefficient outcomes. Nevertheless, three lines of criticism remarks emerge from the contributions. The first is more benign in nature, but the last two are fundamental. First of all, in many cases, recourse to the commentary is necessary to establish that a provision is in line with economic analysis (assuming that the provision is interpreted in line with the commentary). Only in the commentary is the rationale of the rules made explicit; in the light thereof, one can then see that it is consistent with economic analysis. Given that the commentary is less normative than the rules of DCFR, some of the insights currently in the comments could be moved into the text of the rules so as to achieve greater clarity and reduce the risk of inefficient interpretation. Such was the case for Articles II.–3:101 (duty to inform), II.–7:201 (mistake) and III.–3:104 (excuse for non-performance). Article II.–3:101, in particular, is tautological on its face: parties will reasonably expect what the law tells them is indeed reasonable to expect. Of course, the comments provide more precise criteria (referring to the costs of generating the information). Greater attention to law and economics literature in the drafting groups would have allowed for these provisions to be drafted more sharply. Secondly, on a number of occasions, the rules found in the DCFR seem to have been formulated without a complete assessment of their rationales, which economic analysis would have made possible. For one, the provisions on discrimination at Articles II.–2:101 and following do not distinguish between taste-based and statistically-based discrimination, whereas economic analysis shows the difference between the two forms of discrimination and why as a matter of public policy it would be advisable to treat them differently. A similar problem arises as regards liability for interference with contractual obligations, at Article VI.–2:211, where the DCFR fails to distinguish between cases of efficient breach and efficient performance. The rules of compensation for termination of long-term contracts at Article IV. E.-2:303 also ignore the distinction between general investments and relationship-specific investments. Conversely, for the control of standard terms, consumer and business contracts are subject to different tests (Articles II.–9:403 to 9:405), whereas the policy concern underlying the control of standard terms (information asymmetry) is similar. With respect to consumer-related provisions and to non-contractual liability, contributors have also noted that the DCFR does not seem to be based on a clear policy line, whereas the available policy choices were set out in the law and economics literature. Thirdly, the drafters of the DCFR seem to have been oblivious to the ex ante impact of the DCFR. Yet law and economics analysis demonstrates consistently and repeatedly that legal provisions not only enable a normative judgment on behaviour ex post, but also (and

Conclusion

331

perhaps more importantly) that they affect behaviour ex ante, by creating expectations and shaping incentives. These expectations and incentives must be factored into the design of the law, lest the law cause more harm than good. For instance, while in principle economic analysis would support expectation damages for contractual breach (see Article III.–3:702), difficulties in monitoring performance and enforcement mean that it might be necessary to increase expectation damages in order to keep the debtor incentivized to perform the contract. Similarly, while it might seem sensible ex post to reduce excessive penalty clauses (Article III.–3:710(2)), the rule is too broadly formulated and it risks depriving penalty clauses of their signalling function. The rules on termination are especially problematic in this respect. One contributor pointed out that the duty to renegotiate in good faith before requesting a court to terminate a contract (Article III.–1:110(3)(d)) could give an incentive to parties to play wasteful games. Another noted that the rules on termination with notice (Article III.–1:109) do not prevent opportunistic behaviour and cheating. As regards the termination rules for long-term, relational contracts (agency, franchising, distributorship) at Article IV. E.-2:303, a third contributor found the same flaw: in the absence of any requirement that the terminating party behaved opportunistically before the other party is entitled to compensation, the incentives of the parties are distorted. A fourth one found that the distinction between fundamental and non-fundamental breaches, in the specific termination rules of Articles III.–3:502 and 3:503, was hard to justify. In a sense, the work of the Economic Impact Group highlights and documents criticisms which have been levelled at the DCFR – and at a certain conception of private law scholarship – elsewhere in the literature. It has been said that privatists are prone to conceive of their work as ‘technical’ and devoid of policy and political dimension; it can be argued whether this is genuine belief or strategic positioning.4 Privatists also tend to see private law in a vacuum, as a neutral instrument in the hands of private parties, including a few mandatory rules. The interplay between the law and the behaviour of private parties, in particular the way the law influences that behaviour through incentives, expectations, etc. is downplayed. By way of response to these criticisms,5 the drafters of the DCFR have included a separate and more elaborate section on ‘principles’ in the final version of the DCFR.6 According to that section, four principles underpin the whole of the DCFR: freedom, security, justice and efficiency. The ‘principles’ section reads very well, but it sometimes comes closer to an exercise in style than a real discussion of principles and policy orientations. In turn, each principle is shown to inform certain elements of contract law (Books II to IV DCFR), noncontractual obligations (Books V to VII) and property law (Books VIII to X). While it is acknowledged that the principles conflict with one another,7 the presentation is anecdotal, noting that one principle sometimes wins, sometimes loses. So while the ‘Principles’ section is useful to understand the DCFR better, it paints a far too calm and rosy picture of private law, which belies that the section was written after the DCFR had been drafted, and not beforehand.

4

5

6 7

See on this issue M. W. Hesselink, ed., The Politics of a European Civil Code (The Hague: Kluwer Law International, 2006), and in particular the eponymous contribution of the editor at 143-170. And also as a consequence of the work of another CoPECL group led by the Association HenriCapitant and the Société de législation comparée: see DCFR, Introduction at para. 11 and ff. DCFR, Principles at pp. 57 and ff. Ibid. at para. 1.

332

Filomena Chirico, Eric van Damme & Pierre Larouche

From an economics perspective, the most troublesome part of the ‘Principles’ section is the treatment of efficiency as a principle. While the inclusion of efficiency in the list is to be welcomed, the reluctance of the drafters towards ‘efficiency’ is unmistakable: it is “more mundane and less fundamental” than other principles, “but it is nonetheless important and has to be included”.8 When efficiency is discussed in greater detail later on, it is split between “efficiency for the purposes of the parties who might use the rules” and “efficiency for wider public purposes”.9 The former appears concerned with reducing transaction costs, as evidenced by the examples given (minimal formalities, minimal substantive restrictions, efficient default rules).10 The latter type of efficiency is equated with the promotion of ‘economic welfare’. At the same time, efficiency concerns are present throughout the ‘Principles’ section without this being acknowledged explicitly. For instance, the justification for freedom of contract is based on efficiency, and externalities are invoked as the reason why contract law might intervene to restrict freedom of contract.11 Furthermore, the principle of security, as applied to contractual transactions, implies a trade-off between certainty and flexibility in the contractual rules, which is none other than the law and economics debate between rules and standards.12 Finally, the principle of justice is said to imply that parties are not allowed to rely on their own unlawful, dishonest or unreasonable conduct, to take undue advantage of others or to make grossly excessive demands. Here as well, the discussion would have been bolstered and sharpened by pointing to economic concepts such as market power or hold-up and opportunistic behaviour. In sum, the ‘Principles’ section evidences too narrow a view of the significance of economic analysis for an enterprise such as the DCFR. Economic analysis is concentrated under a separate ‘efficiency’ principle, which in turn is given a subsidiary role. The real value of law and economics is not so much at the normative level, but rather at the analytical level. Irrespective of the policy objective pursued, the law can be subjected to an economic analysis: freedom, security and justice can be more or less efficiently achieved, depending on the content of the law. It is the role of law and economics to point to inconsistencies in the design of the law and suggest how it could more efficiently reach its objectives. The above remarks lead back to a key shortcoming of the DCFR, namely the lack of a solid methodological basis. Economic analysis was not used, and neither were the policy choices underpinning private law investigated in depth. In the absence of democratic legitimization for the drafting groups, in the end only a group of high-level legal specialists remain. Their expertise is beyond doubt, but it is not clear how they came to their conclusions and whether they applied the same methods consistently throughout. They produced a momentous work in putting together the DCFR, but it remains a fragile accomplishment. 8 9 10 11

12

Ibid. Ibid. at para. 54. Ibid. at para. 55-57. Ibid. at para. 3. Other types of market failure (information asymmetries, prohibitive transaction costs) are not mentioned. Ibid. at para. 22. Including references to law and economics literature would have enhanced the discussion.