Law and Economics: Volume I: Theory, Volume II: Practice (SAGE Law) [1 ed.] 8132110099, 9788132110095

Law and Economics is a method of analyzing laws, legal processes, and judicial decisions using the concerns and techniqu

1,206 157 6MB

English Pages 406 Year 2013

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Law and Economics: Volume I: Theory, Volume II: Practice (SAGE Law) [1 ed.]
 8132110099, 9788132110095

Table of contents :
Cover
Volume 1: Theory
Contents
List of Abbreviations
Preface
1 - Introduction
2 - Property Rights
3 - Intellectual Property Rights
4 - Public Rights over Property
5 - Contracts and Their Enforcement
6 - Economics of Tort Law
7 - Economics of Crime: Some Preliminary Insights
8 - Economic Laws
9 - Economics of the Judicial System
Bibliography
Index
About the Authors
Cover
Volume 2: Practice
Contents
List of Abbreviations
Preface
Introduction
1 - Rehabilitation of the ‘Project Affected’: Eminent Domain and Just Compensation
2 - ‘Law and Economics’ of Indian Patent Law
3 - Two Cases of IPR Issues in India: Pharmaceutical Patents and Film Industry Copyrights
4 - Economics of Contract Law in India
5 - The Economics of Contract Law: A Business Outsourcing Application
6 - Economic Analysis of Tort Law: Some Conceptual and Interpretative Issues
7 - Efficiency of Liability Rules: An Experimental Analysis in India
8 - Regulation in the Case of a Natural Monopoly: The Electricity Act in India
9 - Initial Public Offerings: Institutions in India
10 - Legal and Judicial Process in India: A Preliminary Economic Analysis of Some Issues
About the Editors and Contributors
Index

Citation preview

Law and Economics

Law and Economics Volume 1: Theory

Shubhashis Gangopadhyay V. Santhakumar

Copyright © Shubhashis Gangopadhyay and V. Santhakumar, 2013 All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage or retrieval system, without permission in writing from the publisher. First published in 2013 by Sage Publications India Pvt Ltd B1/I-1 Mohan Cooperative Industrial Area Mathura Road, New Delhi 110 044, India www.sagepub.in Sage Publications Inc 2455 Teller Road Thousand Oaks, California 91320, USA Sage Publications Ltd 1 Oliver’s Yard, 55 City Road London EC1Y 1SP, United Kingdom Sage Publications Asia-Pacific Pte Ltd 33 Pekin Street #02-01 Far East Square Singapore 048763 Published by Vivek Mehra for Sage Publications India Pvt Ltd, typeset in 10/12pt Adobe Garamond by Diligent Typesetter, Delhi and printed at Saurabh Printers, New Delhi. Library of Congress Cataloging-in-Publication Data Gangopadhyay, Shubhashis. Law and economics/Shubhashis Gangopadhyay, V. Santhakumar.    pages cm   Includes bibliographical references and index.   1. Law—Economic aspects—India. 2. Law and economics. I. Santhakumar, V. II. Title. KNS74.G38   343.5407—dc23   2013   2012047864 ISBN:  978-81-321-1009-5 (HB) The Sage Team:  Rudra Narayan, Aniruddha De, Nand Kumar Jha and Dally Verghese

To Professor N. R. Madhava Menon, Kasturi and Ishanti. —Shubhashis Gangopadhyay To Professor Menon, Abja and Karthika. —V. Santhakumar

Thank you for choosing a SAGE product! If you have any comment, observation or feedback, I would like to personally hear from you. Please write to me at [email protected] —Vivek Mehra, Managing Director and CEO, SAGE Publications India Pvt Ltd, New Delhi

Bulk Sales SAGE India offers special discounts for purchase of books in bulk. We also make available special imprints and excerpts from our books on demand. For orders and enquiries, write to us at Marketing Department SAGE Publications India Pvt Ltd B1/I-1, Mohan Cooperative Industrial Area Mathura Road, Post Bag 7 New Delhi 110044, India E-mail us at [email protected]

Get to know more about SAGE, be invited to SAGE events, get on our mailing list. Write today to [email protected]

This book is also available as an e-book.



Contents List of Abbreviations ix Preface xi 1. Introduction

1

2.  Property Rights

26

3.  Intellectual Property Rights

42

4.  Public Rights over Property

58

5.  Contracts and Their Enforcement

86

6.  Economics of Tort Law

109

7.  Economics of Crime: Some Preliminary Insights

126

8.  Economic Laws

139

9.  Economics of the Judicial System

152

Bibliography Index About the Authors

165 168 173

Abbreviations CAC CCI CNG EEZ FPP GDP MRTP Act OEM PIL RBI SEBI VL WBIDC WTO

Corporate Audit Commission Competition Commission of India compressed natural gas exclusive economic zone full packaged product gross domestic product Monopoly and Restrictive Trade Practices Act original equipment manufacturers public interest litigation Reserve Bank of India Securities and Exchange Board of India volume license West Bengal Industrial Development Corporation World Trade Organization

Preface Getting institutions (including laws) right for development has become a cliché in economic literature these days. Such thinking has encouraged the designing of a number of ‘legal reform’ projects currently being implemented in developing countries funded by multilateral institutions. With the setting up of new generation law schools in India, teaching economics or economics of laws in general, and ‘law and economics’ in particular, have become an accepted practice in legal education within the country. All these developments take place without having an adequate knowledge base that uses economics in analysing laws and legal issues of India and other developing countries. This is a serious lacuna since the need for context-specific material is obvious for any systematic study of institutions (including laws), and the insights provided by the general theory of ‘law and economics’, though useful, are inadequate. During our experiences with teaching, as well as during our public engagements, we strongly felt the need for a book on law and economics which explains theory with case materials from India. The purpose of the current book is, therefore, to communicate the economic insights to a wider audience (including lawyers) who may have no formal training in economics. We include, along with the theory, applications on actual laws and legal issues. In different ways we teach the same subject but to be honest to ourselves, we realized that we need the help of practitioners to be able to drive home the points we were trying to make. Hence, while the first volume focuses only on the theory, the second is edited by us with contributions from others. Shubhashis entered this field through a course he offered at Gothenburg University, Sweden, in the early nineties, on the law and economics of capital markets. Clas Wihlborg, then at Gothenburg University, was instrumental in encouraging him to offer this course and Shubhashis owes him a deep sense of gratitude for getting him started. This developed into a course in law and economics which he has been offering for more than a decade at the Jawaharlal Nehru University, India. So, obviously, his students, both in India and Sweden, have had a large role to play in the contents of this book. Like Santhakumar, Shubhashis also had the good fortune of coming in contact with Professor N. R. Madhava Menon. Indeed, without Professor Menon,

xii  Law and Economics

this book would not have been conceived, let alone written. Shubhashis has also benefited greatly from discussions with his Swedish colleagues, Ulf Petrusson, Stefan Sjogren and others, in the Gothenburg law department. Santhakumar had started working on economic analysis of law when he was a post-doctoral scholar at the Vanderbilt University. His contributions in this area included an economic analysis of public interest litigations in India, and a theoretical analysis (jointly with Mark Cohen) of the mandatory information disclosure. There was a demand for teaching law and economics as part of the M.Phil Course in Applied Economics at the Centre for Development Studies (CDS, where he was working then) mainly because of the persuasion of Professor N. R. Madhava Menon (who founded the new generation law schools in India) while he was the chairman of the governing board of the CDS. On taking up this responsibility, Santhakumar felt the non-availability of Indian applications as a major challenge for teaching the course. He was toying with the idea of developing a number of ‘law and economics’ commentaries of Indian laws and legal issues. Santhakumar wishes to note that his excitement in learning law and economics was primarily ignited by the lectures (which he has audited) by Professor Andrew Daughety of the Vanderbilt University. The different batches of M.Phil students of the CDS who have attended the lectures of Santhakumar, and his doctoral student Indervir Singh who has worked on the economics of Indian contract law have helped to sustain the interest in the subject. Professors Narayanan Nair and Sunil Mani of the CDS were instrumental in assigning the responsibility of teaching law and economics to him. The project of developing such a book became very appealing when Sugata Ghosh of SAGE Publications came forward and offered to publish the book. The proposal for joint work in this regard reduced the burden on each of us and that made the project look more feasible. The response of potential authors of the application chapters for the edited volume was very encouraging. We are happy that the efforts during the last one to one-anda-half years by the authors, contributors of the chapters in the edited volume and the publishers have finally led to this publication. The two ladies in Shubhashis’ life—Kasturi and Ishanti—were mostly unimpressed by this book-writing activity but graciously put up with it. People who suffered Santhakumar’s obsessive compulsive disorder in matters related to academics are Abja and Karthika.

1

Introduction Why This Book? ‘Law and Economics’ has become an important sub-discipline within economics and within law in many countries and, not surprisingly, has become an important topic of teaching and research in the top universities of the world. Substantial research has been going on in this area over the last 30 or 40 years, as is evident in the growth of journals devoted to this subject as well as the proliferation of articles in law and economics in academic journals. It is taught in many undergraduate programmes in economics and standard textbooks are available for use in these courses. What then is the use of producing one more book for Indian readers? Why do we not simply use the standard textbooks available in the market? Or, what is different in the content and form of this book that may make it relevant? Not surprisingly, the empirical or case materials used in existing textbooks are borrowed from the laws or legal cases in the United States or Europe. There are not many books based on the laws, legal and judicial institutions of India. Hence, there are fewer interpretations of the laws and legal processes of this region using a law and economics framework. The need for contextspecific information for the study of institutions (including laws and legal institutions) is somewhat obvious. Moreover, legal or judicial reform (creating enabling institutions for economic development) needs to be seen as an integral part of legal institutions in developing countries such as ours. This requires analysis of the existing rules so as to design more efficient laws and legal systems. For instance, in competition law and its practice, the context in which competitive markets are playing out in India (and other emerging countries) is very different from that in matured markets of the United States and European Union. If one applied the anti-trust laws of matured markets when Coca Cola bought Parle in India, or Unilever bought Kwality, neither of

2  Law and Economics

these acquisitions would have been allowed, as the immediate impact of these acquisitions was a very large increase in market concentration.1 However, if we look at the Indian markets now, both the soft drinks and ice cream markets are very competitive. This is largely because both these markets have been growing and the static definition of concentration usable in matured markets is not that relevant in India’s growing markets. The teaching or learning and the use of law and economics in India is constrained seriously by the lack of adequate applications of this methodology to our context. Even though the theoretical issues can be taught by using US-based textbooks, it would be ideal to have books on theory written with context-specific examples. The lack of volumes with enough applications or case studies from the region is a serious limitation. It is this problem that we try to address through this two-volume book. This book is not written either as a mere textbook aimed at a full-fledged course in law and economics or as a volume carrying research articles and applications of law and economics to various issues in the region. Instead, the main purpose is to communicate to the practising lawyers and economists who may not have separately studied the other discipline. Thus, the basic theory and applications are written with examples from the region in a manner accessible to readers without the help of a lecturer or an instructor (as in a class room). However, this book would also be useful as reading material in a course on law and economics (for the students of either economics or law) as a compendium of applications from India. The instructors could continue to use standard textbooks like Cooter and Ulen (2004) in addition to this one in a full-fledged course of law and economics for the students of economics; for law students on the other hand, this book could independently satisfy the requirement of such a course. This two-volume book deals with the theory of ‘law and economics’ and its applications in the context of India. It is written with an objective to convey the principles and the use of this discipline (based on real-world examples) to lawyers and economists as well as policy makers, analysts and civil society observers. Those who are unfamiliar with this discipline may think that this is a multi-disciplinary affair bringing together the disciplines of law and economics. This is not correct. It is less multi-disciplinary than the name suggests. In reality, ‘law and economics’ is a particular way of analysing the 1 Market concentration is measured by the Herfindahl Index. If x is the share of firm i in i total sales by all companies in that market, then the Herfindahl Index is given by Σi xi2. Note that the market share of firm i is a static concept—the value today rather than what it will be in the future.



Introduction  3

contents of the law, legal process and judicial decision-making. It uses the concerns and methodology of economics to study the economic implications of various laws and legal processes. The first volume of the series describes the basic theory of law and economics in an accessible manner with examples from India. The second volume is an edited compilation of ten articles, each article dealing with one legal issue relevant for India. The first volume is organized into nine chapters including the Introduction. The other chapters deal with property rights, intellectual property, public property and public rights over property, contract laws, tort laws, economics of crime, and legal and judicial processes. The second volume has 10 applications of law in India. Some preliminary understanding of ideas of neoclassical economics is needed for the easy comprehension of the materials in this book. This is what we provide in the remaining part of this introductory chapter.

What Is ‘Law and Economics’? What is the major concern in any economic analysis of law? It is to answer, for instance, questions of the following type: how efficient is a provision in law or a step in the legal process and what could be the economic implications of a judgment? By the economic implication of a judgment, we mean the effects a judgment will have on the economic behaviour of people. Take, for example, the Supreme Court’s judgment a few years back that all commercial vehicles operating in Delhi must be powered by CNG (compressed natural gas) to reduce air pollution in the city. Its immediate impact was that all commercial vehicles had to switch to CNG. This boosted the production of CNG (since its demand increased), of cylinders that held the CNG, as well as the people who serviced these activities. At a more subtle level, economists would want to know whether such a judgment requiring all commercial vehicles to use only CNG creates the right incentives for scientists and technologists to develop fuels that could be cleaner than CNG or engines that use any other fuel more efficiently. Economists have a very strict definition of efficiency and we will have a more detailed treatment of this issue in the next section. For now, it is enough to say that we want to see whether a particular legal action (a law or a step in the legal process) enhances the utility (or welfare or even happiness) of a society to the maximum extent possible, given the level of inputs or resources? Is it possible to increase the utility of some people, without hurting others,

4  Law and Economics

and are such possibilities overlooked by a specific law or legal action? Given that wealth and consumption may play an important role in enhancing such utility (welfare and happiness)—without any presumption that these are the only factors that determine such welfare—does a law or legal step facilitate the maximization of wealth or the present value of consumption over different periods of time, for the given level of inputs? This can be for the society as a whole or, alternatively, for some individuals, only when we can guarantee that it does not reduce the wealth or consumption of others. Economists have a way of using economic analysis in many fields. The branch of environmental economics, for instance, looks at environmental issues, treating them as problems in (negative) externalities, an area in which economists have well-developed methodologies. In public economics we look at government decisions on taxation and expenditure to maximize the wealth of nations through the provision of public goods, which again is an area where economists have worked a lot. These are examples of areas where we feel that the economics methodology is best suited for discussing what these investigations are trying to probe. In law and economics, however, we have a slightly different approach. The study of law has its own objectives and an appropriate methodology that allows it to achieve those objectives. These objectives are differently defined, and measured, from those of economists. While law tries to create a ‘just’ society based on the principle of equality of all individuals before the law, economists, as stated before, want to achieve efficiency. The concept of what is ‘just’ is far removed from the concept of efficiency and, hence, economists cannot claim that they have a methodology that is well suited to study issues in law.2 Wherever the two methodologies give the same answer regarding what to do in a society, that is, the objectives of law and economics are both satisfied there is no problem. Where they do not, the two disciplines need to come together to resolve the conflicting answers. The study of law and economics tries to bring these two aspects to the fore. As an example, let us consider the rights of an individual, which are the traditional concern of the legal fraternity. There are some rights that are in tune with the concerns of enhancing social welfare (or efficiency) as described above. The right to life and to hold property are examples. Here the concerns of economics and law are not in conflict. There may be certain other distributional rights enshrined in the law in specific contexts. Some of these distributional rights are also important for the overall welfare of society, 2 Perhaps that is why we call it law and economics and not legal economics like we do, for instance, when it comes to environmental economics.



Introduction  5

in the way it is understood in economics. As an example, consider the law that provides access to primary education as a right. Such access is important even for the economic development of a society since the lack of primary education for some people can not only be a barrier to the increase in the quality of life of these people, but it can also pull back the overall economic growth of the society for several reasons. When people do not have skills, the overall level of skills in society would be lower; such people cannot earn their livelihood on their own, and society may be forced to spend money on them. Thus, certain rights like access to education are also needed to enhance welfare as it is defined within economics. However, in most societies there can also be other rights which help certain specific groups, but may not be important for the overall economic welfare of the society. Such rights are determined politically when the interests of different groups are resolved democratically or otherwise. This may lead to some privileges for some people or some restrictions on some other people. Restrictions on some people owning land in some societies are of this category. Some of these privileges or restrictions may have some economic benefits (which may require detailed analysis in specific contexts). However, if some rights are granted to some groups legally, just because of distributional reasons, then these may not be ‘efficient’. This is not to say that there should not be distributional concerns. Instead, we ask a different type of question in law and economics: is it better to address such distributional concerns through means other than law? Also, which is the right wording of the law that will achieve the goal at minimum cost? This cost includes not only those required to make and implement the law. Any law is expected to change the behaviour of people, and thus, there are costs involved in trying to follow the law or gains foregone from activities that are now prevented by the law. The use of law for distributing resources can be costly, and there can be cheaper ways of achieving the same level of distribution. Any society wants its members to have a minimum quality of life. An important element of this is to recognize that no one should stay hungry. To ensure that government officials do enough to guarantee this, India wants a statute on food security. If the statute lays down how this is to be achieved, cheaper methods of ensuring it that may become relevant in the future are immediately ruled out. On the other hand, if the government is free to choose the method of achieving this goal as long as it does it well, respective governments will search for the cheapest and most efficient way of achieving this. For instance, if the right context is developed, direct transfer of money to the beneficiaries could in some cases be a much better distributional mechanism than the direct transfer of subsidized food (as is currently done

6  Law and Economics

through the public distribution system or through ration shops). Policy is always more flexible than law. This is a major cost of law—the cost incurred to change a law when the social and economic context that necessitated the law in the first place changes. Indeed, changing an existing law is as costly as, if not more than, passing a law. The discipline of law and economics analyses specific laws and legal processes (filing or non-filing of cases, use of advocates, arguments in the courts, judgment, punishment, etc.) from the viewpoint of economic ‘efficiency’. What is an economically efficient allocation of rights in a given context? Is the existing law or legal provision that assigns that right in a specific context efficient? Will the remedies available in courts, when people approach with a case of violation of these rights, lead to an efficient resolution of conflicts? These are some of the questions taken up in law and economics.

An Economics Primer for Lawyers Since the book uses an efficiency point of view to analyse laws and legal processes, lawyers who read the book need to have a little more familiarity with this point of view. This may be more important for lawyers in India who are unlikely to have a training or exposure to economics during their education. However, this short section is not aimed at teaching economics. Instead, it attempts, with a few simple examples, to illustrate what is efficiency, how it is important and why laws are important in certain cases to achieve efficiency. In particular, it tries to demonstrate how economists think.

Rationality, Markets and Voluntary Transactions In economics we try to maximize the welfare of societies. The obvious first question is what we mean by welfare. The economics methodology depends crucially on how we measure things, and we find it difficult to study what we cannot measure. Therefore, economists needed to come up with concepts and definitions of welfare that are measurable. It is necessary for economists to be able to say that one level of welfare is greater than another. One obvious candidate for welfare is happiness. Everyone should try for a more happy society. However, this raises more questions than an economist can answer. Is it the happiness of a particular individual, a group of individuals



Introduction  7

or the entire society? Can we say that one is twice as happy as before, or than another person? If something makes one happy, does the same thing make another person equally happy? If happiness is measured in ‘hap’ units, can we say that one unit of hap transferred from one person adds a unit of hap to the person who gains from this transfer? How do we measure this hap anyway? Not surprisingly, economists stayed away from trying to maximize happiness in society! Instead, the focus shifted to society’s wealth and how we can add to it.3 For instance, the income of a nation is the amount generated from its current wealth. Maximizing income then became equivalent to maximizing the potential to add to wealth. Income in a particular period was, therefore, calculated as the amount generated in addition to the wealth accumulated until that period. And this was operationalized through the GDP (gross domestic product). Indeed, environmentalists claim that the degradation of the environment is deducting from accumulated wealth and, hence, the GDP must be discounted by this amount. Accordingly, people are trying to develop measures for environmental wealth in the same way that other forms of societal wealth are measured. However, it is not feasible that every time one evaluates an action, or option, one calculates its effects on GDP! Moreover, individuals take their own actions and are not interested in the maximization of overall welfare but their own. The starting point of neoclassical economics is that economic agents, individuals, companies and government officials are all driven by this basic objective of maximizing their own gains. We call this rational behaviour. This puts the problems in economics on a different plane altogether. As an objective, we want societal welfare maximization; as a basic principle we assume that individuals are interested in their welfare only. How do we ensure that people acting in their own interests will also maximize societal welfare? This problem is non-trivial. What is in my best interest may not be in yours. If I listen to loud music, and you are my neighbour, the satisfaction I derive in listening to music will reduce your satisfaction from silent meditation. This is called an externality—an action taken by a person or group affecting another individual or group that was not a party to the original action. The fact that my neighbour and I are two different people is what causes the problem. If I like to listen to loud music and do silent meditation, I would solve this problem by listening to music when I liked it better than 3 The so-called first book on modern economics was aptly titled An Inquiry into the Nature and Causes of the Wealth of Nations by its author, Adam Smith.

8  Law and Economics

meditating and vice versa. Instead, if we are two people—one who likes loud music and the other meditation—we would have to coordinate our actions in such a way that I listen to music when you are not meditating. This understanding may be difficult to achieve and this becomes evident when we consider that such coordination may have to be made not between two, but hundreds of individuals (for instance, which side of the road to drive on). Our starting point is an economy where there are producers, endowed with technology, and consumers who have their own preferences and resources, which we call endowments. One obvious endowment is the ability to work, or what we call labour. Consumers maximize their own welfare (which we call utility, derived from their preferences) and producers maximize profit. We define a competitive market (economy) equilibrium to be a set of prices where, at those prices, consumers maximize their utility; producers their profit; and, in the process, markets clear (i.e., there is no unfulfilled demand for any good or service). Simply put, competition is a situation where no one agent’s actions can affect the price of any good or service. Our first two (also referred to as fundamental) theorems in welfare economics say the following: (a) any competitive economy is Pareto efficient and (b) any Pareto-efficient outcome can be achieved in a competitive economy provided one is allowed to carry out a redistribution of initial endowments. Let us understand these theorems in a bit more detail. First, what is Pareto efficiency, or efficiency as it is better known? One major problem in economics is that we do not want to compare a person’s utility with another’s. So, if I were to give you an additional apple and you like apples, I know that your utility goes up. Alternatively, if I take away an apple that you are considering consuming, I reduce your utility. If I take it away from you and give it to someone else, I do not know whether the other person’s utility goes up by more than, equal to or less than the amount you lose. We call such comparisons interpersonal utility comparisons and in economics we shy away from having to make such comparisons as much as we can. Our concept of efficiency reflects this shyness. Suppose we can divide up the value generated in a society in either one of two ways, A and B. Alternative B is such that some people are better off than what they were under A, while others are no worse off than what they were under A. In other words, allocation B makes some people better off without making anyone worse off. We then describe A as an inefficient allocation of resources (because there is another allocation B that ‘dominates’ it). An allocation is efficient if it is not inefficient. Let us consider a simple and somewhat trivial example. Suppose I had `100 and I divide that between two people X and Y (who together constitute our



Introduction  9

society). I denote an allocation by the ordered pair (x,y) where the first entry x goes to X and the second entry y goes to Y (hence, an ‘ordered’ pair since the order of the two entries is important to know who gets what). One such allocation could be (45,50), meaning I give `45 to X and `50 to Y. Let me call this allocation A. Consider another allocation B, given by (50,50). We immediately know that A is inefficient since allocation B makes X better off without making Y worse off. But is allocation B efficient? If we were to make, say, X better off than what she is getting now, we have to give her more than `50. However, my total resource is `100. If I give X more than `50, I have to give Y less than `50. In other words, there exists no feasible allocation that makes B inefficient; hence B is efficient. So, the search for efficient allocations boils down to identifying those feasible allocations that are not inefficient. A feasible allocation is inefficient if there exists another feasible allocation that makes some agents better off without making anyone else worse off. Any feasible allocation that is not inefficient is efficient. This simple definition of efficiency has a very serious problem­—it generates many efficient allocations. Even in our simple example with the `100, there are too many efficient allocations. In fact, any (x,y) that satisfies x + y = 100 is an efficient allocation. In particular, both (100,0) and (0,100) are efficient allocations. Economists cannot survive as social scientists if they are satisfied with their result that giving everything in a society to one person is efficient and that’s that. This is where the second theorem helps. It says that if you prefer an allocation (60,40) and absolutely hate (100,0), then a competitive economy will get you to the former outcome (or very close to it) if you do an appropriate redistribution of the initial ownership of resources.4 Remember the definition of a competitive equilibrium we have referred to above. Preferences, technology and endowments are given; we are saying that if you can redistribute endowments, you can generate an equilibrium distribution of final value according to your choice if you leave economic agents to carry out their activities under competitive conditions. These two theorems allow the economist to concentrate on two things— find out what makes markets non-competitive and then investigate what the remedies are for non-competitive markets. In other words, we study how to ensure that market outcomes are as close to what they will be if they were competitive markets. Contrary to popular belief in certain quarters, economists were the first to systematically study when market outcomes do 4 For those who are not faint-hearted about mathematics can look at Gerard Debreu’s Theory of Value (1959) for the most comprehensive and general statements, along with the complete proofs of these results.

10  Law and Economics

not maximize aggregate welfare or are inefficient. Further, when it comes to law and economics, economists concentrate on finding out where and how law can keep markets efficient. There are two obvious questions that arise. First, why are economists so caught up with markets? This is because we believe that rational people will voluntarily transact, or trade, with each other only when it is in their interest to do so. If they voluntarily trade, it cannot be that any one of them will be worse off for they can always decide not to trade. Such voluntary trade takes place in what we call markets and hence our obsession with them. Second, we have stated earlier that economists are interested in maximizing aggregate welfare. How is that related to the concept of efficiency? When trying to maximize aggregate welfare, why are our fundamental theorems concerned only with efficiency? This leads us to an important principle called the value maximization principle.

Value Maximization Principle We all know that different people facing the same choices opt differently. Economists are adept at handling this. What makes life difficult for them is that the same person facing the same choices may choose differently at different times. One important reason for this is that the person’s ‘wealth’ often drives the choices made. In other words, a person’s choice is not only determined by the options she faces but also her wealth position at the time she is called upon to choose between different options. In economics, this is termed as the wealth effect. Fortunately for us, while the wealth effect can be significant in many cases, in most cases we can ignore it simply because it plays little or no role. Situations where wealth effect can be safely ignored are characterized by the conditions listed below.5 Consider two options, A and B, faced by an individual and suppose that she chooses A. 1. There exists an amount of money, m, such that if this individual were given this money and told that she could keep it if she gave up A and chose B instead, she would take it.

See Economics, Organization and Management by Paul Milgrom and John Roberts (1992). 5



Introduction  11

2. Consider the two options as before. But now, the individual is given some amount of money before making her choice (with no conditions attached). She then chooses A, which is what she was doing without this additional money. 3. Just as one had to give some money to induce the individual to move from A to B, she is willing to give up some money to move from B to A. We assume that she has enough money to absorb this outflow of cash and hence able to ‘afford’ A. The first condition means that you can always compensate someone with money to make her do what she would not do without the money. This is a condition that we expect to hold in most of the cases we will study in this book. At the same time it is important to keep in mind that this is an assumption and it is not universally true. For instance, it is highly unlikely that a parent will harm her child if we give her some money. (In algebraic language, A denotes ‘not harming one’s child’ and B denotes ‘harming the child’. We are saying that in this situation no amount of money will make the parent shift from A to B, that is, from not harming the child to harming the child and this is even when we make some payment to the parent to harm the child.) For the second condition, suppose that someone chooses a particular type of chewing gum (among the many available) each time he buys a packet. One day, as he is walking to the store, his uncle calls up and says that he is gifting him one of his houses in a swanky neighbourhood. Once he reaches the store, he will buy the same type of chewing gum. The fact that he has a windfall increase in his wealth will not make him move away from his favourite brand of chewing gum. This also is an assumption that may not hold true in certain cases. For instance, suppose you are training to become an economist rather than a painter. You also buy lottery tickets on a regular basis. One day, while you are on your way to class, you realize that you have won many millions in the lottery. You may stop studying economics and start planning on having a studio of your own. The third condition simply means that you have enough resources to afford a choice. So, before buying the chewing gum, you had the choice of not buying it or paying the price of the packet of gum and buying it. You had enough money in your pocket to buy the packet. When will this not be true? Many economics professors dream of driving a Ferrari but few can absorb the ‘shock’ of its price! When all these conditions prevail, we say that there are no wealth effects. As you will observe, most of the situations we will cover in this book will

12  Law and Economics

satisfy these conditions. When that happens, we have the following (value maximization) principle: ‘An allocation among a group of people whose preferences display no wealth effects is efficient only if it maximizes the total value of the affected parties. Moreover, for any inefficient allocation, there exists another (total value maximizing) allocation that all the parties strictly prefer’ (Milgrom and Roberts 1992: 36). This principle allows us to move between total value maximization and efficiency and hence the correspondence between aggregate welfare and efficiency, and the significance of the two theorems in welfare economics.

How Do Economists Think about Law? We now give two simple examples of how economists think laws can help markets perform the way economists want them to. Consider a public corporation—one in which there are many shareholders, the shares are listed on the stock market and people can buy and sell their shares on the stock market. There are many laws that govern shares and the stock market but our special interest here is in a very fundamental law about public corporations; all public corporations are limited liability companies. The latter is required by law; it is not possible for a company that lists its shares on the stock market to have unlimited liability. What it essentially means is that once I buy a share of a listed company, the maximum I can lose is what I paid for the share. The worst possible case is that the company goes bankrupt and my share becomes worthless. As a shareholder, I am a ‘residual’ earner from the asset—I get some value from my ownership only if the asset has value left over after it has met all its financial obligations. For example, suppose I own some shares of a company. The company sells its product and earns revenue. From this money, the company has to pay its workers, office rent, suppliers, etc. The company also has a debt obligation—having taken a loan in the past that is now repayable—and is unable to meet it. That is, the amount left after making the payments referred to above, the company is unable to pay off its debt. It can try to get some money by selling off its machines and other equipment or borrowing against them. However, even then, the money it gets is lower than the value of the debt obligation. In this case, the company could get liquidated, with the creditor taking whatever value the company’s assets generate. Of course, since this value is less than the amount that was to be paid to the creditor, there is nothing left after the creditor is paid and so I get nothing. The value of my share is zero. On the other hand, the creditor



Introduction  13

has not been fully paid but cannot come to me and demand any payment from me to recover the unpaid debt. The law of limited liability prevents the creditor from doing so. First, note that if small shareholders like me do not buy shares, then the investment needed to develop large industries will not occur. Let us explain how this comes about. Suppose there are 100 small investors with `1 each to invest and 10 large investors with `10 each. This society has 110 people. There are 10 projects, each requiring at least 20 units of investment each. This lumpiness of investment is a very important phenomenon; modern projects require a minimum scale of operation and, hence, a minimum amount of investment. If all 10 projects are to be implemented, then the society requires an investment of `200 (20 * `10). This is possible if all investors invest—the small and the large. Suppose a project is started with 20 small investors, each having invested her `1 in an unlimited liability project. The project is uncertain and with a positive probability it fails to make any revenue. But the project has to make its obligatory payments and suppose this is equal to `10. With unlimited liability each investor will have to sell her home, pool the money and then give up `0.50 worth of the value from the sale of her house. If, instead, this project was funded by the 10 large investors, each of them would have paid `2 each. Even with unlimited liability, they would not lose their homes as they have `8 remaining with them. No one wants to lose one’s house and so the large investor will be more inclined to fund the project than the small investor. However, if the small investors do not invest in these projects because all projects have unlimited liability, then only a few of the 10 projects will be funded. If the large and small invested together, then all 10 projects could be funded and the minimum exposure required from each large investor would be `1 for each project. The actual calculations are as follows: (a) each large investor puts in `1 on each project; (b) ten large investors mean `10 for each project; (c) each small investor puts in `0.10 in each project; (d) hundred small investors mean 100 * `0.10 = `10 from small investors in each project; and (e) this gives us a total of `20 for each project. This is risk diversification—instead of `2 each in five projects, large investors can put in `1 each in 10 projects if small investors also invest. Indeed, large investors are less likely to fund five projects that require them to put in `2 per project than when they can put in `1 per project in 10 projects. So, attracting small investors always helps investment through stock markets. Economists believe that this limited liability requirement for all public corporations is a law that enables stock markets to function by encouraging

14  Law and Economics

small investors to enter the market. Suppose this law was not there. Every time I bought a share, I would need to check if the company had limited or unlimited liability. If it had unlimited liability, my loss from owning a share could be more than the price I paid for the share. Creditors could come to me and demand missed payment. If I had a car, or any other asset, I would have to sell it and transfer the proceeds to the creditor. This is because as an owner of the company I was liable to meet all the company’s obligations (at least to the proportionate amount of my ownership in the company, if not more). The possibility of such a loss would have made me less inclined to buy shares. Thus, the limited liability requirement of public corporations is a guarantee to the shareholder that she will not lose more than what she has paid for the share. The fact that it is a law means that the guarantee is credible and, hence, it encourages a small investor to operate in the stock market (either through the direct purchase of shares or through mutual funds). This was a simple example. Let us now consider a slightly more sophisticated example—not in terms of its difficulty but in terms of its use of economic principles. Suppose that A owns a house and has money. The value of the assets (house and money), V0A, is given by V0A = hA + mA where hA is the value that A puts on her own house and mA is the amount of money she possesses. It is important to understand that hA is the personal value to A of the house; at the very least, she must be given this amount of money if she has to voluntarily give up the house in exchange for money.6 This is known in economics as A’s reservation price of the house. There are two other economic agents of interest to us, B who has an asset value V0B = mB and C who has asset value V0C = mC. The aggregate wealth among these three economic agents is given by V0 = V0A + V0B + V0C = hA + mA + m B + m C. One day, while going past A’s house, B sees the house and immediately takes a fancy to it. He wants the house and convinces himself that he is willing to pay a maximum price of hB for it; this is B’s reservation price—any price below that and he will buy it. He knocks on her door and they sit down to bargain about the price. Let us suppose that hB > hA, that is, the maximum price B is willing to pay is greater than the minimum price A is willing to accept. Then one can safely say that unless something goes terribly wrong, 6 This is the first condition of the value maximization principle, discussed above, at work. Whatever may be the preference A has for the house, she is willing to give it up if she is given an amount of money that is at least as large as hA.



Introduction  15

the two will be able to reach some understanding and settle on a price p such that hA ≤ p ≤ hB. If this transaction were to take place, then a number of economic principles would be satisfied. First, let us calculate the total value of assets resulting from this transaction. A now has mA + p; B has mB – p + hB and C continues to have mC. Therefore, the total wealth of the economy is V1 = V1A + V1B + V1C = [mA + p] + [mB – p + hB] + mC = hB + mA + mB + mC > V0 as long as hB > hA, which was true to begin with. Thus, a transaction that transfers an asset to one who values it more than the one from whom it is being transferred increases the total wealth of the society. Thus, any society that allows, or enables, such transfers is desirable.7 Second, how does an asset market (or, in this case, a market for houses) help? In a market, individuals transact voluntarily, that is, an individual will not go through with a transaction unless she benefits from it. Thus, as long as the price of the house, p, is greater than hA or the reservation price of A, she would be willing to sell the house because she gets more than she gives up; similarly, as long as p is less than hB, the buyer pays less than his reservation price or, again, loses less than what he gains. So, both are better off while C stays on where he was before the transaction. So, two are made better off while the third is no worse off than before.8 Now, let us introduce a twist. After all, not everyone goes out to buy a house with a pocketful of money. So, after A and B strike a deal, let us assume that B is given two days to put the money together and come back with it to A. However, in the intervening period, C also comes across the house and immediately wants to possess it; indeed, he values it at hC > hB. He approaches A and offers to buy it at a price q where hC ≥ q ≥ p. Observe that C has to offer a price at least as high as B, otherwise it is not in A’s interest to sell it to C. After the visit by B, A’s reservation price is no longer hA, but p ≥ hA. Alternately, the higher q is above p, the greater is A’s interest in selling the house to C. So, suppose that A gives in to the temptation of a higher price and decides to sell the house to C. Then, when B comes back with the money to buy the house, A refuses to sell to him. An upset B decides to file a suit claiming breach of contract. In this suit, B is the plaintiff complaining about the defendant A. The court has a number of options but let us consider the two extreme ones: (a) it can allow such a breach and throw out B’s suit or (b) the court says that a contract cannot be breached and A has to sell the house to B as had been agreed upon by the two of them. But before it does anything, it asks an economist which of 7 8

This is value maximization at work. Here we have efficiency.

16  Law and Economics

the two alternatives is preferred by the economist. The economist does the following calculations: 1. If C buys the house, then the total wealth of the society is given by V2 (a) = [mA + q] + mB + [mC – q + hC ] = hC + mA + mB + mC > V1 > V0. 2. If the house has to be sold to B, then the total wealth jumps to V1. However, this is not the end of the story. B now knows that C is willing to buy the house and the first thing to do is to check whether he will buy it from B. Why? Because, since hC > hB, any new price q', such that q' is in-between (max{q, hB}, hC ), will be acceptable to both since both will be better off. Thus, the total wealth will be calculated in two steps: V2 (b) = [mA + p] + [(mB – p + hB) + (q' – hB)] + [mC – q' + hC ] = hC + mA + mB + mC = V2(a). Thus, whether the court allows A to renege on her understanding with B or forces her to honour it, the final outcome, in terms of the overall wealth, is the same! However, what changes is the division of the surplus that is generated. The total surplus generated is the difference in the valuation of the house between the initial owner and the final owner, that is, hC – hA. In the first option, (a), the surplus distribution is given by V2A (a) – V0A = q – hA, V2B (a) – V 0B = 0 and V 2C (a) – V 0C = hC – q and in the second option, (b), it is given by V2A (b) – V0A = p – hA, V2B (b) – V0B = q' – hB – (p – hB) = q' – p, and V2C (b) – V0C = hC – q'. So, should economists be indifferent to how the courts decide on such a violation of an agreement? This is where the thinking in law coincides somewhat with the thinking in economics. In a common law system (as in India), a court ruling on a new dispute (that has come before it for settlement for the first time) tells everybody how it will settle similar disputes in future. This is what we call a precedent being set by a judgment. It becomes like a new law governing such disputes. For economists too, this is important—the settlement in this particular case will affect the way parties contract with each other in the future. Thus, if the court allows A to renege on her agreement with B, then in the future all such agreements between people like A and B will become meaningless, for one party could always renege on them with support from the legal system. The ability to enter into binding contracts is efficient and allows many transactions that would otherwise not be undertaken. Therefore, while options (a) and (b) do not have any impact



Introduction  17

on the current value creation, option (a) has serious negative consequences on efficiency for future value creation. Hence, economists will concur with the more ethical, and hence legal, decision of enforcing promises that were voluntarily made.9 So, economists and lawyers will have different reasons for enforcing the promise by A to sell the house to B, but they will agree on the decision all the same.

How Does Efficiency Play Out? Efficiency can be viewed at three levels. First, one could look for its implications at an individual level. Here we are interested in knowing how an individual’s objective of maximizing utility shows up in the following decisions: what and how much to consume, how much leisure time to have, how much time they want to devote to earning an income, and so on. A person’s actions would be termed efficient if they individually and jointly maximize her utility. In other words, given her resources, a set of decisions would be efficient if there does not exist any other set of decisions that is not only feasible but also gives her greater utility. We have an even simpler exposition if we consider the decision of, say, consuming a banana. The consumption of a banana can be increased until the point where the additional pleasure or happiness becomes equal to the additional cost of procuring that item. This depends crucially on two assumptions—one easy, the other a bit more subtle. The easy one says that more bananas cost more. What it means is that the cost of buying five bananas is not less than the cost of two bananas. If bananas come at a price, that is, they are traded in a market, this is most often true. The more non-trivial assumption is that the additional utility from the second banana is higher than the additional utility from the fifth banana (assumption of diminishing marginal utility). Thus, if the individual is consuming five bananas, we can make the following observations: 1. The utility from consuming the first banana is greater than its cost (i.e., price). 2. The additional utility from buying and consuming the second banana is also greater than the cost (i.e., price) of the second banana. However, given the non-trivial assumption of diminishing marginal utility, this addition to utility from the second banana (though higher than its price) is less than the addition to utility from the first banana. 9 Of course, courts can, and do, decide which promises they will enforce and which they will not. We will discuss these in more detail later.

18  Law and Economics

3. What is true for the second banana is also true for the third, fourth and fifth bananas; that is, the additional (marginal) utility from each of these bananas is greater than the price of a banana, though the addition to utility for the fifth banana is less than that of the fourth, which, in turn, is less than that of the third. 4. The addition to utility from consuming the sixth banana is less than its price, and, hence, the consumer does not buy it. Another way of putting all these together is as follows: each time the individual buys an additional banana, it must add to the surplus (marginal utility greater than marginal cost); otherwise, the individual will not buy that additional banana. In general, we say that the consumption of bananas is at the efficient level when the additional utility of the last banana is no less than the cost of the additional banana, or when marginal utility equals marginal cost (given the two assumptions stated above that marginal utility is decreasing and marginal cost is non-decreasing [or increasing]). Second, we may want to ask about the efficient level of production. This notion of efficiency is much easier to understand. A production level is efficient when it uses the minimum amount of resources to produce a given level of output(s) or service(s). Alternately, an efficient production process produces the maximum amount of output(s) or service(s) given a level of input(s). Once again, at the efficient level, the marginal productivity (instead of utility) of an input is equal to its marginal cost. We now go on to the concept of efficiency in exchanges or trades of goods and services. Why do people exchange goods and services? What is the social benefit of such exchange or trade? There are many examples of such trade or exchange even in a relatively underdeveloped village. People buy clothes and vegetables from a travelling salesman or employ a local artisan to repair agricultural implements. People gain (in whatever way they measure or perceive such gains) from such transactions; otherwise they would not take the trouble to participate in such transactions. Consider the example given earlier about the sale and purchase of a house. The concern for efficiency in exchange could lead to the following questions: Are these gains the maximum possible? Is there some way the gains of both can be increased further or is it possible to increase the gain of at least one, without harming the other? An efficient situation is when all the gains or surplus theoretically possible from a transaction or exchange are actually realized by the concerned parties. Such an ideal situation occurs often when people do mutually beneficial transactions or act on their own. However, even that requires a broad



Introduction  19

institutional framework. For instance, if the law and order situation in a region is very bad, and the travelling salesman could lose his wares to a mugger, he will not visit that region and efficient transactions will not take place. If an auto-rickshaw driver fears that he may be mugged at particular parts of the city, he will refuse to take paying passengers to these destinations even when both could gain from such a transaction. Hence, an institutional set-up that maintains law and order is essential. These are broad, and general, contexts in which transactions take place. However, there is a larger set of situations where we may need specific laws, beyond this general maintenance of law and order. These are what we study in this book.

Need for Laws from an Efficiency Point of View An obvious case is that of property rights. In order to make the gains or surplus described in the example stated above (of the sale and purchase of the house), B will not even consider talking to A unless A ‘owned’ the house. For voluntary transactions through a market, one can exchange only what one owns and society must define and enforce property rights. In the absence of such a system, there is no assurance for B that (a) A does have the capability to sell the house to B as well as that (b) once B buys the house, C will not be able to take it away from him without paying a suitable price. In the absence of such property rights, all the gains made by society through the sale of the house will become uncertain and, most likely, non-attainable. A significant part of this book will be devoted to the study of appropriate property rights in different contexts. Some exchanges can take place on the spot or over the counter. That happens when we go to a corner shop and buy a cake of soap or a packet of chewing gum. Here we take the good and pay the money immediately. Observe that all such transactions are two-way—one transfers money, the other a good. Sometimes, one type of good or service is transferred one way and another good or service is transferred the other way. Importantly, many transactions do not happen instantaneously. Continuing with the earlier example of the house, there is usually a gap between agreeing to a transaction concerning a house and its actual sale and transfer of the purchase price. Sometimes, a good is sold first and the money paid after a period of time (e.g., instalment buying). Alternatively, money is paid first and the good is transferred later. How do we ensure that the good is actually delivered if money is paid earlier or the money is paid if the good is delivered earlier? Such transactions are carried out through contracts and such contracts can

20  Law and Economics

create value only if there are ways of enforcing them. In the absence of such contracts (or their enforcement), people may not be willing to enter into trades or exchanges in which there is a time gap between the delivery of goods or services and the payment of money. This is inefficient since the potential surplus or gains from such exchange or trade will not be realized. Thus, the legal mechanisms that ensure the validity of contracts also enhance efficiency. In the contractual exchange described above, it may be possible to detail what is to be done if one of the parties acts in a manner harmful to the other (say by providing a poor-quality good or service). Such clauses in the contract can be enforced in the courts, if needed. However, there are some cases of exchange or interaction where the potential harm and its compensation cannot be negotiated. When a person uses the road, she cannot do so after contracting with all other vehicle drivers about what would be the compensation if she is struck down by a driver in a road accident. Thus, there should be some compensation mechanism(s) for such losses. In the absence of such compensation mechanism(s), she would not be using the road to the extent that she desires, or she may take costly precautions. These are inefficient, since efficiency requires that she should use the road (to the extent she wants) at a minimum possible cost to society; also, the cost of precautions that she takes should be the minimum possible for a given level of safety. Thus, legal mechanisms to compensate for such harm, which cannot be contracted for by concerned parties, are also necessary for achieving efficiency. So far we have gone ahead with the assumption that it is better for people to depend on exchange of goods and services, and the only role of the legal institutions is to protect private property rights, enforce contracts and see that non-contracted harms10 are compensated. However, there may be cases where a collective, or state, may have to interfere with private property rights of individuals. These include cases where one person’s actions in her private property affect the other (and the former does not bear the cost of such actions). Pollution from a factory to surrounding land, water bodies and atmosphere are classic cases of such problems. This may require public intervention into private property rights (on what a private property owner can and cannot do). This can also be viewed as a harm, for which a contract may not be the best solution. 10 This is not to imply that contracted harms need not be compensated, but such compensation will be part of contracts, and hence no legal mechanism other than that for the enforcement of contracts is necessary in this regard.



Introduction  21

For instance, let us think about a marketplace in a rural area. There may be several shops selling vegetables, food, meat, fish, etc. Each shop has some waste to dispose of at the end of the day. It may simply dump this waste somewhere outside the shop, most probably by the roadside. When all or most of the shops dump their waste outside, the whole area begins to stink. This pollution affects everybody, keeps customers away and affects all vendors’ ability to do business. How does one prevent such harm caused by one to the other? First, it is worthwhile to note that often such pollution happens because the shops do not have to get permission from (or pay) anybody to dump this waste in a public area. In India, this is largely due to the feeling that the ‘roadside’ is not owned by anyone and, hence, everybody can use it the way one wants. In other words, though the roadside is owned by the state, it does not enforce its rights effectively. One can view the roadside garbage problem as an outcome resulting from a lack of well-defined, and enforced, property rights over public places. Second, one can view this as part of a non-contracted harm; it is difficult for people (who are affected by the pollution) to enter into a contract with the polluters on the amount they should be compensated for the harm. When such pollution is seen as a property right issue or that of non-contracted harm, it requires other legal interventions. Or, a lesson could be that if there is some pollution for which one can enter into a contract for compensation, then such legal intervention may not be needed. There may be cases where collective goods (like a public park) have to be provided, and this may require public property (or cases where there should be public ownership of property). In addition there may be cases where it is better for the state to intervene legally (or through policy measures) on the process of exchange itself. Some of these instances are given below.11 1. Consider a family that needs electricity but is living in a village that does not have an electricity connection. They could buy a diesel or kerosene generator. In addition to the cost of the generator, they would have to spend some money on diesel or kerosene, say `50 daily, to get electricity for three to four hours every night. If they get electricity from the State Electricity Board, then the expenditure for each household would be much lower than what they would have to spend on buying and running generators. This is due to economy of scale. Thus, there are cases where the per-household expenditure for the same service may come down when a service is provided to 11

This part is an adapted version from Santhakumar (2011).

22  Law and Economics

all by a single agency. This is possible if one private firm provides electricity to all the households in the village. However, only one firm can function here since the operation of a number of firms will work against the cost reduction mentioned above. But this firm will not have any pressure to charge only the minimum required price, or provide as much service as in a situation where a number of firms compete with each other. This creates social losses and hence becomes socially inefficient. Thus, there should be some role for the ‘public’ or the state in compelling this single firm to operate as if it were in a competitive situation. This is a problem similar to that of a ‘natural’ monopoly and may require legal mechanisms to avoid it. 2. The buying and selling of many goods and services are facilitated through legal measures mandated by the state to ensure the use of proper weights, measures and standards. In the absence of such intervention, the seller is unable to credibly communicate the real quality or quantity of the product sold or service delivered. The buyer is unable to secure proper information on (the quantity or quality of) the product sold, and the resultant excessive caution may affect (or reduce the volume of) exchanges or trade. These issues are due to private information that one party in an exchange has, which is not available to the other party. So far we have talked about the ‘need for laws’ for efficiency. However, the efficiency view does not stop here. Even when a law is needed and made, the law so made must ensure that people are able to carry out transactions that achieve an efficient outcome. Efficiency is lost when the cost of a transaction increases and adherence to the law increases the cost of the transaction. For instance, the requirement that every sale must be done through a bill or invoice means one has to incur the cost of paper, time, writing ink, etc. This increases transaction cost and reduces the number of possible trades. Indeed, certain human activities or transactions between people will be reduced in the presence of a law. For example, when pollution by a firm is prohibited by law, it is likely to reduce the availability of commodities whose production creates the pollution. This may impose a cost on the society (as the foregone benefits of production), even when the reduction in pollution is beneficial to society. Thus, along with the reduction in pollution that produces benefits, we need to consider the costs of reduced production of the commodity in question. Thus, different laws or different ways of restricting the polluter may have different costs or foregone benefits. Thus, for efficiency, we need to pick those laws that impose the least cost on society for a given



Introduction  23

level of pollution reduction. Hence, all the laws made for reasons of efficiency as described above need to be evaluated in terms of whether these, or the specific provisions in each one of these, are efficient or not. There are also direct costs, which include the resources spent by the society on those who are needed to make the rules, as well as for maintaining those who enforce them. The legal system needs to guarantee that those who violate the law will have to compensate the affected individual or society. There is unlikely to be a unique enforcement machinery for each law, but all laws together in a society are usually enforced by the state (through the police machinery) and violations of the law are handled ultimately by the judicial establishment. The cost of such enforcement mechanisms should be the minimum required for a given level of enforcement. Thus, it is not just the laws, but the efficiency of the legal and judicial processes that are also important.

The Uses of an Economic Analysis of Law Exposure to discussions in law and economics may help lawyers, policy makers, officials and civil society researchers and activists in many ways. First and foremost, this can improve the making of laws to a great extent. It is not unusual to see laws being made without a clear understanding of what they want to achieve or without any consideration of the cost they impose on society, and without an analysis of legal alternatives that may provide the same benefit but at a lesser cost to the society. Familiarity with law and economics may change that situation. A similar approach is true for the enforcement machinery as a whole and particularly for the legal and judicial establishment. There are many steps by which the social cost (which include not only that for the government but also people who use the courts) of law enforcement can be reduced. These need to be viewed as part of legal and judicial reforms. The quality of such reforms can be enhanced, as well as social attitude towards (or public debates on) them can be made more meaningful, with some understanding of law and economics. This is not to say that the opposition to such reforms by certain interest groups may decline, but society may be in a better position to comprehend the incentives for such opposition with the help of law and economics. Law and economics can also help judicial decisions. It may help the judicial officers to sort out the conflicts reaching the courts in a socially optimal

24  Law and Economics

manner, even without violating the right-based legal provisions. The courts’ decisions or legal remedies can be used to instil efficient behaviour on the part of the contesting parties. Courts may be in a better position to understand the incentives behind the position of one or other parties, and through an appropriate law can change their behaviour in a manner consonant with the achievement of overall social welfare. The earlier example in this chapter of public corporations being mandated to have limited liability is a law that encourages small investors to participate in the stock market and make its functioning more efficient. In other words, much of law and economics focuses on making laws that enable markets. By enabling markets we mean allowing individuals to make decisions on their own. There are many reasons for depending on individuals to make their own choices. First, it is impossible for anyone to know what is best for each individual; they differ in their preferences, in the environments in which they take decisions and, most importantly, in the choices that are feasible for them. Second, the world is uncertain; even if we were to know what is good for everyone today, tomorrow that may change as we will have more information than we have today. This becomes very important in the way laws are written, and it is this writing of the law (its provisions) that can make all the difference in an uncertain world. First, let us consider an abstract example. Suppose there are n commercial activities that we know of but, somehow, we have decided that k of them are ‘bad’, that is, society does not want them. Instances of such activities could be drug trafficking and prostitution. We are not interested in knowing why they are bad; let us simply assume that we as a society have agreed that they should not be allowed. We can write the law preventing them in two ways: (a) only the listed n – k activities are allowed or (b) the listed k activities are banned. It is immediate that as long as we have the complete list of n activities, the outcome is the same regardless of how we write the law. However, in an uncertain world, the complete list of commercial activities is never given. The list keeps getting modified and, more important, added to as we go along. So, what do we do when the (n + 1)st activity comes along? Under (b), someone takes it up because it is not in the ‘banned’ list. Under (a), the innovator has to rush to the courts and policy makers to bring it on to the ‘allowed’ list. This increases the cost of introducing the new activity. Most importantly, it also prevents experimentation with new ideas. Such experimentation is essential in an innovative society. To make the abstract example more concrete, let us consider two actual examples. Recall the ‘license-permit raj’ pre-1991 when anyone setting up a new industrial production facility had to get permissions from various



Introduction  25

ministries. Indeed, the introduction of any new technology also required such permissions, especially if it involved imports (of, say, capital equipment or specialized raw materials not available in India). The net result was that our industry stagnated and we continued as a ‘developing’ economy because everyone else did better and newer things before we could get our policy makers to ‘allow’ us to do them. A more recent example is what happened regarding pollution in Delhi. The use of CNG in commercial vehicles became part of the law in Delhi (thanks to a court judgment) a few years back. The law did reduce pollution; however, it did so by stipulating exactly how to do it. What happens if some young IIT-ian develops a better alternative than CNG in reducing pollution? How will that alternative be developed when the cost of introducing it includes the cost of having to change the law into allowing non-CNG alternatives? So, while the law mandating use of CNG reduced pollution in Delhi, it also reduced the incentives to develop better alternatives to reduce pollution. In the rest of the book, we discuss in detail the principles outlined here. The purpose of the book is not to come up with a set of binding recommendations. Instead, it is an attempt to systematically develop a body of thought that will help us in analysing specific issues in specific contexts.

2

Property Rights Why Do We Need Property Rights? Lack of property rights or a situation where a property can be used by anyone can be economically harmful (Gordon 1954; Hardin 1968). This can be explained with a simple example. Assume that there is a fishing lake in a location inhabited by a community. The people can either fish in the lake or take up some job in the nearby city. Those who work in the city will get `100 a day. Let us assume that no one owns the lake, and hence anybody wanting to fish can do so without seeking permission from anyone. We presume that the only cost of fishing is the time of the fisherman. Then, the fisherman earns a surplus from fishing equal to the revenue from fish sales minus the wage that the fisherman could earn from the city job. To demonstrate the point, Table 2.1 gives the values of additional income from fishing as more and more people do it. Let us think of the lake as an asset that provides value because it has fish. The first person who thinks of fishing in the lake gives up `100 worth of city work to catch fish that can be sold at `300. This person earns a surplus of `(300 – 100) = `200 over what he could by working in the city. As more and more people fish (column A), there is initially an increase in the amount of revenue from the sale of fish (column B). However, after this initial increase, the total revenue from fish sales starts to decline (as the number of fishermen increases from 7 to 8). This could be due to a number of different reasons—oversupply of fish in the local market leading to a fall in prices, congestion in the lake as one fishing boat gets in the way of another, etc. The first reason is a market fundamental, greater supply depresses price; the second is what we call negative externality—one more person’s fishing reduces everybody’s productivity, due to crowding, killing of young fish, etc. So, the first thing to note is that too much fishing, or overuse of the asset, can actually decrease its value.



Property Rights  27

Table 2.1. The Problem of Absence of Property Rights Total Revenue from Fish Sales

Marginal Revenue (additional revenue as one more person enters fishing)

Average Revenue for the Fishermen

Total Surplus (total revenue minus wage cost of the fishermen)

B

C

D = B/A

E = B – (A*100)

1

300

300

300

200

2

700

400

350

500

3

1,050

350

350

750

4

1,320

270

330

920

5

1,500

180

300

1,000

6

1,620

120

270

1,020

7

1,680

60

240

980

8

1,600

–80

200

800

9

1,440

–160

160

540

10

1,000

–440

100

0

11

440

–560

40

–660

Number of People Fishing A

The second thing of note is a bit more subtle. Consider the time-zero situation, when everyone is working in the city and the lake is not being used. Suppose there are 20 people who can potentially fish but are working in the city. They are each earning `100 from their city jobs (a total of 20 * `100 = `2,000) and there is nothing being earned from the lake. Now if 6 of them fish, while the others continue to work in the city, the value generated by these 20 people will be `1,020 (column E against the row with 6 fishermen) in addition to the `2,000. Let us see this in more detail. Since 6 people fish, the remaining 14 get 14 * `100 = `1,400 in the city. The 6 who are fishing are generating `1,620 from the sale of fish. Thus, the 20 people together are getting `1,400 plus `1,620, which gives us a total of `3,020 and that is a net gain of `1,020 (over the `2,000 they were earning from the city alone). In other words, the maximum net value that the lake can generate is `1,020 and this will happen when exactly six people fish in the lake. However, if we assume that all fishermen are equally adept (i.e., they all catch the same amount of fish or the average haul) then more than six workers will want to be fishing. The only thing they are interested in (an assumption economists always makes) is their own earnings or average net income (= B/A – 100) and as long as this is positive they will have an incentive to not work in the city but go fishing. This net surplus per fisherman goes to zero when there

28  Law and Economics

are 10 fishermen. The total then earned is 10 * `100 from the city and another `1,000 from the sale of fish by 10 fishermen (column B against the row with 10 fishermen) and the total earned is `2,000, the same as when everyone works in the city! If the lake is not owned by anyone, people will enter into fishing as long as their income from fishing (average income) is greater than what they may earn by working outside. However, if the lake is owned by someone and this owner wants to maximize the returns from the lake, she will hire only six fishermen at the going wage rate (in the city) of `100. This would maximize the economic rent from owning the lake. Observe that this also maximizes the total surplus from the two activities. In other words, if we assume that everyone looks after one’s own interest, allowing someone to own the lake is consistent with maximizing the surplus. This is what the recognition of private property allows us to do. Private property rights allow owners to decide what should be done on the property (whether it should be used for fishing or boat rides for tourists) and with what intensity (how many fishermen or how many tourist boats at any point in time). We as economists expect these decisions to be taken in a way that will maximize the value of the lake. The obvious question that comes up is why private ownership is the way to do things. For example, in our situation, we know that six is the optimal number of fishermen that should fish on the lake. So, why do we not pass a law that says so? Why allow private ownership and then expect the owner to take in only six fishermen? There are two reasons for this. The first reason is consistent with our assumption about rationality (as discussed in Chapter 1). The owner being rational will take decisions that maximize her returns, provided she can claim those returns for herself. Private property rights guarantee that the owner of the property will extract the value generated from the property. The second reason is associated with the complexities of the real world. Different lakes will be of different sizes, have different fish populations and be at different distances from the fish market, all of which means that they will generate different revenue and cost figures. If we are to make a law or a policy regarding how many fishermen must be allowed to fish in the lake, whoever is making the law will have to know the facts and figures for every lake and make a different policy for each lake. This makes things very complicated. Instead, if we know that the lake owner has the right incentives to maximize the returns from the lake, then we need do nothing but enforce those incentives. Recognizing private ownership is such an enforcement of incentives to maximize the asset value.



Property Rights  29

Of course, this raises a problem. In our example, the owner gets to keep the entire net surplus of `1,020 that is generated from the lake. As we discussed in Chapter 1, societies may have strong preferences on how surpluses are distributed. While each of the 20 people makes `100 from working, the owner of the lake makes `1,020 from (owning and) managing the lake. There are various ways the surplus can be divided. One obvious way is to tax the owner of the lake and distribute the money collected to the city workers and the fishermen. However, it is also important to keep in mind that the rule for distributing surplus could have an effect on the total surplus produced. Let us take a simple example. A chef has a recipe and the skill to produce a great dish. However, the chef does not have the money to buy the ingredients for making the dish (a common problem for students staying away from home when only one, or a few, of them know how to cook). It would be good for both, the person who wants to eat what the chef cooks and the chef, if they got together. The surplus generated is the extra pleasure from eating good food compared to what is given in the student canteen or mess. The student with the money can buy the ingredients, the chef can cook and then the chef and the moneybag can jointly enjoy the dish. The question that comes up is how the dish is to be divided up between the two once it has been prepared. Suppose the distribution rule is that the chef decides who gets how much. The student with the money is immediately sceptical and quite rightly so—it is in the interest of the chef to cook well and then take the major portion for herself and leave a small portion for the student. With such a distribution rule, the student will not want to give the money to the chef and the dish will not be prepared and the surplus will not be realized. The same thing will happen if it is the student who decides who gets how much of the prepared dish. Here, the distribution rule affects the production of the surplus. There are two possible ways to solve our problem of distributing the surplus from the lake. One is property tax. Suppose a property tax Tp is introduced. The owner of the lake will have to pay this amount every year. If S is the surplus generated from the lake, the owner will no longer be earning S each year but (S –Tp). It is immediate that the decision to maximize S is the same as the decision to maximize (S –Tp), since Tp is not affected by how many fishermen are chosen to fish on the lake. If this tax, once collected, is distributed among the fishermen and city workers, then the lake owner gets (S –Tp) and each worker, whether fishing or working in the city, gets T 100 + ( P ). Higher the Tp, greater is the amount distributed away from the 20 owner to the fishermen and city workers.

30  Law and Economics

Alternatively, one could think of a tax on the surplus (or income of the lake owner). Suppose this is a proportional tax, ts,0 < ts < 1; that is, a proportion ts of the income must be handed over to the tax collecting authority. (If the tax rate is 30 per cent, then ts = 0.3.) The take-home (after tax) income of the lake owner is (1 – ts)S and that of each worker or fisherman is 100 + (tsS)/20. Again, observe that the maximization of S is independent of ts and so, one gets the same maximum surplus but different distributions of this surplus depending on the value of ts. In other words, ownership of valuable property allows maximization of this value while there exist tax policies for distributing the surplus. Note, however, that we have not established that all properties should be owned by individuals. We will discuss the situations where state or community can own the property more usefully than individuals in a later section. The importance of private property rights is that it provides the incentive to maximize individual benefits, and also provides appropriate feedback to learn and adapt according to the changing realities.1 Investments in property will not be rewarding if the property rights are ill defined, restricted or not clear and can be forcefully acquired or taken away by somebody else.2

How Do Property Rights Come to Exist? We have seen that there are gains from property rights. But there are also costs of having such rights. What are the costs of protecting such rights? There are institutional costs in measuring, recording and registering the properties owned by individuals and other legal entities. There is cost in seeing that one person’s property is not taken over or used involuntarily by another. This may require investments by the owners, like building boundary walls or

See Anderson and Hill (1975) for a discussion on the evolution of property rights. There are a number of studies that document the gains of such clearly defined property rights. In an interesting study on land, the effect of property rights on investments in the housing market was studied at a location in California. In the 19th century, land was allocated in this area in alternating blocks to the Agua Caliente tribes of native American-Indians and other non-native population. However, there were several restrictions on (sale/mortgage/leasing) the property right given to native people. It was shown that these restrictions led to poor development of housing market/investments in their plots, compared to the ones given to the non-native population who were not subjected to similar restrictions (Akee 2009). If we consider newer properties like wireless spectrum, it is documented that property rights increase efficiency and lower retail prices (ultimately benefiting consumers) (Hazlett 2008). 1 2



Property Rights  31

fences. It was noted that the invention of ‘barbed wire’ has reduced the cost of such protection, and this has encouraged the emergence of property rights (Anderson and Deal 2001). Investments in legal and judicial mechanisms are also needed to take action against those violating property rights. It is the careful balancing of these costs and benefits, that is, the benefits being greater than the costs, which has led individuals or societies to define what can be protected as property rights and how to protect them.3 Thus, when the gains from the use of a natural resource increase or when the cost of open access use is higher than the social costs of maintaining property rights, there will be social demand for such rights, and these are likely to be granted by the state. As a result, some resources that initially may be treated as common property may be allocated with ownership rights as and when they become scarcer or economically important and/or when the cost of maintaining (and enforcing) boundaries become cheaper. Consequently, there are mechanisms currently to allocate space orbits for satellites and spectrum of frequencies for wireless communication. Or, when the congestion by the fishing boats in some parts of the seas is perceived as very costly, there will be informal or formal arrangements to create some form of property rights. The United Nations Convention on the Law of the Sea, which demarcates the exclusive economic zone of each coastal country, was framed due to increasing conflicts over fishing in, and other uses of, the Atlantic Ocean. Different people may have different valuations on the same natural resource. Thus, tribal societies who use forests may have one valuation, and because of this they may continue to prefer common property rights. On the other hand, those who value the same land for the minerals lying below may prefer private property rights. Changes in resource quality or other factors that may increase the relative price of a resource can induce changes in property rights (sometimes from ‘no property rights’ to some form of rights including private property [Kaffine 2010]).

Economic Need to Protect Property Rights Legally Let us consider the following example. A developer or builder, B, wants to build a multi-storeyed complex, MSB. The soil in this site has high clay content and, hence, any tall building on this site requires a deep foundation. The builder needs to excavate down to a depth of 100 metres to get to the 3

For a theoretical argument on the emergence of property rights, see Demsetz (1967).

32  Law and Economics

rocky surface available at this depth. However, this excavation, without adequate protection of the sides, can lead to some caving in of the surrounding land. This could create some cracks in the houses located on the adjacent plots. There are five adjacent houses (with owners H1–H5) as shown in Figure 2.1. The value of each house is `h million. Should cracks develop in a house, an amount of `r million will be needed to repair the cracks and bring the house back to its original value. Alternatively, the builder can construct a protective wall during the excavation that prevents the soil around MSB from shifting and, thereby, also prevent the houses from developing any cracks. The cost of this wall is c. B expects a value of b once MSB is constructed. Suppose these houses and MSB were not adjacent to each other and, hence, MSB had no effect on the houses. Then, the total value of the assets we are considering is 5h + b. But, the way the lands are situated, building MSB will have an impact on the value of the houses and, hence, the total value of these six pieces of land. There are two possibilities for the total value: (a) if B puts up the protective wall at the excavation, it is 5h + (b – c) while (b) if B does not put up the wall, it is 5(h – r) + b. As discussed in Chapter 1, economists are interested in maximizing the total surplus and so we want to ensure that the six entities (the house owners, H1–H5, and the builder, B) generate the maximum possible. Let us assume that the values in rupees of the various parameters are the following: h = 5 million, r = 0.5 million, b = 100 million and c = 3 million. Then we have the following amounts: 5h + (b – c) million = 5 * 5 + (100 – 3) million = 25 + 97 million = 122 million 5(h – r) + b million = 5(5 – 0.5) + 100 million = 22.5 + 100 million = 122.5 million

Main Road

H1

H5

MSB

Narrow Lane H2 Figure 2.1. A Hypothetical Illustration

H3

H4



Property Rights  33

In general, what we want to check is whether or not 5h + (b – c) > 5(h – r) + b, that is, whether or not 5r > c. In our numerical example, 5r = 2.5 and c = 3. The value of the total surplus is maximized when the cost of maintaining the house values is minimized and this happens when the cracks in each house are repaired, not when the cracks are prevented from happening. Observe the first problem in this setting. If we say that private property is enough to attain the maximization of value, we will be making a mistake. In our example, each homeowner has the right to her property and the builder has the right to build its property on its land. But the maximization of value by the builder impinges on the homeowners’ maximization of their values. We have already mentioned such possible conflicts in Chapter 1 (listening to loud music in one’s home versus the neighbour’s right to silent meditation). This is a recurring theme in economics and in law—realizing one’s objective, or exercising one’s rights, interfering with another’s ability to satisfy the same. It is, therefore, essential in law to clearly draw the boundaries of rights, that is, what one can do for oneself, and duties, that is, what one must do to recognize others’ rights. In economics, we look at these as constraints to one’s maximization problem. For instance, a speed limit constrains a person from exercising her right to breeze through from place A to place B in consideration of someone else’s right not to be hurt in an accident. The utility of going quickly from one place to another is constrained by having to follow the speed limit. Indeed, private property is a bundle of rights and duties. For instance, a person has the freedom to use her property in the way she wants or give it to whomever she wants. At the same time she has the duty not to use it in certain ways; a plot of land owned by you cannot be used to grow marijuana or a car owned by you cannot be used to knock down lampposts! This leads us to the second problem in our setting—how to protect the rights. Having observed that B’s activity affects H1–H5, one has to determine whether B can build without considering this or must consider this. Suppose we start with a situation where the law is not clear and B is worried that once it starts building, and cracks develop in their houses, H1–H5 may file a suit in court. How will the court decide? It has two obvious choices: (a) allow B to build and tell H1–H5 to mend their own cracks or (b) tell B that it must ensure that its building activity does not cause cracks in the adjoining houses. Under (a), the total surplus is `122.5 million. What happens under (b)? If B puts up the wall, the total value is `122 million, less than that under (a). B knows this. Being rational, and knowing that the homeowners are also rational, it can approach them and say that it will finance their repairs when the cracks develop. Then, instead of `3 million he will spend only

34  Law and Economics

`2.5 million and save half a million. The only difference between the two solutions is who spends how much. In (a), each homeowner spends half a million and B spends nothing; in (b), B spends 2.5 million and no homeowner spends anything. So, either way, economists are satisfied that the total surplus is being maximized. Suppose we were to change the parameter values a bit. The cost of the protective wall, c, is still `3 million, but the cost of repairing the cracks, r, is now `0.8 million. In this case, it is better to build the protective wall rather than not build it and repair the cracks in each house. If the court orders (a) as above, it is in the homeowners’ interest to get together and convince B to build the wall for which they will pay. If the court orders (b) instead, then B will build the wall. In either case, the efficient solution will once again be realized, independent of how property rights are defined by the court. This is an illustration of a very important result in law and economics. If both the parties can negotiate without any hassle or cost, then an efficient solution is achieved, whatever be the legal right or the prevailing understanding of that right (Coase 1960). It is important to appreciate what this negotiation hassle, or transaction cost as economists term it, is. Suppose we are back in the original example where the cost of repairing cracks is `0.5 million. In (a), no one needs to talk to anybody, the law states what B is allowed and tells what each homeowner will have to do if she wants to recover her original house value. In option (b), the builder has to convince everyone to accept its offer—‘Let the cracks develop and I will repair them.’ If anyone declines, then the builder has to build the protective wall. Getting everybody to agree is a time-consuming affair and requires negotiating skill and effort. In our example so far we have ignored this. This cost will be much higher if, say, one homeowner was away on a longish trek in the mountains with no mobile phone connectivity, forcing the builder to wait for her to come back. In general, there are many ways in which transaction cost arises. Here are a few: 1. If the situation is such that B and H1–H5 do not talk to each other and they cannot discuss and negotiate, then transaction costs are prohibitive. Different homeowners have different and conflicting perceptions of what is a fair right or distribution, preventing them from entering into a smooth negotiation. 2. Let us assume that there are many homeowners in the neighbourhood who are affected by the construction of MSB and suppose that the repair costs are such that it is better to build the wall and the court



Property Rights  35

has decided that homeowners must repair their cracks. If some owners take up the issue with the builder, it may lead to the construction of the protective walls around the foundation trench. Then all the other homeowners, including those who have not participated in the discussion, benefit. However, when it comes to paying their part of the cost of the wall to B, they may say that they were not party to the negotiations. In other words, any negotiation will have to involve every player. 3. An even more serious constraint is the enforcement of the agreement. What if a homeowner, after taking money from B to repair the cracks, objects to the foundation being dug by B? Alternatively, what if B pulls out of the agreement after B has agreed to repair the cracks in the houses? If the cost of enforcement is too high, it may discourage the parties from entering into any negotiation. Let us revisit what we have done so far, but now we take transaction cost into account. Consider the original example where r was equal to `0.5 million. Suppose the cost of any negotiation, N, between the group H1–H5 and B is `0.6 million (N = 0.6 million). If the court feels that homeowners should make their own arrangements to ensure there are no cracks, that is, repair them when they happen, the total cost is `2.5 million (5 * 0.5 million). If the court says that it is the responsibility of B to ensure that the homeowners do not object to the construction of MSB, then B’s only option is to build the wall. B knows that repairing the cracks as and when they appear is cheaper, but to get the homeowners to agree, B has to spend an additional `0.6 million as transaction, or negotiation, cost. His total cost would then be `3.1 million (5 * `0.5 million + `0.6 million). Building the wall by himself would be cheaper at `3 million. So with the transaction cost, if the court holds B responsible for ensuring that adjoining houses are not damaged, we will not be able to generate the efficient solution where cracks in the homes are allowed to occur and then repaired. In this case, efficiency is guaranteed if the court says that B does not have to prevent cracks in nearby homes.

What Does Legal Right Mean? The legal right makes clear what it is that each person can do with her (varied) property and what others cannot do with regard to this property. Even this

36  Law and Economics

clarity itself is likely to reduce the negotiation or transaction costs. When a dispute occurs, each person has some idea of what she is or is not entitled to. Definition of property rights provides that information. But definition alone is not sufficient. This requires enforcement too. If X violates the property right of Y, the cost of violation internalized or perceived by X depends on how good its enforcement is. If it is very weak, then the situation is as good as ‘no property right’. On the other hand, the perception of reasonable enforcement too makes the negotiation easier (and cheaper). Each person will then have a reasonable understanding of what she ‘gets’ if she violates the other person’s right, or what is the remedy one can get if someone else violates her right. These make bargaining positions clear. It is not only that each party knows her own potential gain or loss, but also the other person’s gains from the law. Such information on mutual gains facilitates negotiation. We have seen that when the negotiation or transaction costs are high, legal right matters, and such a right is efficient when it is assigned in a way that the net surplus is maximized. This may vary from case to case. It requires the court or the authority assigning the legal right to know who can maximize the surplus. However, as we have discussed earlier, it is difficult to grant a legal right on a case-by-case basis. Thus, there is a trade-off here. If the cost of acquiring information is not so high compared to transaction costs, the assigning of a legal right based on efficiency consideration is fine. On the other hand, if the cost of such information is too high, the acceptable procedure is to follow the precedent, with the understanding that the precedent would further reduce the transaction costs.

What Are the Legal Remedies Available? When one person violates others’ right, such as the builder in the earlier example, creating cracks in house and damages being incurred by the house owner, what are the likely remedies available to the victim (the person whose right is violated) if she approaches the courts? There are two types of remedies. As a remedy, the courts can give an injunction. This tells that the person who violates the property to stop his action and that he cannot continue with the violation. This is effective when the violation is yet to create any damage to the victim, since such injunction does not say anything about the



Property Rights  37

damage already suffered. But the injunction stops the intention to violate or further violation, if it has already occurred. In one way, the injunction can be interpreted as a ‘mechanical’ application of property right. Whatever the property right in a given context, it is enforced by the courts. If it is so, one cannot ask the efficiency question about the courts’ decision. Then the efficiency issue is with regard to the property right itself. But we have seen already that the efficiency of the legal right (which arises only in situations where negotiation or transaction costs are high) depends on who can take remedial action cheaply. This varies from one context to another. Thus, mechanical interpretation of injunction can lead to inefficiency under certain situations. On the other hand, the injunction can also be interpreted in a ‘creative’ manner. It tells the parties to sort out the matter. It makes the property right clear. It allows the parties to negotiate and arrive at an agreement. Such an agreement would lead to efficiency whatever the legal right. Thus, injunction is efficient in situations that lead to negotiated agreement. However, in this case the transaction costs should not preclude negotiation or lead to inefficient costs. When the plaintiff and defendant can negotiate and the transaction costs are low, an injunction may encourage them to negotiate and this would lead to an efficient outcome. Thus, two people located nearby (and not enemies of each other) are likely to negotiate. They could also monitor each other’s actions relatively cheaply (whether the side wall is being constructed or not as in the example). This would also reduce transaction costs. Hence, the injunction is a useful remedy in private disputes between two parties. On the other hand, higher transaction costs or reasons that prevent parties from negotiating may lead to inefficient outcomes. One interesting option is for the courts to award damages to be paid by the defendant to the plaintiff. This is useful, if the violation of property right has already caused damage to the victim. How do we interpret this in our example? Suppose B knows that transaction costs are high and so he cannot make the homeowners agree. However, if he knows that the court will ask him to award damages to the home owners should cracks appear, it is in his best interest not to build the protective wall and wait for the homeowners to complain. If and when they do so, the court will ask him to repair the damages which he will, at a cost of `2.5 million, less by half a million from the cost of building the protective wall. So, even in the presence of transaction cost, if the courts follow the principle of awarding damages, we will get the efficient solution.

38  Law and Economics

Registering Property Rights With respect to properties like land, all transactions (like X buying it from Y ) are registered. There is also some verification of past deeds and documents at the time that every new deed is registered. This reduces any uncertainty regarding the ownership of land. Thus, it reduces the possibility of somebody (like a tenant) posing as the owner and making sale deeds. However, we do not register all such deeds. When we go to a shop selling used equipment, we undertake transactions in good faith. We do not insist on knowing whether the good in question is legally owned by the shop owner or whether he has acquired it from a thief who has stolen it from the legal owner. Thus, all transactions of properties are not legally registered or we do not bother to verify the legal ownership status of all properties. This is due to the following rationale. Though registration reduces uncertainty, registration is costly. Thus, there is a trade-off, between being certain about the owner and the cost of being certain. Property registration is practised only in the case of those properties which are highly valuable (compared to the cost of registration) like land or cars but not for televisions or refrigerators.

Owning Property without Title There are many reasons why some people may be occupying a property (like land) without formal titles, especially in countries like India. Land and other property could be inherited from previous generations, without formalization of transfers. There are tribal societies who have used forest land for shifting cultivation, and have settled over a period of time in some of these lands, and they may not have taken the effort to acquire formal titles. Moreover, in a country like India, property rights over vast stretches of land were established in different ways. It could be that kings or chieftains brought in vast areas of land under their occupation through attacks on other kingdoms, and then they may have distributed this land to some people with or without formal titles. Thus, many people who occupy specific pieces of land may not have legal title. This is also very common amongst those who live in slums in urban areas and some of the world’s largest slums are in India. There is also a related issue. If a person uses a property for a fairly long time (at least 12 years) she may claim ownership of that property. This issue



Property Rights  39

seems to be the concern for those who wish to rent or lease out a property for a long-term period. Thus, those who lease or rent out may prefer short period contracts. The legal regime dealing with such issues is called adverse possession. The original idea of adverse possession goes back to our basic economic thinking that if asset values are to be maximized, they should be in the possession, and control, of those who can realize this maximum value. In general, if I value your asset more than you do, I will do so only because I can use it in a way that generates more value than you can. Ideally, I should be able to buy the asset off you. The minimum (or reservation) price at which you will sell it to me is the maximum value that you can realize from it. This could be obtained by using the way you want it (this could include an offer by someone else to buy the asset). The maximum (or reservation) price I am willing to pay is the value that I can generate from this asset. If my reservation (maximum) buying price is greater than your reservation (minimum) selling price, then I could pay you more than your price (and less than my price) and take control of (i.e., own) your asset. This is what an asset market does. However, often such markets do not work the way they are supposed to. So, suppose for whatever reason, you refuse to sell it to me. Also, you keep the asset unutilized and society loses the value that it can generate. Seeing this, I may ignore the fact that you own the asset and start using it the way I think it should be used and generate more value than you are generating by keeping it unused. If I do this for a sufficient length of time, then adverse possession says that the asset belongs to me. However, the concept of adverse possession is deeper than this simple exposition makes it out to be. Indeed, of late, the Supreme Court of India has come down hard on the wording, interpretation and enforcement of this law. In general, they want to ensure that dishonest people do not encroach and occupy land belonging to others through this method. In fact, an important element of adverse possession is that it happens with the knowledge of the true owner. Second, the possession must be ‘adverse’ to the true owner; if I am using your property with your consent, I cannot be in ‘adverse possession’ of your property. Third, it must be held continuously by me for at least 12 years and this period begins when the adverse possession starts. Thus, in principle, if I have stayed as a tenant on your property for more than 12 years or have been fighting in court against you for more than 12 years, I cannot claim adverse possession.

40  Law and Economics

What Properties Cannot Be Sold? In many countries, there are restrictions on the sale of kidneys, blood or other body organs. Strictly speaking these ‘parts’ are owned by the persons. Why do we then object to their sale? This could be partly driven by moral considerations. A woman agreeing to have sexual intercourse with someone on payment is seen as selling her ‘body’ and this is frowned upon in many social contexts. The restrictions on trade in organs like kidneys can be costly for the society. People who need such organs badly do not get them even when people are willing to sell. Thus, mutually beneficial trade is not taking place, and this has an impact on social welfare. In countries like ours, there is an important argument for taking a negative view of such trade. When the general law and order and institutional framework sustaining trade is ill functioning, there is no assurance that some people are not involuntarily forced to participate in trade. For example, the negative attitude to prostitution in many parts of the world is also due to the fact that allowing it may encourage illegal groups to forcibly bring underage girls into the sex market. Similarly, opening up organ trade may induce mafia groups to force or even kill people to facilitate such trade. The possibilities of such extreme exploitation is also an important reason for the negative attitude towards the trade of body parts. How do we treat blood donations? Why is donating blood, or body organs, encouraged while their commercial sale is restricted? One hopes that donation of blood, or organs, is out of concern for the patient rather than for commercial interest. Concern for the recipient is an important element in donations. For example, if the potential donor has had a recent bout of hepatitis, she will not donate her blood, as it could be harmful for the recipient. If, on the other hand, she was selling her blood, her interest would be in the price she gets for it and not the recipient’s recovery. This would encourage her to hide, if she could, her recent incidence of hepatitis. So, if the transfer of blood from one to another is for the welfare of the recipient, it is better to transfer from those who have the recipient’s welfare in mind. A market process may actually be harmful here. This is a classic case in economics, that of ‘market failure due to asymmetric information’. The one who is transferring blood has more information about her blood than the recipient or the doctors overseeing the transfer. The cost of detecting wilful non-disclosure of information is very expensive, though detection is not impossible.



Property Rights  41

Economists have another argument for why trade in body organs may be detrimental to society. Consider a situation where commerce in the transfer of blood is not allowed. Good people may come forward and donate blood because they feel good helping others and, more importantly, they feel good that others think them to be good. (This is the main reason why many people have charities in their own name, so that everyone knows whose money is being given away.) However, when blood can also be sold, one never knows if the blood was transferred by a donor or a seller. A donor has to be able to prove that she did not take money for her blood. While one may like being known as a donor, one may also be concerned that others think of her as a seller. Consequently, blood donations may dry up when blood sales are allowed!

3

Intellectual Property Rights In the previous chapter we discussed rights over tangible properties such as land, house and machinery. There are intangible assets too. Ideas, inventions, industrial designs and artistic expressions are intangible properties. What is the need for legal rights over such intellectual property?

Need for Intellectual Property Rights The cost of production of such property can be high for the producer but the marginal cost of copying it can be very low. Thus, the first buyer can resell it cheaply at the cost of copying. In the case of a piece of software, the copying cost can be near zero if it can be downloaded from the Internet or if the software can be copied at the cost of a disk. This could mean that the intellectual product can be used by many (millions of users) without giving any reward to the creator of the product. If someone produces a novel or a painting, this can be reproduced through photocopying, and sold at the copying cost. Thus, the novelist or the painter will not get any reward. This can be true in the case of most inventions. If the ‘tricks’ can be learned and copied, it can be put to use without paying any money to the inventor, if there are no intellectual property rights. This may discourage people from developing inventions, artistic expressions or ideas that can be used to improve existing goods and services or to produce newer ones. How did societies solve this problem traditionally? It could be that the creators of such ideas were paid by other means. Kings used to have literati working in their courts, and they received rewards, although not necessarily based on the products that they created. There were centres of learning that supported scientists and innovators and here too the rewards were not directly linked to the innovation. The other form of protection for intellectual property, widely practised in countries like India, was secrecy. Ayurveda



Intellectual Property Rights  43

practitioners, for a long time, and sometimes even now, kept the knowledge of specific herbs and medicines for specific diseases as family secrets (transmitted only to close relatives), and such secrecy help them receive the rents of their intellectual property. In the modern world too, several intellectual properties are produced and protected through this means. Many companies use trade secrets to protect their intellectual properties. Sometimes technology helps such protection. There can be codes in software that may prevent its free downloading and pirated use. Even when intellectual property rights are available, it is not uncommon to see firms reluctant to use this mechanism due to the possible danger of their proprietary information becoming available in the public domain as part of the process of patent application, which they fear will encourage imitation by other firms, using the loopholes of patent statutes, or firms located in countries that do not have a strong patent regime. It is also noted (Baker and Mezzetti 2005) that sometimes firms have incentives to voluntarily provide information about their inventive activities to patent controllers, since patents are given for inventions beyond what already exists or prior art, and the voluntary provision of information adds to this prior art, which would make it difficult for competing firms to claim patents on the results of inventing activities going on in the information-providing firm at the time of disclosure. However, here we discuss the basic economics of different forms of intellectual property. These include patents, copyrights, trademarks and confidentiality clauses that help businesses protect certain secrets. Patents are awarded to inventions, whereas copyrights are awarded to artistic expressions.

Patent A typical patent gives the inventor the sole right to use the invention or to license it to other users. It is a ‘property’ because it comes with many of the trappings of what we commonly understand as private property. Just as one can use one’s property the way one wants, when one is granted an intellectual property right over some creation, one can use that creation the way the rightholder likes. A house owner can live in the house or rent it out to others to live in it; similarly, when one is granted a patent in an invention, one can use it exclusively (i.e., not let others use it, in the same way that when you live in your house others cannot come and party in it without your permission)

44  Law and Economics

or lease or license the invention to other users (like renting your house). However, there is a problem with treating intellectual property the same way as one would treat other private property, like a house. In the previous chapter we showed how private property rights lead to maximization of asset value and, hence, to social welfare with a market-based distribution rule. We will have to make a similar argument when it comes to patents, and, this is where the trouble starts! The granting of a patent for an idea means that no one else can use the idea to make commercial gains without asking the permission of the patent holder. This gives the patent holder exclusive rights, or a monopoly, to use the idea. According to basic economics, monopoly hampers efficiency. Let us see how it does so. First, efficiency demands that price equals marginal cost. Price is a way of measuring how much society is willing to pay for each unit of the product and marginal cost is the cost of producing an additional unit for society to consume. So, if society is willing to pay for the additional unit more than what it costs, then additional value is generated by producing that extra unit. Alternatively, if the maximum that society is willing to pay is less than the cost of producing the additional unit, society is better off not producing it. Thus, society maximizes its value when the willingness to pay for the additional unit (its price) is equal to its marginal cost. In general, though, producers maximize profit when the additional revenue they get by selling the last unit is equal to the cost of producing the additional unit. For our purposes, we want the willingness to pay, price, to be equal to the marginal cost. That will happen on its own if the price is equal to the marginal revenue of the producers. In a competitive world, where many producers sell a homogeneous product, price does equal marginal revenue. This is because, each producer is small and they cannot affect the price in the market. So, when they produce the additional unit, the ruling price is exactly what they get. However, in a monopoly, the lone producer faces the market demand curve, which is downward sloping (to sell more, price has to be lowered) and hence, it has to allow for the fact that when it sells more its additional revenue will be lower than the ruling price (or, where the marginal revenue is equal to the marginal cost the price, by definition, is greater than marginal revenue). This causes the inefficiency in a monopoly. Figure 3.1 illustrates what we said in the previous paragraph. The aggregate demand curve is the maximum amount society will buy at each price (when we read the curve horizontally); alternately, it indicates the maximum price society will pay for specific amounts of the product (if we read the curve vertically). A monopolist faces this demand curve for every unit that the society buys from this one firm. The price of a product is the cost to the



Intellectual Property Rights  45 Price in rupees Marginal cost Demand or average revenue for a competitive firm

Pm Pc

Aggregate demand or average revenue for the monopolist Marginal revenue for monopolist Qm

Qc

Units of quantity

Figure 3.1. Equilibrium Price and Quantity: Monopoly and Perfect Competition Source: Authors.

consumer of a unit of the product; thus, if 100 units are sold at a price equal to 2, then the average revenue per unit sold is also equal to the price. Hence, the aggregate demand curve is the average revenue curve for the monopolist. The marginal revenue curve corresponding to this average revenue curve lies below the demand curve. Since the average revenue curve is downward sloping, its marginal curve has to be below the average revenue curve. (If the average pocket money in a class of 20 is `100, then when the 21st student (the marginal student) enters the class, the average holding of pocket money in the class will fall only if this additional student comes in with less than `100 in her pocket.) The quantity at which the monopolist maximizes its profit is Qm, where its marginal revenue is equal to its marginal cost. For any quantity less than Qm, or to the left of Qm, the addition to (marginal) revenue is greater than the addition to (marginal) cost, and it pays the monopolist to increase output; to the right of Qm, it similarly pays the monopolist to reduce output since marginal cost is greater than marginal revenue. Given this quantity, the maximum price the monopolist can charge is P m, and it will charge this maximum price to maximize its profit. In a competitive setting, where there are many sellers, the marginal cost is the cost to the industry rather than any one firm. Each firm takes the price as given and each has a small (negligible) impact on the total sales. So, for them, selling additional units has a negligible impact on price, small enough for them to think that approximately they have no impact.

46  Law and Economics

In other words, each small firm faces a demand curve that is flat. A flat demand curve implies no change in average revenue; so then, price equals average revenue equals marginal revenue (see the pocket money example in the previous paragraph). Once again, for profit maximization by each firm, marginal cost must equal marginal revenue. Since marginal revenue equals price for a competitive firm, we have price equals to marginal revenue and the equilibrium output in a competitive market is Qc > Qm, with P c < P m. For the monopoly equilibrium, society is willing to pay more than it costs for an additional unit (P m is greater than marginal cost); for a competitive equilibrium, society is paying exactly what it costs (P c is equal to marginal cost). So, society is maximizing its surplus in the competitive equilibrium but is below the optimum amount of production when there is monopoly. Hence, monopoly is inefficient. The obvious question then is: why do we create a monopoly by granting patents? The standard argument is that people produce goods and services to earn revenue; if this is greater than the cost then it makes sense for them to do so. Similarly, people will produce an idea only when it is profitable to do so. In the production of goods, each time it is produced, and by whomever it is produced, there are costs associated with it; however, once an idea is produced the cost of reproducing the idea is miniscule or zero. So, if the producer of the idea, after a lot of personal sweat, was planning to use it commercially, she may be thwarted in her attempt because others will copy her and drive down the profits from the idea to zero. She, therefore, feels that it is not worth putting in the sweat to produce ideas. Since ideas are valuable, this will harm society because potentially valuable ideas will not be produced. What patents allow her to do is protect her idea and prevent others from copying it without taking her permission (Cooter and Ulen 2004). The patent trades off two important economic aspects—the efficiency cost of a monopoly with the incentive to generate ideas resulting from monopoly rents. To get a patent, it is important that the creator of the idea do two things: first, she must be able to show how it can be commercially used and second, she must completely disclose the idea or, put it entirely in the public domain. The first ensures that people do not rush to patent anything they think of and clog up the patent offices with things that will not help anyone in any way. The second ensures that whenever the idea is helpful towards developing other novelties, people can do so because there is full disclosure of the earlier idea. There are two crucial issues in patents. The first one is the duration of patents. A typical period, as also provided in Indian patent laws, is 20 years. If the period is too small, the reward that the inventor gets is too low, since



Intellectual Property Rights  47

copying can take place early. Thus, the incentive for invention is lower. On the other hand, if the period is too long then the monopoly costs are too large. Longer durations would imply that widespread use of this invention by others in society could take place only after a longer period. The second issue is how similar or different inventions should be for granting a patent. When someone files a patent application, the information on invention becomes publicly available (but legally others cannot use this information). Others can file objections to grant of the patent, saying that the additional innovative element in the invention (for which the patent is sought) is very limited and it is more or less similar to the one that is already patented or existing. This incremental nature of the invention is crucial in patenting. This can also be called the breadth of the patent. A broad patenting rule means that small incremental inventions cannot be patented. On the other hand, these incremental inventions can be patented with a narrow patenting regime. Broad patenting encourages fundamental and basic research, whereas narrow patenting encourages the development of applications. There can also be different durations for different inventions or multiple durations of patents (with pioneering inventions getting full term and minor inventions getting a much shorter period), as practised in Germany and to some extent in India. Minor inventions are granted ‘petty patents’ in some countries. There can also be an increasing annual fee to be paid by patent holders. The increasing annual fee may encourage the inventors to allow the patent to lapse after some years, if their benefits from holding the patent decrease. The patenting rule in a society can be tuned to the specific requirements of that society. In a country where more fundamental research is needed, broad patenting can be encouraged, while a narrow patent regime is more appropriate in a context where more applications have to be encouraged. The principle is that ‘if the social value of investments in fundamental research exceeds the social value of developing applications, then patents should be broadened’ (Cooter and Ulen 2004: 120–21). This is less likely to be the case in developing countries like India, where the social value of developing applications is likely to be high. Thus, there can be a case for narrower patents for countries like India. However, the more basic invention and applications based on them together are joint products, so how firms carrying out basic research and those working on applications share benefits can be determined by the organizational relationships between firms, which is in turn determined by the transaction costs. If transaction costs forbid exchange or contracting between firms who supply basic innovation and who supply application then patent law can attempt to facilitate this exchange. It is noted that

48  Law and Economics

such economic considerations rarely get reflected in the patent acts or court decisions in this regard in the United States (though there has been greater appreciation of economic arguments lately) (Cooter and Ulen 2004). Fundamental and basic research is usually encouraged through public financial support for research organizations and universities and not through patenting rules.1 In reality, there is not much difference in the patenting rules between different countries these days after the integration of world trade and the adoption of rules of international organizations such as the World Trade Organization (WTO). Earlier India allowed patenting only for the processes and not for products. The same product could be produced through different processes. This enabled Indian companies (mainly in the pharmaceutical sector) to reverse engineer internationally patented products and produce them through different processes. Thus, they could produce many pharmaceutical products cheaply, but this may have eroded the gains of international patent-holding firms. However, India too has moved towards a product-patenting regime recently.

Copyright Copyright protection is given expressions of ideas such as those in books and articles or art forms like fiction, poetry, paintings and so on. The main purpose is to ensure a reward to the creator of expressions of ideas or arts so as to serve as an incentive for such creation. In the absence of such rights, the creator’s work can be copied and sold at the expense of copying. What can be copyrighted? Consider the various stories or novels that one has read. There is an amazing similarity in the underlying themes of many of them. A hero and heroine in love struggling against many odds and finally living together, a woman emotionally torn apart between a husband and a lover, etc., appear as the underlying issue in several stories. Similarly, there are umpteen versions of ‘rags to riches’ stories in different cultures written at different points of time. This is because, unlike patents, copyright 1 It is noted that rewards (by the state) is a competing mechanism to patents and the main problem is the need for governments to collect information to calculate rewards (Shavell and Ypersele 2001) as well as the problems associated with likely government failures. Whether rewards can replace patents for all inventions is not clear yet, but which of these two mechanisms would be appropriate for a given type of invention (given the costs and benefits of patents versus rewards) is an important issue.



Intellectual Property Rights  49

protection is granted to the ‘expression’ of an idea or theme, and not to the idea itself. How does one differentiate her expression from that of others? Can one rewrite a story with different names for the characters and places? These are contestable issues but the basic idea is to ensure that copyright does not prevent people from conceptualizing the same idea in different expressive forms. But there should be reasonable differences between two expressions. The duration of copyright protection is longer compared to that of patents. It can be typically the lifetime of the creator plus a few more decades. What is the logic of having longer durations for copyrights than that for patents? Unlike patents, copyrights do not prevent (or such materials are not so relevant for) further development through the dissemination and spread of information. One is not building a new novel or painting by working further on a previous novel or painting. On the other hand, developing a new invention depends heavily on the information and knowledge embedded in previous inventions. The lack of access to them could be detrimental to the development of new inventions. A similar problem does not exist for copyrighted materials. Then why cannot these expressions (of ideas and art forms) be given copyright for infinite periods? There is a cost in giving such infinitely long protection. Each creator of new ideas and art forms has to ensure that she is not copying others’ expressions. There is a search cost. This search cost increases (probably at an increasing rate) as the protection period of copyright increases. Imagine the effort required for somebody, who writes a new novel or story, to ensure that similar expressions do not exist in different parts of the world during previous centuries. Thus, there is a social need to reduce the search cost. This is the rationale for limiting the duration of copyrights to a finite number of years after the death of the creator. Information technology poses serious challenges to the enforcement of copyrights. This is visible in software, music CDs, etc. (see Zentner 2006). Either the regulation has to change in a manner to cope with the changes in technology, or there must be technological solutions limiting access to only those who have license to use the product.

Trademark The purpose of the trademark is fundamentally different from that of patent and copyright. We have mentioned in the first chapter the kind of

50  Law and Economics

information problems that are faced in any exchange. Buyers are not in a position to get information on the real quality of the products. Moreover, sellers may not be in a position to credibly communicate, even if the quality of the product is really good. Why should buyers take it seriously if the sellers say that the product is of good quality (every seller has the incentive to say the same thing irrespective of the actual quality). It is to solve this problem that public- (government-) mandated standards are being used. There can also be mandatory information disclosure provisions (like ‘cigarette smoking is injurious to health’ or ‘mutual fund investments are subject to market risks’). However, many firms do not depend solely on such governmental standards and mandatory information disclosure. Their interest is in building reputation. Past satisfactory experience of their customers may lead to loyalty, and this may enhance the value of their ‘brand’. Thus, there may be a section of consumers who wish to use the product of a particular firm, due to brand value (which incorporates the past satisfactory experience of consumers and the firm’s reputation). Thus, ‘branding’ can be reckoned as a market-based instrument to solve the problem of asymmetric information between buyers and sellers. Such branding needs some legal support. This is because other (not-soreputed) firms may be interested in selling their products in the name of the reputed firm. Thus, they may use names for their products that are similar to the one used by a reputed company or they may name their company in a way that sounds very similar to a well-established company. If consumers are not discerning enough to identify the differences between the two companies, and the pretender sells a low quality product, it leads to a loss of business for the reputed firm. Trademark is necessary to avoid such loss. The firm with reputation wants to communicate to potential customers that a particular product or message is indeed from it, and not from others who are copying it. This is usually done with the help of a specific mark used in all its messages and product packs. This mark is registered with a public registry, and once registered it is a punishable offence to copy such registered trademarks. In economics, one cannot rest one’s case by showing how a particular law helps a company to build its reputation and make profits. After all, economists are supposed to see what maximizes social welfare and not simply the profit of a company! We have to carry the idea one step further and show that such profit making is good for all, producers and consumers. When a company is faced with the choice of producing a better-quality product versus a lowerquality product, it will choose the one that gives it greater profit. Similarly, consumers will be willing to pay more for a better-quality product only if it



Intellectual Property Rights  51

gives higher satisfaction. If consumers think that a higher-priced product is always a better-quality product, then all bad-quality producers will sell their product at the higher price! On the other hand, if consumers need to ‘open up’ a product to gauge its quality, they would not only have to put it back together again, they will also have to be experts at dismantling a product or pay somebody else to do so. This will make the cost of buying a product many times more than its price. If, however, a company has painstakingly built up a reputation of producing and selling high-quality products, consumers only need to know which product is sold by this company. Observe that consumers are willing to pay for this company’s product and the company is willingly producing this product. So, since both are willing, and everyone is rational, they must be doing what is best for both. Trademarks allow consumers to identify the products of a company and enable such markets to perform well. There are no issues of duration or breadth in the case of trademarks, and hence, there is no need to limit the validity period. One entity’s registration of trademark does not prevent others from registering their own trademarks or trying to acquire reputations of their own. It may be noted that trademarks are valuable only in the case of firms that have acquired a certain level of reputation in the markets. New firms may also acquire trademarks, but in the hope of achieving reputation in due course. There are restrictions on using a ‘generic name’ as a trademark. If somebody is selling mineral water using ‘mineral water’ as a trademark, it can be problematic. If such a trademark is registered, it may restrain competition in the sale of this generic commodity. However, it is also possible that some brand names may evolve into popular names for a commodity or service. It is not unusual to use the term ‘xerox’ing for photocopying, whereas Xerox is only one company selling photocopying machines.

Cases from India Consider this book. The two of us (authors) have produced a book for students to better understand economists’ views on how law can help in attaining goals that economists find desirable. However, there is nothing in this book that is really original, or new. Indeed, seldom is there something new in a text book. So, why cannot students copy this book by checking it out of the library and taking it to a photocopying shop? Let us state upfront that were it possible for students to do so, we would not have taken the

52  Law and Economics

trouble to write this book! Of course, some of you, reading this last sentence, may be tempted to say that the world would have been a better place had we not written the book. But, as economists, we believe in the freedom of people to try out various things; if their attempts are not fruitful, so be it. However, if they are fruitful, everyone gains, including those who tried. It is this latter possibility that encourages people to try out things and, hence, we wrote this book. Interestingly, there was a case in India on a similar issue. Eastern Book Company publishes law books. One of its publications, a journal called Supreme Court Cases, publishes all the judgments of the Supreme Court. However, these are annotated judgments and include cross-references to similar or related judgments, as well as summaries highlighting the major facts and arguments of the case. Obviously, they have law experts who prepare the additional notes on each case for the ease of their readers. The experts spend considerable amounts of time and expertise to prepare the notes and they get compensated for their effort. The company has been publishing the journal since 1969. However, they suddenly found that someone had developed a software product called Grand Jurix, which was a faithful copy of the publication. The issue before the court was whether Grand Jurix was infringing on the publisher’s copyright.2 The case is interesting because it is very similar to the example of copying this book given at the beginning of this section. Indeed, Grand Jurix entered the plea that all Supreme Court judgments are in the public domain and, hence, no one has any copyright over them. This is similar to the argument that a student photocopying this book might make; the authors have not said anything new and there are hundreds, may be even millions, of lawyers and economists who knew these things even before the book was written. Hence, it is not an original expression and cannot be copyrighted. Before we reveal the court’s judgment, let us explain how economists would argue this. The journal is making profits for the company (actually a team of entities was involved) and this is the reason why it is being regularly published. Although the court cases and judgments are in the public domain, people are buying this book. It must be because the annotations to each case add value to the publication. Such annotations require time and effort and the annotators expect to be compensated for this; otherwise, they would have gone and watched a cricket match. So, both the producers and the users of this book are benefited. If others were free to copy this, people would buy 2 Eastern Book Company & Ors. Petitioner vs D. B. Modak & Anr., Respondent, 2007(SC3)-GJX-1158-SC.



Intellectual Property Rights  53

the copies, which would be a lot cheaper. It is important to understand why the copies would be cheaper. They would be cheaper, not because of the new medium—computer software—but because the copiers were not paying for the expert annotators’ time. The original publishers would not make the profits they expected to make in the absence of cheap copies and, hence, the case insights developed by the experts will not be available, as they cannot be compensated any more. If they do not produce these insights, the publishers as well as Grand Jurix would not be able to produce something that was of obvious value to society. Hence, Grand Jurix is in violation of the publisher’s right not to be copied. The court’s final decision was the same as that of the economists. To emphasize the point being made here, let us highlight another recent incident. This appears in a book entitled Against Intellectual Monopoly by Boldrin and Levine (2008). The United States’ 9/11 Commission report is called Final Report of the National Commission on Terrorist Attacks Upon the United States and it was published in 2004. The report was published as a book by W. W. Norton at the same time as the government of the United States made the report public on its website, on 22 July 2004. The publisher was given the report as late as possible to prevent leaks of the material before it was officially released to the public. The book went on sale simultaneously with the release of the report on the government website. In other words, the publisher did not get the ‘copyright’ on the book; the government was releasing it free on its website even as the book was being sold at a price. How does this compare with the Indian publication of annotated judgments? Well, in the example from the United States, the publisher was only getting the right to be the first, not the exclusive, non-government entity to publish it. Indeed, the government itself was simultaneously publishing it free on its website and a cheaper (than the publisher’s price) downloaded and bound copy of the report was available from its government publication outlet. The publisher did not get any copyright because the publisher had not developed any of the content; that was prepared, as an ‘expression’ of the Commission members’ findings. So, there was no disincentive for the publisher not to generate the content, which was being generated anyway. The publisher made money on its superior ability to fast-track the publication and distribute the book to a wide audience as quickly as possible. And it made enough money to give all the 9/11 victims’ families a copy of the book and donate US$600,000, which was a portion of its profits from the sale of this book. If the Eastern Book Company team was simply republishing the Supreme Court’s judgments, then Grand Jurix would have been at liberty to publish

54  Law and Economics

the same through computer software. This is because the court’s judgments are not dependent on whether they are going to be published and who is going to publish them. However, the annotations to these cases would not have been available if the Eastern Book Company could not sell the annotated book for a profit. The focus in copyright is not on the act of publishing but on the creation of the book’s content or, expression. To drive home how economists think, let us consider a well-known recent case (see Basheer and Reddy 2008). The case was between Novartis AG and the Government of India and was filed in the Madras High Court.3 Novartis had applied for patent protection for its cancer drug Glivec that it had already patented in 40 other countries. The drug was based on the beta crystalline form of the salt imatinib mesylate. The Indian Patent Office rejected the patent application and the Madras High Court upheld that decision by referring to Section 3(d) of the Indian statute on patents. Section 3(d) disallows patents on minor improvements that do not have a significant impact on the efficacy of the drug. The purpose of this section is to prevent something called ‘evergreening’, whereby drug companies make slight modifications on an earlier patent when its period is running out. This prevents generic manufacturers from producing the drug as the original patent holder now has a new patent on the slightly modified version. This, of course, increases the drug price; when Novartis filed for patent protection, the cost of the drug Glivec jumped from `10,000 for a monthly dose to `120,000! Novartis appealed against the Patent Office’s decision in the Madras High Court, arguing that Section 3(d) goes against the basic idea of Trade-Related Aspects of Intellectual Property Rights (TRIPS).4 As economists, our first task is to see whether a section like 3(d) in the Patents Act, 1970 makes economic sense. Remember that the Patent Office and the High Court maintained that Novartis had carried out a minor modification as measured by the degree of improved ‘efficacy’. The argument for patent protection is to encourage creators of ideas to produce them. With such protection, they will be able to earn monopoly rents; without this protection there will be no returns from creating ideas as others will copy them at no cost (of creating the idea) and, hence, be able to undercut the creator. However, we also know that monopoly is bad as it is inefficient (see the first few paragraphs of the section titled ‘Patent’ in this chapter) because it charges a higher price than what it costs the society to produce at the margin (i.e., an additional unit of the product). 3 4

Novartis AG vs Union of India, 2007-4, MLJ 1153 [Madras High Court]. For more on TRIPS see http://www.wto.org/english/tratop_e/trips_e/intel2_e.htm.



Intellectual Property Rights  55

The efficiency loss in production is traded off against the added benefits of a new value-producing idea. If the new idea is only insignificantly valuable, then the efficiency loss of a monopoly is greater than the value of the idea and, hence, should not be allowed. The issue for economists here is not whether the court and the Patent Office were technically correct in evaluating the benefits of the idea; the economists’ point is that if the technical calculations of the benefit are indeed small, then the monopoly should not be allowed. We close this section by referring to another case. This was not conducted in any usual court but in the Competition Commission of India (CCI). The reason for referring to this case is to highlight the significance of defining patent, copyright and trademark as intellectual property. If you have a house you could sell it, rent it, lease it; you could also hire it out for a few days only to different sets of people as with a ‘guest house’; or, you could keep a paying guest for a year or two, or more. The important thing is that each one of these activities will be differently priced—the per day cost of rent will be different from the per day cost of a guest house which will be different from the cost to a paying guest. In all these cases, however, the person or entity who is paying the owner of the house will be enjoying services that a house provides. The pricing will be different because the extent and type of services enjoyed by each type of contract will be different. In other words, property is a bundle of rights and the owner of the property charges different prices in each of the contract types mentioned here because she sells a different (subset of the) set of rights to each type of buyer of those rights. There is no price discrimination among different sets of people for the same set of services from the house. In economics, price discrimination among buyers of the same product is possible only if the seller is a monopolist, the buyers are segregated and there is no arbitrage possibility (i.e., a buyer buying at a cheaper price and selling it to another buyer at a price greater than this price but lower than what the monopolist is charging the second buyer). In general, (perfect) price discrimination may be efficient, but it reduces consumer surplus to zero and hence, is not allowed by any policy that supports competition for the benefits it gives to consumers.5 In the case before the CCI, Singhania and Partners LLP, a law firm, placed orders for Microsoft Business Vista, a software package under the original equipment manufacturers (OEM) category. The order was placed with a Microsoft dealer along with an advance of 50 per cent of the OEM price. However, the dealer was advised by Microsoft that Singhania and Partners 5

See any standard textbook in on microeconomics for a price-discriminating monopolist.

56  Law and Economics

LLP was not eligible for OEM licensing (of the software) but eligible under the volume license (VL) category.6 The price difference between the two is quite substantial, with the OEM license being about half the price of the VL. The prospective buyers complained to the CCI that this was price discrimination since both the licenses enabled the product (software) to the same set of activities, that is, they were identical products. They argued that Microsoft could do this because it had 80 per cent of the market and, hence, could discriminate on price (like a monopolist). Though it was the dealer with whom the law firm placed the order, it filed the complaint against the dealer and Microsoft because it maintained that the dealer could not act without the permission of Microsoft. Microsoft argued that it has three different ways of selling its product and these were different products (offering different sets of services and having different costs per unit) and hence, had different prices for each of them. OEM licenses were the cheapest, followed by VL and the highest price was for retail licensing, also called full packaged product (FPP) licensing. In OEM, the software is loaded by the manufacturer of the machine (PC) and it is this manufacturer who licenses the software to the customer (in this case, the purchaser of the PC). This means that it is the PC manufacturer who provides support for the software. The consumer’s rights are limited in that should the computer break down, or the consumer discard this computer, the software is no longer available to be put on another machine. VL is available to all those who will buy five or more licenses together and, here, it is Microsoft that licenses the product and, hence, provides all software support. Unlike in the OEM, the consumer has the right to transfer the software from one machine to the other (though she cannot run it on two computers simultaneously). Finally, in FPP licensing, they are sold in individual packages, with the lowest volume of sales for Microsoft and the highest packaging cost. It comes with a media box containing a DVD and a manual and the customer has full rights to transfer and use it on other machines. In other words, each of one of them is a different product (because of the different sets of rights that is transferred by the owner, Microsoft, to the customer). Hence, just as you will charge a different price (per day) if I want to use your house for a party, compared to when I want to take it on rent for a year, Microsoft claimed that it could charge different prices for these three types of licenses. CCI upheld Microsoft’s stand. 6 Singhania & Partners LLP and Microsoft Corporation (I) Pvt. Ltd. & Ors. available at http://www.cci.gov.in/May2011/OrderOfCommission/SinghaniaMainOrder050711.pdf (last accessed on 24 October 2012)



Intellectual Property Rights  57

An Alternative View What we have discussed so far is the prevailing situation and the thought process that rules India and the world. This is the mainstream thought and what has gone into the making of laws and their economic justifications. However, there are some who argue that such limited monopolies, granted by the state, actually reduce welfare in the long run (Boldrin and Levine 2008). Historically, monarchs granted monopoly rights to people they liked; East India Company was granted such a privilege to trade with what was then known as the East Indies. The history of patents is traced to the English Parliament’s 1623 Statute of Monopolies (Boldrin and Levine 2008) that took away the power of the king to grant monopolies and allowed only the Parliament to do so. There are two basic pillars in the arguments against granting monopolies through patents and copyrights. The first is theoretical. Did Shakespeare need copyright protection to create his works? Or, did those who developed the building blocks of the Internet do so because they had patent protection? In other words, while the creators of ideas, music and literary works need money to survive and produce ideas, there is no reason to believe that granting monopoly rights is the best way to generate incomes for them. The second argument against prevalent thinking is based on various empirical evidences in world history. The proponents of the alternative view refer to the giant strides in technology made in the absence of monopoly protection—like the development of the steam engine after Watt’s patent on the original steam engine ran out, as well as the Internet revolution that took root when patent protection had not been introduced for it.

4

Public Rights over Property In Chapter 1 we referred to how society may want to restrict private property rights in various ways, with respect to different types of property, for different reasons. In this chapter, we will discuss these further, building up to situations where property is better held by a group or community, rather than by an individual. In fact, there could be situations where the government can take away property from an individual if it can argue that suspending private ownership is better for society. These are often called government ‘takings’ or ‘acquisitions’ and governments in most countries are allowed by law to undertake such action even when the private owners of these properties do not give up their ownership voluntarily. Of course, when the government can do this is also prescribed in law. Note that by the word ‘public’ in the heading of this chapter, we mean any collective entity like community, residents’ associations or governments at different levels (local, state and national).

Why Restrict Rights? Why should a public entity have rights over some property or over property owned by individuals? Let us consider some examples. 1. Some people living in one part of a village are badly in need of a road (since they use motorbikes to take home-produced goods to the market). These households are ready to put in some effort (or money) to buy land and construct the road. However, their problem is that once the road is built, even those who did not pay for it will be able to use it. Indeed, if one knows that the road is going to be built, and even if she directly benefits from the road and is willing to pay for it, she will find it in her interest not to pay for the road. The



Public Rights over Property  59

main point here is that once the road is built, no one can be excluded from using it (unless it is ring-fenced with controls at points of entry onto the road). Hence, even when the sum of benefits accruing to each potential user is larger than the cost of building the road, the road may not get built because not every potential user is willing to announce her benefit from the road or pay for it. In other words, the market system whereby people pay a price for a good, or service, is unable to solve the problem. The road is called a public good while, so far in this book, while talking about how the market is efficient we have actually been talking about what are called private goods or services.   More generally, goods and services can be categorized in the following fashion. There are two aspects of each good that we consider: rivalry and excludability. A rival good or service is one where if one consumes it, another cannot. So, if you eat a particular apple, another cannot eat that apple. An excludable service or good is one where you can exclude others from consuming it. We thus get a 2x2 categorization of goods and services, as indicated in Table 4.1.   We have already considered the pure private good (an apple) and a pure public good (a road). Non-excludable but rival goods would be the fish in the lake discussed in Chapter 2. In a natural lake, every fish that a person catches another person cannot (rival), but it is difficult to prevent a person from fishing in the lake. All members of a club can simultaneously use the lawn and the swimming pool facilities (non-rival) but non-members can be prevented from using any of the facilities (excludable).   These are conceptual differences among the types of goods and services that economists describe. With respect to the 2x2 classification of goods and services is given in Table 4.1, note that it is entirely possible given the context that a good moves from one category into the other. The most illustrative example is that of a road: it is a pure public good if it is not congested but a common resource if it is; if there is a toll to be paid for using the road, it is a club good. In general, most roads are not private property, except maybe the driveway of a Table 4.1. Categorization of Goods and Services Excludable

Non-excludable

Rival

Private goods

Common property resources

Non-rival

Club goods or services

Public goods

60  Law and Economics

house, the part that goes from the public road to your garage. So, a public property is at the other extreme to a private property—everyone owns it.   The concept of everyone owning the property is not an ‘aggregation’ of everyone having private property rights. Obviously, if a park is public property, you cannot play football and cricket simultaneously unless it is big enough to play both and everyone has agreed on the area where each game is played. Observe that if you own the park as private property, you only need 22 players to participate and nobody’s permission to play in the park. A public property, therefore, also has restrictions on how to, and who can, decide what to do with it, very much like private property. 2. Think about a marketplace in a rural area. There may be several shops selling vegetables, food, meat, fish, etc. Each shop has some waste to dispose of at the end of the day. It may simply dump this waste somewhere outside the shop, most probably by the roadside. The fact that the ‘roadside’ is not owned by anyone and everybody can use it does not mean that one can use it to dump garbage. When all or most of the shops dump their waste outside, the whole area begins to stink. This pollution affects everybody. What aggravates the problem is the following: if one shop owner thinks that it is bad to dump waste like this and tries not to do so, there may not be much improvement in the situation since all others are doing it. So this shop owner may lose interest in not doing it because she cannot make the stink go away by not stinking it up herself. One individual’s action in isolation is not sufficient and hence, some (public) agency (representing all or the majority) has to play a role. This is an issue of negative externality—dumping of waste affects others, and those who do so are not compensating the losses of others who suffer the bad smell and other inconveniences. 3. There are other situations where property may be owned by the public. In many societies, governments provide education and health in publicly owned schools and hospitals respectively. Many argue that basic education and health cannot be efficiently provided by private agencies because they have externalities—everyone prefers an educated and healthy neighbour, quite distinct from the neighbour’s preference to be educated and healthy. This is clearly true for public health issues—if I have an infectious disease and am unable to get treatment because I cannot afford the price of the cure, others run the risk of catching the infection from me.



Public Rights over Property  61

4. In a small beach used by tourists, there would be a messy situation if fishermen were allowed to dry their unsold fish. This would lead to the death of tourism. However fishermen too need space close to the sea to keep their vessels, to auction their catch and for drying their catch. However these two activities—fishing and tourism—cannot take place at the same place on the beach. There are a number of such conflicting activities. For instance, it may not be desirable to have residences and industry in the same part of a city. Under such conditions, it is better to set apart one place for one activity—one beach for tourism and another one for fishermen, one part of the city for residence and another part for industry. Where is the restriction on private property? Even if someone owns a place on the beach earmarked for a hotel, her property rights are restricted in that she cannot use it to dry fish. An example of this in India is Coastal Zone Regulations (Coastal Regulation Zone Notification of Government of India). This is essentially to see that coastal areas (defined as the area covering 500 metres inward from the high tide lines in seas and coastal water bodies) and the ecosystems associated with these areas are not damaged due to construction and other developmental activities. However, for historical reasons, some areas within such coastal areas are already well developed (used for buildings and other residential and commercial activities), especially in cities like Mumbai. Thus, it is costly to introduce restrictions in such areas. Moreover, the natural ecosystems there are already destroyed. Thus, there is no point in imposing coastal zone regulations in such areas that are already developed. On the other hand, there is usefulness in not allowing further destruction of those coastal areas where natural coastal ecosystems still exist. Thus, the coastal zone regulations in India have more stringent regulations against development in undeveloped areas, but less stringent restrictions against development in already developed coastal areas.

Who Should Own Natural Resources Like Minerals? Mineral resources are spread under large stretches of land. The surface land, however, may be divided up into a number of plots owned by different owners. There may be some economy of scale in mineral extraction; many small operators trying to extract minerals below their small pieces of lands

62  Law and Economics

may not be efficient. However, for a single company to mine the mineral requires every landowner to give the company digging rights on all the land. The mining company has to deal individually with each and every landowner and this increases the cost of transaction. Furthermore, once the landowner knows that there is mineral under her land, she realizes that the value she can extract for her land is many times more than what she gets at her current use. This may encourage her not to come to a quick decision and instead, hold back her sale in the hope of extracting as much value as possible. If everyone holds back, waiting for others to make the first move, not only will matters get delayed, but also it is entirely possible that nothing will happen. This is clearly not advisable since the value extracted from the mineral may be many times more than that currently being obtained from the surface of the land. In other words, the relevant question is: who owns the mineral resources under one’s land and, more specifically, does land ownership confer property rights on minerals underneath the land? What complicates the issue is the fact that the information, or knowledge, about the mineral deposits may come after the last transaction with respect to this land. So, the first issue on ownership of mineral resources is how to dissociate ownership of minerals from ownership of surface land. Now let us suppose that some mineral resource has been identified on land that is not privately owned but is under public ownership. If everyone is allowed open access (community resource, discussed in the section titled ‘Why restrict rights?’ in this chapter), then every member of the community will have an incentive to rush in and extract the mineral as much as possible before others. Delaying may leave one with no mineral to extract! This is a familiar problem with common property resource and it leads to overextraction—the mineral will be extracted at a rate that is greater than the socially optimal rate. Moreover, there is no incentive for any individual to ensure that there is no degradation of the surrounding land, for example, that it does not cave in. One alternative, of course, is to say that all mineral resources are a natural, national resource, and are under public ownership. This too raises a number of questions. 1. Which tier of the government—state or central—should own mineral resources? 2. Even if there is public ownership of mineral resources, who—public organizations or private sector companies—should extract and use (market) them for higher efficiency (productivity)? 3. How should mining companies get surface rights from landowners?



Public Rights over Property  63

We have referred to two types of problems—economies of scale and ownership. Currently, in India, there are three tiers of government—local, state and central. Considering the geographical area under which specific mineral stocks are located, and the amount of stock required for viable extraction (taking into account the economy of scale in operation), local government (the way it is defined in India) may not be the optimal tier of government to own minerals. Usually, the land under which resources have been found cover more than one habitation and one would need coordination among different local governments. The central government is too far away from the actual stock and may have insufficient knowledge about local aspirations. Moreover, in a federal set-up, with the focus on decentralized administration, consistency demands that local resources be developed and used by local bodies. The state government seems to be the appropriate level in this regard. It can address both the issues of scale and ownership. For historical reasons however, in India, mineral resources are owned by the central government. This is because immediately after independence, the central government wanted to keep for itself the ability to industrialize the country and distribute growth prospects equally among all the states. So, if the centre owned these resources, it would be easier for it to allocate them among the central public sector units. Also, if it wanted to encourage industrialization in a resource-poor state, it could make available to it the resources from a resource-rich state at concessional prices. Certainly, a resource-rich state government, elected by the people of that state, will have no incentive to give any concession to a poor neighbouring state! One still needs to worry about what will happen when the land above the mineral resources is not owned by the government. This becomes a difficult issue in India simply because the land can be owned by several people as the average size of land holdings is small. There can also be people who use these lands without full ownership rights (like the tribal people using forest land). How do private or public companies acquire rights to use the land to extract mineral resources in this situation? Theoretically, if all such landowners have well defined property rights, they can bargain with the mining company—whether it is private or public—to transfer use rights. Owners of land can receive a pre-agreed compensation. However, there may be two or three (at least) issues that affect the efficiency of this process: 1. Since there are many landowners, the transaction costs involved in bargaining with all of them may be high.

64  Law and Economics

2. The need for investments required for mining on the lands (in this case, the company will make investments on the land for mineral extraction, and hence it may ‘fear’ a demand for ex post renegotiation of contracts by some landowners) may encourage the company to own the lands de facto, rather than entering into contracts with each landowner. 3. If some landowners hold on to their right without allowing the mining company to use their land to access minerals, there may be further complications. At one level, it may increase the cost of operation of mining companies (if contiguous land is needed for efficient extraction). On the other hand, if others allow the company to mine, and minerals are extracted from the neighbouring land, there can be a negative externality on the piece of land that is not surrendered. Many of the issues here are similar to what we will deal with later in this chapter—land acquisition by the government. In most societies governed by the rule of law, the government has the legal right to take land away from the possession of private owners if it can be shown to benefit the public (including those whose lands are being taken away). An economic case can be made by using the standard methodology of evaluating such decisions—will the benefit from land acquisition by the government be greater than that from not acquiring it? If the value from extracting the minerals is greater than the value now being enjoyed from the land and there is a distribution mechanism that ensures that the current owners become better off after the acquisition, then land acquisition by the government may be allowed. We have already argued that if the land is not acquired, then the high transaction costs and other pitfalls of the market process may lead to no transaction at all. In principle, if most of the mineral resources are under lands owned by a large number of owners, and if they perceive the mineral wealth (and not the capital investment required for extraction) as the main source of surplus, they may see the need for jointly owning the company extracting minerals, as part of this contract exigency. Thus, there may be a case for incorporating a public mining company as a way out of this contracting problem. On the other hand, if the capital investment needed for extraction is much higher, and only a few of the surface rights require bargaining with large landowners, a private company owning the lands or minerals may be the preferred option. Public companies may use a political process to acquire lands and compensate the people whose lands are required for mining purposes. This can be part of a redistribution package—with a part of the income from



Public Rights over Property  65

mining being given to these people either as better compensation or jobs in mining companies, or through other welfare schemes. In practice, often the accountability of such a process is best ensured through political means and not through legal instruments. Thus, compensation for surface rights can be structured in the following ways: 1. Private companies acquiring surface rights through bargaining (through market or contract) with landowners. If ownership of the minerals is vested in the state, and private companies have acquired mining rights through licensing, they may pay royalties to the government, which in turn may come back to the people at large (in addition to some going directly to the original landowners) as public goods. 2. Public companies providing compensation (probably better than what they would have given through market- or contract-based transfer of rights) and probably jobs to the landowners. This can be interpreted as a different redistribution policy—a greater part of the income from mining being shared with landowners in the mining area, rather than giving them only a part of the public goods from the public exchequer through royalties or income from mining by public sector companies. If compensation policies are developed so that they encourage transactions in land, they will impact the distribution of gains among different sections in society (people living in mining areas and others elsewhere within the state), and get rid of the inefficiency inherent in the failure of the market to provide a solution.

Regulations to Control Environmental Pollution We have seen earlier that if one person’s action creates a damage for another person (like construction of the apartment building creating cracks in a neighbour’s building), and if they can negotiate with each other without much difficulty, then socially efficient outcomes can be reached whatever the legal entitlements (Coase 1960; see also the section titled ‘Economic need to protect property rights legally’ in Chapter 2). (There may be distributional differences depending on the way legal entitlements are held.) However

66  Law and Economics

there can be several cases when one person’s action affects other peoples’ life (consumption, production or properties) without the former compensating for the losses of the latter. Environmental pollution is one such case. In those cases, the polluter and the victims of pollution may not be able to negotiate with each other for a number of reasons. There may be one polluter (like an industrial firm) but its pollution may affect many people. (Or there may be several polluters like motor vehicles affecting several people in the same locality.) In such cases, victims have to come together or to agree to send a representative to negotiate with the polluter. Some people may not agree, or may not be willing to share the burden, since once the pollution problem is solved through the actions of others, it would also benefit them (without having to undertake any effort, the free rider problem referred to in the section titled ‘Why restrict rights?’ in this chapter). If everybody thinks this way, the problem will remain unaddressed. This is what we call a public ‘bad’; neither does my smelling the stink of someone’s garbage prevent others from facing the same stink (non-rival), nor is it possible for some people to be excluded from getting the stink. As in the case of a pure public good, this situation leads to a market failure because the transaction costs, of bargaining among several sufferers and several polluters, are simply too high. However, the aggregate good from no pollution may be larger than that from pollution. When a negotiated solution is not feasible, legal entitlement is important. What does law do here? It can fix liability. This is the procedure in nuisance cases. If one person’s action is responsible for causing damage to others, the former can be made strictly liable. This allows the polluter to either continue the pollution by paying compensation for the damage, or to stop it since it becomes uneconomical due to the need for paying compensation. The ‘polluter pays’ principle fixes such liability. In this case, victims do not have to wait for any collective action, but can claim compensation, with or without the support of the judicial process. However, there are cases where such compensation may not be that easy or feasible. It may be relatively easy to pay compensation if the damage is already done and there is no ongoing or further damage foreseeable in the future. The assessment of compensation for past damage is relatively easy. On the other hand, courts have to fix a recurring compensation for future damage. This is relatively less complex when there are only a few victims. But it is very difficult in cases when pollution affects a large number of people. It is not easy to identify the victims and to assess the damage that they may suffer due to pollution. Identification of victims and assessment of damages may incur substantial (transaction) costs. Thus, there may be cases where it is legally efficient to control pollution in a way that does not



Public Rights over Property  67

require victims to be identified. This takes us to the regulation of activities that may cause pollution. Regulation essentially puts limits on what owners can do with their own, others’ and/or public property. Mandating that factories should not pollute the river—which is a public property—with water containing sulphuric acid of more than 100 parts per million units is an example. One cannot construct a house very close to the boundary of the site—this is another example of regulation. All these are public restrictions on what can be done with one’s own private property. There are different ways of implementing such restrictions. One way is to say that private individuals or firms should conform to a particular standard. Requiring that a newly built house be at least two metres inward from the boundary road is one example. Insisting that no firm releases hot water from the factory with a temperature of more than 100 degrees centigrade into the river or water bodies is another example. There can also be technical standards. Laws mandating that all cars should use a particular emission standard so as to minimize air pollution is one such case. There may be similar rules that mandate the height of a chimney in a factory. Broadly, these are called standard-based regulation. In the case of those who violate this standard-based rule, they are likely to face prosecution, and if proven guilty, they may have to pay punishment or face imprisonment, independent of the damage actually caused. The other form of regulatory instrument is taxes. Such a regulation does not insist on specific behaviour or prohibit certain activities. Instead, it requires a firm to pay, say, `1,000 for each kilogram of sulphuric acid released into the river. Thus, a firm may pollute by paying the corresponding taxes. Economists generally prefer such taxes. The basic idea is that if tax per unit of pollution can be fixed at a rate that corresponds to the additional social cost due to that pollution, these taxes would help to compensate for, or mitigate, social losses. It is expected that taxes have some advantages over standards. First of all it gives the freedom to the polluter to decide whether to pay taxes or follow standards. (It may be noted that since taxes, by definition, equal the social cost of pollution, payment of taxes or following standards will not make a difference to the social impact of pollution.) For some firms, the additional cost of controlling pollution by one more unit may be higher than the taxes (since this could be a firm using older technology), and they may get the freedom to pay the taxes and pollute. On the other hand, for other firms, controlling greater pollution (than what is desired while fixing the standard) may be more economical than paying taxes (and these may be using newer

68  Law and Economics

technology in their factories). This may reduce pollution to lower levels (than expected through fixing standards). This, by itself, shows that a regime of taxes may encourage the firms to adopt less-polluting technologies, compared to a situation where standards are in place since they will not gain by reducing pollution beyond or lesser than the standard. However, this is true only if such types of pollution create a purely negative externality. Given the increasing awareness among people of the ill effects of pollution, many take the stand that they will not buy polluting goods or invest in polluting firms. This shows up in people willing to pay a little bit more for an environment-friendly product even when an otherwise identical but environmentally-unfriendly product is available at a cheaper price. In economics, we would characterize it as a part of the externality being internalized by consumers who think like this. The first paper in economics (Arora and Gangopadhyay 1995) that analysed such a situation showed that setting minimum standards of pollution does a lot better than imposing taxes on pollution. Moreover, it is not always easy to enforce taxes. In order to design taxes, we need to know the social cost imposed by each unit of pollution and/or additional cost to be borne by each firm to control each unit of pollution. It is not easy to collect this information. Thus, though taxes have some attraction in terms of economic incentives, these are not widely used. The most widespread form of environmental regulation is standard setting (emission standards or technical standards). Indian environmental regulations too follow emission standards (e.g., in the case of commercial motor vehicles in Delhi, technical standard is followed) and the use of environmental taxes is very limited in India. There can also be other policies to control pollution. There are measures to subsidize firms to take pollution control measures.1 Of late, a new legal mechanism has emerged. This is to mandate firms to provide information on the pollution created by them. This is like other legal mandates of information disclosure (such as the audited statement of companies listed on stock exchanges). What is the use of this information disclosure? It may help in many ways. It may help public enforcement authorities to locate polluting firms. It may help vigilant citizens living near the firms to alert public enforcement officials. But there can also be some non-public (without the help of public enforcement authorities) actions facilitated through the mandatory information disclosure. Some people living near a polluted firm The paper by Arora and Gangopadhyay (1995) shows why subsidies are optimal (rather than taxes) when consumers are willing to pay more for environment-friendly products. 1



Public Rights over Property  69

may decide to take action (like moving out), even if the pollution levels are lesser than the prescribed standard. (For example, they can be more sensitive to certain chemicals than others.) Similarly some others may move into the locality by knowing that pollution is higher and hence land prices are lower there. This may facilitate ‘selection of neighbours’, based on the willingness to accept pollution. It is also seen in the developed world that firms creating too much pollution (as per the information disclosed) are valued less in share markets—this may be due to concerns of citizens regarding pollution or due to a perception that the profitability of such firms in future may be lower since they are likely to face regulatory action by the government. The products of highly polluting firms can also be valued less by concerned citizens. Thus, there are multiple ways in which mandatorily disclosed information is used by the people. In countries like India, enforcement of environmental regulations is relatively weak. This encourages people to take actions on their own, resisting the location of potentially polluting factories or taking civil disobedience actions against currently operating polluters. The Indian legal system has also created a provision for ‘public interest litigation’ (PIL) and PILs are used liberally by people affected by polluters. Cases are filed before courts as PILs regarding actual or potential pollution or harm (by proposed factories or projects), and courts respond by either directing the enforcement authorities to act or by fixing liabilities. There are costs and benefits from such citizens’ actions to protect the environment, especially in countries like India where there are other institutional problems like long delays in getting court decisions and poor enforcement of general law and order (Santhakumar 2003).

Public Ownership of Property The problem of ‘no property right’ was understood by human societies early on and subsequently, different forms of rights emerged. The first response to this problem was to keep some forms of property under the ownership or control of communities. Thus, traditional fishing communities all over the world developed certain institutional arrangements to see that their fishing effort does not lead to destruction of the stock of fish (Baland and Platteau 1996; Plattea, 1989). Similar community rights existed over forests and grasslands. Social norms and community punishment mechanisms helped in enforcing such rights. These rights and the social norms that sustain them were passed down by custom through generations.

70  Law and Economics

When economic opportunities to use these resources widened, and when one community’s action inevitably affected the others, different arrangements were tried out by the larger society or the state. There was an explicit recognition that some form of state intervention was needed in the case of resources such as forests and fisheries. Thus, ownership of the state (local or national) came about in the case of some of these resources. There are certain forms of state ownership or intervention, which are both unavoidable and appropriate. For example, the nation state has internationally accepted rights over the waters within an EEZ (exclusive economic zone). Thus, a foreign fishing vessel needs permission from the state to fish within 200 nautical miles of a country’s coast. Moreover, the state manages the general law and order and judicial mechanisms, which serve as the base on which any property right system, whether community or private, can be built. However, for state ownership to be effective, a costly mechanism of monitoring and enforcement and a machinery of efficient administration are needed. Moreover the transaction costs of contracting or licensing resource use are likely to be greater under state ownership than when such contracting is carried out by the private sector (Cheung 1970). Thus, the effectiveness of state intervention is generally found to be poor. In some cases the state does not restrict access among its citizens, whereas in other cases it does so on paper though actual enforcement is lacking (Feeny et al. 1996: 188). Thus, the outcome of state ownership in certain cases may be closer to an open access regime. It is not uncommon to see failure on the part of state agencies resulting in the failure of fisheries.2 There are, however, strong incentives for political and administrative systems to demand the continuation or expansion of the State’s role in the management of resources.

Why Should Governments Acquire Private Land? All over the world, governments have retained the power to acquire land for ‘public purposes’. The economic logic of government action for ‘public purpose’ is closely related to the definition of ‘market failure’ in economics. It is costly to exclude somebody from using some goods or services like roads. 2 Feeny et al. (1996: 197) elaborates on the state’s failure that led to the collapse of Californian sardine fishery. According to them, ‘too many government bodies competed with each other for political resources for any of them to account meaningfully for such diffused, intangible, or trans-generational values as were at stake in the sardine controversy’.



Public Rights over Property  71

The additional cost of having a few more people using them may not be high. The inability of a firm to exclude non-payers from using these goods prevents it from extracting a price for using them. This, and the benefit in allowing others to use them since the marginal cost of additional consumption is zero or minimal, would imply that some entity representing society’s collective interest, such as the government, has to play an active role in the provision of these public goods.3 Most infrastructure services like roads, ports, airports, etc. satisfy the properties of being a public good (see the section titled ‘Why restrict rights?’ in this chapter). When a private firm provides these services, it may not adequately take into account the social demand, and hence a role by the government is justified (though it is not necessary for the government to provide the service directly). For large infrastructure projects such as airport or roads, if one or a few persons are unwilling to sell parcels of required land, then the viability of such projects in specific locations will be affected. Similarly in the case of certain other projects, even if some landowners are unwilling to move out, their existence in the location can be affected once the project is made operational. The reservoir for an irrigation project or a hydroelectric dam is such an example. Thus, it is argued that some collective, and sometimes coercive, effort is needed to acquire land for large investment projects. It is for these ‘public purposes’ that governments are expected to acquire or take over land or private properties. Economic analysis provides some insights, discussed below, on how efficiency can be achieved with regard to government acquisition of land or assets.

There Should Be ‘Adequate’ Compensation There was a debate in economics on whether there can be an efficient means or level of compensation for private properties taken over by the government (Blume et al. 1984). The doubts about the efficiency of compensation were because such compensation may provide an incentive to the landowners to over invest in lands that are likely to be taken over by the government. However, it was argued even then that a rule, which pays compensation only for those land uses in effect prior to any discussion of a government 3 Government intervention in a marketplace may be needed in certain other cases too; for example, it may be cheaper to have one large airport in a city rather than a number of small and competing ones.

72  Law and Economics

investment, would be efficient.4 In other words, if agricultural land is being taken over to build an airport, while the value of the land may be many times more than the value generated by keeping it for agricultural purposes, the compensation to the original owner should be in accordance with the land’s agricultural value and not its value resulting from its use as a landing strip. Since efficiency is a primary concern for economists, let us see exactly why government acquisition is required. In most cases of this type, the land required stretches over a large geographical area with a number of different owners. Lands owned by these owners have to be aggregated into one continuous stretch to generate value for the public good. To make matters simple, let us take a specific example. Suppose there are three identical and adjacent plots of land owned by A, B and C respectively. Each of the current owners, given her use of the land, values the plot at `100. A road connecting one extreme end of these plots to the other extreme end will generate a total value of `350. Of course, this additional value will be distributed over a large number of users of the road and the road is such that its use is non-rival and no one can be excluded from using it; it is a pure public good. Clearly, if all the three plots of land were aggregated and the road was built, the value generated (`350) would be greater than the current value `(300 = 100 + 100 + 100) being generated from these assets. For argument’s sake, suppose that the road builder, the local government or its appropriate department, has already bought the land from A and B; C is the last person with whom the buyer must negotiate. Suppose the price paid to A and B for the land was `110 each, A and B voluntarily transfer their land because `110 is greater than 100. So, when the buyer goes to C, the buyer has already spent `220. If the negotiations with C break down, the buyer would lose `220 because an unfinished road, regardless of what proportion of it is unfinished, gives zero value. If C is aware of this, then C is in an excellent bargaining position. For, suppose C makes a take-it-or-leave-it offer of `200; that is, if the buyer pays at least `200, C will sell, and not otherwise. The buyer has two options—to accept and pay `200 for C’s land or reject C’s offer and not build the road. If the buyer accepts, the buyer gets the value of `350 from the finished road at a price of `420 (`110 each to A and B and `200 to C). This means the buyer makes a loss of `70 (420 – 350). If the buyer reject’s C’s offer, the buyer gets zero value from the unfinished road but has already paid `220 4 Such a payment is also necessary, it is argued, to avoid social losses due to the costly actions taken possibly by individuals to fend off the ‘risk’ of a public decision (which is likely to be a non-random and strategic action) to take over land without compensation.



Public Rights over Property  73

to A and B; the buyer’s loss is then `220. In both cases the buyer is making a loss and it makes sense for the buyer to choose the option that minimizes his loss. So, the buyer should accept C’s offer. A simple generalization of the argument immediately leads us to the following conclusion: an aware C should charge a price as close to `350 as possible for her land. For example, an offer price of `349, if accepted, means a loss of `219 (350 – [110 + 110 + 349]) to the buyer; if rejected, it means a loss of `220, which is a greater loss than `219. In other words, if the lands were to be aggregated in a way that allows for voluntary transactions, there are two possibilities. The first one we have just illustrated—the road building authority always makes a loss. The second is that everyone wants to be the last seller, C, since it gives a greater advantage to the original owners. Since everyone wants to be among the last to sell, there is no first seller and the process of land aggregation does not even begin, let alone completed. This, in economics, is called the ‘hold-up’ problem. Market transactions are not possible where hold-up is prevalent and, hence, the properties of efficient market equilibria are not applicable. An alternative is for the road authority to forcefully acquire the land and compensate each owner by an amount of `100 at least. Giving adequate, but not exorbitant, compensation is necessary for yet another reason. Suppose I know that in the not-too-distant future, a road will have to be built on land that is currently owned by a large number of small owners. If the government were to acquire this land and give compensation much higher than the value of the land’s current use, I will buy the land from the current owners and hold it until the government is ready to acquire it. This leads to two problems. First, people with money start speculating on land and this may result in land price bubbles that are inefficient. Second, and something more sinister, powerful land grabbers may induce the government to build the road on lands being held by them. In principle, compensation should not give incentives to the owner of the land or asset to make investments in the hope that it will be acquired and compensated. This can lead to over investments in land and over utilisation of capital (especially financial). If compensation is enough to cover the lost profits and capital costs undertaken by the owner on the land, it would not have any of the distorted incentives we mentioned in the last paragraph. Hence, many economists argue that the compensation should be a sum equal to the value of the land in comparable sales of similar properties, interpreted by lawyers ‘as the value of a fair sale in an open market’ (Burrows 1991). However, there is another aspect of the compensation that needs to be considered. When you buy land, you may be worried about a possible takeover

74  Law and Economics

by the government. Since this is a risk, and people do not like risk, the value of land will be less than it should be.5 Ideally, whenever there is such a risk, insurance markets, should they exist, are efficient. However, such insurance markets do not exist and hence, governments, while compensating, should add this risk premium. How does it work? You want to sell your land to me, where I want to build my house; however, I am giving you a lower price than I would otherwise be willing to give since I am afraid that after I have built the house, the government might acquire the land. This lowers the current value of your land. Economists argue that the nature of risk in acquisition is such that compensation by the government is the ‘appropriate’ insurance in this regard since governments can spread the risk through the base of all taxpayers (Burrows 1991; Cooter and Ulen 2004). This is especially so, considering the incentives for the government. In the absence of such compensation, it can become an incentive for the government to over-acquire, or it might use acquisition of private property as a way of generating resources for the government. (Financing the state through acquisition of private property rather than taxes has negative effects on economic efficiency.) Finally, compensation based on a simplistic use of the concept of ‘market price’ can create problems of inefficiency if the geographical position of the land attributes monopoly power to the owner, or if the market price itself is influenced by the decision of the government to invest there. The market price of an asset is the current discounted value of future income streams from the asset (it has little or nothing to do with past value). Monopoly ownership of land inflates the value and this may lead to less than socially optimal levels of government investment if the original owner has to be compensated the land’s market value. This happens because once the government announces the acquisition of a particular plot of land, its market price jumps to the value that might be generated from it from the government’s proposed use. One way to reduce such inefficiency is to decide compensation by comparing it with a plot ‘whose characteristics resembled those displayed by the acquired plot before government project plans were promulgated’ (Burrows 1991: 62). In general, compensation for land acquisition should be decided by (a) providing just protection to asset losers and (b) minimising the incentives for asset losers to indulge in inefficient behaviour. The basic guideline emanating from the literature on law and economics is the following: ‘In general, the government should only take private 5 If you want the land to build a house, but the land may be acquired by the government, you would want to pay a lower price for the land than what you would have paid if you were sure that the land would not be acquired.



Public Rights over Property  75

property with compensation to provide a public good/infrastructure when transaction costs preclude purchasing the necessary property’ (Cooter and Ulen 2004: 177).6

Regulation versus Acquisition In acquisitions, governments take over lands for public goods and infrastructure by paying compensation. Regulation is also used by the governments for the purpose of enhancing public welfare. Insisting that property owners in a location cannot construct multi-storey buildings beyond a particular height, or that people cannot use property located in the residential zone in a city for building factories, etc. also limit the scope of private property. This is done without paying compensation for the losses, if any, due to the restrictions imposed on them. For example, people in Kerala cannot convert a paddy field into a plot for the construction of a house. This is done with a public purpose—conversion of paddy fields into house plots may reduce the areas for water storage during monsoon, causing floods in other areas and/or reduction in ground water storage. Such a restriction reduces the value of that plot of land, but this loss incurred by the property owner is not compensated by the government. On the other hand, if government acquires the paddy fields to keep it idle (without using them for construction) so that more space is allowed for water storage, then it may have to pay compensation. This difference between regulation and acquisition may provide perverse incentives to property owners and the government. On the one hand, the need for paying compensation as part of the acquisition may encourage the government to be careful (as it has only limited resources). The lack of need to pay compensation may encourage the state to over-regulate the private property (since any level of restriction on the use of property does not add to government’s cost). On the other hand, if the government compensates fully, then property owners do not have the incentive to be careful with their investments, mindful of potential regulations. These investments would be wasted if regulations are forthcoming, and this is social waste. Alternately, no compensation by the government makes the property owner mindful of

6 One strand of literature on the law and economics of ‘takings’ is not considered here: regarding regulation (restricting the activities of private properties for some public benefit, without paying compensation) and takings together, and how it can be decided whether regulation or taking is most appropriate in a given case.

76  Law and Economics

potential regulations in the immediate future, since she may suffer losses (as there is no compensation). Thus, in a context where people are likely to invest more in expectation of compensation, there is a desirable role for regulation. On the other hand, if state is concerned about compensation, then acquisition is better as it will put rational limits on its action.

Land Issues in India Of late, we have seen a number of headline-grabbing court cases regarding land acquisition along with separate instances of allegations of land-grabbing by politicians and their henchmen. Starting with Singur in West Bengal and culminating in the court judgments on land acquisition by the government of Uttar Pradesh, there have been many debates, discussions and agitations on how to tackle this problem. The dilemma faced by India is that there is a great need for industrialization and urbanization but there is very little land that is not privately owned where such developments can take place. So, the government is confronted with the problem of determining how to transform agricultural land into land for alternative uses. One thing that is immediately noticeable is that much of the dispute (and protest) concerns land that has been (or is being) acquired by the government for what it describes as ‘public purpose’. Recall that economists have an argument for why some land needs to be publicly owned (see the section titled ‘Why restrict rights?’ earlier in this chapter); public purpose, however, is a much broader concept. For instance, in Singur, the land that was acquired by the state was not to build publicly owned and accessed infrastructure but to give it (at a price that the government was unwilling to disclose) to a private company to build a car factory. The court judgment against the state of Uttar Pradesh was similarly concerning land taken from farmers to be given for housing and commercial spaces to other private parties. The public purpose concept used here was that a car factory in Singur and urbanization in Uttar Pradesh would generate employment and overall development of the region.

Laws Governing Land Acquisition in India Originally, the Indian Constitution guaranteed the right to property as a fundamental right. However, immediately after India adopted the



Public Rights over Property  77

Constitution, the Parliament clashed with the judiciary on this matter. The Congress government had an aggressive land reform policy as an important part of its political agenda and the fundamental right to own property had to be compromised. The landowners, aware of this, went to court. The court declared the land reform laws unconstitutional. The Parliament responded by listing all land-related matters in the Ninth Schedule of the Constitution. The Ninth Schedule, introduced as the first amendment to the Constitution as early as in 1951, listed matters on which Parliament could pass laws without any judicial review. The next milestone in land laws came during 1971 and 1972. The Parliament gave itself the absolute right to amend the entire the Constitution, including the parts that guaranteed the fundamental rights of the citizens. Hence, disputes regarding land reforms and land acquisitions by the government could no longer come under judicial review.7 Almost all countries allow the concept of ‘eminent domain’. This gives the government the right to seize private property for public use in exchange for fair market value (compensation, as discussed in earlier sections of this chapter). Thus, the acquisition of land by the state for public purpose is not unique to India. It allows the building of roads, for example, which are used by the general public. It also allows the building of dams for power generation that usually leads to moving people out of their current possession of land. The real difference across countries is the use of the concept of public purpose. For roads, or the provision of public goods, where there is clear public ownership, public purpose or use is easy to understand. However, what happened in both Singur and Poletown (see below) was the takeover by the state of private property to transfer it to another private party—General Motors in Poletown and Tata in Singur. Interestingly, a case came up in the Supreme Court of India in 1971 where a property owner challenged the acquisition of his land by a state government to build a factory. In the first instance, the land was earmarked for the setting up of a public-sector unit. However, for various reasons, the public-sector unit could not be set up and the land was given to the private sector to develop. The challenge in court was that this is not for public use but for private gain. The court’s position was that public purpose was decided by the legislature. This is enshrined in law by the amended Land Acquisition (Amendment) Act, 1984 whereby government could transfer private land 7 It was not until 1978 that the fundamental right to hold property was omitted from the Constitution. More recently, the Supreme Court has maintained that everything is under judicial review if it goes against the basic nature of the Constitution and thus, the Parliament’s unfettered right to amend the Constitution has been held back by the Supreme Court.

78  Law and Economics

from one to another party once the new activity or project could be justified as serving a public purpose. Hence, this case was decided in favour of the state government. Before we proceed any further, it may be a good idea to compare two instances of similar transactions—Singur in India and Poletown in the United States—and the issues that developed from each instance.

Poletown and Singur Around 1981, the United States car company General Motors decided to move its Cadillac plant away from the Detroit region and relocate it to a place in the south of the country. Its main contention was the plant could stay in Detroit only if the city of Detroit made extensive infrastructure improvements. This prompted the cities of Detroit and Hamtramck to collaborate in a grand plan to bring industry back to what was perceived as a ‘dying’ city.8 The two mayors and their offices felt that if the new plant was built in Detroit it would arrest the drop in employment and generate tax revenues for the city, allowing it to mobilize resources for other developmental and public projects. According to this plan, land would be given to General Motors to build a state-of-the-art plant, crossing the border with Hamtramck, which would include the neighbourhood of Poletown. The plant was expected to generate 6,000 direct jobs and an unspecified number of indirect jobs through the setting up of ancillary units supplying to the plant. A total of 1,300 homes and 600 businesses had to relocate for the plan to go through. Since it was a ‘depressed’ neighbourhood, many homeowners readily agreed. Their decisions were greatly helped by the government’s compensation offers. The homes were valued at a minimum of $6,000 and averaged about $13,000. An additional amount, up to a maximum of $15,000, was to be given to cover the cost of a new house bought by anyone who was being dispossessed of her current home. Furthermore, a bonus of $3,500 was given to all households who moved within a stipulated time. The government’s cost of acquiring the property and developing it was estimated at $200 million, but the price charged by the government to General Motors was $8 million. However, importantly, not everyone was willing to move. The Poletown neighbourhood association and some other individuals filed a suit in Wayne Circuit Court. 8 Recall former Prime Minister Rajiv Gandhi describing Kolkata in the late 1980s as a dying city!



Public Rights over Property  79

Back in India, on 18 May 2006, the Tata group of companies announced that it was going to set up a car factory in West Bengal. The Tata automobile plant was to be the first major step in the reindustrialization of the state. Over the years, industry has been leaving West Bengal, citing labour union activities and lack of infrastructure as major concerns. The government, which had just won a popular mandate, was determined to woo industry back to the state as a means of generating employment and income. Like in the United States, the local government’s stated purpose of acquiring the land was to kick-start a depressed economy. There were 10,437 hectares of farmland in Singur block, out of which in 9,099 hectares more than one crop was produced. Like in most parts of India, land holding was skewed towards small farmers and 9,020 farmers owned a total of 635 acres of the land that was to be acquired. Paddy was the major crop produced by the farmers. The government’s calculations showed that the revenue from paddy cultivation was `9,940 per bigha (roughly one-third of an acre) of land. Given a cost of cultivation equal to `5,500, farmers earned an annual income of `4,400 per bigha from paddy. The government offered a compensation of `100,000 per bigha. This price was apparently 30 per cent more than the market price prevailing in the region. If the buyout price were invested in safe assets (like bank fixed deposits), then the farmers would have obtained an annual return of `7,000 (interest of 7 per cent per annum). An additional 10 per cent of the price was to be given to those who surrendered their land within the stipulated period and without any dispute. Much of the land was cultivated by tenant farmers and the government offered a lump sum payment of ` 250,000 to them. The company stated that the automobile factory would generate 2,000 jobs within the plant and 10,000 outside the plant through its ancillary industries. However, given that the target amount of land to be acquired was about 1,000 acres, less than 60 per cent of it was voluntarily surrendered. Three groups supported the transfer of land—absentee landowners, registered sharecroppers and those losing a part of their land. This was largely because of the policies that the ruling party (in power since 1977) had earlier initiated in which tenant cultivators, once registered, could not be evicted as long as they shared a government-determined proportion (50 per cent) of the value of their produce with the landowner. The government’s offer to buy out the land benefitted the landowners who got full value of their land as against 50 per cent of the revenue from the land in perpetuity. Tenants similarly got value for an asset that they really did not own! Three groups opposed the land transfer—unregistered sharecroppers, daily-wage earners associated with agriculture and those losing all their

80  Law and Economics

land. For these groups, it was a loss of livelihood, with the first two groups not being compensated at all. The government was quick to respond to the grievances of the non-registered sharecroppers and offered to give them the same compensation as the registered ones. However, being unregistered, and hence without records, there was a strong feeling that only activists and sympathizers of the ruling party would benefit as unregistered tenants. By the beginning of February 2007, 700 locals had been inducted as factory guards by WBIDC (West Bengal Industrial Development Corporation), 2,792 as daily-wage labourers had been hired for civil work in and around the factory site and a couple of rudimentary workshops were set up to train women in sewing (to stitch uniforms) and the men in elementary machine work. Once again, the people who were left out of these jobs were the erstwhile agricultural labourers who had not owned land or been tenants (registered or otherwise).

Position of the Law in Poletown In Poletown, the use of the power of eminent domain to acquire one’s property to transfer it to another private entity was challenged, with the local government claiming that this was to boost the economy and hence, was for a public purpose. The contention of the homeowners was that General Motors was the primary beneficiary of this transfer. The government’s contention was that the acquisition was to create an industrial site, which would generate employment, economic activity and add to the state’s coffers. The benefit to General Motors was purely incidental. The court found for the defendant. Of interest to us is the dissenting judgment by one of the judges. We present his major points below. In Michigan (the state where Detroit is situated) there have been many instances of acquisition of private land for public use. They have been justified as attempts to improve the ‘instruments of commerce’ and through this process various highways, railroads and canals have been built. The dissenting judge maintained that there were three characteristics of such acquisitions: ‘(a) public necessity of the extreme sort, (b) continuing accountability to the public, and (c) selection of land according to facts of independent public significance.’9 He felt that these conditions were not satisfied with respect to the General Motors automobile plant. 9 For the dissenting judge’s quotes, see http://www.law.berkeley.edu/faculty/rubinfeldd/ LS145/poletown.html (last accessed on 23 October 2012).



Public Rights over Property  81

Private property is an essential concept in market-driven, rule-of-law economies. Thus, any suspension of such property rights must have a solid argument. One must be able to argue that the objectives of a market economy will be better served in a specific instance if private property rights are suspended. The dissenting judge maintained that while there is a sound theoretical argument in certain cases for the involuntary transfer of private ownership to public ownership, there are no sound arguments for the government forcing a private party to give up land to another private party. In other words, why should the government act as a nonmarket intermediary in what is essentially something that land markets are supposed to do? His second point was extremely foresighted. Suppose that the creation of new employment is a desired public purpose goal. Who is going to implement this? If at some time in the future, General Motors, in its bid to maximize profit, decides to shed labour, will the government hold it liable because it had promised 6,000 new jobs? If not, who satisfies the stated public purpose of increased employment? If, on the other hand, the government holds General Motors accountable, it will interfere in its market activity and this is against the basic tenet of a market economy based on free enterprise. In other words, the position is either one of zero accountability or one where the government becomes an interventionist entity. According to him, ‘[w] ith this case the Court has subordinated a constitutional right to private corporate interests’.10 The 1982 judgment opened the floodgates. It gave a huge incentive to local governments to use eminent domain to take over land from private parties in the name of development. It encouraged companies and developers to make wild claims of what would happen after the takeover. Indeed, the Poletown project itself did not live up to expectations. Many claim that in the ultimate analysis, it destroyed more jobs than it created. In India also, we see this happening. The unbridled powers of the government to take away land from private parties and give it to other private parties (in a very loose interpretation of public purpose) have led to unprecedented numbers of land scams, nepotism and other corrupt activities through collusion between administrators and developers. In a rare move, on 30 July 2004, the Michigan Supreme Court reversed the Poletown decision. It said that for forceful acquisition by government, it is not enough to argue that an entity’s profit maximization contributed to 10 http://www.law.berkeley.edu/faculty/rubinfeldd/LS145/poletown.html (last accessed on 23 October 2012).

82  Law and Economics

the health of the general economy, an argument that was essentially made in the Poletown case. At the same time, the Michigan Supreme Court decided another important eminent domain case. In this case, the Detroit Wayne County Stadium Authority had forcefully acquired the property of a certain Ms Frida Alibri, promising her that it would not be given to a private party but would be used to build a public stadium. However, the property was transferred to a private corporation. Alibri filed to get her property back and the court upheld her complaint. Recall that this is essentially similar to the 1971 case in the Indian Supreme Court where the plaintiff objected to land being taken for a public sector unit and then handing it over to a private corporation. It is uncanny that there should be two near-identical cases in the two largest democracies concerning the government’s ability to forcefully acquire land. In India, everything went in favour of the government and against the private citizen. In the US, while initially the state was given precedence, the position was finally reversed and eminent domain was not allowed to be used to transfer land from one private entity to another using the argument of ‘public purpose’.

Why Is a Third Party Necessary? The question that still remains is whether we need a third party—like the government—to effect land transfer between private parties when such transfers may lead to overall development but where the new activities are not owned and operated by the state. Note that this involves not only land for new industry, but also the acquisition of land by municipal bodies to be handed over to private owners for residential and/or commercial use in the drive towards increased urbanization. In short, is it possible to find a pure market solution to the problem where suppliers voluntarily surrender their property to the buyer in exchange for monetary compensation? There are at least three reasons why market-driven voluntary transactions may not work. First, since the transactions have to be voluntary and this is a one-off sale (in the sense that parcels of these lands are not routinely bought and sold), the buyer has to negotiate the price with each landowner. These costs can become prohibitively high. Second, the negotiations are between small owners with limited resources and a big company with a vast amount of resources. These resources are not simply financial but include property lawyers, accountants and networks of people. This gives rise to great disparities in what the literature terms as bargaining strength between



Public Rights over Property  83

the suppliers and the buyer. This may result in sellers getting a much lower price than what they would have fetched if the resource disparities were not so great. In other words, one may require an agency to negotiate on behalf of all the farmers together so that the disparity in bargaining strength can be mitigated. Thus, instead of multilateral transactions bilateral negotiations where an agency or association of sellers interacts with the buyer in the marketplace may be required. The third problem is the hold-up problem, which we have already dealt within the sub-section titled ‘There should be “adequate” compensation’ in this chapter. Thus, a pure market solution may not work, or be desirable, given these three problems—high transaction costs, disparate bargaining strength and the ability of sellers to hold up all negotiations. Interestingly enough, this is not a unique problem. Similar issues arise when a publicly traded company is being acquired. When an acquirer wants to take over a company, she needs to buy at least 50 per cent of the outstanding shares. The problem again is two-fold—large numbers of small shareholders and inability to generate value from the takeover unless majority shares are obtained. Again, we have high transaction costs (large number of small shareholders), disparate bargaining strength (small shareholders versus a large acquiring company) and a hold-up. The hold-up problem is slightly different here because not surrendering the shares is the best strategy for the small shareholder; remember that in the land case not selling is not the best strategy but selling last is. Nevertheless, in both cases transactions do not take place. In the takeover of publicly traded companies, markets and institutions that govern them (stock exchanges and corporate governance laws) have addressed and solved the problems through the mechanism of the ‘takeover code’. We are not saying that something similar to the takeover code has to be applied to land acquisition for privately held commercial use, but we can certainly start thinking about alternatives that are based more on proper incentives to surrender land, rather than by vesting arbitrary powers in the government to bring about such transfers.

The Land Acquisition Bill Of course, in a democracy, governments have to listen to people. Given the flurry of court cases and street protests, the government and the civil society are debating the clauses of an appropriate legislation for land acquisition. There are two major points in this regard.

84  Law and Economics

The first issue is the amount of compensation. We have already mentioned that the historical value of an asset, regardless of how high or low it is, has no significance on its future commercial value, and hence, its current price. Therefore, whether it should be six times or five times or even half of its last registered sale price has no meaning, unless we are confident that should the lands be aggregated and acquired, that will be the value it generates in the future. Indeed, if any form of restructuring of the use of land increases its value it should be done; what the government needs to worry about is whether the institutional mechanisms exist for enabling such value-enhancing transactions. There are two immediate problems with fixing levels of compensation as the current bill does. First, suppose that a particular modification to the use of rural land will increase the land value by four times. If a compensation of six times is required, one of two things may happen. The value-enhancing transaction may not be carried out because it will lead to losses in the government exchequer—it will pay out six times when it can generate value of only four times, and therefore, though such transformation of the land is value-enhancing it will not be undertaken. Worse, it will encourage corruption though the acquisition of not-so-valuable land at exorbitant prices if these lands are owned by the decision makers in the government—politicians, bureaucrats and powerbrokers. One may say that if one looks at all such transactions in the past, land values have multiplied more than six times. Indeed, over the following years, as this law is used, such valuations may be justified. However, if this process goes on, say for 5–10 years, it is not inconceivable that we would have had enough industrialization and urbanization so that any additional transformation will generate less than six times the value. Should not an improvement of five times the value be undertaken? Well, the law passed today will not allow that to happen, and law has this habit of binding one to the nature of all future actions! The second point is the use of public purpose. The bill continues to prevaricate this issue. In its current form, it allows for the transfer of land from one group of private owners—farmers—to another group of private owners—homeowners and industries—through government intermediation. The standard argument people make for affecting such transfers is increase in employment, increase in incomes, increase in government tax revenues for the provision of public goods, etc. Observe that when I open a shop, I do so for greater income; I employ people in my shop and generate employment; I pay taxes on my profit, which the government uses for providing public goods. Indeed, any private economic activity does all the things that one



Public Rights over Property  85

is referring to here when defining a public purpose. In a market-based, free-enterprise society this is what everyone does as a natural activity within the rules of the markets and the legal institutions that define contracts, fair practices and fraud. Why are we asking the government to play the larger role of an intermediary in such private land transactions?

5

Contracts and Their Enforcement Need for Contracts Contracts are agreements between two or more parties to carry out a predetermined action or set of actions (probably at predetermined times or in given sequences) with predetermined rewards (to be given at specified times). These agreements may be tacit, oral or written. By this definition, there is some element of contracting that is involved in all economic transactions. Suppose you buy a kilogram of potatoes from a grocer. There is a tacit agreement between the shopkeeper and you that the quality of potatoes you are buying has a minimum standard. It is tacit because you do not ask, nor does the shopkeeper say anything about the quality. You expect the price differences of different bags of potatoes to correctly reflect their quality (and the quantity in the bag). However, of interest to us in this book are only those contracts that are legally binding. In other words, if you do not like the quality of potatoes, you do not and usually cannot, lodge a complaint against the grocer in a court of law. You simply go on to the next grocer the next time you want to buy potatoes. Of course, part of the agreement with any grocer involves a legal commitment by the grocer that the weight of the potatoes in the bag is what it says on the bag. Why do we need contracts? If we want a commodity or service, why can we not simply go to the market and buy it (by taking the good or service and paying the price)? In any transaction between two individuals, both give, or transfer, something to the other party. However, there is nothing that says that this two-way transfer must happen simultaneously. For example, if you are planning to throw a party on your birthday, you will most likely place orders with the caterer a few days before the party. Caterers usually take some money in advance (part of the total bill payable for the food and other services at the party) at the time of ‘booking’ orders but, at that time, give you nothing more than a slip of paper that records how much you have paid



Contracts and Their Enforcement  87

as advance. Such a transaction is not completed at the time of placing the order but is stretched out from the time you place the order to the time the food is delivered on your birthday. There is, therefore, a time lag between your advance payment and the caterer’s delivery of the food (which may or may not happen simultaneously with your paying the remainder of the bill). There are many transactions of goods and services in which there is a lag between their delivery and the payment for them. A loan, for instance, is a transaction of this type—the lender gives the money upfront but receives payment in the future until the loan is fully repaid with interest (if any). Furthermore, it may not always be money first, goods later. Every time you turn on the light, you are taking electricity from the power supplier but will pay for that electricity only when the next bill is delivered to you. Two issues arise immediately. First, why are some transactions stretched out over time? Second, why is law relevant for such types of transactions? We consider them one by one. Contracts are usual in transactions that are specific to the parties, that is, the products or services being traded are not readily, or perfectly, substitutable. Many restaurants are uniquely positioned—at busy intersections or conveniently placed where people visit regularly because each is close to their place of work or neighbourhood. Different chefs in different restaurants have uniquely different ways in which they produce different dishes. You want your party catered by a particular restaurant because it is in your neighbourhood, has built a reputation by providing dishes that you have always liked or, has been written about in newspapers and food magazines. You cannot substitute easily the dishes produced by one restaurant with those of another. Let us take the example of the caterer and your birthday party. You have given a customized menu for your invitees; you have also specified the number of guests. Suppose the catering company is a restaurant that services parties outside the restaurant. Your menu, with each guest’s individual likes and dislikes is particular to your set of friends and the restaurant does not cater to them on a regular basis. Hence, there is a need for a customized contract between you and the caterer. The restaurant prepares food in keeping with what it thinks people will order on an average day—butter chicken as opposed to palak paneer or idli vada. Your customized menu is not in keeping with the composition that the restaurant usually prepares on a given day. The restaurant owner is worried that if he produces your menu and you decide to cancel your birthday party because your favourite friend cannot come to the party on that day, he will be left with more food than he can sell to his restaurant customers or, at least, with a composition of items that does not reflect the wants of

88  Law and Economics

his average customers. So, he wants you to share with him the risk that you may want to cancel the party at a later date but before the actual day of the party. He does so by taking an advance; should you want to cancel, he owes you nothing. In general, a client may require a good with certain specific features (of quantity and quality), and for this purpose, the producer may have to make investments specific to the requirement of the client. Once such investments are made, or the product is made, there may not be enough (or any) alternative clients for the same product. In such a case, the producer may be reluctant to make those investments (called relationship-specific investments) unless he is credibly assured that the products made will be bought by the client. If money is paid beforehand, then the producer is assured, because if the order is cancelled the client also suffers a loss (the advance payment), which is avoidable. However, what about you? You too require a credible commitment from the catering company. While the caterer is assured of your seriousness in wanting to throw the party, you also need to be assured that the caterer will deliver on the day of your party. This is where the law comes in; if the caterer does not deliver, you can go to (consumer) court and the strong arm of the law will ask the restaurateur to return your advance and compensate you for your loss of satisfaction in seeing your friends enjoy the feast that you had planned. The main point being made here is that in contract settings, what is ex ante (before, or at the time of, the agreement) value enhancing for both parties, may not be ex post (i.e., any time after the agreement but before all the sequential details of the agreement have been carried out) value enhancing for a particular party. In the example of the catering for the birthday party, at the price at which the caterer agrees to deliver food, both you and the caterer are making a surplus. Your happiness from the party is more than your loss of happiness from paying the agreed-upon money that you finally pay to the caterer; the caterer’s monetary gain from what you pay him is more than his total cost of providing you with food for the customized menu. This is what we call ex ante value enhancing; both expect to gain from the agreement than from not coming to an agreement. However, consider a time after you have paid the advance but before the party. If the advance paid was more than the surplus that the caterer would enjoy at the completion of the contract, it pays for the caterer to forget that he ever agreed on anything with you and not turn up at your party. This way he makes more money than if he had lived up to his agreement. In other words, ex post, the caterer is better off not keeping his part of the agreement. Or, suddenly, your worst enemy, with the knowledge that you



Contracts and Their Enforcement  89

have organized a birthday party and not invited her, walks up to the caterer and offers a larger surplus for an alternative party on the same day and at the same time, inviting all your other enemies and the caterer does not have the human resources to cater to both parties. Once again, it is in his best interest to cancel the agreement with you. So, ex post, after you have paid the advance and before the contract is completed, the caterer may breach the agreement or renege on the contract. This is what we call ‘opportunistic behaviour’ and in any contract this is always a possibility. You may wonder why, when law is based on ethical principles, why economists are worried about economic agents breaking their promises. Most people will say that if the caterer refuses to deliver after taking an advance, and does so as a deliberate decision, the caterer is cheating you and should not be encouraged to do so. Economists, as individuals, may not like people to cheat each other, but economists, as social welfare maximizers, must have a better argument than merely stating that cheating is morally or ethically undesirable. If the caterer takes your advance and does not deliver, you have lost your money and the caterer has gained that amount. Whatever you have lost the caterer has gained, and there has been a simple transfer of money from you to the caterer. In other words, there has been no net loss, or addition, to total surplus and economists should not be concerned with such redistribution of resources between two individuals. But economists are concerned because if reneging on agreements is allowed, you will never make a deal with the caterer where you pay first and the caterer supplies later. If that were the case, the extra happiness you get from organizing a perfect party and the profit the caterer gets from supplying food to the party will not happen. Since both would make a gain if the agreement were carried out as planned, conditions where such agreements cannot be made lead to inefficiency in transactions. Hence, economists want the help of the law to encourage contracting. Let us consider a credit transaction to see how the law plays its role in economic contracting. Consider a fisherman, who needs `1,000 a day to buy diesel for his boat and other materials to fish in the sea. He is expecting an income of `3,000 on that day. He does not have `1,000 in hand. Thus, he cannot generate the surplus of `2,000.1 Suppose he borrows this money from a lender and goes out fishing. Suppose, also, the lender asks for a 10 per cent interest on the loan. The borrower should return `1,100 to the lender and suppose he does. Then, the net surplus to the lender would be Strictly speaking, `2,000 cannot be taken as the surplus. In addition to `1,000, many other items, including what he would have earned by doing some job other than fishing on that day (his ‘opportunity cost’ of fishing) may have to be deducted to get the actual surplus. 1

90  Law and Economics

`100 (1100 – 1000)2 and the net surplus enjoyed by the borrower would be `1,900 (3000 – 1100). The lender and borrower together have generated a surplus of `2,000 and the economists are happy. Alternatively, suppose after returning from the fishing expedition, the borrower refuses to pay back the loan that he took from the lender. Then what happens? If we were to consider this two-person society, the lender has transferred out `1,000 from the total amount of money she possesses and the borrower has generated an income of `3000 for himself. The lender loses her `1,000 and the borrower gains `3,000, implying a total net gain to this duo of `2,000 (3000 – 1000). There is no difference between when the borrower repays the loan and when the borrower does not repay the loan in terms of the total gain to society from fishing. What does differ is the distribution of the gain of `2,000 between the borrower and the lender. This is the first observation. However, the more serious observation is the one we have already made (see the house purchase example in Chapter 1). If the lender knows that the borrower may not return the loan, she will be less inclined to give a loan and the fisherman will not be able to go out fishing and the societal gain of `2,000 will never happen. Thus, one of the parties has to be convinced that the other party will not cheat. Simply saying ‘I will not cheat you’ is not sufficient. Even if somebody is honest and wants to abide by the contract, there is a problem for her in convincing the other party. How can she credibly communicate her real intention to abide by the contract? Signing a contract, which will be enforced by a third party (ultimately the state and judiciary), which has authority over both the parties, is a way of assuring one of the parties that the other will not cheat. It helps even those parties that were planning to abide by the conditions of the contract honestly. Signing the contract becomes a credible way of communicating that one is indeed interested in sticking to the contract (under normal conditions). Thus, agreeing to be governed by the law makes legal contracting possible and this is the primary purpose of contract law. However, whenever two or more parties draw up a contract, they are looking into the future and the future is always uncertain. In the catering example, heavy rains lash the city on the day of your party and the streets are flooded, making it impossible for your friends to come to the party. The caterer is also unable to buy the special ingredients needed to cater to your 2 Actually, the net surplus would be this amount minus the maximum he could have earned by giving the money to someone else. Here we are assuming that he had no other alternative but to hold on to the money.



Contracts and Their Enforcement  91

menu. Taking these things into consideration, it is better for both you and the caterer to cancel the party and all the arrangements for the party. What happens to the contract then? We will consider such situations later in the chapter. What is special about this situation is that both parties are equally unaware, at the time of contracting, of the possibility of heavy rains on the day of the party. In other words, it may be entirely possible that after having signed a contract to carry out specific actions on a future date, both parties find it in their interest not to carry out their promises when the future date arrives. When this happens, the parties can sit down and decide to annul their agreement. Should law, charged with enforcing contractual agreements allow promises to be broken, simply because all contracting parties have had a change of heart? If the purpose of contracts is to increase efficiency and that of law is to make promises made in the contract credible, then there is no reason why law should step in and force the contract when it is no longer efficient to carry out those promises. In the catering example, both you and the caterer are better off not going ahead with the party when it rains and efficiency demands that the contract is allowed to lapse. Further, unless one party complains to the court that the contract has been breached by the other party, the court should stay out of enforcing the contract. Of course, the court can still play a role in determining what the parties need to do in case the contract has to be rewritten, but more about that later in the chapter. A more serious consideration is one where it is in the interest of one party to break the contract, but not another’s. One approach to this problem is, of course, a moral or ethical one—a promise once made must not be broken. The other approach is to ensure that appropriate compensation is paid to the aggrieved party (who is adversely affected by the broken promise) by the party breaking the promise. This is the path we suggested when considering the breach of house owner A’s promise to sell the house to B. The law can and does play a role in determining what the compensation should be. Finally, there is a third possible approach, where the court allows the promise to be broken. This possibility is not a desirable one because it reduces the credibility of any promise and credibility is essential to take any promise seriously, which in turn is necessary to encourage people to sign value-enhancing contracts.

What Are Acceptable Contracts? So far, we have been making a simple point. Promises made by different parties to each other increase aggregate value provided such promises are kept. While a promise made by A encourages B to commit to a future action

92  Law and Economics

should A keep her promise, once A has kept her word, B may find it better not to keep his. This is what we call ‘opportunistic behaviour’ by B. This is what A is afraid of and law steps in to assure her that B will be made to keep his promise. Hence, value-enhancing voluntary agreements are entered into by both parties. However, we still need to clearly define what, indeed, a value-enhancing voluntary agreement is. Consider the following situation. I put a (loaded and cocked) gun to your head and say ‘Your wallet or your life’. If you give me your wallet, you get to live and you value your life more than the money in your wallet. So you gain more by giving up your wallet. I gain the money in your wallet. Now, after I have left with your money you report to the police that I have mugged you. When the police catch up with me I tell them (and show them a videotape that I have recorded for evidence) that you handed over your wallet voluntarily after I promised that I would not kill you if did so. I had kept my promise not to kill you after you gave me your wallet, according to the terms of the agreement. So, where did I go wrong? The answer lies in understanding what enhances value. When we say something increases value, we have to be able to define the starting value from which the value is being increased. The general answer is: the starting value is that value had the contract not been agreed upon. However, this raises the question of what the starting point of the proposed agreement in the above example is—before I put the gun to your head or after I do so. The answer is ‘before the gun was put to your head’. How do we do that in a way that is a general principle and can be followed as a dictum in all similar cases? This is the essence of law as a discipline; it has to be based on certain very basic legal principles and things cannot be decided in an ad hoc fashion by wise heads sitting together. So, the law asks itself the more fundamental question: what should be allowed as contracts? Its answer is that agreements entered into ‘under duress’ cannot be considered voluntary agreements, or contracts. Economists love this approach because they can make an efficiency argument for this. Consider a society that allows contracts under duress. You will not carry any money in your wallet, you will not be able to buy what you want, producers will not be able to sell what you are willing to buy and the general level of welfare in the society will be rather low. So, not allowing any agreement that we made with my gun pointed at your head is good for economic efficiency. Indeed, for courts to agree to enforce promises made in a contract, they have to clearly state what they mean by contracts. When will they recognize that an agreement is a contract? Consider the case where instead of putting a gun to your head, someone plies with you a hallucinating drug, or an alcoholic



Contracts and Their Enforcement  93

drink, before you are asked to sign away your wallet to that someone. The courts would say that you were not in a position to make the agreement to give up your wallet. Under normal circumstances you would not agree to give up your wallet and making you ‘voluntarily’ do so when you are under the influence of some intoxicating substance is not considered a contract by the courts. Similarly, if you ask a child to do something for you in return for a candy and the child refuses to do so after you give the candy, the courts will not find the child to be at fault. The courts refuse to accept as contracts agreements made with a minor.3 For economists, transactions have to be voluntary because individuals know what is good, and hence what is better, for them. That is why voluntary transactions are value enhancing for all parties to the transaction. In other words, for transactions to be voluntary and desirable, people must be in a position to know what is good for them. So, it is good if the courts disallow as contracts agreements made with a child or with a person who was unable (at the time of making the agreement) to think clearly. In 1980, the Supreme Court of India decided on an appeal against the judgment in a case (Daya Shankar vs Smt. Bachi And Ors) in the Allahabad High Court. A person had filed an appeal against the orders of a lower court that had refused to acknowledge his ownership of a house. The plaintiff (in the appeal) maintained that he had received the house as a gift from the erstwhile owner. The defendants, who were the surviving children of the erstwhile owner, who was by then deceased, maintained that the plaintiff had obtained the gift deed when their father was unable to make such a decision. Earlier, when the owner was alive and had been taken ill, the plaintiff had come forward and got him admitted to the hospital and taken care of him. It was during the hospital stay that the gift deed was made and the patient came to know of it only after being released from the hospital. The defendants could successfully convince the court of their stand because they provided evidence that their father had made another gift deed after coming out of the hospital and after learning of the gift deed he apparently signed while in the hospital. In the second deed, he gifted the house to the defendants and annulled the first.

3 The court’s refusal to accept something as a contract has to be distinguished from its decision regarding what to do with such an agreement. If the court’s stated position is that it will enforce all contracts, then one way it can allow breach in an agreement is if it can argue that the agreement was not a contract. Observe that, refusing to accept agreements with minors as contracts, the court is NOT prevented from ENFORCING agreements that are beneficial to the minors.

94  Law and Economics

In Hamelo (Deceased) Appellant vs Jang Sher Singh (Respondent) 2001-(AR2)-GJX-0400-P&H, a widow complained that her land had been fraudulently taken away from her. The plaintiff had inherited agricultural land from her deceased husband and used it for self-cultivation. However, when she was ill and unable to work on the land, she had leased the land for one year to the defendants. She was asked to sign the lease papers and, being illiterate, she affixed her thumb impression to the lease deed. She was given some money by the defendants for signing this lease. After a year, when she wanted the land back, she was told that she had signed a 99-year lease. The court found in her favour because she was able to show that she was illiterate and unable to understand what she had signed and that the money given to her for the lease was too low to justify it being a 99-year lease. In other words, the court decided that what she thought she had signed was not what she actually had signed on and, hence, the extant (lease) contract was not valid. The defendants were asked to return the land to the widow. Here, the courts upheld a subtle but important point. The lease contract produced by the defendants was not considered valid because it did not satisfy the most important thing in a contract—agreement. If I agree to ‘A’ and you agree to ‘B’, we have no agreement between us; we misunderstand that we have an agreement because I mistakenly think that you have agreed to ‘A’ and you mistakenly think I have agreed to ‘B’. In other words, a contract that is one-sided, the terms of which were disliked by one of the parties but she was forced to agree, will not be enforced and that is efficient in economic terms. Going back to the ‘gun to the head’ example, the question to be asked is whether any party to the contract had the option of walking away from the contract. Observe that when I put the gun to your head and demanded your wallet, you were given only two choices—give up the wallet and live or do not and die (and possibly still lose your wallet because if I can take a wallet from a person alive I can certainly pick it off you when you are dead). The choice that was not given to you was whether or not you wanted to choose between these options at all. In other words, before deciding who should do what, you must be able to decide who should be the parties to a contract that you are willing to sign. When I put the gun to your head, I did not give you that choice. This is the crux of the cases just described. Were the parties signing the contract in a position to assert their interest rationally at the time of signing the contract? Normally, it is expected that children are not in a position to assert their interest and thus the courts invalidate all contracts signed by them. Similarly the validity of contracts signed by people who are mentally challenged is not upheld, and contracts signed by people under the



Contracts and Their Enforcement  95

influence of certain medicines or heavy doses of intoxicating agents while signing the contract, or aged people not in a position mentally to assert their interest, etc., are not enforced against them (or their interest). Excluding such contracts from the purview of enforceability is justified in terms of economic efficiency. The enforcement of a contract in court can be challenged for a number of reasons. First, the defendant can argue that there is no contract (‘formation defence’ is the legal terminology) or that the agreement between the parties does not conform to a contract in the legal sense. Yet another argument could be that while there exists a contract, the defendant is not liable since she could not perform as per the contract due to reasons beyond her control (‘performance excuse’ in legal terms). We need to analyse an economically appropriate approach to these arguments (formation defence and performance excuse). Let us take contracts signed under duress. The gun-to-the-head threat means if you do not participate in the contract, I will take way your life, something that you already have, and does not belong to me. Thus, in the event you do not agree to the contract, you would be in a worse position compared to the situation before starting the negotiation. You are, of course, worse off after the negotiation (you either are dead or you have lost your wallet). Thus, if such contracts were allowed, they would be a dead-weight loss to society. In the above case, one of the parties is put in a vulnerable condition by the other to extract a promise and form a contract. However, someone may be in a vulnerable condition to begin with; what about contracts signed by them? A person living in a remote village falls sick, and there is no ambulance or taxi service easily available at that time to take the person to the hospital. The neighbour has a car, but considering the vulnerable position (or necessity) of the sick person, extracts a promise from the latter’s family that they will pay her a large sum to take them to the hospital. How do we see the enforceability of such contracts? In this case, we need to take into account the incentive of the neighbour too. She need not offer to take the sick person to the hospital. Thus, if she demands excessive compensation (say, much higher than that for a taxi or ambulance for a similar service) as part of the contract, denying her the ‘extra’ compensation can be socially costly. This is because she (or others) may not offer such service in future, and for society it is desirable to have such people offering this service since the alternatives (like taxi or ambulance) are not available. Depending on the need for such emergency services, some people may have to be prepared to help others in need, and there may also

96  Law and Economics

be costs of such preparation. Suppose that a professional emergency service or ambulance is kept ready for such service in the area. Given the low frequency of such service, one may have to pay a higher charge for such an ambulance since it cannot be used for other purposes, and it has to be kept ready without actual use. The issue here is that invalidating contracts of this kind can be costly for the society as a whole; however, the compensation or the damages awarded in the event of breaking the promise can be adjusted to take care of such ‘necessities’ that may arise in future, and the cost of a socially desirable mechanism. However, one thing needs to be made clear. The question here is whether a person is in her ordinary state of mind and with all her faculties intact at the time of making the contract. It is not a matter for the courts to decide whether a person is doing the ‘right’ thing by that person. The law can ensure that no one forces you to enter into an agreement that is bad for you; it cannot ensure to protect you from your losses. For instance, if some of your best friends cannot make it to the party that you had planned to be served by the caterer, you cannot unilaterally decide to cancel the party and nullify the contract that you have already signed with the caterer (without forfeiting your advance). Suppose you feel that a piece of property is being sold very cheaply and decide to buy it so that you can sell it for a higher price later. You based this decision on a wrong judgement of how the property market will pan out; you made a mistake. Once you determine this, and it happens after you have bought the house or after you have made a down payment to prevent the current owner from selling it to anyone else, you cannot ask for the money that you have already paid to be returned to you. The courts will not protect you from suffering the consequences of wrong decisions. How would economists justify this? Recall the distinction between ex ante and ex post. Parties signing a contract are looking at the extra value they will get in the future. The future is uncertain and so they base their judgements on their expectations. The contract will be termed voluntary and ex ante efficient if all parties expect positive value enhancement. Ex post, or after the uncertainties are resolved, some parties may wish they had not signed the contract. Take a specific example. You promise to buy shares tomorrow of a particular company. You make this promise to a broker and both of you decide on a price today that you will pay tomorrow and the number of shares the broker will give you at that price. Your reason for signing the contract is that you think the price of the specific share in the stock market tomorrow will be above the price you have agreed upon today with the broker. Thus, by agreeing on the price today, you will save some money tomorrow. The broker, on the other hand, thinks that the price tomorrow will be lower than



Contracts and Their Enforcement  97

the price agreed upon today. So, he will gain by buying the shares at the lower price tomorrow and selling you at the higher price that has been agreed upon today. Both act based on expectations of making gains and thus, voluntarily make the contractual arrangement. However, note that tomorrow’s price cannot be both higher and lower than the price agreed upon today. So, you or the broker will certainly make a loss at the same time that the other makes a gain. The courts will not protect either one of you from making a loss. For economists, this is a good thing. Such trades are essential for stock markets to function efficiently and efficient stock markets are good for investment and growth and that increases aggregate value in the economy. In other words, if both sides to the contract take decisions based on common information, or common lack of knowledge about the future, it does not matter. But, suppose one person has more information. The broker, for instance, has done a careful analysis of the shares he trades in and has superior knowledge about the share price tomorrow. Should he then not trade with you? If he does not, then he has no gain from investing his time and effort in collecting information. Once again, if there is more information among market players, markets tend to be more efficient and so the broker should be allowed such incentives, even if this means he has more information than you. If you want to counter that, you can yourself invest time and effort to get more information. This brings us to insider trading, a term usually associated with stock market deals where a person accused of insider trading is charged with buying or selling shares based on information that is not public and has often been acquired by that person in the course of activities that the person had been contracted to perform. For instance, an employee of an institution working on mergers and acquisitions gets information about the possible effect on a client’s share price owing to a merger activity and trades on these shares because of the gains that can be made with this information. Similarly, the top management of the client, with knowledge of the merger, may also make some gains by trading their company’s shares before the merger decision is made public. This is insider trading and is not allowed by law. First, given the person’s own interest in gains from share trading, decisions may no longer be objectively taken. Second, the gains made are not due to time and effort put in to obtain this information but the result of activities unconnected with information gathering. Both these destroy the confidence of investors who are not privy to such information; this seriously and adversely affects the efficiency of the stock market. Getting back to the informational aspects at the time of signing a contract, an important point to remember is that it is entirely possible that what looked

98  Law and Economics

like a good deal at the beginning may not look as good as time unfolds. We have already stated the position when it comes to future contracts on share trades. Now consider a situation when both parties to a contract find that what they agreed upon, if carried out in full, would destroy values for both of them. Or, as in the example of the sale of a house in Chapter 1, the saving to one party from not carrying out the contract is larger than the loss to the other party. If A breached the agreement with B, the loss to B (own valuation minus the price promised to A) is less than what A gains (the difference between C’s and B’s price). We consider these issues in the sections that follow in this chapter.

Ex ante and Ex Post Efficiency Let us take a very simple contract. A and B agree that A will deliver something (of a particular quantity and/or quality) at a specified time in the future in return for a specific reward that B will give at a particular time. Here A may be called the promisor and B the promisee. If either A or B fails to do what is agreed upon, there will be clauses in the contract penalizing them. For example, as per the contract, B pays money in predetermined instalments to a building company, A, and it agrees to deliver an apartment constructed on a particular plot with certain specifications at a specified time. What happens if the company fails to deliver on time? Six months before the stipulated delivery time, the builder encounters a situation that had not been foreseen, such as unexpected shortage of some materials or labour unrest in the area or something similar. This leads to disruptions in the construction work for two months. It becomes clear that under normal conditions the builder cannot deliver the apartment in the remaining four months. However, if A hires more workers to work around the clock, it may be able to deliver the apartment on time. This, of course, entails additional costs, costs that had not been foreseen at the time of making the contract. As a rational entity, it may consider the consequences of delaying the delivery of the apartment by a few more months. Suppose the contract stipulated that A has to pay a fixed amount of money to B for each day that A was late in delivering the flat. Such clauses are often termed penalty clauses. Suppose, without additional hire, A will be two months’ late in handing over the apartment and the penalty would then be `x. The cost of hiring additional workers to finish the apartment on time is `y. If x > y, it pays for A to hire the additional workers; otherwise, it will allow the delivery to be delayed and pay the penalty.



Contracts and Their Enforcement  99

This is privately efficient for the builder, since it solved his problem cheaply. Is it efficient for the buyer of the flat? In this case, we should presume that it is so, since the buyer has agreed beforehand that the penalty for any delay is the payment of a specific amount of money per day for the period of delay. If the loss to the buyer due to the delay is compensated through the penalty paid by the builder, B has no complaints. Let us be more specific about this example. B is in a rented house at the time of signing the contract. So, if A delays delivery, B has the additional cost of paying rent during the period of delay. When the contract was drawn up, B was not expecting to pay rent beyond the date on which the apartment was to be delivered to her. The penalty clause x could, therefore, be that A will compensate the additional rent B would have to pay. Observe that whether A pays this rent or B pays this rent is a matter of pure transfer; there is an additional rental cost and they have to decide between themselves who has to bear the cost. Ex post efficiency demands that if the cost of rent is lower than the cost of hiring more workers, additional workers should not be hired. Observe that this is independent of what the penalty clause states; the penalty clause only decides what happens if there is delay, not what is efficient. Both law and economics would want the penalty clause to be such that efficiency is maintained. How does one ensure that? One of the biggest problems in contracting is that no one can be sure how the future will unfold. In particular, neither A nor B knows whether there will be a workers’ strike. Nor do they know whether there will be a materials’ shortage or any of the other things that can cause a delay. What can be said, however, is that A, who deals with labour and materials on a daily basis, may have more information about these possibilities than B. Consequently, A should be in a better position to plan for these exigencies and be held accountable for inadequate planning. However, what if the cause of the materials’ shortage is the breakdown of transportation lines because of a political unrest, say a riot, somewhere else. This is something that A could not have anticipated or is no better at anticipating than B. Of course, the penalty clause could be more refined and state that a penalty is due from A to B if the delay is due to situations that A was better at anticipating; otherwise, there is no penalty. The more fundamental question is whether penalty clauses should be allowed in a contract. The penalty clause was voluntarily agreed upon by both parties, in the hope that courts would enforce it. B is assured by A that there will be no avoidable delays and this encourages her to sign the contract which is value enhancing. However, one thing is immediately obvious. Suppose A is pretty confident that there will be no delays and so is agreeable to any amount of penalty that B wants included in the contract. B wants the

100  Law and Economics

penalty to be high to ensure that A has absolutely no incentive to delay. Let the penalty be z and let us assume that z > y > x (see above). Recall that x was the actual cost suffered by B due to the delay and ex post efficiency demands that A should pay this amount to B. However, given the penalty clause z, A will make every effort to prevent the delay and it can do so by incurring the cost y < z. This will result in an outcome that is ex post inefficient because society is undergoing the cost y when x would have sufficed. Courts are often nervous about penalty clauses and they provide an additional reason for this. Observe that with a penalty clause of z, B actually gains if A delays, by an amount (z – x). So, B would pray for the riot or, worse, may quietly instigate one to make the gain! This will cause huge losses to society and not simply by the amount (y – x). In other words, law points out that a penalty clause is meant to avoid delays; it should not become an incentive to cause delays! The law wants to resolve this issue by saying that it is the courts that should decide on what the penalty should be and not the parties signing the contract. There are two opposing reasons that say that penalty clauses should be allowed even when z > y > x. While x is the cost of rent that B has to pay in the event of a delay, there is another cost that she has to bear. Individuals do not like fluctuations in their consumption and the possibility of delay is causing this fluctuation—the utility of living in one’s own house from a specific date versus that of living there from a later date. The rental cost does not meet this additional cost (due to what economists call risk aversion). So the actual cost is greater than x, and z could just be reflecting that. Alternately, suppose B is a new builder and, hence, cannot depend on past reputation. It is willing to commit to a higher penalty to signal that it is confident of delivering the apartment at the specified time and the buyer knows that if the courts implement this penalty in case of a delay, the builder will lose a lot of money. Since the builder has still agreed to it, it must be because the builder has taken adequate precautions and is signalling its confidence in delivering on time (Cooter and Ulen 2004). There is yet another reason why the cost of delay to B could be greater than x. Suppose B had organized an elaborate housewarming party immediately after the contracted delivery date and she had paid the caterer an advance to enter into a contract for the delivery of a customized menu. Now that the apartment has not been finished, the party has to be cancelled and she has lost the advance that she had paid. A final point needs to be made when discussing ex ante and ex post efficiency, which is, what happens when what was considered ex ante efficient becomes ex post inefficient and both parties do not want the promises made by either party to be carried out? For instance, after the



Contracts and Their Enforcement  101

contract is signed with the builder, it is found that the land on which the apartment complex is to be built is disputed. That is, there is a previous owner of the land who claims that he has not sold his land to anyone. Both A and B, when they come to know of it, would much rather not have the contract and move on to something else. What should a court that has promised to enforce contracts do in this case? The answer really is very straightforward—nothing. Remember, in our legal system, courts mostly deal with disputes. If there is no dispute, the courts do not get involved and there is nothing to settle. In this example, if A and B mutually agree to annul the contract, no one will run to court complaining that the other is not carrying out the contract.

Information Issues in Contracting As has been already discussed, in a contract, one party may have ‘some’ information that the other party does not have. The seller of a plot of land may know that it is flood-prone, but the buyer may not have that information. The seller of a used car may know the repair and maintenance history of the car while the buyer may find it difficult to get this information unless the seller tells her. A buyer of a plot of land may know that a new commercial centre is coming up in the neighbouring plot (which may increase the value of the first plot), and the current owner of the plot may not have that information. A buyer of an old artefact may know that it could be sold at the international market at a very high price, but the seller may not have that information. When such parties contract, and one of the parties later comes to know that the other had information which was not divulged at the time of contracting, she may want to approach the courts for remedy. How should the courts address this issue to enhance economic efficiency? There are different types of information. Not revealing certain information can be costly not only to the other party signing the contract, but also to society at large. This is so since such a practice would encourage people to invest heavily in getting to know more information or verifying what is provided, and in certain cases, if such information-seeking costs are very high, parties may be reluctant to get into contractual exchanges. For economic efficiency, contract law should facilitate mutually beneficial exchanges by reducing transaction costs including information-seeking costs. In this context, recall the argument made in an earlier section of this chapter regarding insider trading.

102  Law and Economics

There are some obvious cases where contract law or its enforcement by the courts should encourage the revealing of information (or set aside contracts or award damages for not revealing information). The seller of a product should reveal related safety information and failure to do so should be a ground for awarding damages against the contract. The seller intentionally hiding information that would reduce the value of the contract for the buyer may be another case. The seller not revealing the breakdown history of a car may be an example. In this case, buyers of such cars may have to invest heavily to get such information from other sources, or such potential costs to get information may inhibit mutually beneficial exchanges. However, there are some cases where the courts may not insist on disclosure of information for the sake of economic efficiency. Assume that an investor spends money and effort to acquire information on selling antiques in different foreign markets. He procures antiques from India. Someone from India gets into a contract to sell the antiques to this merchant without knowing its international price. Later on, the seller comes to know that the merchant has sold the antique in the international market at a price much higher than what was given to him. The seller sues the merchant for not revealing the actual information on the international price of the antiques. Should the court take any action against the merchant for not revealing information (just like in the case of safety information)? The merchant has invested time and/or effort to procure information so that the product can be passed on to consumers who value them more (it is enhancing the value or surplus through the exchange process). Economic efficiency requires contracts to facilitate surplus-enhancing exchanges. There will not be any return on the information-acquiring investment made by the merchant, if he freely reveals the information that he has procured. Then he will not have any incentive to procure information. Contract law should facilitate surplus-enhancing investments. In this case, the merchant’s information-acquiring investment must be rewarded. Thus, the greater price that he obtained by selling the antiques in the international market can be seen as a genuine reward for his information-seeking investment. However, we need to be careful about rewarding all information. Assume that a government official comes to know (much ahead of others, since he was involved in decision-making process) that an airport is going to come up at a particular location. He starts buying properties in the vicinity of the airport, knowing the possibility of a drastic price increase after the implementation of the airport project. This is very similar to insider trading and should not be allowed. In a sense this information is not productive. Such information merely causes redistribution—the current landowners near the airport would



Contracts and Their Enforcement  103

have benefited from price increase, and instead, the official benefits at their cost. Similarly, with insider trading, some shareholders, who do not have the private information that the insider trader has, lose what the trader gains. There is no economic efficiency in carrying out this mere redistribution. Conversely, there can be costs to the society. If society encourages such acquisition of private information (which enables only redistribution), then many people may invest their money and effort to get such information, diverting scarce resources away from productive or surplus-enhancing activities. This is not efficient and has to be discouraged.

Remedies for Breaching Contracts In a number of cases, there may be clauses in the contract itself defining what each party should do, if she cannot perform according to the terms of the contract. It is also very likely that there may not be such specification in the event of many contingencies in the contract. In general, there always exist such gaps in any contract; to economists, such contracts are incomplete. There could be two reasons for this. First, many contingencies are often not visualized by the contracting parties and hence, their agreements do not specify what happens when the contingencies occur. Second, even when a contingency is visualized, there may be a need for elaborate negotiations to arrive at contractual clauses on what each party should do in the event of such a contingency. For example, suppose the car in which your contracted caterer was bringing the food to your house caught fire due to a design fault in the manufacturing process of the car. We know this can happen but should we negotiate how to write a contract that distributes responsibilities in such a situation? Let us say that the probability of the delivery car catching fire is 0.00001. The cost of negotiating what should be done in such a situation is the value of time of each of the parties. Let us say that you will have to take half a day off from work to meet with the caterer. You are part of the new breed of daily labourers in the service sector (or consultants) who get paid by the hour and half a day means a lot of money lost in wages. Thus, the transaction cost of trying to make the contract ‘complete’ on this contingency is too high (i.e., the loss of time spent negotiating is much more than the gain you expect from a successful negotiation since the probability of the event is singularly insignificant). So, you take the risk and convince yourself that it is not worth your while to worry about not having contracted for the possibility of the delivery car catching fire.

104  Law and Economics

Of course, should such a thing happen and you and the caterer cannot come to an amicable settlement, the courts must be standing by to settle your dispute. In short, a contract can be breached both by a deliberate act of one of the parties or because of unforeseen circumstances outside any of the parties’ control. The first thing the courts can do if the breach is a deliberate one is to require one party to keep its promise. In this case, the court orders the party that has broken its promise to carry out precisely what it was supposed to. In most cases, such specific performance would suffice, but for the fact that there is a time lag between when the promise was supposed to have been kept and when it was actually enforced by the court. This time lag may be costly to the party that was to benefit from the promise being kept at the right time and hence, may need some compensation in addition to the specific performance. Consider the sale of the house example considered in Chapter 1. When A refused to sell the house to B even after having agreed to do so, the courts could take the view that specific performance is called for and A should be made to sell the house to B. However, in that example, something happened after the contract was signed between A and B, namely, the appearance of C. This made the ex ante efficient contract between A and B ex post inefficient. In fact, as the discussion of the example showed, in the presence of transaction costs, specific performance would make the overall situation inefficient. This brings us to try and understand exactly how specific performance works. A court order of specific performance can be interpreted as an advice, or encouragement, for negotiation between the contracting parties. Indeed, it facilitates negotiations. Both A and B know what will happen if they go to court; both know how much each will gain from specific performance; and, most importantly, they both know that there is a dead-weight loss to them from specific performance. Remember that specific performance will mean A sells to B followed by B selling to C which has a greater cost to all three than if sold directly to C. By reducing the cost, they will have more surpluses for all of them. However, specific performance helps define the minimum B must get to keep her from going to court and complaining about A. This actually helps in the negotiations because it identifies to all parties B’s fallback option. Thus, it is efficient if the order of specific performance and its interpretation are aimed at negotiation between the parties. However specific performance is unlikely to work in the case of already breached clauses in the contract (where damages have already been inflicted). Therefore, as remedies for such breaches of contracts, courts may have to decide on compensation to be paid for breach. How are such damages estimated?



Contracts and Their Enforcement  105

Three Types of Damages Instead of specific performance, the courts can make the promise breaker compensate the aggrieved party. What is the damage to the aggrieved party when the other does not perform according to the contract? There are multiple ways of conceptualizing this damage. First is the loss to the victim, compared to a situation where the other party has performed as per the contract. Let us take the example of the builder and the buyer of the apartment again. The builder has agreed to deliver the apartment by a specified date and there was no penalty clause for late delivery in the contract. The apartment is not delivered on the specified date and there is a breach of the contract. Given what has been discussed before, the compensation for breach is to be assessed keeping in mind the gain the flat owner expected had she received the flat on the specified date. The damage due to the delay in this case may include the rental costs to the flat owner of staying in a similar flat (with similar features) from the delivery date agreed upon in the contract to the actual date of delivery. Add to that the extra joy from living in one’s own home compared to rented accommodation. This is called expectation damages; she was expecting to stay in the apartment from the agreed upon delivery date, but could not do so, and hence she is entitled to damages that include the rental cost of staying in a similar flat during the period of delay. There are also alternative concepts of damage. This is based on thinking about what the flat owner would have done if she had not entered into the contract with this particular builder. Assume that she approached this particular builder after approaching other builders for a similar apartment but the most that these alternative builders could offer was to deliver the flat at a later date than the one promised by this builder. That would mean if she had not entered into the contract with this builder, the next best alternative for her was to contract with another builder with the earliest delivery date being after the one promised by the one she signed up with. Suppose the next best alternative would have given her a similar apartment in a similar complex two months after the one she signed up for. If the apartment she contracted for is delayed by three months, then her compensation should be the additional month’s rent that she now has to pay (three months minus two months) plus the difference in cost between the two builders. This would be compensation based on alternative costs, and is called opportunity cost damages. Alternately, suppose the builder decides to abandon the project after some time. Assume that the apartment buyer was to pay `2 million in four quarterly instalments. For this, she had taken a loan of `1 million from the bank at an interest cost of 10 per cent. Now since the project has been

106  Law and Economics

abandoned halfway through, the buyer has ended up with no apartment and a debt she owes to the bank. She has to return the loan and if she wants to pay back the loan before it is due to save on the interest burden she has to pay an additional charge. Banks usually charge a pre-payment penalty and this is an additional cost (though less than the 10 per cent interest she would have to pay if she continues with the loan). Let us take her back to the time when she had not signed any contract. The difference between then and the abandoned project is that she has taken and repaid the loan incurring the cost of the loan. She has incurred this additional loan cost relying on the builder’s ability to deliver her the apartment. The builder may be asked by the court to return the money she has already paid (instalments that have already been met) plus the additional costs due to the loan she took from the bank. These are reliance damages—the cost incurred for relying on the contract. It is rather easy to see that expectation damages would be the highest, followed by opportunity cost damages, and reliance damages would be the minimum. Expectation damages makes the victim indifferent to the performance and non-performance of the contract. Thus, she gains whatever surplus she is expecting from the best contract that she has entered into, even when the contract is broken (but with compensation equivalent to expectation damages). If the compensation is based on opportunity cost, it is likely to be less desirable than expectation damages. This is so, since the compensation is based on the next best alternative to the contract relationship (and since she has opted for the contract, it is evident that the agreed upon option was better for her than the available alternatives). Reliance damages would be least preferred, since it compensates her only to the extent of making her indifferent between contract and ‘no contract’. The next best alternative to the current contract would have given her more surpluses compared to the ‘no contract’ situation.

Default Rules As mentioned earlier, it is difficult to make ‘complete’ contracts and contracts are likely to be incomplete. Continuing with the example of the builder, suppose that after the building has been handed over to the apartment owner, it is hit by an earthquake, leading to immense structural damages. Now what happens? Given the rarity of an earthquake (this is in a region that does not face a high risk of earthquakes), the buyer had not asked for a specific clause in the contract concerning this and neither had the builder thought it necessary to safeguard itself from such a disaster. There is a gap in



Contracts and Their Enforcement  107

the contract on how the loss should be allocated between the builder and the apartment owner. The apartment owner feels that the builder was supposed to have handed over an apartment that developed no structural damages for at least 30 years. So she sues the builder for unsafe construction. The court could tell the builder that apartment owners paid expecting 30 years of no structural damage, so the builder should repair everyone’s apartment. Alternately, the court could tell the apartment owners that the earthquake is outside the control of the builder and since there was nothing specified in the contract about this eventuality the builder has no further responsibility. Since this is a rare event, economists have a ready answer for this. If the court holds the builder responsible and it knows that this is what the court will say, it will insure itself against such a calamity and include the insurance premium as part of the cost of the apartment. Similarly, if court ordered that apartment owners should pay for the earthquake damage, and they knew it, they would have bought earthquake insurance. Indeed, economists argue that transferring risk from those who are more averse to it to those who are less averse to it is efficient (Varian 1987). Insurance companies can bear more risk because they can diversify their risk across many insurance buyers. In other words, as long as the court has a consistent stance, and the contracting parties know about it, there is no problem. However, when events are not rare (i.e., they are not very low probability events), and/or those wanting insurance face correlated risk (when the earthquake hits, everyone’s apartments are damaged, as well as those in other neighbouring buildings whose owners may have all bought insurance from the same company) then the insurance market is not a solution to the problem. Here, the courts need to take a more specialized stand. Again, the best thing to do is to see what is least costly to society as a whole. Suppose the builder could design the building to be earthquake resistant and the cost of doing so at the time of constructing the building is much lower than what anyone has to pay to repair the damages after the earthquake hits the building when it is not earthquake resistant. For instance, if it costs `0.2 million to make the building earthquake resistant and the earthquake causes a damage of `10 million to an ordinary building, then it makes sense to fortify the building if the probability of an earthquake is 1 in 50. Also, a single apartment owner cannot fortify only her apartment if the whole building is not made earthquake resistant. Simple logic demands that the court needs to encourage the builder to build an earthquake-resistant building even when this is not explicitly stated in the contract. Thus, efficient default would allocate the loss to the builder, if the purpose were to have an efficient contract law.

108  Law and Economics

There is another dimension of efficiency here. Assume that the existing default rule is inefficient and the apartment owner bears the full loss of the earthquake. She will foresee this possibility and may insist on having a clause that covers her for this while signing the contract. Thus, she may insist on filling this gap in contract ex ante, even though it is rational to leave the gap considering the transaction costs involved in allocating the risk ex post, as mentioned earlier. In principle then, even when a contract is silent on a particular event, should the dispute come to court, the court will allocate the liability for that event to the one who was the most efficient, or had the cheapest means of preventing a loss. Given this principle, it becomes the default rule that everyone accepts without spending undue time haggling about what should happen when such an event takes place.

6

Economics of Tort Law Someone is building a multi-storey house and a brick falls on to the road where people are walking and injures someone. A car driving down the road hits a pedestrian. In both cases, society suffers a loss equal to the injuries caused. Society also suffers a loss if the car does not hit a pedestrian, but a lamppost that ceases to work. Both the injured people and the destroyed lamppost have to be fixed so that things could go back to where they were before each of the three events. The building contractor and the person injured by the falling brick had no prior agreement regarding who bears the cost of the injury; nor did the car owner, or driver, with the pedestrian or the owner of the lamppost (municipal authority). There are a number of obvious issues that come up in each of the examples cited above. First, of course, is who is responsible for each of the events or how to determine who caused the damage. Second, who should bear the cost of the injuries to the persons or the damage to the lamppost? Obviously, one should find answers to these questions in such a way that, if feasible, they would reduce the incidence of such events as much as possible. However, if society wants to reduce the occurrence of such events, it may be better to approach the problem in a slightly different way rather than simply trying to apportion blame. There are two immediate difficulties in obtaining straightforward solutions. First, one has to determine what one means by causing harm. Did the contractor cause the injury by allowing the brick to fall or did the person on the street cause it by walking beside a building being constructed? Second, was it an accident or a deliberate act? That is to say, we cannot always be absolutely sure that if something were done, no accidents would occur. Regardless of what the contractor does, there is always the possibility that some bricks will fall and one can only hope that it misses everybody on the ground. It may be noted that the damages for harms like those mentioned here cannot be decided according to what we have studied so far—contract or property law. Contract law cannot be used, since there was no pre-existing

110  Law and Economics

contract between the vehicle driver, or the contractor, and the accident victim. It is almost impossible to have a contract between vehicle drivers and potential victims, ex ante, on what each of them will do in the event of an accident. Given that signing contracts is costly, this would make all activities extremely costly to undertake. Property law can be used, if we treat the injured leg or hand or any other part of the victim’s body as a property of the victim, and the accident as a transgression into that private property. Since this is a bit unusual, we need other legal means to sort out conflicts regarding such accidents. Tort law fixes liabilities; very simply put, if some loss has occurred, who should pay for that loss. There is a subtle distinction between what are usually covered by tort law and what are not. The good news for economists is that they can come up with a reason for this distinction. So, suppose that someone misses a step while walking down the stairs, falls down and breaks a leg. The leg can be set right and made as good as new (or at least, set to what it was before the person fell down) by doctors in a hospital. The person, however, suffers a loss. A tangible way of measuring this is, of course, the hospital bill. You could add to that the loss in wages on account of being in a cast, especially for a daily labourer. If the person is not compensated for the bill, she bears the loss. If hospitals dispense treatment free of cost, as is the case in many countries, then all taxpayers compensate the costs of this one person. Everyone finds this acceptable because no one knows who will be the next person to fall down the stairs and so everybody shares the risk. This solution can be ‘bettered’ if instead of government-run hospitals we had private hospitals and perfectly competitive (health) insurance companies. This last point follows in a perfect world because insurance companies are, then, competitive and risk-neutral. The real choice here is between the government and private insurance providers. In the first case everyone pays taxes and the risk of breaking a leg is covered by taxes. In the second case, each individual buys insurance and the insurance company covers the risk. Observe that we have not referred to tort law but discussed two methods of addressing the costs of such accidents as, essentially, two different policy options. Why have we not considered tort law as a tool for apportioning the liability for such an accident? We could say that if there were no stairs, the person would not have fallen. The labourers who built the stairs did so because the housing contractor asked them to do so. The contractor built the stairs because the architect designed the stairs. The architect designed the stairs because the owner of the house wanted a structure that has more than one storey and did not want an elevator shaft. Which one of these various persons has caused the damage to the person who fell down the stairs? There are two



Economics of Tort Law  111

reasons why we do not want to follow this path of trying to identify which particular action led to the accident. First, it is extremely difficult to pin blame because many persons, taking various actions, are involved. This obviously makes it very difficult to apportion blame. Second, even if everyone took the utmost care, people do fall down the stairs every year for reasons that are beyond anyone’s control. After all, that is what we call a pure accident—something that cannot be perfectly foreseen and, hence, cannot be completely avoided. So, unless, someone had used a material that makes stairs slippery, or thrown water without warning those using them that the surface was wet, we really cannot say that someone committed an error. In other words, one does not need tort law in cases where the probability of an event is outside anyone’s control; some form of insurance usually does the trick of attaining efficiency. Economists studying tort law are more interested in seeing how they can prevent accidents, or losses in society, in an efficient (least cost) manner.

Socially Desirable Levels of Precaution Economists are interested in the use of tort law because it can be used to efficiently avoid events that cause loss. Unfortunately, ‘there is no free lunch’. If accidents are costly and a value-conscious society wants to avoid them, it must face the costs of avoidance. These costs are essentially the levels of precaution that one must take to avoid accidents. Before we come to who causes what, let us take an ‘objective’ view of things and recognize that accidents can often be avoided by both the victim and the perpetrator. The contractor can cover the building where work is going on and cordon off a part of the road to prevent pedestrians from straying into the danger zone and the car driver can drive more carefully. Similarly, a person can cross over to the side of the street where the construction activity is not taking place; a pedestrian can look right and left (and right again) before crossing the street. As economists, we know that these things do not happen by themselves simply because economic agents do not like undertaking costs. The contractor has to pay for the cover and the barriers that cordon off the space under the building, while the pedestrian loses time and utility by crossing over to the other side. Let the level of precaution taken by the potential perpetrator be y and that by the potential victim be x and their respective costs be Cy (y) and Cx (x). Let the costs incurred by each increase with the level of precaution taken, that is, Cy(y2) > Cy(y1) if y2 > y1 ≥ 0 and, similarly, Cx (x2) > Cx (x1) if x2 > x1 ≥ 0.

112  Law and Economics

Given the levels of precaution taken by each agent, the chance of an accident happening is given by p(x,y) where p(.) denotes probability. The probability function satisfies the following properties: 1. 0 ≤ p(x2, y0) ≤ p(x1, y0) < 1, where x2 > x1 ≥ 0 and y0 ≥ 0 2. 0 ≤ p(x0, y2) ≤ p(x0, y1) < 1, where y2 > y1 ≥ 0 and x0 ≥ 0 3. p(0,0) = p– ≤ 1 and p(x, y) ≥ 0 Property (1) says that the probability of the accident goes down as the potential victim takes more precaution and (2) says the same thing about the potential perpetrator. Property (3) simply reiterates that p is probability. Let the cost of the accident, say hospital bill, equal A. Then for any pair of precautionary levels, the expected cost from the accident is p(x, y)A. Society would like to minimize the cost from the accident as well as that arising from the levels of caution of each of the agents. Formally, the problem is to minimize p(x, y)A + Cx (x) + Cy (y) Suppose this gives us a solution (x*, y*), that is, the potential victim should take precaution at the level x* and the potential perpetrator at the level y*. There could be a third level of precaution. Pedestrian crossings may be well marked, street lights and walk signs may be added and police may be made to monitor traffic to ensure that cars do not jump lights and pedestrians do not jaywalk. In this case, the probability of an accident is dependent on a third variable, say z, the level of costs incurred by road authorities. This cost can be denoted by Cz(z) with Cz(z2) > Cz(z1) if z2 > z1 ≥ 0 and the probability function as q(x, y, z) with q(x0, y0, z2) ≤ q(x0, y0, z1) if z2 > z1 ≥ 0.1 The relevant expression to be minimized is then given by q (x, y, z)A + Cx(x) + Cy(y) + Cz(z) and the solution by (x', y', z' ). This is the model economists would like to use to analyse the efficiency of tort law.

Implementing the Solution through Tort Law Observe two important aspects of the approach outlined in the previous section. First, we are not assuming that an accident can be entirely avoided, 1 With respect to x and y, the probability function q(.) exhibits the properties given in (1) and (2), suitably modified to take into account the fact that the probability is now dependent on three, rather than two, variables.



Economics of Tort Law  113

that is, we are not implying that there exist levels of caution to be taken by one or more entities such that the probability of an accident falls to zero. Indeed, we are not even minimizing the probability of the accident; instead, we are trying to minimize the costs associated with the accident—the cost of prevention plus the cost, should the accident happen. The reasons for doing this are (a) to acknowledge the fact that accidents do happen, we can only try and avoid large costs associated with them and (b) even if accidents can be entirely prevented, the costs of prevention should be sufficiently lower than the cost due to the accident. For instance, road accidents can be completely avoided by ensuring that there are no cars on them. But, the efficiency cost to society of not having any cars is far greater than the efficiency gains from not having any road accidents in this case. Second, we are allowing for the fact that the potential victim too can be held responsible for the harm an accident causes. For instance, suppose the pedestrian runs out into the street when the car has a green light and is going through it. Or, the person walking on the street looks up, sees the brick falling, and then deliberately runs underneath it. Our approach makes these questions irrelevant and focuses more on who had the last chance to avoid the accident. The economists’ view on tort law helps develop a framework of incentives that encourage an efficient way to avoid losses. The model described in the previous section is a general model. To understand its implications for tort law, a law that fixes liabilities (who should pay for the damage), we will take special cases of this general model. The first special case we have already discussed, the case where accidents happen regardless of precautions taken by anyone. In this situation, the probability function q(.) is independent of who does what. People fall down the stairs regardless of the general levels of precaution by the building authorities or owners, as well as the person using the stairs. In such cases, we have insurance companies or a government-owned group insurance plan to tackle the issue. Which should be used is a matter of economic analysis and policy and we will not be concerned with them here. So, let us move on to situations where probabilities are affected by levels of precaution to see how the law can affect outcomes. It is immediate from the model that whoever has the cheaper cost of avoidance should take that preventive measure. In other words, suppose that the probability of an accident can be reduced from q0 to q1 in two different ways. The potential victim can take some precaution at a cost of c'x, while the potential perpetrator can do so at a cost of c'y. If society wants the probability of the accident to be brought down to q1 from q0, it can either ask the potential victim or the potential perpetrator to take precautionary

114  Law and Economics

measures. Clearly, if c'x < c'y, it makes (‘efficiency’) sense for the potential victim to take the precautionary measure. Will that happen? Well, let us see how economists answer this question when there is no law. Taking this additional precaution saves the potential victim an amount equal to (q0 – q1)A for a cost c'x. As long as (q0 – q1)A > c'x, the rational potential victim will undertake the cost without any prodding. In this case, with or without any law, the probability of the accident will come down and the society will save on costs by the amount (q0 – q1)A – c'x > 0. Observe that, if by some measure, the society could force the potential perpetrator to take the precaution needed to bring down the probability, that could also be a saving to society if (q0 – q1)A – c'y > 0 but it would not be efficient because c'x < c'y implies that (q0 – q1)A – c'x > (q0 – q1)A – c'y > 0. Without any law, the potential victim suffers the cost of the accident and hence, makes a personal savings of (q0 – q1)A – c'x by taking the preventive measure and this amount is exactly equal to society’s savings. Individual welfare matches with societal welfare and rational agents will by themselves generate the social optimum. Law need not play any role here. Consider a highway where people live on one side of the road but their workplace is on the other side of the road. There is a subway that allows people to walk from one side to the other side and this takes a little bit more time than walking across the highway. The additional cost of using the subway, rather than sprinting across the highway, is c'x. Sprinting across the highway generates a probability q0 of an accident. The probability of an accidental injury in the subway (while, say, going down the subway stairs) is 0 ≤ q1 < q0. Alternately, the potential perpetrator in the car can slow down to prevent hitting anyone crossing the highway and this slow speed comes at a cost c'y > c'x ≥ 0. If there is no tort law, people wanting to go from one side to the other side will use the subway if (q0 – q1)A – c'x > 0 and we will have an overall efficient system. Observe that the cost c'x is crucially dependent on the existing subway being close to where the people want to cross. If the subway were miles away from where the pedestrians have to cross, then the value would be way above what it is in this example. Here is where the city or highway authorities play a significant role in achieving efficiency. This role is captured by the variable z in the section titled ‘Socially desirable levels of precaution’ in this chapter. Welfare economics and public finance economics determine the role to be played by z which, unfortunately, is outside the scope of this book but this observation underlines the importance of carrying out the analysis of efficient law keeping in mind the local context in which the laws are to be implemented. Borrowing a set of laws from other societies where the context could be entirely different is not the best way to achieve efficiency in any society!



Economics of Tort Law  115

Let us now consider the situation where (q0 – q1)A – c'y > (q0 – q1)A – c'x > 0; that is, it is better for the potential perpetrator to take the precaution since c'y < c'x. Observe that here also it pays for the potential victim to take the necessary precaution and, in the absence of any laws, she will take it since it adds to her savings. However, this will not be efficient since the savings are higher if the potential perpetrator takes the preventive measure. Will the latter do so? The answer is no. Without any tort law, the cost of the accident is borne by the pedestrian, not by the car that hit her. Thus, any preventive measure, in this example, does not generate any expected savings for the potential perpetrator since he does not bear any part of the accident cost. So, any preventive measure taken by him is a pure cost to him without any corresponding gain. Thus, left to them, it is the potential victim who will take the precaution and, given the parameters, that would be inefficient. This is a classic case in economics; individuals maximizing their respective welfares are generating an aggregate outcome that is not socially optimal. The law can step in and generate the efficient outcome. Let us see how. Suppose there is a tort law that says that the car that hits the pedestrian is liable, that is, the cost of the accident, A, has to be borne by the owner of the car. Immediately, we observe that what was preventing the efficient outcome in the last paragraph is no longer true. Now, the entire cost is borne by the potential perpetrator, for the potential victim gets reimbursed by him for the cost of fixing her up in the hospital and for all other associated costs. Suppose the precaution we are considering is that of driving within the speed limit or making sure that there are no pedestrians at the zebra crossing in the path of the car. If the car is not careful then the expected cost to the car owner is q0A; if it is careful, the cost is q1A + c'y and by assumption this is less than q0A. The savings in cost due to the reduction in probability of the accident is more than the cost of reducing the probability. Tort law has generated the right incentive for the entity whose precaution generates the efficient level of caution. It is important to note that we are not looking at who ‘caused’ the accident but at who is the best person to avoid the accident and, by best, we mean the entity with the highest net gain. What tort law has done is assign the liability in a way that gives the incentive to the person who has the cheaper cost of avoidance. Both generate the same savings, (q0 – q1)A, but the potential perpetrator has the lower cost, c'y < c'x. The other point to note is that law helps when the potential perpetrator who, by definition, does not ‘suffer’ the accident, is also the entity that has the least cost of reducing the chance of an accident. Now, consider a city street with zebra crossings (pedestrian crossing zones), traffic lights and speed limits. In this situation, all three actors referred to in

116  Law and Economics

the model are involved. The city authorities have tried to reduce incidences of accidents by putting in traffic lights and zebra crossings for pedestrians (z' ); pedestrians must take care and cross only when the walk sign is on and at zebra crossings (x' ); motorists must drive within the speed limit and stop before the stop line and keep the zebra crossings free for pedestrians (y' ). All three together are now involved in reducing accidents and the degree of caution each must take is specified. Suppose everyone is taking their respectively specified precaution. The first rains of the season have come and, as happens, the dust and oil that have accumulated on the road during the dry season have made certain parts of the road very slippery. (People are advised to drive extra carefully during the first rains for this reason.) Drivers have no way of knowing which part. Observe also that the first rains and the oil and dust build-up on the road are entirely outside their control, or the pedestrians’ or the city authorities’. The city authorities can ban all motor vehicles when the rains come, wash the streets to ensure that last season’s dust and oil are no longer on the street before they allow traffic back on to the road. However, this is very costly to society given that the probability of cars skidding as they come to a stop at the lights is very small, though positive. So, society decides not to put this highly costly measure in place; hence, it is possible that some accidents will happen but since every car is ready to stop at traffic lights, and pedestrians are willing to wait for walk signs, these accidents will happen with very, very low frequencies. But accidents will happen and tort law will still have to decide how to ensure that people have the right incentives to prevent accidents. For instance, cars can be extra careful, stay well below the speed limit and be prepared to ‘ride’ the skid and still stop before the stop line at crossings. Pedestrians can wait for cars to stop and not start walking as soon as they see the walk sign assuming always that cars in motion will stop. Still, accidents may happen. In this case, of course, insurance starts playing a role as people are taking chances where the probability of loss is beyond their control. Tort law can decide who buys the insurance by deciding who is liable under such circumstances. Suppose the law says that if drivers have stayed within the speed limit and tried to stop before the walk area they are not liable. Then pedestrians will have to decide whether they will buy accident insurance, for should an accident happen under such circumstances, they would have to pay their own hospital bills. Alternately, if cars are still held liable, they will buy insurance to cover these costs. In most societies, cars are held liable for all accidents unless there is gross negligence on the part of the pedestrian—running out on to the street when she does not have the walk sign. In some societies, even



Economics of Tort Law  117

then cars are held liable. For instance, in some cities of Sweden, people can cross the city streets at all times and at all places and cars are held liable if there is an accident. This obviously makes cars very expensive to drive and the city authorities like that because they want to limit the use of cars in the city and want to encourage people to take public transport to cut down on pollution and congestion. The general reason for making cars liable is because the cost of insurance is a small part of the cost of the car and can be afforded by all those who want to buy cars. Many pedestrians, on the other hand, may not be able to afford insurance and need to be protected from having to pay for such accidents. This latter concept is a very important one in economics. Consider a person who earns `100,000 a month and another who earns `10,000. Let the hospital bill, should an accident happen, be `20,000. First, we expect that the 100,000-per-month income earner is more likely to have a car than the 10,000-per-month income earner. Second, `20,000 for the lower income earner is more ‘valuable’ than to the higher income earner. You may be able to eat out when you are a student with little pocket money but you would much rather use the money to buy a book for the course that you are taking. In other words, the ‘utility’ loss of spending a certain amount of money is higher for a person who has less money than for a person who has more money. This ‘diminishing marginal utility of money for a person with more money’ implies that a richer person will suffer a lower loss in paying the insurance premium compared to a poorer person. Thus, if anybody should pay the insurance premium to cover the loss to society due to an accident, it should be the richer person. The richer person here is the car owner and he will pay for the insurance if he is made liable for the accident. That is the reason why cars are held liable for motor accidents to a pedestrian. We still have an unresolved problem. What does it mean to say pedestrians are not liable in an accident? Your friend is driving your car and gets into an accident. Are you liable as the owner of your car? Is your friend liable since he was driving it at the time of the accident? Suppose he is, but is not carrying insurance and does not have enough money to cover the cost of the accident. If he were rich, he would probably have his own car and not need yours. Now who pays for the accident? Your friend has every incentive not to be careful since even if he is liable, he has no money and, hence, nothing to lose! Instead of your friend, suppose it is someone who you have hired as a driver that gets into an accident? Chances are very high that he will not be able to afford the insurance premium required. So, if he is held liable observe what will happen. You will not care who is driving your car since you are never held liable for harming the pedestrian. Your hired driver will not care

118  Law and Economics

because he has no money to lose. Of course, the law can make it mandatory for all drivers to buy insurance. That would discourage potential drivers from taking up that particular livelihood. Or, the occasional driver, your friend, will have to first buy insurance for that day before he takes your car out. In short, the transaction costs for driving your car will be so great that your car will remain idle more often than not. All this will have a negative effect on the welfare of society. So, it is the car owner who is held liable. You should be able to ‘lay down the law’ when you hire the driver and make sure that he is law-abiding; you know your friends better than others and so should give your car keys only when you are sure of your friend’s driving habits.

Is Tort Law Sufficient? In the previous section, we have introduced the concept of ‘diminishing marginal utility of money’. Let us explain this in a bit more detail. Suppose the money cost of taking a particular precaution by the car owner is c. For instance, if the tyres have become worn out, new tyres should be put in, as the new treaded tyres are better at preventing skids and hence, accidents. This c is the cost of new tyres. However, the car owner is a very rich person. If she gets into an accident and her insurance company covers the cost, her insurance premium will go up. Given her wealth and money resources, this increased premium is too small for her to really care. Her disutility (loss in utility) from the increased premium is much less than the utility from being lazy and not checking the tyres. She will then take her chances and not take proper care of her tyres. Since she has a lot of money, the increment in utility from an additional rupee (or, equivalently, the disutility from losing a rupee) is so small that her liability, and the resultant increase in the premium, is not sufficient for her to take adequate precaution. This destroys our analysis of how to use tort law to achieve optimal levels of precaution. So, tort law goes another step ahead and says that in addition to the liability for the accident costs, the owner of the car is liable to be prosecuted and fined for not driving with proper tyres. In other words, the total liability of the car owner is being increased by an amount that is more than the actual cost of the accident. This additional part can be interpreted as the punitive part of the damages that the car owner has to pay; the punitive part acts as a deterrent that encourages the owner to take the necessary precaution to prevent an accident. In general, tort law does not involve punitive damages except when it is used specifically as a deterrent. Punishing the car owner for not having proper



Economics of Tort Law  119

tyres does not prevent the accident (it has already happened!) and, even if the fine is given to the victim it is a pure transfer from one person to another and does not add to the total welfare. If, instead of a fine, the person is sent to jail, then there is a dead-weight loss to society; no one gains from the car owner going to jail and he is definitely worse off in jail than out of it. So, in either case, it is not clear why someone should be punished unless one can strongly make the case that the punishment acts as a deterrent. Punishment is usually given when the action of the injurer is close to wilful, intentional, oppressive, fraudulent, etc. If there are elements of intentional behaviour or wilful neglect in causing an accident, punitive damages are awarded to address this additional issue. It may be noted here that these cases are closer to crimes, which we consider in the next chapter. Punitive damages are important also in a context where law enforcement is relatively weak. Assume that A has done harm to B, by not taking a precaution, causing a damage of `10,000 to B. However, the law enforcement is weak (legal fees are high and it is a long and arduous process for the courts to finish proceedings), such that only 50 per cent of such cases are brought to the courts and awarded compensation. Once proved, A will be forced to pay full compensation. However since the probability that compensation will have to be paid is ½, A’s expected cost of accident is only ½ * `10000, that is, `5,000. A will have less incentive to take adequate precaution for he now saves only `5,000by avoiding the accident. If the cost of precaution is `9,000, that is more than the amount saved and so A is better off not taking the precaution. If, however, there was strong enforcement, the cost of the accident will be `10,000 rupees and this amount will be saved if the precaution were taken, creating a net gain of `1,000. (Here, we are assuming that without the precaution, the probability of harm is one; with the precaution the probability is zero.) This brings us to a very important issue in tort law—fixing the amount of compensation. While one may argue against punitive compensation, one still has to convincingly argue that the compensation amount is, indeed, the right one. We all know that it is difficult to arrive at a compensation amount when a person is injured, or has lost the use of a limb in an accident or, worse still, killed. When a person dies, one way of looking at the problem is to hold (what is now considered an extreme and erroneous position) that the person requires no compensation for one cannot give back life to the dead victim. This is erroneous because it generates a perverse incentive of ensuring that one’s victim is dead rather than alive! The accident and the resultant death can cause severe hardships (both emotional and material) to the victim’s family and they need to be compensated. While hospital bills are easy and tangible, compensation for death is very difficult. Societies, at

120  Law and Economics

different times, value lives differently and such valuation often shows up in the compensation allowed by courts in cases of accidental death. This is more a problem for judges and philosophers as it has a lot more emotive content than economists can handle!

Other Liability Issues Suppose Anup has a car and, one day, as he is driving it, it catches fire and Anup suffers some superficial burns before he can get out. His car is destroyed by the fire. Anup loses the value of the car and has to pay the hospital for his treatment. The car may have caught fire because of an electrical short circuit resulting from an error committed in the factory; or, it may be due to a design error. What does Anup do about this? What is society’s view of Anup’s problem? Suppose there is no law regarding this. Other people, hearing of Anup’s plight, may become nervous about buying cars. Before buying one, they would like to hire design engineers and automotive engineers to examine any car they think of buying. Once the car is built, this becomes a highly costly affair and may cost as much as the car itself! Clearly this is not good for the automobile industry, as this will vastly increase the cost of buying a car. Of course, it is possible for a company to advertise that its car does not catch fire on its own but how many buyers are willing to accept all the claims made by an advertiser? How can car buyers be made to trust a car producer? Without that, the automobile industry will operate at scales lower than optimal and that would imply a low total welfare to society. Tort law can bring back efficiency in the automobile market by holding producers liable in such cases. Thus, if the car catches fire, law holds the producer responsible and asks it to pay the owner her hospital bill and for the loss of the car. If the car buyer knows that this is what the law requires, she knows that a rational car company has the right incentive to design and build a car that does not catch fire. This helps her ‘trust’ the car company and make the purchase, if it suits her budget and taste. This is producer’s product liability; it has to compensate losses due to faults in the product. In most cases, tort law fixes such liability on the producer rather than on the consumer. The economic argument for this is straightforward. As pointed out in the last paragraph, it is very costly for the consumer to check out the unit she is buying; by costly, we mean the cost of verifying the (safety) qualities of the product may be large compared to the price of the product. On the other hand, for the producer who produces many of these cars, the



Economics of Tort Law  121

cost per car to design and build a car that does not catch fire would be a lot less. The producer has the resources, and the expertise, to set up research and development facilities that can test a new car design and assembly lines. This takes us back to what we have discussed already—tort law must set the incentives in such a way that the entity that is best suited to prevent an accident should undertake that activity. In tort law, as discussed so far, the threat of liability seems to lead to efficient precaution. The precaution levels in the algebraic model involved both the potential victim and the potential perpetrator. How are these to be addressed after the accident? One way of interpreting these is as follows: 1. If both have taken adequate levels of precaution (x' for victim and y' for perpetrator) then the perpetrator is not liable and the victim bears the loss. 2. If the victim has been careful, but not the perpetrator, the latter bears the full liability. 3. If the perpetrator has been careful, but not the victim, then the victim is liable. Of course, as mentioned in the case of certain cities in Sweden, the pedestrian is never liable in a road accident; that is, there is no specified level of precaution (other than deliberately throwing oneself in the path of a car) for the pedestrian. This still leaves a fourth possibility when the victim is supposed to have taken some precaution, as well as the perpetrator, and both have not been careful. This is referred to as contributory negligence—both have simultaneously contributed to the accident by their respective negligent behaviour in failing to prevent the accident. Here, the courts attribute less than 100 per cent liability on the perpetrator unless one can show that the perpetrator did actually have the opportunity to avoid the accident. For instance, suppose a pedestrian tries to cross the road when the light is green for oncoming traffic. The car, which is at a distance, has the right of way, speeds up but remains below the speed limit and hits the pedestrian. Here, if it can be argued that the car saw the pedestrian and could have avoided the accident if it slowed down (or at least, did not speed up) then the perpetrator can be held to be fully liable. To economists, this approach by the law makes eminent sense. After all, it is not important who was right and who was wrong; of greater significance is the fact that accidents are costly and whoever has the ability to avoid this cost should have the incentive to do so. A hundred per cent liability for the perpetrator gives the right incentive in this case.

122  Law and Economics

Finally, tort law seems to kick in after the harm, or damage, has happened. However, this is not the full story. Certain incidents are ‘actionable’ offences, even if the accident has not happened. In other words, tort law need not operate in a vacuum; alternately, there are many other aspects of law that help tort law achieve efficiency. The police can catch a driver for speeding even when there has been no accident. Speed limits generate a standard of safety and cars must conform to these standards. Such standards are especially significant for legal safety norms in products. Thus, tyres on cars must be built to specific safety standards even if no accidents have happened on a car that uses tyres that do not satisfy these specifications. Recall what we mentioned earlier about how laws can help in generating trust in the safety of products that consumers buy. If they are confident that companies not producing safe tyres will be prosecuted even when there are no accidents, they will not have to spend huge sums of money to scientifically check all the tyres they buy.

The ‘Irrational’ Economic Agent The assumption so far has been that people make reasonable judgements about the probability of events and their expected gains and losses. The usefulness of strict liability or no liability depends on the assumption of such rational estimates. However, it is noted by a number of experiments (one such study is reported in Chapter 7 in Volume 2) that often people may not respond in the way theory predicts. It has been noted that people may underestimate the probability of events as low and consider these as something that may never happen. On the other hand, people may overestimate the probability of well-publicised catastrophic events like nuclear accidents. Such incorrect estimation of probabilities may encourage people to take inappropriate precautions. For example, people may neglect to take precautions in certain cases, even when the probability of an accident is positive (though low). On the other hand, they may be excessively cautious with regard to certain other events—like the location of a nuclear plant. This knowledge can also be used to change the liability under tort law. For example, consider the case where people do not take adequate care in handling certain consumer durables using electricity. A negligence rule is in place with the understanding that both the producers and users of this equipment take certain precautions. In particular, it wrongly assumes that users take adequate care in handling these products. Clearly, tort law would



Economics of Tort Law  123

not lead to efficient prevention of accidents. In such cases, one would like to see a slightly more proactive tort law that takes into account the behavioural aspects of the users. In particular, tort law may increase the requirements for care to be taken by the producer of the product to compensate for the users’ behavioural shortcomings. Again, the issue is not one of who is right and who is wrong, but how to impose the liability in a way that is efficient. If the manufacturers can cheaply design the equipment in a way to minimize the accidents for users, even if they use it somewhat carelessly, then that is what the law will implement. The behavioural shortcoming of users may not be irrational at all but simply due to lack of information. The best example of this is smoking; earlier, people did not know of the damages to health caused by cigarettes. Now, law requires cigarette companies to announce the dangers of smoking on the packets and sellers can be charged if they sell cigarettes to minors or near educational institutes.

Cases from India Consider the following actual case, 2002 ACJ 780.2 A Gujarat State Transport Corporation bus was involved in a head-on collision with a Matador van. It was a serious accident where most of the harmed party were in the van, including three people who died. The van belonged to a state-funded research institute carrying a group of researchers, along with some of their family members. The dispute covered both issues: compensation for the victims and contributory negligence. Since two vehicles were involved, there was an obvious issue of who was responsible and hence, who should pay compensation and how much. The case is interesting for these reasons as well as the fact that both were government organizations. The other interesting fact about the case was that the accident occurred in 1981, but the issue was finally decided in 2002! Recall we had said that strict enforcement of tort law makes it easier to achieve efficiency—a 20-year process with 10 appeals where both sides ultimately represent the same entity certainly does not help. The case was decided by a tribunal, but later appeals were made regarding both who should pay and how much. The accident occurred on an ‘s-curve’ on a highway. Both sides claimed that the other side was negligent. However, there was enough evidence to suggest that the bus was driving fast round the bend. As a result, it crossed 2 J Bhatt GSRTC vs KamlabenValjibhai Vora, www.indiankanoon.org/doc/1543088/ (accessed on 24 October 2012).

124  Law and Economics

on to the wrong side of the road and at the point of impact the bus was on the other side of the road. The tribunal held the bus fully responsible for the accident. This was upheld in the appeal. However, the court was not happy with the compensation provided to the victims. It is interesting to see how the compensations were calculated. First, the court reiterated that an accident could not cause a windfall gain to a victim or the victim’s family. The compensation should not be more than what it should be; that is, victims or their families should not be (materially) better off after the accident than before it! Second, using various precedents from India and abroad, they made a distinction between compensation due to death and other injuries. In particular, the court distinguished between death and permanent disability. Both affect the victims’ family if the victims were the sole earning members. Therefore, the compensation given to both must be the loss in lifetime income. This must take into account promotions and increment in salary, the number of years of service, number of years of pension post retirement, as well as extra income from odd jobs after retirement. For the dead victim, one must subtract the expenses that are no longer necessary for the dead person’s own upkeep from the total earning compensation; for the permanently disabled this expense is not deducted from the lifetime earnings. Funeral expenses for the dead are included in the compensation; all additional expenses for taking care of the disabled and hospital bills for the injured are to be added to the compensation amount. Third, a particular individual who was injured had a longish stay in the hospital and a long period of recovery at home and had various members of the (extended) family take care of the victim. Though the family did not pay any money to the family members who provided these services, the court calculated a monetary amount for these services so that the perpetrator may not benefit from such gratuitous provision by family members. Fourth, they also allowed an amount for ‘pain and suffering’. Another very interesting case is 1997-BI2-GJX-0203-SC.3 An accident occurred at an unmanned railway crossing. A bus that was crossing the railway tracks stalled and was hit by a train travelling at 75 miles per hour, leading to the death of 40 persons in the bus, including the driver. The Motor Accidents Claims Tribunal held that the passengers were entitled to compensation from the bus company or the insurance company with whom the bus was insured, as well as the Indian Railways. The case came to the Supreme Court 3 Union of India, Appellant vs United India Insurance Co. Limited and Other Respondents.



Economics of Tort Law  125

on appeal because the railways felt that they could not be held responsible for the accident. This is an interesting case because of the judgment. Apparently, the bus driver was from Tamil Nadu and the bus had started from there. The accident, however, happened in Kerala. The bus had started late because the driver was late and the passengers had had an altercation with the driver on this issue. The driver was very angry with the passengers and had even threatened to abandon the bus in a forest to teach them a lesson. When he came to the crossing, everyone could see the train coming but since everyone was in a hurry, having started late, the driver felt that he could go through before the train arrived. As luck would have it, the bus stalled on the tracks and could not be started up again before the train hit it. There were some interesting points made by the railways, challenging the Tribunal’s observation that they could be held responsible. One of the arguments was that the passengers had ‘goaded’ the driver to drive fast and it was for that reason that the accident had happened. Apparently, when the bus came to the crossing, the train was about a kilometre away and given its speed it would have arrived at the crossing in less than a minute. On seeing the bus, even if it tried to stop it would not have been able to do so and would have certainly derailed many of its coaches and that too would have had disastrous consequences. Both the Tribunal and the Supreme Court found this argument of holding the passengers responsible specious. The court maintained that the bus driver was the one driving the bus and its passengers can never be held responsible for what he was doing. (If a hotel guest throws something out of the hotel window, then the hotel authorities cannot be held responsible for any damage to a person on the street; if, however, a hotel employee did that, the hotel was liable.) The other argument was that the railways are not required to have gated crossings of public roads unless the central government asks them to, and that most people are supposed to exercise adequate caution while going through unmanned crossings. However, the court maintained that if some activity (running of trains) endangers the lives of others, the controller of that activity (here, the railways) must take adequate precaution. Given the volume of vehicular traffic crossing every day (about 300), and the number of high-speed trains going through every day (5), the court felt that it was the ‘common law duty’ of the railways to make this a manned and gated crossing. Hence, they were liable to be sued by the victims of the accident for negligence. Economists would agree with the court’s decision. It may be instructive to construct the argument showing that aggregate welfare is better served if railways take adequate precaution.

7

Economics of Crime: Some Preliminary Insights A detailed analysis of the economics of crime or criminal law is beyond the scope of this book. However, economists have never been known for their ability to stay away from issues that concern them. Further, many criminal offences are actually ‘economic’ crimes; examples are frauds as in the famous Satyam episode, tax evasion and, of course, insider trading. Hence, some basic insights from the ‘law and economics’ literature are unavoidable for a book of this nature.

Why Do We Need Criminal Law? One of the most common crimes one can think of is theft. Stealing is a crime as it violates property rights. Laws on property rights tell us that one cannot take away, or destroy, someone else’s property. What if we were to say the same thing every time someone is caught stealing someone else’s property; it has to be returned (or, if destroyed, repaired to the state it was in before it was destroyed)? This sounds very much like tort law, where the victim of tort is compensated by the perpetrator of the tort. However, there is a very important difference between tort and crime. Recall the case referred to in the previous chapter where the Indian Railways were held liable in tort when a stalled bus on an unmanned level crossing was hit by a fast moving passenger train. No one contests that railways perform a useful function; they allow movement of passengers and freight, and both improve welfare in society. The focus of tort law is not to prevent the operation of the railways but to ensure that the railways ‘internalize’ the costs of their operations. In an accident involving the railways, the victims are often people not involved with railway activity, as was the case with the bus passengers. So, the railways would, in general, not consider this as a cost



Economics of Crime: Some Preliminary Insights  127

in their operations unless the law asked them to compensate victims and their families. The operation of the railways remains as a welfare-enhancing activity even after the railways include the costs of an accident in their overall cost; tort law simply gives an incentive to railways to minimize the cost of an accident. Theft, on the other hand, is beneficial to the thief but not to the person from whom the ‘property’ is taken and does not produce any surplus to society. It transfers resources from one person’s possession to another’s, with no increase in overall economic welfare. Thus, it affects the distribution of resources, not the total amount of resources. Activities subject to tort, through the proper fixation of liabilities, or costs, generate optimal amounts of welfare-enhancing activities. Theft, on the other hand, is not welfare enhancing. Suppose we view theft through the eyes of tort law. Somebody picks your wallet, which has `100 in it. Then two things could happen. First, the thief gains this `100 and your wallet; you lose `100 and the wallet. The net gain to society, what the thief gains minus what you lose, is zero. (If your wallet had a photograph of your very special friend and the thief has no utility from staring at that photograph, there would be a non-economic cost with no corresponding gain to anyone.) The second thing that could happen is that the pickpocket is caught, tort law kicks in, and your wallet with the money is returned to you (along with the photograph). Then you lose nothing and the thief undergoes some wasted effort. Society again gains nothing. For the thief, it is a ‘dominant strategy’ (economists’ terminology) to try to pick your pocket; if caught, she returns the wallet and money; and, if not caught, she gains the wallet and `100. If there is a probability that the thief will get away without getting caught, her expected gain from stealing is positive. You, on the other hand, lose with some probability and gain nothing with some probability; that is, your expected loss is positive in the presence of pickpockets. You would, therefore, move around with extra caution to prevent yourself from becoming a victim and this effort is costly to you and an additional loss to society. So, simple tort law would not be able to control theft, which as an activity is not desirable. Recall that the purpose of tort law is to fix liabilities in an efficient way. Liabilities are measured as the amount of costs incurred in an activity whose objective is not to generate that cost as a deliberate activity. The activity of picking someone’s pocket, on the other hand, is a deliberate act. Compensation does not protect the right of persons not to be affected by the crime. We have seen in the Chapter 1 that voluntary exchange of goods and services enhances overall welfare. On the other hand, if somebody forcefully takes away, or steals some goods from others, that reduces welfare. Thus, the protection from acts like theft enhances economic welfare. Crime is an

128  Law and Economics

act against society and controlling crime has benefits for society beyond the person concerned. Torts are unintentional harms while crimes are intentional. Therefore, unlike in tort, there is, usually, always an element of ‘punishment’ for the act. Very simply put, punishment is the amount that is over and above the cost suffered by the victim. Let us take the case of theft. A thief, A, has the following choices before committing a theft. She can either steal, or not steal. If she does not steal, her net gain is zero—no gain and no loss. If she steals, then with some probability p, she is caught and with some probability (1 – p) she gets away. Her expected net gain from stealing `1 from you is then (1 – p) * 1 + p * (1 – f ), where f is the amount she loses as punishment, when caught. Since she is caught after having stolen `1, she has `1minus whatever she has to pay as fine. For her not to steal, she must be worse off stealing than not stealing, that is, (1 – p) * 1 + p * (1 – f ) ≤ 0 which, on simplification, yields pf ≥ 1. Observe immediately that if p < 1, that is there is some positive probability that she will not be caught (or, (1 – p) > 0), then f > 1 is necessary to prevent the thief from stealing. For all f ≤ 1, pf < 1 if p < 1 and the thief will always take her chances at stealing. But f > 1 means that the thief must pay a fine that is greater than the amount stolen (namely `1). This is the essence of why crime always has an element of punishment while tort law does not. (Remember that in tort law, a perpetrator may gain by avoiding costs of precaution, but never gains from the act of the tort itself.) An important question is if the thief is asked to pay a ‘fine’, which is more than the amount stolen, who gets the additional amount? First, note that you should not be given that extra amount. Then you gain more if `1 is stolen from you than when it is not; that gives you a perverse incentive to become the victim of a crime. Second, in most instances (though not always), a thief steals because she has no money. While she may return your stolen `1, she will find it hard to come up with the extra amount of the fine. Society punishes the thief, in kind, by incarcerating her, making her do community work, banishing her from her preferred neighbourhood, etc. In other words, whether the criminal is punished in cash, or in kind, the victim of the crime is usually not entitled to the extra amount paid by the criminal.

What Can Be Done to Reduce Crimes? The conceptualization given above tells us about possible ways of reducing crimes, and its implications for public policy. First of all, it may be noted that



Economics of Crime: Some Preliminary Insights  129

there are two ways to improve deterrence, either by increasing the amount of punishment f, or by increasing the probability of punishment p. Each of these steps would increase the expected cost of punishment (pf ) and, if the expected cost of punishment is higher than the gain, the analysis above suggests that there would be a disincentive to commit the crime. Thus, with an increase in the expected cost of punishment, either the seriousness of the crime of an individual or the aggregate crimes in a society can be expected to decrease. This may be due to the reduction of crimes committed by some criminals or due to the fact that those who were contemplating a career in crime decided otherwise. The quantum of reduction in the number of crimes with every increase in the expected cost of punishment depends on the ‘elasticity’ of the supply of crime. In certain cases, there may not be much reduction (where we may say that the increase in punishment does not have that much influence—crimes of passion) or in other cases, there may be significant reduction. The elasticity may vary in different socioeconomic contexts. It is possible that in certain contexts these other variables may have an overbearing influence, and this influence cannot be moderated much by changing the economic incentive. An obvious way to increase p is to improve the enforcement of criminal law. This may need investments in police machinery, mechanisms that lead to more accurate judgments through trial and the speed with which justice is delivered. On the other hand, the cost of punishment can be increased by enhancing the fine, f, or by increasing the number of years of imprisonment. Since either of these two ways may improve deterrence, the obvious challenge to economists is to find the more cost-effective one. Increasing fines is not that costly, since the cost of collecting fines does not change much with the amount of fines. However, increasing the years of imprisonment would need higher capacity in jails and increased costs to maintain prisoners. Thus, increasing punishment will be costly, if imprisonment is the main form of punishment. On the other hand, as hinted before, we cannot use fines in all circumstances; for instance, some people who commit crimes may not have the ability to pay high fines. Keeping fines low has a problem too. Recall that in Chapter 6 we introduced the concept of diminishing marginal utility of money. A millionaire being asked to pay a fine of `100 has a much lower ‘utility’ cost than when a much poorer person is asked to pay the same fine. Thus, a low fine may not deter crimes committed by a very rich person. In certain countries, like those in northern Europe, speeding violations attract fines that depend on the margin by which the limit was exceeded and the wealth of the person violating the rule. Richer speed violators simply have to pay more.

130  Law and Economics

In general, most societies have a tacit understanding on what form (and amount) of punishment is suitable for each type of crime. Thus, cutting off one’s hand for picking a pocket is not considered an appropriate punishment even if it were to totally eliminate the act of picking pockets! Moreover, there is a sound economic argument for why a more severe punishment may not be that desirable. Suppose that the punishment for theft is death. An armed thief is being chased by the police. The thief knows that if she is caught, she will be put to death. On the other hand, being armed, she can fight off her pursuers and get away. If she is caught, she dies; if she stays and fights there is a probability that the injury she inflicts on her pursuers will prevent them from catching her. This increases the possibility of a more serious crime than theft. That is the reason why people say that the ‘punishment should fit the crime’. There is some instrumental logic in not meting out very severe punishments for not-so-severe crimes. The first objective of punishment is deterrence. However, when considering the degree of punishment, there are many other issues like retribution, rehabilitation and incapacitation that are also important. Retribution is instilling a cost to the criminal in proportion to the seriousness of her crime. Here, in certain cases, the shame as well as other costs associated with imprisonment may be needed to match the seriousness of the crime. Hence, fines alone may be not sufficient. Imprisonment may also lead to some rehabilitation. This is the objective of, say, juvenile homes. Jail sentences or detentions can be used to help detainees learn productive skills (increase their gains from alternative use of their time otherwise spent on committing crimes) or undergo counselling so that they can reflect critically on their mistakes, and, for habitual criminals, the longer a person stays in jail, the fewer the opportunities they get to commit crimes! The economic view of crime leads to an interesting, and often criticized, conclusion—there could be an optimal (positive) level of crime in a society. The criticism is because the economic view suggests that it is acceptable to have some crimes being committed. As economists, we owe an explanation of what we mean. Suppose a society has a million adults and the local law and order system, or police, is considering how to limit crimes. A wise and experienced person comes up with a brilliant idea. Let us divide the population into two parts, he says. Randomly choose one group and admit everyone in that group into the police force. Let this group be called X. Each member of the police is then randomly assigned to a person in the other group, Y. This society will then have half a million people in the police force and another half a million working as teachers, lawyers, economists, plumbers, carpenters, judges, motor mechanics, drivers, factory workers, etc. Each member of the



Economics of Crime: Some Preliminary Insights  131

group that is being overseen contributes to the society’s GDP (gross domestic product). Part of this goes to pay the police force that does not contribute anything to the society’s GDP. Instead, they follow the GDP-producing persons to make sure that they commit no crime. This will certainly prevent crimes in this society, but is it the best way to organize society? Let us pose the problem in a slightly different fashion, says a brash economist, or a thoughtful law student who has just finished a course in law and economics. She says that it is well known that, everything else the same, a larger police force has better chances of preventing crime than a smaller police force. As the number of crimes goes down, the loss to society goes down and the savings in these costs is the incremental benefit from a larger police force. So, suppose we consider the cost of an additional police officer. Being a member of the police means she does not produce as a plumber or as the CEO of a high-tech company. Let this loss in GDP be denoted by x. This additional member of the police force reduces the number of crimes by a certain amount. This reduction in the number of crimes reduces the loss to GDP caused by these crimes. Let this be denoted by y. If y > x, then the additional member of the police saves more in lost GDP than it costs society to take her out of productive activity. If, however, y < x, then this addition to the police force is inefficient. If y ≥ x at all positive levels of crime, zero crime is the optimal level for this society. Alternately, if y becomes less than x at some positive level of crime and stays that way at all lower levels of crime, then that positive level of crime is the optimal level for this society. Observe that what the economist, or the law student, is saying applies only to the pecuniary costs of crime prevention. A society may have non-pecuniary benefits of zero crime that are much higher than the loss in GDP resulting from a large police force. This is a choice that society needs to make but the basic approach remains the same—what is the net value of a zero crime society compared to a society with some small, but positive, level of crime? What are some of the components of the social costs of crime? If a person steals `10,000 from a household, this per se need not be a social cost (the amount has moved from one person to another, and hence only redistribution has taken place). However while stealing that money, the thief may have destroyed something—locks, windows, or may have attacked or killed the person resisting and so on. This damage is an important part of social cost, but that is not the only one. People who expect such thefts need to take precautions—spend money on security systems or not go on vacations leaving the house locked. There are additional costs expended in controlling crime. These include the investments that individuals make (effort to be vigilant, investments in

132  Law and Economics

personal or home security, etc.), the efforts individuals take to participate in police and judicial processes when subjected to crimes, the investments and recurring costs of police and judicial machinery, and the costs of jails and other forms of punishment. The costs required to see that the crime rate is reduced further when it is already low can be very high, whereas the costs for improvements when the crime rate is high can be low. When a place is infested by the rowdy behaviour of some people, regular patrolling by the police or imprisonment of a few can reduce the crime. However reducing crime to near-zero levels would require massive investment. Increasing punishment, or the probability of prosecution, is not the only means to control crime. It is well known that socioeconomic factors influence the crime rate in a society. Such factors can be broadly divided into two. The first is related to the economic development and public policy of a country. Thus, unemployment,1 poverty, lack of social security, policies related to the use of drugs, alcohol, etc. may influence the rate of crimes. The role of unemployment and poverty can be interpreted in the following way: the gains from committing criminal acts are more than the alternative opportunities available to those people who are poor and unemployed. Thus, with the same level of law enforcement against crimes, an increase in poverty or unemployment may lead to increase in crimes, since private gains from crime increase. However, a general situation of poverty and unemployment may not lead to this situation. This is so since even when poverty and unemployment increase, the benefits of crimes, that is, the opportunities to make gains from crimes decline (since others’ income levels are also low). But in any economy, there can be seasonal unemployment, and there are some people who fail to acquire a decent income from participating in labour markets. The presence of a viable social security system for those people may have an important bearing on the crime rate. In addition, the appropriate policy regarding alcohol and drugs is a contested issue even in developed countries. It is acknowledged that prohibition of the use of drugs has not been that effective in preventing drug use in many countries. Probably, a combination of restriction and policies that enable addicts to counter their habits could be more useful. The other set of factors affecting crime rates are connected to the psychology of persons, social norms and the functioning of family, and social capital. There can also be an interaction between the psychology of individuals and social norms—the latter trying to moderate psychological

1

For evidence in this regard from the United States, see Ihlanfeldt (2006).



Economics of Crime: Some Preliminary Insights  133

deviations harming social order. It has been noted that some communities do not encourage the growth of crime even when they face deprivations in terms of income, employment, etc. Thus, social norms, culture and family values too can play an important role. However, there can be changes in the way social norms influence the behaviour (Wilson and Herrnstein 1985). As is well known, the sustained influence of social norms may need repeated interaction. Such interactions may decline in contexts where there is frequent in- and out-migration, and where livelihood depends on exchanges between anonymous or lesser-known individuals. This may be the reason why crime rates are higher in urban surroundings even in those countries where cultural or social factors keep the crime rates lower in rural settings. However, such social norms and family investments continue to play a hugely important role in countries like India, where there are problems with the regard to the effective enforcement of formal laws against crimes. The role of social norms in affecting criminal behaviour in countries like India itself is an interesting topic requiring more elaborate treatment. It can be noted that social norms need not be benign in all circumstances. Though social norms and their enforcement may constrain theft or robbery in rural settings, there can be violence-enhancing social norms, and the influence of such norms may even reduce the effectiveness of formal laws against violence and crimes. It is well known that though certain actions like seeking dowry are treated as illegal formally, they continue to prevail in India. There is also a high incidence of violence (leading to explicit crimes) within families as part of dowry in India, and this is sustained by social norms. Here, social norms sustain a situation that breeds violence and crime, despite the fact that formal law prohibits the practice that leads to this situation. This may also be true in the case of honour killings—killing of girls for marrying men in caste groups not favoured by norms of specific community or caste groups. Along with social norms, social capital and associational networks too may influence the propensity to carry out crimes (Buonanno et al. 2009). Social capital can enhance enforcement of social norms (and then the effect would be the same as that of positive impact of social norms.) Social capital can also enhance the returns from non-criminal activities, as it may improve the availability of other income-earning opportunities. Economically this would mean a reduction in the gains from crimes. However, social capital can also help criminals. There would be better communication among the criminals, and there may be social covers that protect them from police investigation and criminal proceedings. This reduces the probability of punishment. Thus, the net effect depends on the context and some analyses, like Buonanno et al. (2009), have shown that social capital has a negative effect.

134  Law and Economics

Religion also plays the role of social capital, social norms and beliefs in punishment by god, which may be perceived to add to the punishment received by the courts. Some studies noted religion to have a moderating effect on crimes but the ones that followed such as Jenson (2006), which addressed some of the data or statistical problems of the previous studies, show only a negligible effect. Moreover, in this era of terrorism, driven by fundamentalist religious positions, indoctrination of such positions may add to suicide attacks and similar crimes. Thus, the role of religion cannot be predicted ex ante.

People Who Discount the Future Heavily In developing countries like ours, we hear stories in which the poor or those suffering from extreme famine or starvation, forcibly take away food kept in stores or in affluent households. We see cases where people indulge in violent behaviour, causing damage to assets and others, when they are agitated over some serious issues. We also see people influenced by heavy doses of alcohol or drugs losing their ‘moderation’ in terms of rational behaviour. Teenagers in most societies are likely to indulge in mindless activities, and sometimes such actions may evoke serious social concern. A common thread in all these actions could be the heavy ‘discounting’ of future costs. Suppose you are given the following options: you could get `100 now tomorrow. It is most likely that you would want the `100 today. Suppose, instead, you are asked to choose between `100 today and `100 a year later. You would certainly choose `100. To understand better the significance of these alternatives, suppose you are given a choice between `100 today and `1,000 a year later. While you may choose `100 today over `100 a year later, when faced with the second choice you may actually choose `1,000 a year later to `100 today. Between these two extremes, there could be a situation where you are indifferent to `100 today and, say, `120 a year later. In other words, `120 a year later is the same as `100 today; in economics jargon, the discounted value of `120 a year later is `100. We know that for the ‘rational’ criminal that we have discussed so far, the gains from committing the crime occur immediately after the crime. The costs, on the other hand, if any, especially in societies where enforcement is slow, could come much later. Thus, there is a possibility of discounting future costs. For some reason, if the person discounts future costs heavily (the current value of the cost of `120 at a later date is not `100 as in the example just given, but, say, only



Economics of Crime: Some Preliminary Insights  135

`80), then the present gains may outweigh the future costs of committing the crime. This can encourage such people to commit crimes. It is noted in environmental economics that people living in extreme poverty may have a higher discount rate. This may be because the present gains in terms of survival may be much more important than future costs (as is the case in environmental impacts) for such people. People who encounter extreme life-threatening events like famine may also face a similar situation—they may not see themselves living to encounter future costs, and for the present, survival is the only important objective. There are others who do not face such extreme events or poverty, but who may indulge in behaviour unmindful of future consequences. Such people are said to have variability in moods, and under certain moods they may choose a high discount rate. Teenagers and their alcohol-induced street violence or rowdy behaviour causing inconvenience or damage to others (which is visible even in some developed countries) is one such category. The violent incidents on campuses, including those connected with ragging, can also be considered as part of this behaviour. The students or teenagers, who take part in such actions, could not control their emotions then and there and that led to the criminal actions. But they may feel bad about what they have done soon after. For this category of people, enhancing future punishment arrived after an elaborate and time-consuming judicial process may not be that effective. Or, such enhancement may not deter their actions. On the other hand, consistent and relatively swift actions to restrain them may be more effective. However, this is also a case where public policy should address factors that make people discount the future heavily. Reduction of poverty and swift measures to enhance the chances of survival during extreme events like famine are more important here. Similarly, measures that limit the impact of alcoholism or drug addiction are more important if we wish to control the criminal activities of such people. Enhancing the level of punishment may not be that effective in this regard.

What Private Individuals Do to Control Crimes Sustaining a social environment that is less prone to crimes in any context is not the job of only the police or the judiciary. In fact, a greater part of the effort to control crimes in almost all societies is made by individuals themselves. Use of locks in houses and cars, gated compounds, keeping valuables in safe custody (including vaults or lockers), electronic gadgets

136  Law and Economics

including those sending alarm signals, efforts to avoid certain places at certain times for the fear of crimes, etc., are some of the investments that individuals make to control crimes. Controlling crime requires ex ante and ex post actions. Ex post actions are to prosecute and punish criminals, and we expect this to have a deterrent effect on future crimes. There are ex ante actions, to see the crime does not take place at all. There may be a social trade-off between ex ante and ex post actions, where under certain circumstances, one type of action may lead to lesser social cost than the other, and then the cheaper one may be preferred. In most countries where there is a reasonable law and order system, ex post actions are taken swiftly by public authorities. (In other cases individuals may take revenge actions with or without the help of gangsters or mafia, against those who have harmed them criminally. This is private ex post action, but this is not considered appropriate for social order.) Public ex post action is desirable, since it is society or the state that represents the society with the coercive power to punish some individuals. Granting coercive power to the state, while simultaneously depriving individuals of such power, is a fundamental characteristic of the emergence of a modern state. There are ex ante actions carried out by public authorities also. Patrolling by the police in public spaces is part of this ex ante action. However, a greater part of the ex ante action is taken by private individuals. In fact, certain types of crimes are better controlled ex ante through such self-vigilance. Consider the potential crime that the spouse in a marriage may inflict on the other. In this case, power for the public authorities to exercise greater ex ante control may be in conflict with the privacy needs of the individuals. Some individual effort to avoid crime against oneself is unavoidable in any context. All such personal efforts may have certain social implications. In certain cases, one person’s care or investment in safety may be beneficial to others as well. For example, if some people use security persons in one area, that may deter thieves from approaching the whole area. If some households use automatic alarms, and potential criminals do not know ex ante which households have alarms, then this may discourage thieves from the entering the neighbourhood, affecting all households (those with and without alarms). On the other hand, if some households have visibly higher compound walls and locks, potential criminals may try to enter those households without such visible security mechanisms. Though effective prosecution and punishment may deter criminals to some extent, it is evident from the experiences in different parts of the world that other socioeconomic and cultural factors like unemployment, poverty, stability of family, mental health, alcohol and drug use, also influence the



Economics of Crime: Some Preliminary Insights  137

crime rates.2 Lowering crime rates requires public efforts in terms of job creation, social security, family counselling and measures to control the use of drugs and alcohol. However individuals and families too play an important role in encouraging children to get education and employment, providing security and instilling norms that help healthy social behaviour.

Economics of the Death Penalty For economists, the death penalty is difficult to justify. Death deprives a person of life and the whole purpose of economics is to find ways by which social welfare is maximized so that everyone is better off. Expelling someone from society certainly does not help in this. To top it, the death penalty is a punishment and economists do not support any punishment unless one can argue that it acts as a deterrent. In fact, precisely for this reason, economists are usually also against life imprisonment without parole. When a labourer works for her employer, she produces value that sufficiently compensates her for the disutility of labour; the employer gets a positive value even after paying the labourer for her work. In other words, the value produced by the employer and employee is more than the losses suffered by each of them (disutility of effort by the labourer and wage cost paid by the employer). That is, the value of the activity is sufficient to compensate both actors for their losses. The death penalty compensates no one for the death of the person thus punished! There are empirical investigations on the social benefit of death penalty, that is, whether such punishment deters murderers. This is not yet settled in the literature. The reason for the continuing debate is because it is very difficult to devise the methodology for empirically determining whether the death penalty deters criminals or not. In principle, one should consider two otherwise identical societies that differ only in one aspect—one with the death penalty and the other without. If people supporting the death penalty are right, the society with the death penalty should have a lower crime rate; if they are wrong, this should not be evident in the crime rate. Unfortunately, it is impossible to get two such societies. The closest we can get is in a comparison between the United States (which has the death penalty) and European countries (that do not). The crimes (that attract the death penalty

2

See studies such as Dilulio (1994) and Wilson and Herrnstein (1985).

138  Law and Economics

in the United States) are much lower in Europe than in the United States. Indeed, European countries also have more comfortable jails and punish less heavily than the United States for similar crimes. On the other hand, European countries have very different social and cultural norms than the United States, as well as very different social security covers for their citizens, and are less fragmented societies. Remember, we said that more than fines, punishment and enforcement affect the incidence of crime. We are stuck in the realm of philosophy and morals while deciding whether we should have the death penalty or not. The answer, however, derived from the economics methodology is a clear no.

8

Economic Laws Why Do We Need Economic Laws? When we say ‘economic’ laws, we mean laws that allow transactions to take place in an efficient manner. So far, we have been studying laws that govern markets; more precisely, we have been studying situations where people, left to themselves, with a little help from legal institutions, will transact with each other in a way that adds to society’s welfare. It is somewhat like the rules in any game. In football, for example, everyone wants to see a ‘good’ game, exciting and thrilling. The rules of the game are made in such a way that if the players play according to the rules, then spectators could watch a beautiful spectacle for 90 minutes. On the field, players show a lot of skill and imagination to win the game for their team; the rules define what are legitimate moves and what are not; they do not tell the players how and to whom to pass the ball (as long as they do not use their hands, of course). Individual players, trying to win for their team, use all their resourcefulness to play a wonderful game. A tearaway forward running with the ball towards the goal is confident that the referee will enforce the rules if any opponent trips her from behind and the opponent’s team will be penalized. Our ‘field’ is the marketplace and the various laws we have so far considered are the rules that govern the actions of the economic agents in the marketplace. The objective in all sports is to encourage players to show their skill and ‘think on their feet’. The rules are written so as to enable them to do precisely that; sport is competitive and good players want to win the matches they play. If every tennis player wants to win every match they play, then they will try to be more and more skilful and the tennis match organizers are happy. The only difference between sports and the market is that in sport, if one wins, the other loses. In economics, on the other hand, if one gains, the other does not necessarily lose. In the market transaction of buying and selling, the seller gains because he covers his costs and makes a profit, the

140  Law and Economics

buyer gains because her pleasure from what she buys is more than the price she pays. So, one of the major objectives in economic law making is to enable such voluntary transactions. Consider, for instance, contract law. We know that people will enter into contracts—a set of promises—with each other if it is in their ex ante interest to do so. Ex post, if it is in all parties’ interest to redraft their promises, they will do so. All that law wants to do is enable them to make credible promises—promises that are enforced by a third party (the law) to cut out ex post opportunistic behaviour by parties making the promise. So, one of the fundamental principles in law and economics is that the law should be enabling. Instead of telling economic agents how to maximize their interests, the law can simply lay down the rules of exchange; do what you want, provided there is ‘fair play’. The way fair play is defined is very important. Let us give an example. Suppose that there are n economic activities that a society can undertake. For various reasons, society has decided (through referendum, democratic voting, etc.) that k of these activities, though commercially profitable, are harmful to society. These bad activities could be drug trafficking, extortion, ‘cooking the books’ (the Satyam fraud), etc. Society can, therefore, draw up a law to prohibit of some activities. They are harmful because even if they result in gains for some, the losses suffered by others outweigh such gains. There are two ways of going about it: (a) the law can list the (n – k) activities that are considered ‘good’ and say that only these are permitted and (b) the law can list the k activities that are considered ‘bad’ and ban them. Recall that law affects not only what are on the ground today but also all future activities. In a certain world, both (a) and (b) serve the same purpose and are, hence, equivalent ways of achieving the same objective. However, the future is uncertain and new activities develop all the time. (For those too young to remember, ask your parents about how life was when mobile phones and call centres were unheard of.) What happens to the society if a person finds that there is a possibility of a (n + 1)st activity that is commercially viable and has no harmful effects on society? Under the legal regime prescribed by (b), the activity will be tried out; under the regime (a), the person has to move the law-making authorities to make an amendment to the law so that she can carry out the activity. This obviously increases the transaction cost of starting up the (n + 1)st activity and that is inefficient. Simply put, a list of banned activities is more enabling than a list of permitted activities. Of course, the (n + 1)st activity may also be harmful. Under (b), it will be tried out, people will immediately come to know of the harm it causes and armed with this evidence, it is easier to add this activity to the list of



Economic Laws  141

banned items. In principle, the uncertain future can throw up a new activity that is more like the k bad activities or, more like the (n – k)good activities. Why then is law regime (b) more enabling? In general, it is easier to ban an activity that can be demonstrated to be harmful (as is the case if regime (b) is being followed) and the (n + 1)st activity is harmful; it is more difficult to convince everyone that an activity may have bad consequences without being able to demonstrate it, as would be the case if regime (a) is being followed. The license raj (ask your parents, if you do not know) that we followed prior to 1991 was regime (a); we shifted to regime (b) after 1991. Let us take another example that emphasizes what we mean. At the turn of the century, the Supreme Court, responding to a public interest case (referred to as PIL, or public interest litigation), ordered all commercial vehicles plying in Delhi to use compressed natural gas (CNG) as fuel.1 This was to cut down on the very high levels of pollution prevalent then in Delhi. Since CNG was the court-ordered fuel, use of any other fuel became illegal. Suppose a new fuel, more environmentally friendly than CNG, is discovered [the (n + 1)st activity]. It cannot be used until the Supreme Court modifies its order on CNG. That will take time and legal experts will have to indulge in acquiring scientific and engineering expertise. On the other hand, if the courts passed a stricture on how much emission was permitted from the engine of a commercial vehicle, this new fuel could be tried. The current order is less enabling than if the court had put a legal threshold on emissions. At the beginning of this chapter, we remarked that, unlike in sports where if one side wins the other loses, in economic transactions it is possible for both sides to win. Indeed, again unlike in sports, it is possible in economic transactions for both sides to lose! Consider residents in a city-colony getting their water supply through a distribution pipe. The water is available for a specific period twice a day. Residents collect water in their respective tanks (on the roof) and then use the tank water throughout the day. Usually, the pressure in the distribution pipes is sufficient to carry the water to the roofs of every resident. However, sometimes for various reasons, the pressure could be low and some residents, therefore, install a booster pump to extract more water from the pipes. When one puts in a booster pump, residents beyond the pump get less water than they otherwise would have obtained. So, they also install a booster pump. Of course, if everyone installs a pump, then everyone gets the same water they would have got if no one had the pump! Here is a case where everyone loses; in their bid to get more water for themselves 1

648.

Supreme Court of India—July 28-1998: M C Mehta v Union of India (1998) 8 SCC

142  Law and Economics

they get the same water as before, but now have to pay the additional cost of the pump and the electricity to run it. In trying to ‘win’ more water, each resident loses. To prevent such wasteful expenditure, most cities have made attaching booster pumps to the public distribution system illegal. Alternately, suppose there is no rule regarding which side of the road one must drive. You want to go to a place on the right side of the road and you are on the left side. With no rule, you move over. This blocks traffic on the right side of the road. Similarly, people on the right, wanting to get to the left, move over to the left side of the road. The net result is complete chaos and no one goes anywhere. Law therefore stipulates that you must always allow people moving in the opposite direction to be on your right. In both these cases, law is enabling; it allows activities that would otherwise not be valuable at all. In the chapter on tort law (Chapter 6) we talked about producers’ liabilities and how they create the trust necessary for consumers to buy their products. Legally specified standards play the same role that standards of precaution do. The safety standards prescribed by law that producers must follow assure consumers that the procedures of safety have been laid down by people who have more expertise than they have and that the law will enforce these standards. Use of allowable preservatives in food packaging is an example of this. Prescribing standards for electrical wiring in buildings to prevent electrical fires is another example. Without such laws and their enforcement, markets will not develop and we have the classic cases of market failure.

Economics of Competition Law Suppose there is only one shop in your neighbourhood. You could buy your requirements from this shop or go to another neighbourhood to buy the things you need. However, in the latter case you would have to pay the cost of travel and time, and the effective cost of buying from another shop could be more than what your neighbourhood store is charging. You buy from your neighbourhood store for two reasons: (a) the price you pay for any item is less than what you are willing to pay (net utility from the product—utility in money terms minus the price—is positive) and (b) the price is less than the price plus cost of getting it from another shop. Your neighbourhood store makes a profit because the price it charges is more than its cost of procuring, storing and selling the item(s). Being a neighbourhood monopolist, the shopkeeper makes a sizeable profit, p.



Economic Laws  143

Suppose someone sets up another shop, selling identical items at identical prices. Since no one in your neighbourhood has any particular preference for one shopkeeper over another, they randomly allocate themselves as buyers. Thus, half the people buy from one shop, the other half from the other shop. If the costs are the same to both shopkeepers, each now services half the market and the erstwhile monopolist makes a profit of (p/2), as does the new one. The new shopkeeper realizes that if she charges a slightly lower price, not only would her customers be willing to buy more but customers going to the other shop would prefer to buy from her at the lower price. If the first shopkeeper does not do anything, he will lose all his customers and the new shopkeeper gets slightly less than p (since she is charging a slightly lower price). If the original monopolist matches her price, he will continue to serve half the market. In this case, his profit per sale is lower (lower price but same unit cost) but he sells more than half of what he was selling originally as a monopolist. At the lower price, total sales by both shopkeepers are more than under monopoly. Obviously, it pays the erstwhile monopolist to match the price of the new shopkeeper. If each matches the other’s price, each sells half of what the market buys at the lower price. Equilibrium occurs when both shopkeepers have no further incentive to sell more; since each is matching the other’s price, any further reduction in price reduces per unit profit by more than it increases sales and hence, the overall profit is less. This is the classic duopoly outcome. Let the profit made by each duopolist be p'. We know from economics that the sum of duopoly profits is less than the monopoly profit, that is, 2p' < p. We also know that consumers are getting a better deal. Indeed, what we do know is that the improvement in the surplus enjoyed by the consumers (because of the lower price in a duopoly) is more than the profit lost in the market. If the consumer surplus changes from CS to CS', then CS' – CS > p – 2p', or the total surplus in the society from a duopoly is greater than that from a monopoly, or CS' + 2p' > CS + p. This is the essence of what competition does and why it is more efficient. The example highlights an important point. As we move from a monopoly to a duopoly, consumer surplus goes up but producer surplus goes down (sum of duopoly profits is less than monopoly profit). This continues as we go on increasing the number of shopkeepers, from one to two to three, and so on. Thus, producers as a group do not like competition and consumers as a group love it. Economists, on the other hand, love competition because the sum of producers’ and consumers’ surplus is more under competition, and not because they are driven by any specific love for consumers. There are two basic assumptions for the arguments above and they form the basis of competition policy and law. First, we assume that simply by

144  Law and Economics

observing the profit made by a monopoly shopkeeper, the second shopkeeper enters the market. There is nothing that prevents the ‘entry’ into the marketplace of the second seller. This freedom to enter is the essence of competition. It allows consumers a choice; they can buy from one or the other, or as we say in economics, the consumer can ‘vote with her feet’, moving away from the seller she does not like to the one she does. If society ensures that entrants can enter at minimal cost then competition is encouraged. Entry costs tend to be high if it is difficult to start up a new business venture, giving undue advantage to the incumbent business entity. Second, we assume that the two shopkeepers do not collude but compete. Observe that the two shopkeepers, by competing, get p' each; if they collude and agree to charge the monopoly price, and each agrees to sell no more than half the monopoly output, then each gets (p/2) and as we have noted before, p/2 > p'. Let pn be the profit made by each shopkeeper in the presence of n identical shops, n ≥ 1, and let CSn be the corresponding consumer surplus. Economics tells us that (a) p1 > ∑ni=1 pi for all i > 1, (b) CSi < CSi+1 for all n ≥ i ≥ 1 and (c) p1 + CS1 < pi + CSi < pi +1 + CSi +1 for all i > 1. These conditions are simply generalizations of the comparison between the monopoly and duopoly cases. Given (a), producers (regardless of their numbers) would prefer to collude than compete. If they successfully collude, then the aggregate outcome in the market is no different than that in a monopoly. In other words, while an increase in the number of producers is necessary for greater competition, it is not sufficient; the sufficiency breaks down if they are allowed to collude. Much of competition policy is (and the associated laws are) driven by this basic idea. There are other strategies sellers use to prevent competition. We give an example in a setting more dynamic than what we considered above. Consider two sellers competing with each other; one is a cash-rich seller, also known as a ‘deep pocket’ and the other is not. They are otherwise identical; in each period each sells an amount qd (the subscript ‘d’ denotes duopoly) of a homogeneous product, whose cost is c, and each charges a price p > c. The profit in each period is pd = qd (pd – c). They expect to get this profit over a long period of time, and let the present value of this profit stream be Πd.2 The cash-rich seller has cash, or money, equal to m; the cash-poor seller has no initial money. Suppose the cash-rich supplier chooses a current price, p', such that p' < c < pd. The cash-poor seller has two selling options: (a) continue charging a price greater than c or (b) 2 The present value of a future profit stream is the discounted sum of the profit in each period.



Economic Laws  145

match the price p'. If she follows the strategy (a), she sells nothing and suffers a loss (the cost of stocks or inventory purchases); if she follows (b) she makes a per unit loss equal to (c – p' ). Since she has no money, she cannot undergo a loss and hence, finds it in her best interest to get out of the market. She shuts shop and lets the cash-rich seller become a monopolist. Let q' be the market demand at the lower price of p'. The current loss to the cash-rich seller is (c – p' )q' and being cash-rich, the amount of money m is greater than this amount. He undergoes this loss but becomes a monopolist in the next period. His profit as a duopolist was Πd; his profit from following this strategy is Πm – (c – p' )q', where Πm is the present value of the future monopoly profit stream and  is the discount rate. (The cash-rich seller suffers a loss today and from tomorrow earns the monopoly profit stream; because he earns the monopoly profit from tomorrow and not today, one has to apply the discount rate.) So, if Πm – (c – p' )q' > Πd it pays the cash-rich seller to follow this strategy. This is called ‘predatory’ pricing; charge a low enough price to make the competitor suffer a loss, force her out of the market to become a monopolist and then start making monopoly profits. The real problem here is ex ante prevention of such activities. The competition authority has to be extremely careful and sophisticated to understand these strategies by various companies and step in with fines and other penalties to discourage them from using them. However, the difficulty of observing such practices until after they have worked themselves out can often lead to regulatory authorities becoming too interventionist in their approach. For instance, it is entirely possible that a price reduction by one seller is a competitive strategy to gain greater market share and hence, greater profit. Price reductions increase consumer welfare and so there is no reason for authorities to prevent this. What adds to the authority’s problem is that it is very difficult to correctly estimate the costs of a particular activity since the use of resources leading to these costs are trade secrets that companies do not always want to publicize. Dealing with all aspects of competition law is outside the scope of this book. What is to be noted is that modern economic theory has come a long way in understanding competitive and anti-competitive strategies by firms. Simultaneously, the rich global experience from case laws has also added to our understanding. In India, the Competition Act, 2002 has replaced the earlier MRTP (Monopoly and Restrictive Trade Practices) Act, 1969 and it has incorporated many of the more-recent advances in economic theory. MRTP focused more on the size of the industry, whereas the approach in the Competition Act is to evaluate the behaviour and actual outcome,

146  Law and Economics

rather than the size3 of the company, to determine acceptable strategies. Thus, the possibility that some firms can be bigger due to efficiency and not necessarily due to monopoly, an insight from the economics of competition, is internalized in current Indian legislation. This focus on behaviour is evident from the following clauses in the Act: Any agreement entered into between enterprises, or associations of enterprises or persons or associations of persons or between and person and enterprise or practice carried on or decision taken by any association of enterprises or association of persons, including cartels engaged in identical or similar trade of goods or provision of services, which (a) directly or indirectly determines purchase or sale price, (b) limits or controls production, supply, markets, technical development, investment or provision of services (c) shares the market or source of production or provision of services by way of allocation of geographical area of market, or types of good or services, or number of customers in the market or any other similar way (d) directly or indirectly results in bid rigging or collusive bidding shall be presumed to have an appreciable adverse effect on competition. Provided that nothing contained in this sub-section shall apply to any agreement entered into by way of joint ventures if such agreement increases efficiency in production, supply, distribution, storage, acquisition or control of goods or provision of services.

All arrangements like tie-in, exclusivity clauses in supply or distribution agreements, refusal to deal, resale price maintenance, etc., are considered illegal restrictive trade practices under the Indian legislation, if they cause, or are likely to cause an adverse impact on competition. In general Indian legislation has not made any arrangement per se illegal. This gives enough scope for judicial review. Moreover, if the parties can prove that they adopted such agreements or practices for improvement in efficiency, then the Competition Act may not treat them as illegal. However, much will depend on how precedents are set by case law. The Competition Act is relatively new and we will have to wait and see. This is noted by T. C. A. Anant. See http://www.cuts-international.org/documents/ Day1/Competition%20Policy%20in%20India1.ppt#256 (1, Competition Policy in India: An Overview). 3



Economic Laws  147

Laws to Address Information Issues Increasingly, people are becoming aware of the environment. Accompanying this awareness is the increasing desire of individuals to participate in environmentally friendly activities. These take various forms and shapes; some people buy environmentally friendly cars, some buy shares of companies that spend money on cleaning up the environment, while others buy organic food because they do not like chemical fertilizers. Suppose one such person is your friendly, neighbourhood economist. Having spent all her time trying to learn economics and trying to keep up with new developments in the discipline, she has not had the time to learn how to identify environmentally friendly products (and activities) from those that are not. This requires a certain degree of expertise and knowledge that she does not have the time to acquire. So, how does she do all these desirable things and feel happy about it? This is a real issue for a very simple reason; satisfying one’s desire to be environmentally aware is costly. The cost of environmentally friendly cars is greater than that of other cars; companies that spend on the environment do so using the money that would otherwise go to shareholders; organic fertilizers have lower yields than chemical fertilizers and this increases the cost of the crop. She is willing to pay a higher price for all of these only if she is confident that these are indeed environment friendly. Since she herself cannot examine every activity she depends on a credible body or entity to certify that they are. Such certification has a cost and the certifying authority must be paid. There are two ways to do this. The certifying authority can open shop and wait for our environmentally aware potential buyer to walk in, ask her questions, pay a fee and get the answer. The problem with this model is that once she pays and finds out, all her friends who trust her will find out from her without having to pay the certifying entity. This is certainly not a good business model for the certifying entity! The second option is for the company to pay the entity and use the certificate as an advertisement for its product. The end customers (and their friends if they want the product) still pay, since this becomes a part of the total cost of the product and the company adds (a part or all of) this to the price. Observe that the customers depend on the entity to have produced the right certification. Who guarantees this? The company pays the certifying entity to obtain a ‘green’ certificate and if the certificate wrongly classifies the company as green, the company has no incentive to challenge it. On the contrary, if a certifying entity is very lenient in certifying a company as green, every company will pay to get a certificate

148  Law and Economics

from this entity. (Indeed, if there are competing certifiers, a more lenient one can attract more companies to pay for its certificate and hence, make more profit than a stricter enforcer of green standards.) The customers, of course, have no way of knowing how good a job the entity is doing. In other words, there is nothing in this set-up that convinces the customers that the certifying entity is doing a good job. An important consideration in such certifying mechanisms is the conflict of interest. Lowering the standards of certification attracts more companies seeking to obtain certificates from such an entity. If the ultimate customers of these companies are aware of this but do not know which certifying agency is more lenient, they stop taking these certificates seriously and the value of certification goes down. Hence we need an independent mechanism that cannot gain from ‘inflating’ the actual quality of the product. This is evident empirically from regulations that ensure that incorrect claims are not communicated through advertising. A study of the political economy of such regulation in the United States shows that it was desired because it furnished a mechanism through which firms could improve the credibility of advertising (Hansen and Law 2008). In other words, without a third party that has no stake in the certification process ensuring that correct standards are followed by the certifying entity, the entire business of certification becomes incredible. What we just described is a classic case of market failure; left to themselves, economic agents are unable to generate an activity that creates value through market transactions. In economics we term such instances cases of asymmetric information. The sellers know the true worth of their products; buyers do not. Buyers know that sellers have an incentive to overstate the quality of their products and hence, are sceptical of claims made by them. Buyers can be convinced of the claims regarding quality if these claims are checked by a third party who has no incentive to make false claims on behalf of the sellers. Since the sellers pay for the certification, it is difficult for customers to believe in the third party. Certifiers are, therefore, brought under supervision by a regulatory structure that punishes them if they do not do an honest job. Such informational issues, or asymmetric information, are most prevalent in financial markets. A company trying to sell bonds to the public will overstate its expectations about the company’s future prospects; a small investor wanting to invest in the company does not have the expertise to monitor the sector and will stay out of investing unless she can be assured by a credible entity that the company is making acceptable claims. The corporate bond market will fail to start if potential investors cannot believe in what the companies are saying. The bank that invests your money in various projects



Economic Laws  149

may not be doing a good job and you do not have the time say, as a lawyer, to follow each and every loan the bank makes. As an investor, you depend on financial market regulators, SEBI (Securities and Exchange Board of India) and RBI (Reserve Bank of India) in the two cases, to oversee the companies and banks. The job of these regulators is not to assure you a good return or take away the risk from investments but to ensure that those who have better information about the actual risk in a project are, indeed, sharing this knowledge with others in the market.

Auditing: A Case of Mandated Information Provision The corporate frauds that unfolded in the case of Satyam Computers in India and Enron in the United States shed doubts on the system of auditing of companies and the independence of auditors as providers of financial information about companies. There has been a growing perception that auditors are not performing the due diligence that is required of them. They could be acts of omission such as not looking at the right documentation of company financials, or acts of commission such as collusion with a fraudulent company. These reflect the incentive problems inherent in the relationship between corporate and auditing firms that we referred to above. Since the auditing firm is hired and its charges paid directly by the corporate entity, enough perverse incentives may develop to make the auditors ‘meet’ the requirements of the management of the company she is auditing. Currently, checks and balances in this regard are enforced through peer bodies such as the Institute of Chartered Accountants of India. However this peer group pressure does not seem to override the perverse incentives. Hence the auditors may ‘dance to the tune’ of the company, given its power to appoint the auditor and decide on the charges. To some extent, the incentive system described with respect to auditors is designed along the lines of those for lawyers. The client has the right to appoint the lawyer and the fees are determined through negotiation between the two. The minimum quality and ethics of the service of the lawyer are determined broadly by the peer group known as the Bar Association. However, such self-regulation by a peer group seemed to have failed in the auditor-company relationship. This is important since ‘third parties’ (those beyond the operating owners of the company) too use the information (on the financial health of the company) provided by the auditors. The non-executive

150  Law and Economics

owners (including small shareholders or small holders of units of mutual fund companies having shares of the company), potential lenders and the clients of the services provided by the company, etc., are all using the information provided by the auditors. Obviously, there has been a lot of debate and discussion on how to develop a better system of auditing. There seems to be general agreement on what needs to be retained from the old system. First, the auditing must be carried out by private (or competitive) auditing firms, so that efficiency and effectiveness of auditing services are ensured. This efficiency is not only with regard to the direct cost of auditing but also in relation to quicker assimilation of practices of auditing that maximizes social benefits. Second, the cost of auditing has to be borne by the company concerned, so that it should not be allowed to externalize some part of the cost of its functioning. What are some of the changes that people are discussing? One solution being offered is to constitute a Corporate Audit Commission, independently, or under other related bodies like SEBI. The job of this commission is not to directly audit the accounts of the companies, but to hire auditors to audit each and every company. The hiring can be on the basis of technical quality and cost bidding. The companies have to bear the cost of auditing (which can be indexed on some variables such as revenue or turnover) but fees have to be paid through the Corporate Audit Commission (CAC). Companies should pay also for the functioning of CAC. Alternatively, there could be supervisory auditing to ensure that the primary auditing of specific companies is carried out appropriately. This can be on randomly selected transactions or heads of accounts. Supervisory auditing can also be assigned to audit firms on a competitive basis, excluding the firm that conducted the primary audit. The information provided by the supervisory auditor can also be used to rate the firm that carried out the primary audit. The number and nature of discrepancies found during supervisory auditing (after normalizing for the total number of transactions) can be used to measure the primary auditing firm’s inclination for errors. There should be provisions to see that the discrepancies recorded after the supervisory auditing are not frivolous. Moreover, there should be appeal mechanisms through which a primary auditor can respond to the critical remarks of the supervisory auditor, and sort out the grievances, if any. Through this process, the real discrepancies of primary auditing can be finalized, and these can be an input in the rating of auditing firms. This rating can be considered by the CAC while hiring firms for further assignments, with some firms debarred if their rating is below a specific level. What are the possible loopholes in these mechanisms? One is, of course, that if auditors and those audited can collude, so can primary and supervisory



Economic Laws  151

auditors. Such collusive practices may discourage supervisory auditors from bringing to light the errors, if any, in primary auditing. Thus, it may be necessary to encourage effective competition among the supervisory auditing firms. Larger numbers of such firms and supervision of primary audits being randomly allocated to them each year is likely to make collusion difficult. The urge to compete with each other for better ratings may encourage them to be careful with primary auditing, and to be watchful of others’ mistakes if they undertake supervisory auditing.

9

Economics of the Judicial System Why Go to Court? Suppose you are driving your car and somebody hits you from behind. There is no serious injury to anyone, but both cars are damaged. The problem is deciding who should pay whom. The car behind you says that you stopped suddenly, he was unable to stop in time and so hit you; the fault is yours. You claim that someone stepped out on to the road in front of you without any warning and to avoid hitting her, you had to come to a stop. Had the car behind you kept a safe distance, and not been ‘tailgating’ you, it could have also stopped and not hit you; the fault, therefore, is that of the car behind you. If it is your fault, you pay for the damages to the car behind you (as well as, of course, pay for the damages to your own car); if it is the other car’s fault, you collect damages from him. Unfortunately, each one of you wants to blame the other and we have what can be described as a dispute. It is this dispute that the courts have to solve. The court has four options: (a) ask you to pay for the damage to the other car; (b) ask the other car to pay for the damage to your car; (c) find the person who stepped out in front of you and hold her at fault; and (d) maintain that this was a pure accident and ask each one of you to collect from your respective car insurance companies. Suppose in the society you both live in, neither had anyone experienced such an event nor is there any documentation or record of such an event ever happening. Further, since no one had thought of such a possibility, the motor vehicle rulebook has no mention of what happens in such a circumstance. So, the two of you take your dispute to the court (usually, the motor accidents claims tribunal in India). The court listens to the arguments made by your respective lawyers and uses the various principles governing tort law to decide in favour of one of you. The court says that the person hitting you from behind is responsible for the damage to the cars and should pay you to fix the damage your car has suffered. In deciding in your favour,



Economics of the Judicial System  153

the court makes the observation that while you did a good job in avoiding a genuine accident, the car behind you could have easily avoided hitting you if he maintained a safe distance from you. The owner of the car who hit you appeals against the judgment to a higher court. This court upholds the earlier decision. Once the court has found in your favour, the message goes out to everybody and everyone knows that should the same incident occur in the future, the car that hit from behind will be held liable for damages. Then it is equivalent to the driving rulebook saying that it is up to a car following any other car to avoid hitting the car in front if it suddenly slows down or comes to a stop. This is how law is made by the courts in the common law system—while deciding a dispute, the judgment creates a ‘precedent’ and that becomes the law. The next time such an incident occurs, the car that hit from behind (the second car) would have to come up with a very good reason why it is not liable. Given the decision made in the first incident, any car hitting another car from behind and not wanting to pay damages will have to convince the court that her hitting the car in front was not the same as the first incident. In other words, the ‘facts’ of the case in the second incident are very different, in a logical sense, from those of the first incident. That is, the first decision cannot be taken as a precedent in deciding the second case. This means that the decision in the second case will define whether hitting from behind always creates a liability or only in the circumstances surrounding the first incident. For instance, suppose in the second incident the car in front tried to avoid a street dog rather than a person. Then what happens? Or, the car in front suddenly realized that it had to take a left turn that was passing by and so rapidly slowed down to be able to take the turn. Would the car at the back still be held liable for hitting the car in front (that slowed down)? This is where the clarity of the judgment, and the reasoning behind it, becomes very important. The car in front (the first car) using the earlier case would like to argue that it should get paid by the second car who would, in turn, like to argue that in this particular case, the circumstances are different from those in the first case that has been decided by the court. In other words, this second incident would go to court only if either of the two car owners feels that she has a chance of winning. Suppose the damage to the first car is d1 and that to the second car is d2. Let the cost of going to court be, respectively, c1 and c2. If the court finds in favour of car 1 (the first car), car 2 (the second car) will have to pay d1 to the owner of car 1 and cover (a certain part of) the costs of the owner of car 1 being dragged to court, c1. However, these costs are not simply the observed costs but also include the utility costs of going to court, being cross-examined, dealing with lawyers,

154  Law and Economics

etc. So, only a proportion, , of the actual costs can be covered; thus if the owner of car 2 loses, she will have to pay c1 of the costs, where  ≤ 1. Let p1 be the probability of winning (according to the owner of car 1) and (1 – p1) be the probability (in her opinion) that the court will find in favour of the owner of car 2. Thus, when the owner of car 1 goes to court, she already has a damaged car whose cost of fixing is d1, and has to pay the court costs, c1. With probability p1, the owner of car 1 will win the case and be reimbursed an amount equal to d1 + c1. In that case, the payoff for the owner of car 1 will be: 1.  –(d1 + c1) + (d1 + c1) = –(1 – )c1. With probability p2, the owner of car 1 will lose to the owner of car 2 and she will have to pay damages to the latter, and a part of her actual cost. Then, the payoff for the owner of car 1 will be: 2.  –(d1 + c1) – (d2 + c2) = –(d1 + d2) – (c1 + c2). Figure 9.1 gives the payoff for car 1’s owner, while Figure 9.2 shows the payoff for car 2’s owner (worked out in a fashion similar to that of car 1’s owner).

p1

(1 – p1)

–(d1 + c1) + (d1 + c1) = –(1 – )c1

–(d1 + c1) – (d2 + c2) = –(d1 + d2) – (c1 + c2)

Figure 9.1. Payoff to Owner of Car 1 from Going to Court

(1 – p2)

p2

–(d1 + c2) – (d2 + c1) = –(d1 + d2) – (c1 + c2)

–(d2 + c2) + (d2 + c2) = –(1 – )c2

Figure 9.2. Payoff to Owner of Car 2 from Going to Court



Economics of the Judicial System  155

When car 2’s owner asks car 1’s owner to pay up or come to court, the former has the option of paying up quietly; if she does so, her payoff is the sum of damages to the two cars: –(d1 + d2). If she refuses and decides to go to court, her payoff is: 3.  –p1(1 – )c1 – (1 – p1) [(d1 + d2) + (c1 + c2)] So, the owner of car 1 will go to court if and only if: 4.  –p1(1 – )c1 – (1 – p1) [(d1 + d2) + (c1 + c2)] ≥ –(d1 + d2) Rearranging the terms, we get: 5.  p1(d1 + d2 + c1) ≥ c1 + (1 – p1)c2 This expression is very easy to interpret. The left hand side of the inequality is the expected return from going to court; the owner of car 1 wins with probability p1, gets reimbursed the damage to her own car, d1, and part of her cost of going to court, c1, and saves on having to pay damages to the owner of car 2, d2. This saving is important to understand; by going to court and winning, car 1’s owner saves herself the burden of having to pay car 2’s owner. If she does not win, it could be for two reasons: she did not go to court and agreed to the demand of car 2’s owner or went to court to contest the demand of car 2’s owner and lost. In both cases she would have to pay damages, d2. So, if she wins, this amount is saved. The right hand side of the inequality is her cost of going to court. She has to bear the cost, c1, and an additional cost, c2, if he loses, according to probability (1 – p1). A similar analysis for the owner of car 2 (by symmetry) tells us that she will go to court if: 6.  p2(d1 + d2 + c1) ≥ c2 + (1 – p2)c1 These two inequalities are important elements in our understanding of how law plays its part in improving aggregate welfare in a society. Remember we have experienced the second instance of a car hitting another car from behind and we have one instance of an earlier judgment where a car that hit another from behind was held liable. First, observe that if neither car owner goes to court and one pays the other for damages suffered (irrespective of who pays whom) the total loss suffered by the two car owners is d1 + d2. If the owner of car 1 pays the owner of car 2, then the

156  Law and Economics

former bears the total cost (her own plus the other car owner’s) and if the latter pays the former, then her total cost is again d1 + d2. Also, if neither pays for the other’s damages, then also the total cost is the same, each bearing the cost of the damage to their respective cars. Second, if the case goes to court, then regardless of the judgment, the total cost is d1 + d2 + c1 + c2. In other words, the total cost of going to court is greater than the total cost of not going to court. Therefore, if the judicial system is structured such that the two car owners do not come to court and agree on a decision by themselves, welfare is higher. Thus, an ideally efficient judicial system is one where opposing parties correctly anticipate the court’s decision by studying past judgments (here, the first accident) and carry out a simple negotiation of distributing the savings from not going to court, namely the costs c1 + c2. If each decides to shake hands and leave, the owner of car 1 saves c1 and the owner of car 2 saves c2. Consider the extreme case where both believe that the court will decide in favour of the car in front. Since both think alike, p2 = (1 – p1) and, in this case, p1 = 1, p2 = 0. Then, with positive court costs, the inequality (5) always holds unless c1 is very high and inequality (6) never holds. So the owner of car 2 will never go to court and hence, the parties will settle ‘out of court’ with the owner of car 2 transferring d1 to the owner of car 1. What makes the decision for the owner of car 2 easier is if she refuses to pay, then car 1’s owner will always go to court and then she will lose much more, namely, an additional amount c1 + c2. In a rule-of-law society, courts follow legal principles to arrive at decisions and people learn about them, analysing past cases and corresponding judgments (or their legal counsels inform them). As economists, we want to see two basic principles being followed: (a) clear and transparent laws that encourage out-of-court settlements and (b) the law deciding the dispute in a way that encourages efficient driving on the roads. In other words, we first want to see what is more efficient: holding the car in front, or the one at the back, liable. Second, we want the law to clearly inform the public about this decision through case laws. (In this second aspect, a statutory law by the relevant authority that says, that thou shall under no circumstance hit the car in front, is very helpful for we know that courts uphold existing statutes.) Every time a case comes to court, the entire judicial system spends time, energy and effort. Thus, if the courts can give signals that help people settle disputes by themselves, then welfare outcomes are better. In practice, of course, many things can go wrong. For one, there may not be a clear understanding of the court’s decisions. Of course, if the motor vehicles statute has a clear statement on this, it is less likely. If that is not



Economics of the Judicial System  157

available, but the courts have been very clear about this on past occasions then, again, there is no problem. If both parties to a dispute think they can win, then they will go to court. If court decisions are not consistent over time, then more and more disputes will end up in court. Given scarce resources, it will take longer for cases to be heard and decided and this will create backlogs leading to delays in justice. Just as money today is better than (the same amount) tomorrow, compensations tomorrow for costs incurred today is less desirable than compensations received today! Observe that what we want is one probability being close to one at the same time that the other is close to zero. In our case, the owner of car 1 must have a credible threat of going to court and winning to encourage the owner of car 2 not to dispute her demand for compensation. This is very important; we are not saying that both inequalities (5) and (6) should not hold, but that one should hold and the other not. For instance, if both do not hold (this will be the case if both c1 and c2 are very high) then neither will go to court and the negotiated outcome between the two parties will be now completely dependent on the bargaining strength of each party which often degenerates into a ‘might is right’ policy (as is evident from the instances of ‘road rage’ on Indian city streets). There is no reason to believe that this is an efficient policy! All these are largely dependent on how courts carry out their business. This involves allocation of cases to courts, what the procedures for filing and arguing cases will be, what concessions will be given to the party against whom a charge has been filed, how many days the court works, etc. It also involves the type of training received by the judges and lawyers, body of laws and how they are written, resources in the courts (number of judges and other legal staff), etc. In other words, the efficiency of the legal system is very complicated and needs a clear understanding of the objectives and hence, the process of law.

Incentives for Lawyers So far, we have assumed that the ‘subjective’ probabilities of winning the case and the costs of going to court are both exogenous, or given. The fact of the matter, however, is that better lawyers can argue better the cause of their clients than average lawyers, and better lawyers cost more. In effect then, in many instances, the probability of winning is dependent on the amount spent, or the p’s are functions of the respective c’s. The obvious question then is: can the lawyer’s fees be associated with the probability of winning?

158  Law and Economics

In one sense, it already is. One lawyer is considered better than another because she has won more cases than the other. So, the reputation built through her performance in previous cases is reflected in the amount a lawyer charges. However, we know from economics that this is not a foolproof system, especially when a client and a lawyer are not going to work together repeatedly. So, if you are involved in a one-off case and you hire a lawyer who has built a reputation in arguing such cases, you may think that this lawyer will be good for you. The lawyer, on the other hand, may feel that you will never come back to her for advice regardless of how you fare in this case because, as stated earlier, this is a one-off case. In other words, her future income from you is independent of the outcome of this case. This may not encourage her to put in that extra effort to argue your case. On the other hand, if you face such cases on a regular basis, then the lawyer knows that if she puts in that extra effort and wins the case for you, you will come back to her again in the future. Of course, a part of this problem could be rectified if one knew the complete history of the lawyer—type of cases tried, number of positive findings in favour of her clients, etc. Then, just as you decide to bet on a boxer depending on the number of matches, type of opponents, number of bouts won by knockout and the number won on points, you will choose a lawyer based on her past performances. This still has a problem because, unlike the boxer who cannot always choose his fights, the lawyer can always decide which cases to take up and which to deny. In particular, the lawyer may decide to take up only those cases that she will definitely win and hence, create a 100 per cent record for herself. So, she may refuse your case not because you do not have a point, or do not need to be defended, but because it is difficult to win and you cannot afford the higher cost. This last bit is very important. What this implies is that the poorer litigant is always at a disadvantage with respect to a richer opponent. The richer opponent can hire a better lawyer and since this improves the chances of winning, more judgments will favour the richer people in a society. Of course, this situation will be less important if the probability of winning is more objective than subjective, that is, depends on the merits of the case rather than the merits of the lawyer arguing the case. In the first section of this chapter (titled ‘Why go to court?’), this is what we assumed and that assumption is justified as long as laws are transparent, clear and swiftly enforced. There are many who feel that the self-interest of lawyers, rather than interest towards their clients, adds to the delay in the delivery of justice in the country. The quality of service provided by lawyers, just like any other service, is influenced by the price. One can perceive three types of payment to the lawyers. Payment for work for a particular duration (payment per hour or per day) and payment for a service or for an appearance before the court are well



Economics of the Judicial System  159

known. A third type is where the lawyer is paid conditional on the outcome, for example, a share of the compensation awarded to her client. The incentive of the lawyers to work differs under these different payment systems. When the payment is for duration, the lawyer may have an incentive to delay the court’s decision, especially for one-off clients. When the payment is based on procedure, the lawyer may have little incentive to avoid unwanted procedures. Under both these cases, the lawyer may not have strong incentives to be concerned about the success or failure of the case (other than its positive or negative impact on her reputation). Thus, the lawyer need not give a correct opinion on the ‘win-ability’ of the case, if she is paid on the basis of duration or procedure. This would lead to more cases coming up in courts and clogging up the system of justice. When the lawyer gets a share of the net compensation (compensation minus the costs of going to court) awarded to her client (as in cases of tort or accidents), the incentives of the lawyer are, to some extent, aligned with those of the client. Unwanted spending of time, postponement of hearings, appearances and procedures may be avoided. However, even this form of payment is not altogether problem-free. The lawyer would then have an incentive to take up only those cases where the possible awards are large rather than small. There is one payment system that is a slight variation of the share payment (or percentage contract). This is when a claim is ‘sold’ to the lawyer for a ‘price’. Assume that a person, who is expecting an award of `150,000, sells the claim to a lawyer for a price of `120,000. (This may even be decided based on an auction, where many lawyers compete to get the claim, and selling the claim for the highest ‘price’.) Since the lawyer has paid, or agreed to pay, a fixed amount to the person selling the claim, she has an incentive to get the maximum award from the court, since she retains the whole of the additional amount. However, such ‘sale’ of claims is not allowed in many countries, including India. Thus, it is very difficult to have a payment system for the lawyer that will induce her to have the same objective as the client. This is the well-known principal–agent problem in economics, where the client is the principal and the lawyer the agent.

Incentives for Judges We have discussed the incentives of lawyers and those who want to use the judiciary. What about the judges? The judges are usually appointed by the government and their tenures are fixed (with a fixed retirement age) while

160  Law and Economics

their compensation for work done is totally independent of the case outcomes. This latter is to ensure that their judgments are in no way governed by the possibility of higher compensation depending on what they decide (other than it being correct in law). Also, the removal of judges is usually carried out by the higher level courts (in the case of lower court judges) or through a long-drawn process of impeachment in Parliament requiring a two-thirds majority. This compensation scheme is to ensure that judges’ incentives are not related to the cases that they consider, or the decisions that they take. It is also to ensure that the political system does not tamper with the judicial process (thus making punishment of a judge for a specific decision very difficult). This encourages courts to give judgments against the state, or the legislature, if these are necessary in law. However, the same incentives that make them independent of all other institutions need not necessarily ensure the efficiency of the judges, in terms of, say, the number of cases disposed of within a given time, or the quality of their judgments. For that, one depends on peer pressure to discipline judges. There are many professions where such peer pressure (and not external monitoring) is used to achieve efficiency. The academic profession is such an example. In addition to the fact that judgments should not be governed by market forces (findings in favour of the person who can pay the highest), there is another reason for depending on peer group monitoring of judges—the asymmetry of information. It is difficult for an outsider to judge the quality of performance of a particular judge; it is relatively easy for more experienced judges (and those who are likely to be sitting at higher level courts) to decide on such matters. They are in a better position to take decisions on punishment and the provision of positive incentives (like promotion to higher positions in the judiciary). However, such peer pressure may have some limitations: decisions that may have an impact on others in the profession, like adversely affecting the reputation of the judiciary, may not be taken that swiftly. For example, action against malpractices, or corruption, may tarnish the image of the whole judiciary if such actions are publicly observed and acted upon. We may contrast the incentives for the judges with those for the lawyers. Lawyers are paid by their clients to argue their cases. Thus, revealing information that strengthens the cases of their clients and to some extent, concealing information that may work against the interests of the clients, are considered part of normal responsibility of lawyers. We do not expect advocates to follow the principles of full disclosure about their clients. While lying in court is unethical, not revealing information that helps one’s client, is not considered an unethical act. It is through the cross-examination of



Economics of the Judicial System  161

evidence presented by the lawyers of each party that information about the case is revealed in the court. The incentive for the judges, on the other hand, is to get to the truth of the matter and then use legal principles to come to a decision, and, they must use objective evidence, not subjective hunches or biases, to come to a decision.

Errors in Court In the introductory section of this chapter (titled ‘Why go to court?’), we talked about the incentives for going to court. Remember that in most cases, people go to court because there is a dispute and they want that dispute to be settled in court. In going to court, the litigants commit to accepting the decision arrived at by the court. For the settlement of disputes, a lot of information is necessary, not only for the disputing parties but also for the courts. The litigants, for instance, require information on the probability of winning and the costs they will have to bear to go to court. The courts, on the other hand, require, at least, three types of information—what exactly is the nature of the dispute, in what context did it arise and what would be the best settlement. For the first two, if both lawyers have the right incentives, enough information for making the right decision will be brought out during the trial. Indeed, one of the major objectives of court procedures is to enable the establishment of the ‘truth and nothing but the truth’. The third part, namely, which settlement is the best, is a bit more complicated. As economists, we want the settlement to be efficient. What do we mean by that? As we have discussed before, once a decision is taken by the appropriate court, all similar disputes in the future will be settled in the same way. Disputes arise because of two reasons—one party undertakes some opportunistic action (breaks a promise by, for instance, breaching an agreement in a contract) or something happens that the disputing parties had not anticipated and hence, had no agreement on how to behave in the situation that has arisen, in which what one does affects the other. (Continuing with the example of the contract, what should be done under certain situations was not contractually specified and the contract was left ‘incomplete’. The ‘uncovered’ situation occurs, and the contracting parties cannot agree on what to do in an event that no party had conceived of when writing the contract. So, they move the court.) In the second instance, as long as the court implements what is efficient under the circumstances, the economists have nothing to contribute. What the court implements depends

162  Law and Economics

crucially on the legal principles followed by the society and accepted in law. In particular, laws based on socialist ideals are very different from those based on efficiency, especially in the long run. Recall that in the early chapters of this volume we have dealt with this in detail. For the first type of dispute mentioned in the previous paragraph, we want the court to ensure that promises are not broken or, if broken, the aggrieved party is suitably compensated. So, the type of error that can happen in the system is that the courts miscalculate the amount of the compensation. The difference in what the amount should be and what is actually granted is not the result of a difference in the objectives of the court and the economists. It is an error in precisely calculating the unobservable elements of personal costs. In other words, it is very difficult for the courts to precisely, and objectively, arrive at the efficient value of the compensation in each case of dispute. This is largely because the disputing parties have conflicting views about what the compensation should be—the plaintiff wants a high compensation and the defendant wants a low compensation. So, neither has the incentive to reveal the true loss to the plaintiff. The judges have to estimate what the true loss is and estimates are never entirely accurate. Going back to the example given in the beginning of this chapter, if the efficient solution is to make the second car (the one behind) pay for the damages to the first (the one in front), the economists need to convince the judiciary that this is indeed so. However, even if the courts agree to this, they may still make an error in figuring out what the actual damage to the first car, or d1, is. This has little to do with the court system and more to do with the fact that it is difficult to credibly reveal what is not objectively, and precisely, verifiable by a third party, the court. ‘Verifiability by a third party’ is more than ‘observability’. For example, if someone slaps another in the privacy of someone’s home, the slap may be observable by both the one who slaps and the one who is slapped. But it is very difficult for a third party to verify, someone who was not present at that time, as is the case with the judge trying to be convinced that someone was slapped. Going back to the example of the accident claim, the cost of fixing the damage depends on which garage does the repair or to what level of satisfaction the repair is carried out. A more difficult proposition is that it also depends on how new the car was and how badly the owner feels in having a damaged car (since a repaired car is never the same as a car that has never been damaged). In other words, full efficiency is never possible and this is costly to society. This error gives rise to two types of problems. First, there is a ‘distortion’ in efficient care. The car in front, knowing that it may not get the full compensation, will take more than efficient care to avoid being hit from



Economics of the Judicial System  163

behind. Alternately, if the compensation is overestimated, the car behind will take more than efficient care. Either is costly to society. On the other hand, if extra effort is made to get closer to the actual damage, more resources have to be spent, both by the litigants and the judicial system, and that too is costly to society. So, there is a trade-off to society between these two costs. The courts, with help from economists, can make an effort to ensure that the lesser of the two costs is what society pays. In other words, if the cost of getting more accurate information is larger than the distortion cost due to the compensation amount falling short of the efficient solution, let there be an efficiency loss; if the cost of arriving at a more accurate compensation level is lower than the cost of moving closer to the efficient level, spend resources to get more accurate information. The courts usually address this problem by using specified rules, or formulae, for calculating compensation in different cases. The approach here is not to consider the extent of loss (due to miscalculation of the compensation amount) in a particular case but the average loss across all similar incidents. This formula-driven compensation has an important positive impact. It reduces unnecessary efforts at arguing what the compensation should be in each case; instead, more time can be spent on whether one should compensate the other. Recall that we had said that it is important for courts to be transparent and decisions made by them to be stable. We had said that if litigants know what the judgment of the courts will be in certain types of disputes, they would be less inclined to go to court with these disputes and save on court fees. In compensation disputes, there are two dimensions in which there could be differences of opinion between litigants—who is to blame and the amount of liability of the party being blamed. The formula approach reduces the uncertainty in the second dimension. Because each dispute has specific circumstances, for full efficiency, one needs a customized solution (for instance, for the same damage, the costs would be different for different individuals). However, given the lack of objective information, such customization is not possible. So courts decide, based on specific legal principles, what the average compensation would be in such cases. This obviously means that in particular cases, parties may not be happy with the decision since neither of these parties may be involved in the ‘average’ dispute. To understand this further, let us take a special example. Suppose that the compensation level could be D, or d, but there is no objective way of determining which plaintiff has which of the two costs. Hence, if D > d > 0, both types of plaintiffs will ask for D. Full efficiency demands that those with D should be compensated with an amount D and those with d with an amount d. This is, however, not feasible because the

164  Law and Economics

courts cannot identify who is which type. Suppose they decide on D for both types (i.e., they follow the rule D or always award this amount in all such disputes). All D-type plaintiffs will be happy but the defendants of all d-type plaints will be very unhappy. More importantly, the compensation being higher than what they consider is the damage, the plaintiffs will take more than optimal care (i.e., the care corresponding to each type getting their respective costs). Let this cost, the amount by which aggregate welfare is less, be denoted C1. Alternatively, the courts can decide to allow only the amount d as compensation. Then some of the plaintiffs (namely those with cost D) will take more than optimal care and this will result in an aggregate welfare loss of C2. A third alternative could be that the compensation allowed is D + (1 – )d where the courts know that proportion (1 – ) has the higher cost and the proportion has the lower cost. Once again there will be a welfare loss and this is given by C3. Obviously, efficiency demands that the courts decide on a formula that minimizes the cost or the minimum of {C1, C2, C3}. If any of the first two is the minimum, one group is happy and the other group is not. If the third cost is the minimum, neither group is happy! But, the courts are more interested in the overall efficiency and not the efficiency in a particular dispute. Hence, even if the formula is incorrect in a particular case, to make an argument against it, one has to look at the costs incurred because of that formula in the average future dispute!1

1 An example is the O. J. Simpson case in the United States, in which a famous sportsman and actor was charged with the murder of his wife. The jury acquitted him, while a large section of the people thought that they should have found him guilty. Indeed, many felt that he was acquitted because of the way the judicial process is designed and not because he was innocent. In particular, they held that the jury system was flawed. A pressure was built up to change the way such cases are decided and most said that the jury system should be abolished. The judiciary opposed this and continued with the existing system saying that on an average, the jury system works better towards the objectives of the United States’ legal system. We are explaining that argument here.

Bibliography Agarwal, A. and M. A. Chen. 2008. ‘Do Analyst Conflicts Matter? Evidence for Stock Recommendations’, The Journal of Law and Economics, 51 (3): 503–538. Akee, R. 2009. ‘Checkerboards and Coase: The Effect of Property Institutions on Efficiency in Housing Markets’, The Journal of Law and Economics, 52 (2): 395–410. Anderson, T. L. and D. R. Deal. 2001. Free Market Environmentalism. New York: Palgrave Publishing. Anderson, T. L. and P. J. Hill. 1975. ‘The Evolution of Property Rights: A Study of the American West’, The Journal of Law and Economics, 18 (1): 163–179. Arora, Seema and Shubhashis Gangopadhyay. 1995. ‘Voluntary Overcompliance of Emission Standards’, Journal of Economic Behaviour and Organization, 28 (3): 289–309. Baker, S. and C. Mezzetti. 2005. ‘Disclosure as a Strategy in the Patent Race’, The Journal of Law and Economics, 48 (1): 173–194. Baland, J. M. and J. P. Platteau. 1996. Halting Degradation of Natural Resources: Is There a Role for Rural Communities. Rome: FAO. Basheer, Shamnad and Prashant Reddy. 2008. ‘Ducking TRIPS in India: A Saga Involving Novartis and the Legality of Section 3(d)’, National Law School of India Review, 20 (2): 131–155. Baumol, W. J., J. C. Panzar and R. D. Willig. 1982. Contestable Markets and the Theory of Industry Structure. New York: Harcourt Brace Jovanovich, Inc. Blume, L., D. L. Rubinfeld and P. Shapiro. 1984. ‘The Taking of Land: When Should Compensation Be Paid?’, Quarterly Journal of Economics, 99 (1): 71–92. Boldrin, M. and D. K. Levine. 2008. Against Intellectual Monopoly. Cambridge, UK: Cambridge University Press. Buonanno, P., D. Montolio, D. and O. Vanin. 2009. ‘Does Social Capital Reduce Crime?’, Journal of Law and Economics, 52 (1): 145–170. Burrows, P. 1991. ‘Compensation for Compulsory Acquisition’, Land Economics, 67: 49–63. Cheung, S. N. S. 1970. ‘Contractual Arrangements and Resource Allocation in Marine Fisheries’, in A. Scott (ed.), Economics of Fisheries Management: A Symposium, pp. 97–108. Vancouver: University of British Columbia Institute of Animal Resource Ecology. Coase, R. H. 1960. ‘The Problem of Social Cost’, The Journal of Law and Economics, 3: 1–44. Cohen, M. and V. Santhakumar. 2007. ‘Information Disclosure as a Regulation: A Theoretical Analysis’, Environmental and Resource Economics, 37 (3): 599–620. Cooter, R. and T. Ulen. 2004. Law and Economics. New Jersey: Pearson Addison Wesley. Debreu, G. 1959. Theory of Value: An Axiomatic Analysis of Economic Equilibrium. New Haven: Yale University Press. Demsetz, H. 1967. ‘Towards a Theory of Property Rights’, American Economic Review (Papers and Proceedings), 57 (2): 347–357. Dilulio Jr., J. J. 1994. ‘The Question of Black Crime’, Public Interest, 117 (Fall): 3–24.

166  Law and Economics Feeny, D., S. Hanna and A. F. McEvoy. 1996. ‘Questioning the Assumption of the “Tragedy of the Commons” Model of Fisheries’, Land Economics, 72 (2): 187–205. Gordon, H. S. 1954. ‘The Economic Theory of a Common Property Resource: The Fishery’, Journal of Political Economy, 62 (2): 124–142. Hansen, Z. K. and M. T. Law. 2008. ‘The Political Economy of Truth-in-Advertising Regulation during the Progressive Era’, The Journal of Law and Economics, 51 (2): 251–270. Hardin, G. 1968. ‘The Tragedy of the Commons’, Science, 162 (3859): 1243–1248. Hazlett, T. W. 2008. ‘Property Rights and Wireless License Values’, The Journal of Law and Economics, 51 (3): 563–598. Heaton, P. 2006. ‘Does Religion Really Reduce Crime?’, The Journal of Law and Economics, 49 (1): 147–172. Hoyt, R. E., D. B. Mustard and L. S. Powell. 2006. ‘The Efficiency of State Legislation in Mitigating Moral Hazard’, The Journal of Law and Economics, 49 (2): 427–450. Ihlanfeldt, K. R. 2006. ‘Neighbourhood Crime and Young Males’ Job Opportunity’, Journal of Law and Economics, 49 (1): 249–283. Jenson, G. F. 2006. ‘Religious Cosmologies and Homicide Rates among Nations: A Closer Look’, Journal of Religion and Society, 8. Kaffine, D. T. 2010. ‘Quality and the Commons: The Surf Gangs of California’, The Journal of Law and Economics, 52 (4): 727–744. Kamit, R. E. 2010. ‘Analyzing the Effects of Temporary Anti-Trust Immunity: The AlohaHawaiian Immunity Agreement’, The Journal of Law and Economics, 53 (2): 239–262. Lovenstein, M. C. and V. Y. Suslow. 2006. ‘What Determines Cartel Success?’, Journal of Economic Literature, 44 (1): 43–95. McGee, J. 1958. ‘Predatory Price Cutting: The Standard Oil (NJ) Case’, The Journal of Law and Economics, 1: 137–169. Milgrom, P. and J. Roberts. 1992. Economics, Organization and Management. Prentice-Hall. Peltzman, S. 2005. ‘Aaron Director’s Influence on Anti-Trust Policy’, The Journal of Law and Economics, 48 (2): 313–330. Philipson, T. J. and R. A. Posner. 2009. ‘Antitrust in the Not-for-Profit Sector’, The Journal of Law and Economics, 52 (1): 1–18. Platteau, J. P. 1989. ‘Penetration of Capitalism and Persistence of Small-Scale Organisation Forms in Third World Fisheries’, Development and Change, 20 (4): 621–652. Santhakumar, V. 2003. ‘Citizens’—Action for Protecting the Environment in Developing Countries: An Economic Analysis of the Outcome with Empirical Cases from India’, Environment and Development Economics, 8 (3): 505–528. ———. 2011. Economic Analysis of Institutions: A Practical Guide. New Delhi: SAGE Publications. Seldeslachts, J., J. A. Clougherty and P. P. Barros. 2009. ‘Settle for Now but Block for Tomorrow: The Deterrence Effects of Merger Policy Tools’, The Journal of Law and Economics, 52 (3): 607–634. Shavell, S. and T. van Ypersele. 2001. ‘Rewards versus Intellectual Property Rights’, The Journal of Law and Economics, Part 1, 44 (2): 525–547. Smith, A. 2008. An Inquiry into the Nature and Causes of the Wealth of Nations: A Selected Edition. Edited by Kathryn Sutherland. Oxford, UK: Oxford University Press. Stigler. 1964. ‘A Theory of Oligopoly’, Journal of Political Economy, 72 (1): 44–61. Taylor, J. E. 2007. ‘Cartel Code Attributes and Cartel Performance: An Industry Level Analysis of National Industrial Recovery Act’, The Journal of Law and Economics, 50 (3): 597–624.



Bibliography  167

Varian, H. 1987. Intermediate Microeconomics. New York: W.W. Norton. Wilson, J. Q. and Herrnstein, R. 1985. Crime and Human Nature. New York: Simon and Schuster. Zentner, A. 2006. ‘Measuring the Effect of File Sharing on Music Purchases’, The Journal of Law and Economics, 49 (1): 63–90.

Index access to education, as right, 5 acquisitions, 58, 64. See also land acquisitions, by government regulation versus, 75–76 ‘actionable’ offences, 122 adverse possession, concept of, 39 annual fee, to patent holders, 47 asymmetric information, 148–149 auditing, 149–151 ‘barbed wire’, invention of, 31 blood donations, 40–41 branding, 50 breadth of patent, 47 broad patenting, 47 certification, mechanisms of, 147–148 coastal zone regulations in India, 61 common property rights, 31 Competition Act, 2002, 145–146 competition law, economics of, 142–146 competition, prevention of, strategies for, 144–145 competitive market equilibrium, 8 constraints to one’s maximization problem, 33 contract acceptable, 91–98 as agreement enforceable by law, 89–90 agreements under duress, 92, 95 annulment of agreement by parties, 90–91 breach of, remedies for, 103–108 default rules, 106–108 deliberate act and, 104

expectation damages, 105, 106 opportunity cost damages, 105 reliance damages, 106 specific performance, order of, 104 compensation on breaking of, 91 definition of, 86 enforceability of, challenging, 99 ex ante value enhancing, 88, 96 ex post and ex ante efficiency, 98–101 information and signing of, 99–98 information issues in, 99–98, 101–103 mental capacity to, 93–96 by minor, 93, 94 need for, 86–91 opportunistic behaviour, 88–89, 92 penalty clauses in, 98–100 promisor and promisee, 98 purpose of contract law, 90 stock markets and, 96–97 and surplus-enhancing investments, 102 two-way transfer in, 86–87 value-enhancing voluntary agreements, 92–93 contract law, 140 contributory negligence, 121 copyright challenges to, by information technology, 49 duration of, 49 for expressions of ideas, 48–49 purpose of, 48 search cost, 49



Index  169

Corporate Audit Commission (CAC), 150 courts dispute settlement by, 152–153 errors in, 161–164 incentives for going to, 152–157, 161 crime rate psychology of persons and, 132–133 religion, role of, 134 social capital, effect of, 133 social norms in, role of, 133 socioeconomic factors, influence of, 132 criminal law death penalty in, 137–138 discounting of future costs by people, 134–135 investments in reducing crime, 131–132 need of, 126–128 punishment in degree of, 130 increase in expected cost of, 129 objective of, 130 reducing crimes, ways of, 128–134 ex ante and ex post actions, 136 role of private individuals, 135–137 and tort law, difference between, 126–128

need of, 139–142 objectives in, 140 economic transactions and sports, comparison of, 139–142 EEZ (exclusive economic zone), 70 efficiency, 8–9 efficient level of production, 18 in exchange of goods and services, 18–19 at individual level, 17–18 need for laws for, 19–23 eminent domain, concept of, 77 emission standards, 68 endowments, 8 environmental economics, 4 environmental pollution, 65 compensation for damage, 66 negotiated solution, non feasibility of, 66 regulation of activities causing, 67 mandatory information disclosure, 68–69 PILs, filing of, 69 standard-based regulation, 67 taxes, 67–68 expectation damages, 105, 106 externality, 7

Daya Shankar vs Smt. Bachi And Ors, 93 death penalty, 137–138 default rules, 106–108 ‘diminishing marginal utility of money’, 117, 118, 129 discounting of future costs, 134–135 duopoly outcome, 143 duties, 33

goods and services, categorization of, 59 government takings, 58

economic analysis of law, 3–6 uses of, 23–25 economic laws, 139

fair play, 140 financial market regulators, 149 formation defence, 95

Hamelo (Deceased) vs Jang Sher Singh, 94 hold-up problem, 73, 83 illegal restrictive trade practices, 146 imprisonment, 129 informational issues, addressing of, 147–149 information, disclosure of, 102

170  Law and Economics injunction, 36–37 insider trading, 97, 101, 102 intellectual property, 55 copyrights, 48–49. See also copyright patents, 43–48. See also patent protection of, 42–43 trademarks, 49–51. See also trademark intellectual property rights. See also intellectual property case studies on, 51–56 Eastern Book Company case, 52–53 Novartis AG and the Government of India case, 54–55 Singhania and Partners LLP and Microsoft case, 55–56 United States’ 9/11 Commission report publication, 53 granting monopolies, arguments against, 57 need for, 42 use of, 42–43 judges incentive for, 159–161 peer group monitoring of, 160 removal of, 160 Land Acquisition (Amendment) Act, 1984, 77 land acquisition bill, 83–85 land acquisitions, by government, 64 adequate compensation in, providing, 71–75 case studies Poletown in United States, 78, 80–82 Singur in India, 79–80 laws governing, 76–77 legislation for, 83–85 for public purposes, 70–71 and land issues, 76

third party, need of, 82–83 land occupation without legal title, 38 law and economics, 2–6, 140 enabling law regime, 140–142 law of limited liability, 13 lawyer, payment system for, 157–159 loan, 87 marginal cost, 44–46 ‘market failure due to asymmetric information’, 40 mineral extraction, land ownership and, 61–62 central government ownership, 63 compensation for surface rights, 65 mining company, bargaining with, 63–65 public ownership, 62–63 Monopoly and Restrictive Trade Practices Act, 1969, 145 narrow patenting, 47 negative externality, 26, 60 negotiation, for efficient solution, 34, 36 Ninth Schedule of the Constitution, 77 ‘no property rights’, 31, 36, 69 ‘opportunistic behaviour’, 88–89, 92 opportunity cost damages, 105, 106 organ trade, 40–41 out-of-court settlements, 156 Pareto efficiency, 8 patent, 43–48 breadth of, 47 duration of, 46–47 granting of, considerations in, 47 monopoly to patent holder, 44 creation of, reasons for, 46 inefficiency in, 44–46 patenting rule in society, 47–48 product-patenting regime, 48 sole right to inventor, 43–44

peer pressure, 160 penalty clauses, in contract, 98–100 performance excuse, 95 ‘polluter pays’ principle, 66 poverty, and crime rate, 132 precedent, 16, 153 predatory pricing, 145 price, 44 price discrimination, 55 private property, 33 property. See also property rights investments in, 30 legal right to, 35–36 ownership of, 30 owning without title, 38–39 registration, 38 sale of, restrictions on, 40–41 property rights changes in, 31 common, 31 cost for protection of, 30–31 emergence of, 30–31 lack of, 26–28 legal protection of, 31–35 for maximizing surplus, 28 need of, 26–30 private, 28, 30, 31 registration of transactions, 38 remedies on violation of, 36–37 awarding damages, 37 injunction, 36–37 rule for distributing surplus, 29 tax on surplus, 30 property tax, 29 prostitution, 40 public corporations, limited liability requirement of, 12–14 public economics, 4 public interest litigation (PIL), 69 public purpose, concept of, 76, 77 public rights over property community ownership, 69 environmental pollution and, 65–69 mineral resources, 61–65

Index  171 need of, 58–61 state ownership, 70 punitive damages, 118–119 rational behaviour, 7 RBI (Reserve Bank of India), 149 registration, need of, 38 regulation and acquisition, difference between, 75–76 rehabilitation, 130 reliance damages, 106 retribution, 130 rights, drawing boundaries of, 33 search cost, 49 SEBI (Securities and Exchange Board of India), 149 social norms, violence-enhancing, 133 societal welfare maximization, 7 stock markets, trading in, 96–97 supervisory auditing, 150–151 tort law ‘actionable’ offences, 122 behavioural aspects of users and, 123 case studies, 123–125 compensation amount, fixing of, 119–120 contributory negligence, 121 damages for harms in, 109–110 desirable levels of precautions and, 111–112 diminishing marginal utility of money, concept of, 117, 118 liability in, fixing of, 110, 127 negligence and liability in, 125 producers’ liabilities in, 120–121, 142 punitive damages, 118–119 use of, for preventing accidents, 111, 112–118 trademark generic name as, use of, 51

172  Law and Economics for products identification by consumers, 50–51 purpose of, 49–50 registation of, 50, 51 reputed firm and, 50 transaction cost, 34–36

valuations on natural resource, 31 value maximization principle, 10–12

unemployment, and crime rate, 132 United Nations Convention on the Law of the Sea, 31

zero crime society, 130

wealth effect, 10 welfare, 6 World Trade Organization (WTO), 48

About the Authors Shubhashis Gangopadhyay is currently the Research Director of India Development Foundation, Director of the School of Humanities and Social Sciences, Shiv Nadar University, and a Visiting Professor at the University of Gothenburg, Sweden. He obtained his PhD in Economics from Cornell University, USA, in 1983 and joined the Indian Statistical Institute where he became a Professor in 1991. In 2003, he became the founder-Director of India Development Foundation, an independent research institution. He was awarded a doctorate (honoris causa) by the University of Gothenburg, Sweden, in October 2006. In 2008, the year of the global financial crisis, he was appointed Advisor to the Finance Minister, Government of India. He is the Chief Editor of the Journal of Emerging Market Finance and has been on the editorial boards of Journal of Financial Stability and Review of Development Economics. He is a member of the Board of the Centre for Analytical Finance (Indian School of Business, Hyderabad) and International Development Enterprise India. He has been a member of the Bankruptcy Task Force of the Initiative for Policy Dialogue (IPD), Columbia University; the International Advisory Board of the Centre for Law and Economics of Financial Markets (Copenhagen Business School); the South Asia Chief Economist’s Advisory Council of the World Bank; and the board of Industrial Reconstruction Bank of India. Currently, he is also an advisor to the Competition Commission of India. He is also the founder-President of the Society for the Promotion of Game Theory and Its Applications. He has published widely in international journals and has a number of books to his credit on economics and finance. His published work (books) includes Waiting to Connect (co-authored), 2008; Economic Reforms for the Poor (co-edited), 2000; Counting the Poor: Where Are the Poor in India (co-authored), 1998; The Institutions Governing Financial Markets (edited), 1996; Enabling Financial Institutions (coauthored), 1997; Economic Theory and Development Policy (co-edited), 1992; and Economic Development and Policy (co-edited), 1991. V. Santhakumar is currently Professor at Azim Premji University, Bangalore. He had been on the faculty of the Centre for Development Studies,

174  Law and Economics

Trivandrum, for 15 years from 1996. He has held post-doctoral scholarships at the Wageningen Agricultural University (The Netherlands) and Vanderbilt University (USA). He has also carried out a number of consultant assignments for the Asian Development Bank, United Nations Environment Programme (UNEP), and United Nations Development Programme (UNDP) in different states of India, and in Tajikistan, Palestine, Laos, etc. He has received the research medal and the outstanding research award of the Global Development Network during its initial years. He specializes in analysing institutional issues in environment, natural resources, energy, infrastructure, and so on. In addition to publication in international journals, he has authored books titled Analysing Social Opposition to Reforms (SAGE, 2008) and Economic Analysis of Institutions: A Practical Guide (SAGE, 2011).

Law and Economics Volume 2: Practice

Edited by Shubhashis Gangopadhyay V. Santhakumar

Copyright © Shubhashis Gangopadhyay and V. Santhakumar, 2013 All rights reserved. No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage or retrieval system, without permission in writing from the publisher. First published in 2013 by Sage Publications India Pvt Ltd B1/I-1 Mohan Cooperative Industrial Area Mathura Road, New Delhi 110 044, India www.sagepub.in Sage Publications Inc 2455 Teller Road Thousand Oaks, California 91320, USA Sage Publications Ltd 1 Oliver’s Yard, 55 City Road London EC1Y 1SP, United Kingdom Sage Publications Asia-Pacific Pte Ltd 33 Pekin Street #02-01 Far East Square Singapore 048763 Published by Vivek Mehra for Sage Publications India Pvt Ltd, typeset in 10/12pt Adobe Garamond by Diligent Typesetter, Delhi, and printed at Saurabh Printers, New Delhi. Library of Congress Cataloging-in-Publication Data Gangopadhyay, Shubhashis. Law and economics/Shubhashis Gangopadhyay, V. Santhakumar.    pages cm   Includes bibliographical references and index.   1. Law—Economic aspects—India. 2. Law and economics. I. Santhakumar, V. II. Title. KNS74.G38   343.5407—dc23   2013   2012047864 ISBN:  978-81-321-1009-5 (HB) The Sage Team:  Rudra Narayan, Aniruddha De, Rajib Chatterjee and Dally Verghese

Contents List of Abbreviations vii Preface ix Introduction xi Section One: Property Rights 1. Rehabilitation of the ‘Project Affected’: Eminent Domain and Just Compensation Malabika Pal 2. ‘Law and Economics’ of Indian Patent Law V. Santhakumar 3.  Two Cases of IPR Issues in India: Pharmaceutical Patents and Film Industry Copyrights Sushma Kindo

1 19

37

Section Two: Contracts 4. Economics of Contract Law in India Indervir Singh  he Economics of Contract Law: A Business 5. T Outsourcing Application George S. Geis

51

77

Section Three: Tort Law 6.  Economic Analysis of Tort Law: Some Conceptual and Interpretative Issues Satish K. Jain 7.  Efficiency of Liability Rules: An Experimental Analysis in India Sanmitra Ghosh and Rajendra P. Kundu

96 111

vi  Law and Economics

Section Four: Regulation  8.  Regulation in the Case of a Natural Monopoly: The Electricity Act in India V. Santhakumar  9. Initial Public Offerings: Institutions in India Padma Kadiyala 10.  Legal and Judicial Process in India: A Preliminary Economic Analysis of Some Issues V. Santhakumar

132 153

173

About the Editors and Contributors 191 Index 193

Abbreviations ADR BSE CCI DIP Guidelines DTS EMR FAR FIIs FIs ICA IPO IPR LDC NCE NGOs NMDC NRP NRRP NSE PAN PILs QIBs R&D SEBI SEBs SEC SEZs SLA SOEs SRA STU TRIPS TSC WTO

alternative dispute resolution Bombay Stock Exchange Controller of Capital Issues Disclosure and Investor Protection Guidelines diagnosis, therapy or surgery exclusive marketing rights floor area ratio foreign institutional investors financial institutions Indian Contract Act initial public offering intellectual property rights load despatch centres New Chemical Entities non-governmental organizations New Delhi Municipal Council National Rehabilitation Policy National Rehabilitation and Resettlement Policy National Stock Exchange permanent account numbers public interest litigations qualified institutional buyers research and development Securities and Exchange Board of India state electricity boards Securities and Exchange Commission Special Economic Zones service level agreement state-owned enterprises Specific Relief Act state transmission utility Trade-Related Aspects of Intellectual Property Rights total social costs World Trade Organization

Preface Getting institutions (including laws) right for development has become a cliché in economic literature these days. Such thinking has encouraged the designing of a number of ‘legal reform’ projects currently being implemented in developing countries funded by multi-lateral institutions. With the setting up of new generation law schools in India, teaching economics or economics of laws in general, and ‘law and economics’ in particular, have become an accepted practice in legal education within the country. All these developments take place without having an adequate knowledge base that uses economics in analysing laws and legal issues of India and other developing countries. This is a serious lacuna since the need for context-specific material is obvious for any systematic study of institutions (including laws), and the insights provided by the general theory of ‘law and economics’, though useful, are inadequate. During our experiences with teaching, as well as during our public engagements, we strongly felt the need for a book on law and economics which explains theory with case materials from India. The purpose of the current book is, therefore, to communicate the economic insights to a wider audience (including lawyers) who may have no formal training in economics. We include, along with the theory, applications on actual laws and legal issues. In different ways we teach the same subject but to be honest to ourselves, we realized that we need the help of practitioners to be able to drive home the points we were trying to make. Hence, while the first volume focuses only on the theory, the second is edited by us with contributions from others. Shubhashis entered this field through a course he offered at Gothenburg University, Sweden, in the early 1990s, on the law and economics of capital markets. Clas Wihlborg, then at Gothenburg University, was instrumental in encouraging him to offer this course and Shubhashis owes him a deep sense of gratitude for getting him started. This developed into a course in law and economics which he has been offering for more than a decade at the Jawaharlal Nehru University, India. So, obviously, his students, both in India and Sweden, have had a large role to play in the contents of this book. Like Santhakumar, Shubhashis also had the good fortune of coming in contact with Professor N. R. Madhava Menon. Indeed, without Professor Menon,

x  Law and Economics

this book would not have been conceived, let alone written. Shubhashis has also benefited greatly from discussions with his Swedish colleagues, Ulf Petrusson, Stefan Sjogren and others, in the Gothenburg law department. Santhakumar had started working on economic analysis of law when he was a post-doctoral scholar at the Vanderbilt University. His contributions in this area included an economic analysis of public interest litigations in India, and a theoretical analysis (jointly with Mark Cohen) of the mandatory information disclosure. There was a demand for teaching law and economics as part of the M.Phil Course in Applied Economics at the Centre for Development Studies (CDS, where he was working then) mainly because of the persuasion of Professor N. R. Madhava Menon (who founded the new generation law schools in India) while he was the chairman of the governing board of the CDS. On taking up this responsibility, Santhakumar felt the non-availability of Indian applications as a major challenge for teaching the course. He was toying with the idea of developing a number of ‘law and economics’ commentaries of Indian laws and legal issues. Santhakumar wishes to note that his excitement in learning law and economics was primarily ignited by the lectures (which he has audited) by Professor Andrew Daughety of the Vanderbilt University. The different batches of M.Phil students of the CDS who have attended the lectures of Santhakumar, and his doctoral student Indervir Singh who has worked on the economics of Indian contract law have helped to sustain the interest in the subject. Professors Narayanan Nair and Sunil Mani of the CDS were instrumental in assigning the responsibility of teaching law and economics to him. The project of developing such a book became very appealing when Sugata Ghosh of SAGE Publications came forward and offered to publish the book. The proposal for joint work in this regard reduced the burden on each of us and that made the project look more feasible. The response of potential authors of the application chapters for the edited volume was very encouraging. We are happy that the efforts during the last one to one and a half years by the authors, contributors of the chapters in the edited volume and the publishers have finally led to this publication. The two ladies in Shubhashis’ life—Kasturi and Ishanti—were mostly unimpressed by this book-writing activity but graciously put up with it. People who suffered Santhakumar’s obsessive compulsive disorder in matters related to academics are Abja and Karthika.

Introduction This book is the second volume of a two-volume series on Law and Economics in India. The purpose of the series is two-fold. First is to develop an accessible writing on the basic concepts of the theory of ‘law and economics’ which can be used by all those who do not have a proper training in economics, including lawyers. Second purpose is to develop a set of applications, which include ‘law and economics’ commentaries on Indian legislations. This volume consists of these applications. There are 10 chapters in this (second) volume. Three of them deal with property rights, out of which one is on governments’ power to acquire land and the policies to provide compensation. It deals with the policies for rehabilitating and resettling the people affected by the compulsive land acquisition in the country. This is an important property-right issue subjected to political debates in India currently. Two more chapters deal with property rights, but these are on intellectual properties. Third chapter is a law and economics commentary of the recently adopted patent act in India (in place of an older act). Two court cases—one that deals with patent and another one on copyright—that attracted considerable public attention in the country are analysed in another chapter. There are two chapters discussing contracts. One chapter assesses Indian contract Act based on the theoretical debates within law and economics. The other one discusses the practices and challenges in contracting with regard to outsourcing—an activity very important for the recent economic surge in India. The two chapters on torts are not typical applications. In general, the tort legal practice is less developed in India, except those dealing with motor vehicle accidents. One chapter analyses theoretically some of the assumptions behind the law and economics of torts, and show that these insights need not be valid with a different set of assumptions. The other on torts provides the results of an economics experiment, where the results are not in tune with theoretical predictions in law and economic literature. Though these two chapters are not applications in one sense, it may be useful for teaching purposes.

xii  Law and Economics

The following two chapters deal with some specific issues of what can be called economic regulation. Natural monopoly in the provision of certain public services makes governments to intervene. One chapter analyses this issue in the context of electricity supply in India. More specifically it evaluates the Indian Electricity Act passed in 2003. The last chapter analyses the legal and institutional provisions that regulate the initial public offers of shares in companies which wish to start trading its shares in stock markets. This deals with the need for regulation, and compares Indian practices with those of United States of America. The final chapter takes up some issues of legal and judicial process in India. Delay in court proceedings and to get final judgment is a major problem of Indian judiciary. This delay may have implications for the incentive to use out of court settlements and alternative dispute resolution mechanisms. Public interest litigation (PIL) is another notable feature of the legal process in India. The merits and demerits of PIL especially in the context of the delay in judicial process need to be analysed. These are the issues taken up briefly in the final chapter. A little more detailed account of the insights of each of these chapters is given below.

Property Rights The Land Acquisition Act currently in use in India vests wide ranging powers on the state. It can use this coercive power to acquire land for what is called public purpose. The government can acquire land for private companies for the setting up of industries. The newly drafted bill has narrowed the scope by limiting the use of acquisition for private companies only when 70 per cent of the land is bought by the company on its own (through market) route. Though this is a change in right direction, there is considerable confusion over the ‘public purpose’. This does enable governments from acquiring land even for those uses beyond public good and infrastructure, where there is some rationale for government intervention, and for those cases where continuous stretch of land are not needed, or where market purchase does not lead to much transaction costs or where the hold-out problem does not create much social cost—cases where law and economics literature sees a legitimate reason for the use of land acquisition legislation. The compensation for land acquisition is dealt with in detail by Malabika Pal. There is considerable theoretical debate on when shall compensation

Introduction  xiii

be paid (and need not be paid) for land acquisition for enhancing social welfare. The framework in this regard by Michelman (1967) brought in the concept of demoralization costs—which are caused by unpredictable redistributions caused by the state. This cost consists of ‘disutilities’ arising from the realization that no compensation is offered, and also the lost future production caused by the demoralization of uncompensated losers. There is also resettlement cost required to reach compensation settlements adequate to avoid demoralization costs. Thus if the net benefit of project is lesser than the demoralization cost (or the settlement cost to avoid such demoralization) then the project with land acquisition should not be undertaken. Government should undertake the project when net benefits are greater than demoralization costs/settlement costs, but compensation needs to be paid only when settlement costs (avoiding demoralization) need than the cost of such demoralization. There have been important theoretical developments furthering/critiquing Michelman’s framework. An important insight of all these papers is the need to take into account demoralization costs, and not to argue that compensation is inherently ‘good’ in all contexts. Malabika Pal uses this framework to analyse the National Resettlement and Rehabilitation Policy (NRRP 2007) of India. She finds that none of the provisions in Indian legislation adequately addresses the issue of demoralization costs. Moreover, there are instances of land acquisition where government acts as the facilitator for getting land for the private companies. This may lead to greater demoralization, since the acquisition can be seen as an ad-hoc process with lesser predictability on future government behaviour. Moreover, when private companies are asked to pay ‘just compensation’ (based on market price or such yardsticks) for the land acquired for them (through the facilitation of the government), one needs to ask whether these companies bear the demoralization costs imposed on the society. Law and Economics of intellectual property rights like patents and copy rights are the focus of the following two chapters. Santhakumar analyses the key provisions of the Indian Patent Act (as amended in 2005). The act incorporates a number of clauses which are efficiency enhancing. The definitions of inventive step and new invention used in recently amended Indian act are closer to international definitions. The provisions for multiple durations, compulsory licensing, well laid out opportunities offered to contesting parties to be heard, etc., are enabling provisions. However, the provision that excludes ‘mere discovery of a new form of a known substance or the mere discovery of a new property or new use for a known substance or the mere use of a known process, machine or apparatus unless such known process result in a new product or employs at least one new reactant’ from patenting can be problematic. Since any

xiv  Law and Economics

new firm can explore the new properties or the new uses of a known (patented) substance, and since patenting makes the information on the substance available in public domain, this restraining clause in the patent act cannot be justified as one aimed at limiting monopoly power. Moreover, the more appropriate inventive activity for a developing country may require a narrow patenting, and the restraining clause is against this requirement. Moreover, the other provisions of the act could have been used to avoid ever-greening, if that were the concern, and this does not require such a restrictive scope for patenting. As noted in the case of land acquisition act, there seems to be an effort to address some redistributive concern through the patent act, and this may not be an appropriate intervention for such redistribution. The social debates on patents in India are influenced excessively by the concern on drug prices. This may not be a desirable situation as a patent system to address this problem can be costly for some other sectors of the domestic economy. This is the general issue of distorting economic laws with the objective of redistribution (rather than using other instruments for such redistributive ends). Though India may have benefited from a regime of weak patenting (mainly through the growth of pharmaceutical industries and also the domestic availability of relatively cheaper drugs), we need to see whether the current socioeconomic conditions would require a similar or different patenting regime. It is interesting to see the changing political economy in this regard, wherein the domestic actors who may have preferred a weak patent regime, may over a period of time become advocates of stronger patents, along with an increase in their R&D capability and the losses to national economy due to trade sanctions, if any, for not having a TRIPS-compliant patent system. Sushma Kindo takes two recent court cases regarding Intellectual Property Rights in India. First is the well-known case filed by the Novartis (a Swiss pharmaceutical company) in India regarding the drug ‘Glivec’. The contested issue was the rejection of the patent application for the said drug by the Indian patent office, for it being ‘only a new form of a known substance’. The position of the Patent office is held valid by the respective High Court of India too. As mentioned in the previous chapter, the amended patent act in India excludes a mere discovery of a new form of a known substance from patenting, unless its efficacy is proven to be enhanced. One can view the court judgment in two ways: On the one hand it can have beneficial impact by preventing frivolous patenting and ever greening. On the other hand, it can prevent patenting of incremental innovations, which are very important for a developing country such as India. The second case that Sushma deals with is with regard to copy rights in film industry. Novelist Barbara Taylor Bradford filed a suit against Sahara

Introduction  xv

Media Entertainment in Kolkata High Court with the allegation that the television series namely Karishma produced by the latter was based on an unauthorized use of her book. The High Court (and later the Supreme Court of India on appeal by the Novelist) ruled that there was no case of plagiarism in this instance, since television series was based on an idea of ‘a woman making it from rags to riches’ and there cannot be copyright over such ideas. Like the copy right laws prevailing in other major countries, Indian law assigns copyrights only on the ‘form, manner and arrangement, and the expression’ of the ideas, and not on ideas per se. Copyrighting ideas, as noted in the literature, can increase the search cost to the authors, but it has little effect on the cost of copying. Thus the court verdict in this case can be said to promote economic efficiency.

Contracts Indervir Singh analyses the efficiency implications of breach remedies provided under Indian Contract Act and Specific Relief Act. The Indian contract law is based on English law, and shares many rules regarding the award of the breach remedies. The law and economics literature has different views on the efficiency of breach remedies. One strand of literature argues that common law approach is inefficient in the context of penalty clause and specific performance, and emphasized the need for adopting a more liberal approach towards penalties and making specific performance a routine remedy. The chapter argues that Indian contract law needs to be more flexible in awarding specific performance and providing parties-agreed damages when the parties involved are capable enough to take informed decisions. Despite that Indian law is based on English common law, they differ significantly on some points with regard to contract act. In Indian law, the damages agreed by the parties in the contract act as a ceiling for the amount of damage which can be awarded to the plaintiff, and the promisee cannot sue for the court-calculated damages. It is argued that Indian law is efficient in this regard, because the parties know more about their expected loss from the breach, and court’s assessment of the damages (which are often calculated as the difference between the market and the contract price) may be higher than the actual loss due to the difficulty in furnishing evidence on profits. Indian law allows the voluntary surrender of the right to claim damage at the time of contracting. This rule may be efficiency enhancing based on the argument that parties, especially the capable ones, are in better position

xvi  Law and Economics

than the court to decide the provisions in a contract, and this may include a provision for ‘no damage in the case of a breach’. Hence, such voluntary surrender of damages should be allowed until there is a chance that the contract provisions are the result of unequal bargaining and/or asymmetric information among parties. The third provision, where Indian law departs from the English law, is by allowing the promisee to accept a consideration lesser than that envisaged in the original contract. The chapter argues that this provision may be useful in avoiding indirect loss. However the fact that the plaintiff has to suffer indirect loss may provide the defendant an opportunity to compel the plaintiff to accept lesser consideration. The article by George S. Geis looks at outsourcing contracts from a ‘law and economics’ perspective. It may be noted that outsourcing has become an important driver of recent economic growth in India. It is seen that outsourcing relationship is structured through a number of overlapping contracts. This plurality of contracts reflects the transaction costs involved in negotiating and drafting detailed contract terms. These transaction costs include the upfront costs needed to bargain over clauses and to document a contract, and ex-post costs of adjudicating a contractual conflict in the event of a break down in the relationship. As is well known, there can be some trade off between these two costs, which may lead to some gaps in the contract at the time of its formation, which may be filled up by the courts in the event of a dispute. However this contracting is done in the context of a legal assurance that the breached-against party can recover adequate damages. This is much more important for outsourcing relationships where in both the parties may have to make substantial relationship-specific investments. The relationship between contract and outsourcing is much more complex. The cost of contracting has an important bearing on the decision of whether to outsource or to have a vertically integrated production process. Outsourcing contracts have to overcome the inherent principal-agent problem with different levels of asymmetric information. How do the multiple and overlapping contracts made for outsourcing overcomes these inherent problems is the main focus of this chapter. He discussed the following specific strategies: (a) staged commitment through inter-locking multi-contractual framework, (b) the use of redundant agents or the retention of duplicate activity within the firm, (c) incentive compatible compensation, (d ) explicit monitoring and control rights, and (e) for ‘cause’ and for ‘convenience’ exit rights. Regarding exit rights, there is a trade off involved in providing liberal exit rights for the principal (when the outsourcing agency can exit easily from the contract while seeing some ‘opportunistic’ behaviour on the part of the outsourced firm). However, the relationship-specific investments may make agents reluctant to enter into such

Introduction  xvii

contracts with liberal exist rights, or the agents may need costly inducements. This may require some voluntary surrender of liberal exit rights by the principals (outsourcing agency, in this case). George S. Geis has demonstrated some of the contracting problems in outsourcing relationships with the help of a few of such real-world contracts.

Torts The two chapters on torts, unlike other chapters in this volume, are not applications in a strict sense. The first one by Satish K. Jain is a theoretical critique of the mainstream formulation—that the common law is by and large efficient—in the domain of tort law, especially the efficiency of liability rules. His argument is that the established efficiency of negligence rules depends on the notion of negligence used. Negligence is defined here as the short fall from due care. Under this notion, negligence rules (of different versions) lead to efficient outcomes, which is one basis of the argument on the efficiency of common law thesis. Citing literature, he argues that courts do not use such a concept of negligence. Instead, courts reckon negligence on the basis of whether a party has failed to take some precaution ‘which would have resulted in averting loss of an amount greater than the cost of precaution’. He uses such an alternative notion of negligence to consider the efficiency of liability rules. When the negligence is seen in terms of the existence of cost justified untaken precautions, the negligence rules need not be efficient under all circumstances. There are two other dimensions on which he extends his criticism of the thesis of efficiency of common law. Under the two different notions of negligence discussed here, there may not be incentives for the parties to invest to reduce the cost of care. Or it need not pay to be ‘dextrous’, or there can be perverse incentives to hide one’s dexterity in this regard. Even if we go by the standard argument on the efficiency of liability rules, there can be several such rules which are efficient. But in reality only a few are chosen. This leads to the question of whether there is any additional criterion that is used to select a few from a wider set of efficient rules. Thus efficiency cannot be the sole explanation for prevalence of tort law (or law in general). The chapter by Sanmitra Ghosh and Rajendra P. Kundu presents the results of an experiment conducted in India on the efficiency of liability rules (in tort situations). The studies in experimental economics testing the theories of ‘law and economics’ are not many (and there are only very few

xviii  Law and Economics

such experimental studies conducted in India). A similar test was conducted by Kornhauser and Schotter (1990) in the context of unilateral care. They found that the care exercised by the parties under negligence rule corresponds to the predictions in theory, especially where the care standard is set way above the socially optimal level. The chapter by Ghosh and Kundu is on two-party interactions in which only the injurer’s choice affects the expected loss of accident. The experiment is to test hypotheses including the following: Under no liability rule an injurer will always choose no care; an injurer will always choose efficient care under the rule of strict liability. The experiments were conducted with 42 subjects, recruited from the population of graduate and undergraduate students from a University in Kolkata. The result indicates that the average care chosen under no liability was not zero (and is significantly positive). Subjective information increases the average care level substantially. Under strict liability average care is always higher than the efficient level. The authors of the chapter argue that the results of the experiment do not support the efficiency claims of the rules of no liability and strict liability in unilateral care models. However there could be disagreements by others even when they do not see the validity and framework of such experiments. Subjects can also be motivated by altruistic or communion feelings or norms. Can we interpret the efficiency-based analysis or advocacy for the design of rules by ‘law and economics’ scholars as their rejection of any altruistic or norm-based behaviour on the part of individuals? (If they do so, what is their legitimacy in doing so?) Would there be an increased reliance on such norm-based behaviour in contexts such as India where formal enforcement tort laws is weaker? Since the rule of strict liability enhances choice of care in the experimental results too, isn’t there a movement in the direction of the theoretically predicted behaviour, as the rule changes from ‘no liability’ to ‘strict liability’? Since aggregate behaviour could be a reflection of many factors (which are the subjects of rational choice analysis and those beyond), could it be used to negate the predictions of economic theory? Barring these questions, the model building and the design and testing of experiment, and the possibilities of such experiments as part of studies in economics, as evident from this chapter, are useful for researchers and practitioners.

Regulation What can be the desirable role of a law in the case of public services having the features of ‘natural monopoly’? This is the core issue analysed in the

Introduction  xix

context of Electricity Act of India (2003). It is well known that there are multiple institutional mechanisms to address the natural monopoly. These include government ownership of the company providing such good/service, government regulation of private monopoly, ex-ante competition for the license to serve as monopoly, and encouraging competition wherever it is appropriate. Moreover, the natural monopoly is not a static phenomenon. It is dynamic, and depending on technological developments, and socioeconomic change, certain services which showed natural monopoly in the past may cease to be so over a period of time or in specific contexts. Thus the law used to address natural monopoly should be flexible enough to internalize the prospects for this change. The Electricity Act (2003) incorporates many provisions for achieving economic efficiency. However some of the ‘conflict of interest’ issues are left unaddressed. The provisions that that give power to the vertically integrated utility at the state level (which runs transmission and distribution business) to influence the decision on wheeling charge and surcharges for providing open access (to those consumers who want to buy electricity from another generator/distributor) can prevent the realization of open access. Moreover, the law is not very clear with regard to the continuation/abolition of cross-subsidy, which may also influence the efficient functioning of the system. However, the act does not impose a particular form of intervention (government ownership or regulation or competition) in all the contexts, but at the same time, it does not prevent the adoption of any one form. Thus the act can be seen as an umbrella provision which facilitates the adoption of appropriate institutional intervention, depending on the specific socioeconomic and technological factors of the given context. It is argued that, this is a desirable feature of the Electricity Act (2003). Padma Kadiyala takes up the legal and institutional provisions that facilitate initial public offering (IPO) in India, and compares them with those of the United States of America. State intervention is needed in this area to see that the investors are not cheated by fraudulent practices of the firms. However, excessive intervention can discourage firms from the process of equity issuance and this may lead to a need for excessive dependence on bank-financed capital. This may increase the cost of capital and can affect the growth of the economy and also result in inefficient allocation of capital. This shows the importance of an appropriate form of facilitation of the process of IPO. There were legal changes in this regard in India in the early 1990s—by replacing the Controller of Capital Issues with a regulatory body, namely Securities and Exchange Board of India (SEBI)—which provides guidelines for the issue of IPOs. Over a period of time SEBI has tightened eligibility norms for the companies

xx  Law and Economics

to issue IPOs, regulated the book building process to set the offer price, and tightened restrictions on share allocations by the underwriters. The chapter elaborates the theoretical explanations and also presents evidence regarding under-pricing—a phenomenon (with an average IPO increases in price on the first day of trading) widely observed in regard to IPOs. There is some indication of asymmetric information influencing underpricing in India, with IPOs that receive high grading by the credit rating agencies experience less under-pricing. Probably under-pricing can also be a mechanism used to encourage over-subscription, which in turn is used to avoid allocating large chunks of shares to a few investors (which may enhance the potential to create conflicts over control with the current management). The author argues that Indian law makers have probably developed a more appropriate institutional support system—probably less intrusive and more facilitating than US legislations including Sarbanes-Oxley Act of 2002. (The learning from the experience of the developed world may have facilitated the making of relatively more effective regulatory mechanism in India.)

Legal and Judicial Process Indian courts are known to take substantial time to deliver final judgments. The large number of pending cases is an indication. How does this delay influence the incentives to settle out of court and to use alternative dispute resolution mechanisms is one issue discussed in the chapter. Public interest litigations have become a notable feature of Indian judicial system. There are some important advantages due to the PIL. However, these can also be misused. Moreover the delay in getting court judgments may create perverse incentives to misuse PIL and a few of these can become nuisance suits. Some of these situations (especially those related to the control of pollution) are discussed in detail in the chapter. Moreover the possibility of delay itself can create an incentive to use appeals as a way of delaying the implementation of verdicts given at lower courts. There are efforts being made to reduce the delay in Indian courts as part of the ongoing reforms. However some of these reforms are opposed by interest groups such as lawyers’ associations. A ‘law and economic’ analysis, as the one attempted here, can bring out the incentives of such interest groups to oppose the reforms. The cases of corruption are not so rare in Indian judiciary. Broadly there are two types of measures proposed for tackling corruption in judiciary: (a) those that give more autonomy to the sitting judges; (b) measures that

Introduction  xxi

make executive and legislature more powerful in dealing with judicial corruption. The merits and demerits of these two approaches are discussed in the chapter. Due to the delay in court decisions, one can see people using community norms or informal (and sometimes illegal) instruments to enforce contracts and other such deals. The delay encourages the use of such instruments of enforcement in two ways. The excessive time taken to get the court-awarded remedy encourages the affected parties to use non-legal means to punish the defaulters. Secondly those who carry out such punishments (illegally) are not punished by the courts in time, and this too encourages these illegal actions. This volume has picked up only a few cases for demonstrating the use of a ‘law and economics’ analysis or commentary of Indian laws or legal issues. It is hoped that many such issues will be taken up in future by scholars and a ‘law and economics’ analysis would become a common practice in the legal discourse in India.

References Kornhanser, Lewis and Andrew Schotter. 1990. ‘An Experimental Study of Single Actor Accidents’, Journal of Legal Studies, 19(1): 203–233. Michelman, Frank I. 1967. ‘Property, Utility, and Fairness: Comments on the Ethical Foundations of “Just Compensation” Law’, Harvard Law Review, 80(6): 1165–1258.

Section One: Property Rights

1

Rehabilitation of the ‘Project Affected’: Eminent Domain and Just Compensation Malabika Pal

Introduction The preamble to the National Rehabilitation and Resettlement Policy (NRRP), 20071 in India states: Provision of public facilities or infrastructure often requires the exercise of legal powers by the state under the principle of eminent domain for acquisition of private property, leading to involuntary displacement of people, depriving them of their land, livelihood and shelter; restricting their access to traditional resource base, and uprooting them from their socio-cultural environment.

The principles of compensation when eminent domain is used need to be analysed in the Indian context, particularly since there is an attempt to review2 the land acquisition and rehabilitation policies that India has had so far. In recent times, the focus is as much on the circumstances3 when eminent 1 This policy was approved by the Union Cabinet on 11 October 2007. It was published in the Official Gazette and came into force with effect from 31 October 2007. 2 The Rehabilitation and Resettlement Bill, 2007 was proposed so as to give statutory backing to the NRRP, 2007. The Land Acquisition (Amendment) Bill, 2007 was proposed to align its provisions with the goals and objectives of the NRRP. Both bills were passed by the Lok Sabha on 25 February 2009 and tabled in Rajya Sabha on 26 February 2009. However, both bills lapsed with the dissolution of the 14th Lok Sabha. The Land Acquisition, Rehabilitation and Resettlement Bill, 2011 is under consideration in Parliament at present. 3 It is interesting to note that while the draft of the National Rehabilitation Policy (NRP), 2006, released by the Ministry of Rural Development, Department of Land Resources, Government of India had stated: ‘Provision of public facilities or infrastructure sometimes requires…’, the NRRP, 2007 replaced it by ‘often requires’. (Emphasis added.)

2  Malabika Pal

domain is being used as on the compensation package that is offered to those affected in case it is exercised for a legitimate purpose. Compulsory acquisition of land by the state has long been a subject of debate. From the enclosure movement in 17th-century England4 where people were evicted from their traditional lands to, the present day where land is being taken for industrial and infrastructure purposes across the world, acquisition of property without the owner’s consent has been widespread. In India, the issue has become important again due to the large-scale acquisition of land for setting up Special Economic Zones (SEZs).5 The justification provided by the state is the power of eminent domain (as the above quote reveals clearly). Countries have enacted laws allowing for eminent domain (United States), compulsory purchase (England) or compulsory takings (Australia, New Zealand, Thailand). The Fifth Amendment to the United States’ Constitution explicitly added the eminent domain (takings clause), which states: ‘nor shall private property be taken for public use, without just compensation’. In India, this authority is granted by the Land Acquisition Act, 1894 (hereafter referred to as the Act). When property owners are informed by a government agency that their property will be taken by eminent domain, they have no legal power to resist. Section 6(3) of the Act states that a declaration of the government that the land is being acquired for a public purpose is conclusive evidence that the land is needed for public purpose and unless it is shown that there has been a colourable exercise of power the courts cannot go on to examine whether it is a public purpose or not.6 In the United States, the exercise of eminent domain can be challenged on the grounds that it is not for a legitimate public purpose. In most countries, at most, a claim can be made for just compensation. Although private enclosures had been going on for centuries, it was after 1801 when the first General Enclosure Act was passed that the movement gained momentum. It has been estimated that whereas in 1700 about half the arable land of the country was still cultivated on the open field system, by 1830 almost all of it was enclosed. 5 Granting of mining rights over tribal lands to private corporations is another important acquisition issue in India today. Proposals to produce biofuel, if approved, would give rise to land requirements much greater than those envisaged for SEZs. 6 In Somavanti and Others v. State of Punjab and Others (And Connected Petitions) 1963 AIR (SC) 151, it was observed by the Supreme Court: 4

Now whether in a particular case the purpose for which land is needed is a public purpose or not is for the State Government to be satisfied about. If the purpose for which the land is being acquired by the State is within the legislative competence of the State the declaration of the Government will be final subject, however, to one exception. That exception is that if there is a colourable exercise of power the declaration will be open to challenge at the instance of the aggrieved party. (Available at http://www.indiankanoon.org.)



Rehabilitation of the ‘Project Affected’  3

With the state being vested with the power to compulsorily acquire private property and individuals having limited recourse to judicial intervention, the issue of abuse of power becomes an important one. While the law requires just compensation for outright use of the power of eminent domain by which it takes physical possession of private property, the law does not say anything when the state uses regulation to restrict citizens from doing what they want with their property. In fact, courts have granted compensation when tangible things were actually taken by the government, but have refused compensation when there was a reduction in the value of property due to a regulation. While the former is categorized as ‘possessory takings’, the latter have been termed ‘regulatory takings’. Here, a distinction is often made between the state’s use of its power of eminent domain and police power. The focus of this chapter is on the former. The chapter is divided as follows: in the first section the theoretical literature on the issue of just compensation is explored in the context of the state’s use of its eminent domain power. In the takings literature, there is a debate about whether compensation should be paid at all. In particular, the section focuses on the utilitarian approach of Michelman (1967), which has dominated the academic discussion on the takings issue for decades. The next section analyses the principles governing compensation in the Act as well as the provisions of the NRRP, 2007 in terms of Michelman’s framework to bring out the obstacles that stand in the way of truly ‘rehabilitating’ those affected by the state’s use of its power of compulsory acquisition. The final section provides the concluding remarks.

Theoretical Underpinnings of the Compensation Issue When private property is taken for public use, a universal requirement is that just compensation be paid. The economic argument is that if the government has to pay for the resources it gets efficiency is promoted (see, among others, Baxter and Altree 1972 and Epstein 1985). Without the provision of just compensation, individuals would not undertake productive investment, expecting the government to take away their property. Compensation protects private entitlements and at the same time restrains the power of the state from taking property too often. In fact, two critical limitations put in place to avert the creation of a Hobbesian sovereign were that the eminent domain power be allowed only for public use and just compensation be paid. Hobbes, emphasizing the dangers of unbridled self-interest, had

4  Malabika Pal

argued for a comprehensive contract by which all individuals surrender their liberty and property to a sovereign. However, there are costs associated with the requirement of just compensation. Many studies have shown that the government’s willingness to acquire land is sensitive to whether it has to pay the full price. Enhanced compensation could, therefore, deter governments from undertaking projects.7 There is a large body of work on what are the tests that distinguish between compensable and non-compensable takings. In a very influential article on the issue, Michelman (1967) provides a somewhat formal framework in which he argues that the only ‘test’ for the requirement of compensation is the test of fairness: is it fair to put into effect a particular social measure without compensating for the private loss inflicted in the process? Drawing upon the utilitarian property theory of Bentham, he argues whether collective allocational decisions that are accompanied by particular distributions are compensable or not would depend on whether they are ‘critically demoralizing’. Bentham defined property as a collection of rules that govern the exploitation of resources and therefore form a basis of expectations. Human motivations that result in production will not operate without secure expectations about the future enjoyment of product. Unpredictable redistributions would therefore destroy this well-being and would affect morale. However, not all redistributions are totally demoralizing and a line can be drawn between compensable and non-compensable collective impositions. According to Michelman’s utilitarian standard, in deciding whether a government action was a taking that requires compensation, three factors need to be considered: efficiency gains, demoralizing costs and settlement costs. Efficiency gains are: The excess of benefits produced by a measure over losses inflicted by it, where benefits are measured by the total number of dollars which prospective gainers would be willing to pay to secure adoption, and losses are measured by the total number of dollars which prospective losers would insist on as the price of agreeing to adoption. (Michelman 1967)

This can be denoted by B – C. Demoralization costs (D) consist of the total of (1) the dollar value necessary to offset disutilities which accrue to losers and their sympathizers specifically8 from the realization that no 7 Cordes and Weisbrod (1979) found that after the federal statute Uniform Relocation Assistance and Real Property Acquisition Policies Act, 1971 had been in effect for four years, there was a reduction in real highway construction outlays of US$2.2 for each dollar of compensation paid. 8 This means that these costs do not overlap with the cost term C in the calculation of efficiency gains. They arise due to the lack of compensation and disappear if compensation is paid.



Rehabilitation of the ‘Project Affected’  5 compensation is offered, and (2) the present capitalized value of lost future production (reflecting either impaired incentives or social unrest) caused by demoralization of uncompensated losers, their sympathizers, and other observers disturbed by the thought that they themselves may be subjected to similar treatment on some other occasion. (Michelman 1967: 1214)

Settlement costs (S) are ‘measured by the dollar value of time, effort, and resources which would be required in order to reach compensation settlements adequate to avoid demoralization costs’ (Michelman 1967). In other words, demoralization costs are those that arise when there is no compensation, settlement costs arise when compensation is paid. According to Michelman, a government measure whose dollar benefits exceed costs should nevertheless not be adopted if the net benefit is exceeded by both demoralization costs (D) and settlement costs (S). Symbolically, a utilitarian does not undertake the project if (B – C) < min (D, S). If net benefits are positive and greater than either settlement or demoralization costs (or both), the lower S or D should be incurred. That is, the government should pay if (B – C) > S and S < D. On the other hand, the government need not pay if (B – C) > D and D < S. The source of the demoralization costs lies in the distinction that can be made between strategically determined losses and natural hazards. Michelman (1967: 1217) writes: Society has not yet placed itself under any systematic discipline designed to assure people of compensation for all economic losses inflicted by forces regarded beyond social control, such as earthquakes or plague. If then, the just compensation requirement is supposed to rest on strictly utilitarian grounds— if, that is, it is supposed to rest on a purpose of forestalling demoralization which impairs the output of goods—there must be at work a tacit assumption that losses which seem the proximate results of deliberate collective decision have a special counterproductive potency beyond any which may be contained in other kinds of losses.

What is emphasized is that compulsory acquisition being ‘self-determining and purposive’ would have a greater disincentive effect compared to other loss-producing forces which are ‘randomly generated’. He further elaborates that demoralization costs would arise in case of uncompensated government actions in which for instance, the efficiency gains of the project are doubtful, or where settlement costs are low so that compensation can be provided easily or if losers think that they have to bear a disproportionately large burden. Also, situations where there is a feeling that the burden for which compensation is sought is not a rare or peculiar one and there would be similar ones in the future or there is uncertainty about reciprocal benefits or there is a lack of political

6  Malabika Pal

influence to extract mitigating concessions on the part of losers would lead to demoralization costs. Apart from the direct disutility of those whose losses are not compensated to their satisfaction, Michelman also takes into account the disappointment of the losers’ sympathizers and costs associated with social unrest. Scholars have interpreted settlement costs to mean the transaction costs of using eminent domain. These include procedural costs incurred in assessment of property values, services of experts and litigation costs, and dead-weight losses from taxation to finance just compensation. Also included are the costs of moral hazard, i.e., the prospect of compensation can induce landowners from behaving in inefficient ways. When calculating demoralization costs, if the reduction in investment and productive activity caused by incomplete indemnification is included and property owners are assured that they will be fully compensated when their property is needed for public use, then landowners may over invest to enhance the value of their property in an attempt to increase compensation awards. Beginning with Blume et al. (1984), this sort of inefficiency became the principal basis of academic questioning of the basic advantages of paying just compensation. They arrived at the controversial conclusion that one way to induce landowners to invest efficiently in the face of a takings risk was to pay no compensation at all.9 One criticism of this ‘no compensation’ result is that the government responds to incentives just like any other economic agent and if required to pay too little or nothing at all for the resources it uses, it will take too much. Full compensation is therefore necessary to prevent this type of government moral hazard. Blume and Rubinfeld (1984) offer an analysis of compensation as insurance and suggested that courts should distinguish their awards of compensation on the basis of the risk preferences of those affected by takings. Compensation by government agencies is viewed as a last resort only to be awarded to those who could not buy private insurance.10 Kaplow (1986) argues that the first component of Michelman’s demoralization costs appears to be directly connected to the economic conception of risk, because the concept of risk aversion captures the disutility to losers. The second component representing the disincentive to invest arising from uncompensated takings conflicts with the first factor because 9 The result came to be advanced as the basis for a substantial re-evaluation of the compensation practice in American law. 10 In a country like India, where the market for insurance is relatively underdeveloped and many landowners are poor, this analysis suggests full compensation.



Rehabilitation of the ‘Project Affected’  7

compensation distorts incentives. While Michelman suggests that both the factors of risk and incentives favour compensation, Kaplow regards compensation as primarily serving the function of insurance. While analysing legal transitions like tax reform, deregulation and takings, he argues that uncertainty regarding government action is in many respects like other types of uncertainty. Since the market is normally relied upon to provide a balance between dealing with risk and preserving incentives, the market would be able to address the effects of legal transitions more efficiently compared to provision of transitional relief, like compensation in case of takings, by the government. He, therefore, reaches the conclusion that the moral hazard problems that Blume et al. (1984) identify are more effectively controlled by private insurance. One can conclude from Kaplow’s analysis that the main thrust of his argument was to undercut the case for just compensation in so far as it rests upon the concept of demoralization costs. However, as emphasized by Fischel and Shapiro (1988) (and Michelman himself), the source of the demoralization costs is that they are distinct from ordinary risks, like natural disasters. Kaplow (1986: 534, note 66), while recognizing this distinction made by Michelman, rejects it as unpersuasive. Fischel and Shapiro (1988) argue that considering the risk of a taking and the risk of a natural disaster to be analogous led Blume et al. and Kaplow to consider moral hazard as a component of demoralization costs when it logically belonged to settlement costs. This is because landowner moral hazard arises from the expectation that compensation will be paid. Also, one reason for the neglect of Michelman’s distinction between random hazards and government taking was an assumption that government actions seem like a naturally caused random event, rather than deliberate and purposive. Government is ‘an unimpeachable benefit-cost machine’ (Fischel and Shapiro 1988: 285). In such a world, demoralization costs would be zero. Since settlement costs are always positive, due to the moral hazard problem, S > D for all projects. This gives the ‘no compensation’ result. Fischel and Shapiro (1988) conclude, therefore, that Michelman’s framework can accommodate the problem of moral hazard. Regarding the question of whether compensation should be paid, Michelman’s utilitarian framework provides a method of evaluation which is consistent with the principles of economic efficiency. Moreover, it addresses the central normative question of compensation, which is when to respect individual property entitlements in a realistic set-up of strategically determined government actions. This has not been captured by the insurance rationales for compensation since they do not contain demoralization costs. Despite

8  Malabika Pal

alternative formulations, Michelman’s approach continues to be the most influential of the dominant approaches to the compensation question.11 Epstein (1985), in an important work on takings, analyses the level and form of compensation under eminent domain in detail. He argues that in principle, ‘the ideal solution is to leave the individual owner in a position of indifference between the taking by the government and retention of the property’. According to him, it is the value of the property, not its cost, which determines the amount of compensation to a party. One measure is market value, that is, the price a willing seller would receive from a willing buyer. However, there would remain the offer–ask disparity, since people demand more to voluntarily surrender an entitlement they possess than they would pay to obtain an identical entitlement that they were initially assigned. The basic difficulty of the market value formula has been identified to be the denial of compensation for real but subjective values. One alternative measure of compensation is replacement cost, but even if the owner can regain the subjective component of value, he has no incentive to reveal it and may simply not construct the substitute facilities. Offering a bonus value over and above the market value has been suggested in view of the infringement of autonomy due to the forcible nature of the exchange and the systematic underestimation of value in the market value test. Although payment below market value meets with unanimous criticism, a number of doctrines set benchmarks for compensation below market value, especially when cost or previous value is used as a benchmark. Citing the case of Penn Central Transportation Co. v. City of New York,12 Epstein argues that it was incorrect to think that the state’s obligation is discharged whenever it permits the owner of private property to enjoy a reasonable return, for in the above-mentioned case, the property owner was deprived of all or part of the appreciation in market value between the time of his original acquisition or improvement and the date of condemnation. Miceli (1991) suggests that compensation could be set at the full market value evaluated at the efficient level of investment.

The Indian Context Using Michelman’s vocabulary to analyse the issue of just compensation in India, it is found that none of the provisions that lay down the broad 11 ‘Michelman (1967) is the most frequently cited (in law journals) article on the takings issue’ (Fischel and Shapiro 1988: 277, note 15). 12 4380 S 104 (1978).



Rehabilitation of the ‘Project Affected’  9

framework for the government’s acquisition of private property and compensation principles addresses the issue of demoralization costs incurred by property owners. The right to property has proved to be the most contentious both during the framing of the Constitution in India and after its commencement. The consensus reached in the Constituent Assembly was embodied in Articles 19(1)(f) and 31 of the Constitution. In particular, Article 31(2) stated: No property shall be compulsorily acquired or requisitioned save for a public purpose and save by authority of law which provides for acquisition or requisitioning of the property for an amount fixed by such law or which may be determined in accordance with such principles and given in such manner as may be specified in such law; and no such law may be called in question in any Court on the ground that the amount so fixed or determined is not adequate or that the whole or any part of such amount is to be given otherwise than in cash…

However, the clauses relating to property rights were so controversial that they were omitted by the Constitution (Forty-fourth Amendment) Act, 1978 by which the right to property ceased to be a fundamental right. Article 300-A was included in Part XII of the Constitution, which provides that no person shall be deprived of his property save by authority of law. Also, the obligation of the state to pay compensation for property acquired or indemnification of property under Article 300-A or other public purpose was obviated by this amendment. This was explicitly stated in Jilubhai Nanbhai Khachar v. State of Gujarat by the Supreme Court:13 It would thus be clear that acquisition of the property by law laid in furtherance of the directive principles of State policy was to distribute the material resources of the community including acquisition and taking possession of private property for public purpose. It does not require payment of just compensation or indemnification to the owner of the property expropriated. It is very negation of effectuating the public purpose. Payment of market value in lieu of acquired property is not sine qua non for acquisition.

It was further stated that the adequacy of the amount paid cannot be questioned in a court of law. However, the validity of irrelevant principles in the determination of the amount is amenable to judicial scrutiny. The general principles determining compensation in India have been laid down in Sections 23 and 24 of the Act. There has been elaboration 13

AIR 1995 SC 142; Supp (1) SCC 596; (1994) Supp 1 SCR 807.

10  Malabika Pal

and expansion of these principles by the judiciary,14 which has followed the practice of case-by-case adjudication15 in deciding the compensation issue. Section 23 of the Act provides that in determining the amount of compensation the market value of the land on the date of publication of the notification should be taken into account. The market value is the price the property may fetch in the open market, if sold by a willing seller unaffected by the special needs of a particular purchase. In the absence of any direct evidence, the courts may take recourse to other known methods. These are: 1. Comparable Sales Method 2. Capitalisation Method 3. Valuation by Experts It has been held that the best evidence of market value is the sales of comparable properties, which are bona fide transactions, proximate in time to the date of acquisition, similarly situated and possessing similar advantages.16 Where there is no definite evidence of sales of similar lands in the neighbourhood at or about the date of notification under Section 4(1) or otherwise, the courts will resort to the capitalization method which is interpreted as the method used to convert future benefits to present value, discounting such future benefits at an appropriate rate of return. It is the process of converting the net income of property into its equivalent capital value. The determination of the reasonable rate of return with respect to investment in various types of properties seems to be the most difficult part of this method. The courts usually calculate ‘a number of years purchase of the actual or immediately prospective profits of the lands acquired’.17 Normally, this method is used when no other method is available. Further, it has been held that these methods do not preclude the courts from taking into account any other special circumstance into consideration, the requirement being always to arrive as near as possible to an estimate of the market value. Two important principles laid down in cases are: (a) the land acquired has to be valued not only with reference to its condition at 14 In State of West Bengal v. Arun Kumar Mitra, (1984) CLJ 220 (DB), it was held that Section 23 is not exhaustive. 15 Union of India v. Pramod Gupta, 2005 AIR SCW 4645; JT 2005 (8) SC 203; 2005 (7) Scale187; AIR 2005 SC 3708; (2005) 12 SCC 1. 16 Ravindra Narain v. Union of India, AIR 2003 SC 1987; (2003) 4 SCC 481; 2003 AIR SCW 1491; 2003(2) Scale 509; 2003(2) Supreme 893; JT 2003(5) SC 160. 17 Special Land Acquisition Officer v. P. Veerabhadarappa, AIR 1984 SC 774.



Rehabilitation of the ‘Project Affected’  11

the time of the declaration under Section 4 of the Act but its potential value also must be taken into account18 and (b) the prices of small plots of land cannot be taken as criteria for fixing the prices of large plots. There are several problems in the determination of market value. There is an absence of an independent valuation process; there is no clarity in land titles. Also, to change the land use, a farmer has to obtain a non-agricultural use clearance (NAC) from the state government. Morris and Pandey (2007) argue that this restriction costs the farmer a regulatory arbitrage that could be as high as 140 per cent. This has been the cause of agitations in several acquisitions. The permission is only given when the farmer has a clear non-agricultural project to implement on the farmland. Another factor that operates to keep land prices low is the restriction on non-farmers buying agricultural land. Further, the high rate of stamp duties in most states makes it profitable for parties to quote a much lower price than the actual traded price in the transfer documents. Since the comparable sales method is commonly adopted for valuation of land, if the set of recorded prices is low it will lead to a general lowering of the compensation amount. Section 23 further provides that the damage sustained by the person interested by reason of taking standing crops or trees which may be on the land at the time of the Collector’s possession should be taken into account. The value of the land in the case of an orchard will be arrived at by ascertaining the income from the profits of the orchard. There may, of course, be exceptional cases where the overwhelmingly high intrinsic value of the land caused by recent development and growth may have to be considered by the courts, but the general rule is otherwise. Any attempt to separately value the land and add to it the income from the orchard would inevitably result in ‘duplication of values’.19 Moreover, Section 24 of the Act provides that any outlay or improvement on, or disposal of, the land acquired, commenced, made or effected without the sanction of the collector after the date of publication of the notification under Section 4(1) shall not be taken into consideration by the courts in awarding compensation. Due to this provision, the incentive to over invest in the land in the expectation of obtaining a higher compensation is restricted by the fact that the permission of the Collector would have to be taken. Thus, the hypothesis by Blume et al. (1984) would not be applicable as an argument against granting full compensation in India. Another aspect that is vital in deciding the amount of compensation is the damage sustained because of severance of such land from the seller’s other 18 19

Metal Corporation of India Ltd. v. Union of India, AIR 1970 Cal 15. Collector, Varanasi v. Ajit Pratap Singh, 1979 All LJ 1009.

12  Malabika Pal

land or by injuriously affecting his other property, movable or immovable, or his earnings. In Triveni Devi v. Collector of Ranchi 20 it was held that ‘when a portion of the land is acquired and a large portion is left out, there would be diminution in the value of the land left for which some compensation has to be allowed. 5% allowed by the Tribunal Commission is reasonable’. In the case of Cawasjee Nasarwanjee Dinshaw v. Special Land Acquisition Officer 21 it was stated that no further construction could be made on the severed land as per the rules of the Municipal Council although it could be used as open space or for growing fruit trees or for cultivation. In view of this, the compensation for damage on account of severance at the rate of 75 per cent of market value was deemed to be quite justifiable. The measure of damage by severance is the diminution in the value of the un-acquired land. A related issue is the enhancement of the value of the retained land as a result of the acquisition, in which case often a ‘set-off’ is recommended, an amount to be adjusted from the compensation. Although present in English law, there is no corresponding provision in Indian law. In the presence of a provision for ‘set-off’, the compensation payable would become nil in many cases. The compensation for loss of earnings is for the interregnum only, i.e., till such time that the claimants are able to restart their business. Apart from the above components, Section 23 of the Act stipulates that in determining the amount of compensation other factors must be taken into account. These include reasonable expenses incurred if a person is compelled to change his residence or place of business and bona fide damage (if any) resulting from the profits of the land between the time of publication of the declaration and the Collector’s possession of the land. The amendment to the Act, brought into effect by Act 68 of 1984, mandates that in addition to the market value of the land the courts shall award in every case an amount of 12 per cent per annum on such market value for the period commencing on and from the date of publication of notification to the date of award by the Collector, or the date of taking possession of the land, whichever is earlier. Further, in addition to the market value of land, the Act provides that the courts shall in every case award a sum of 30 per cent on such market value, in consideration of the compulsory nature of the acquisition. This provision, which can be seen as an attempt to address the demoralization of property owners, can in principle bridge the gap between the market value compensation and the value to owners, thus moving the level of acquisition 20 21

AIR 1972 SC 1417; (1972) 1 SCC 480; (1972) 3 SCR 208. Pune, 1990 Bom. CR 604.



Rehabilitation of the ‘Project Affected’  13

towards the efficient level—the level of acquisition that takes into account the true cost to owners. This requirement of payment in excess of market value is similar to the statutes enacted by many states in the United States regarding mills during the 19th century. However, since there is a systematic bias in the calculation of market value, the provision of solatium does not serve its intended purpose in a significant way. Section 24 of the Act specifies that in determining the amount of compensation the matters that should not be taken into consideration include ‘any disinclination of the person interested to part with the land acquired’ and ‘any increase in the value of the land acquired likely to accrue from the use to which it will be put when acquired’. Both these provisions militate against the objective of complete indemnification.22 The power of eminent domain means that property can be taken without consent. However, the prospective return from the project for which the property is being acquired is known and the losers can be made recipients of a part of the gains. This would reduce the demoralization costs incurred both on account of the acquisition being a setback to their investment-backed expectations and the unfairness that owners perceive because the takings burden is concentrated and disproportionate. Although the NRRP, 2007 recognizes that acquisition of private property by use of eminent domain has ‘traumatic psychological and socio-cultural consequences on the affected population’, there are no specific provisions to address the demoralization costs. The principles of the policy also apply to the ‘rehabilitation and resettlement of persons involuntarily displaced permanently due to any other reason’ (emphasis added). People affected by compulsory acquisition and those affected by other reasons like natural calamities cannot be treated on the same footing. Since the taking of private property and the resultant displacement of people are deliberate, planned events, the compensation due to them should be thought of differently from that to be paid to the victims of natural disasters. While the state is not obliged to pay the subjective value in the latter case, in the former it must pay just compensation or else it would become the Hobbesian sovereign. The demoralization costs that accompany takings without compensation must be addressed separately when eminent domain is used. While highlighting that demoralization costs are different from ordinary risks, Michelman (1967: 1217) argues: ‘…even though people can adjust satisfactorily to 22 In Baroda Prosad Dey v. Secretary of State, 25 CWN 677; (1921) 49 Cal. 83, it was held that the intention of ‘sec 23 of L. A. Act taken as a whole is to provide complete indemnity to a person whose land is being compulsorily acquired.’

14  Malabika Pal

random uncertainty, which can be dealt with through insurance, including self-insurance, they will remain on edge when contemplating the possibility of strategically determined losses.’ The focus of NRRP, 2007 is on resettlement, with monetary stipulations for land development, construction of cattle shed (both of which are fixed) and transportation cost (on actual cost basis). For an affected person who is a rural artisan, small trader or self-employed person, he shall get a one-time financial assistance for construction of work shed or workshop. Given the delay in receiving the package, the monetary amounts would in no way be just compensation for what the affected person has had to give up, loss of goodwill being just one component. The novel part of the new policy is the pledge to provide employment to affected persons who lose their employment due to the project, subject to the availability of vacancies and suitability of the affected person for the employment. If the project is mining, many landowners may not want to be employed in such hazardous jobs. Further, in case of SEZs, the labour requirements would be of a skilled nature and therefore the ‘suitability’ of the farmers to be profitably employed is ruled out at the very outset. The availability of vacancies depends entirely on the discretion of the Requiring Body. In case of not being given agricultural land or employment, there is a provision for 750 days of minimum agricultural wages and in addition 20 per cent of this amount (which can be raised to 50 per cent at the discretion of the appropriate government) will be given in the form of shares or debentures in the Requiring Body, in such a manner as may be prescribed, and in lieu of land-for-land or employment, the affected person shall be given a site or apartment within the project. The issue of reciprocity is couched in many uncertainties like ‘subject to availability’, ‘suitability’, ‘discretion’, etc. It is found that the provisions for future benefit sharing are not adequate to enable the affected persons to recoup the loss and the demoralization costs arising due to this would remain. In the current phase of land acquisition, the government is acting as a facilitator23 for private companies. This may appear to be, in Michelman’s words, ‘unprincipled redistribution’, making the efficiency of the project disputable and therefore entail greater demoralization not only among the actual losers but also their sympathizers. The fact that similar acquisitions are going on all over the country makes people anxious that they too could 23 The Land Acquisition, Rehabilitation and Resettlement Bill, 2011 provides for Government acquisition of land ‘with the purpose to transfer it for the use of private companies for public purpose’ (Chapter 1, Clause 2 (1) (b), available at http://rural.nic.in/sites/policies.asp).



Rehabilitation of the ‘Project Affected’  15

be subjected to uncompensated taking of private property.24 These ‘secondorder effects’ are important because they form the basis of public perception of the compensation practice followed by a government. As Fisher (1988) argues, these effects, which have been taken into account under the rubric of demoralization costs, can have important implications in shaping public opinion.25

Conclusion In the above sections, an attempt has been made to draw upon the takings literature to throw light on the recent public scrutiny of land acquisitions in India. An analysis of the theoretical background of the issue of compensation shows that of the tests that try to distinguish between compensable and non-compensable takings, Michelman’s demoralization-settlement costs framework appears to be most relevant. The Land Acquisition Act, 1894 and the National Rehabilitation and Resettlement Policy, 2007 do not address the demoralization costs incurred by people affected by the state’s use of eminent domain. At the heart of the current controversy and conflicts regarding land acquisition in India is the compensation problem. One aspect of the problem is who funds the compensation package. If the acquisition is for public projects like building or broadening roads or making way for railway lines, the returns from the projects cannot cover the costs and tax revenue will have to be used to fund the compensation.26 However, with the state acting as a mediator, private companies will have to bear the compensation package. 24 On 20 August 2010, the National Alliance of People’s Movements, along with many other movements that came together under the banner of Sangharsh, staged a protest against government policies towards forcible land acquisition and compensation. This, however, is not an isolated incident. Struggles against what are felt to be unjust acquisitions have been intensifying across the country. 25 Padel (2009) undertakes an in-depth study of the Konds of Orissa and comes to the general conclusion that in the name of India’s development what is going on is a ‘national sacrifice’ of its tribal minority. 26 On 22 August 2008 the New Delhi Municipal Council (NMDC) decided to stop paying monetary compensation for the land it acquires from private owners for development purpose—widening Master Plan roads, building pedestrian pathways, bridges and parking lots, and instead give property owners the benefit of floor area ratio (FAR), by allowing them to build vertically on the rest of the plot. The decision was taken after the NDMC had to pay `80 million for less than an acre of property in Janpath to extend Tolstoy Marg.

16  Malabika Pal

The companies would have to be provided with incentives so that they take into account not only the cost component C, as is done routinely, but also the costs due to the compulsory nature of the purchase of land, D and S, in deciding whether to undertake the projects.27 This can be done by ensuring that the affected landowners receive just compensation. Given that the value to the owner is largely subjective and difficult to estimate, a move can be made in the right direction by correcting for the underestimation of market value. The comparable sales method so commonly used has some fundamental flaws that must be recognized. This is an area that needs further research and clear guidelines so that a move is made towards arriving at as close as possible to the value a private owner attaches to his property. In the final analysis, one may argue that a democratic society cannot afford to coerce its citizens to part with property without just compensation and still continue to retain its legitimacy. The continual expansion of the definition of public purpose28 to suit the needs of changing circumstances will meet with criticism if the process is not accompanied by corrective measures to resolve the compensation issue. The ‘project affected’ will have to be given a share in the gains on a long-term basis to avoid the demoralization costs and ensure that they are truly ‘rehabilitated’.

Bibliography Baxter, William F. and Lillian R. Altree. 1972. ‘Legal Aspects of Airport Noise’, The Journal of Law and Economics, 15 (1): 1–113. Bhaduri, Amit. 2006. ‘Economic Growth: A Meaningless Obsession’, B. N. Ganguly Lecture, CSDS, Delhi, November. Available at http://www.india-seminar.com/2007/569/569_ amit_bhaduri.htm.

27 Dipankar Dasgupta (2008) had argued that the price of the Nano car be raised by `10,000 to recoup the money that could then be paid to the landowners from whom the land was acquired in Singur for the Tata Motors factory. 28 The Supreme Court of India on 5 September 2008 further extended the scope of public purpose, stating: ‘If the project taken as a whole is an attempt in the direction of bringing foreign exchange, generating employment opportunities and securing economic benefits to the State and the public at large, it will serve public purpose’ (in the case: [1] Sooraram Pratap Reddy and Others; [2] Suraram Krishna Reddy And Another; [3] V. Krishna Prasad; [4] A. L. Sadanand; [5] Malla Reddy And Others; [6] Bandari Pentaiah and Others v [1] District Collector, Ranga Reddy District and Others; [2] District Collector, Land Acquisition And Others; [3] Government of Andhra Pradesh and Others; [4] District Collector And Others 2008 (9) SCC 552, available at http://www.indiankanoon.org).



Rehabilitation of the ‘Project Affected’  17

Blume, L. and D. L. Rubinfeld. 1984. ‘Compensation for Takings: An Economic Analysis’, California Law Review, 72: 569–628. Blume, L., D. L. Rubinfeld, and P. Shapiro. 1984. ‘The Taking of Land: When Should Compensation be Paid?’, Quarterly Journal of Economics, 99 (1): 71–92. Burrows, Paul. 1991. ‘Compensation for Compulsory Acquisition’, Land Economics, 67 (1): 49–63. Calabresi, Guido and A. Douglas Melamed. 1972. ‘Property Rules, Liability Rules, and Inalienability: One View of the Cathedral’, Harvard Law Review, 85 (6): 1089–1128. Coase, Ronald H. 1960. ‘The Problem of Social Cost’, The Journal of Law and Economics, 3 (October): 1–44. Cordes, J. J. and B. A. Weisbrod. 1979. ‘Government Behaviour in Response to Compensation Requirements’, Journal of Public Economics, 11 (1): 47–58. Dipankar Dasgupta. 2008. ‘A Solution for Singur’, The Times of India, 6 September. Available at http://articles.timesofindia.indiatimes.com/2008-09-06/edit-page/27950428_1_singurland-prices-tatas-nano. Epstein, R. A. 1985. Takings: Private Property and the Power of Eminent Domain. Harvard University Press. Fischel, W. A. 1995. Regulatory Takings: Law, Economics and Politics. Cambridge: Harvard University Press. Fischel, W. A. and P. Shapiro. 1988. ‘Takings, Insurance and Michelman: Comments on Economic Interpretations of “Just Compensation” Law’, Journal of Legal Studies, 17 (2): 269–293. Fisher, William W. III. 1988. ‘The Significance of Public Perceptions of Takings Doctrine’, Columbia Law Review, 88 (8): 1774–1794. Hermalin, Benjamin E. 1995. ‘An Economic Analysis of Takings’, Journal of Law, Economics and Organization, 11 (1): 64–86. Innes, Robert. 2000. ‘The Economics of Takings and Compensation When Land and Its Public Use Value are in Private Hands’, Land Economics, 76 (2): 195–212. Kaplow, L. 1986. ‘An Economic Analysis of Legal Transitions’, Harvard Law Review, 99 (3): 509–617. Kaplow, L. and S. Shavell. 1996. ‘Property Rules Versus Liability Rules: An Economic Analysis’, Harvard Law Review, 109 (4): 715–790. Merrill, T. W. 1986. ‘The Economics of Public Use’, Cornell Law Review, 72: 61–116. Miceli, Thomas J. 1991. ‘Compensation for the Taking of Land under Eminent Domain’, Journal of Theoretical and Institutional Economics, 147 (2): 354–363. Miceli, Thomas J. and Kathleen Segerson. 1994. ‘Regulatory Takings: When Should Compensation Be Paid?’, Journal of Legal Studies, 23: 749–776. Michelman, Frank I. 1967. ‘Property, Utility, and Fairness: Comments on the Ethical Foundations of “Just Compensation” Law’, Harvard Law Review, 80 (6): 1165– 1258. Ministry of Rural Development, Department of Land Resources. 2006. Draft of the Proposed National Rehabilitation Policy, 2006. Government of India. Available at http://rural.nic. in/sites/policies.asp (accessed in June 2008). ———. 2007. The National Rehabilitation and Resettlement Policy. Government of India. Available at http://rural.nic.in/sites/policies.asp (accessed in 2008). Morris, S. and A. Pandey. 2007. ‘Towards Reform of Land Acquisition Framework in India’, Economic and Political Weekly, 2 June: 2083–2090. Padel, F. 2009. Sacrificing People. New Delhi: Orient BlackSwan.

18  Malabika Pal Palit, A. and S. Bhattacharjee. 2008. Special Economic Zones in India: Myths and Realities. New Delhi: Anthem Press. Sarkar, P. K. 2007. Law of Acquisition of Land in India including Requisition and Acquisition of Immovable Property, 2nd ed. Kolkata: Eastern Law House. Sax, Joseph. 1964. ‘Takings and the Police Power’, The Yale Law Journal, 74 (1): 36–76. Tideman, N. 1988. ‘Takings, Moral Evolution and Justice’, Columbia Law Review, 88 (8): 1714–1730.

2

‘Law and Economics’ of Indian Patent Law V. Santhakumar

Introduction This chapter analyses certain aspects of the Patents Act, 1970 (as amended in 2005) of India with the insights from the theory of law and economics. Some may doubt the relevance of such an exercise since there exists a substantial body of literature in India on what may be called the economics (economic commentary) of intellectual property rights.1 This literature has brought to light what the author considers the impact of (specific types of) patent rules on different sectors of the economy, as manifested in the prices of products, market share of domestic as well as foreign firms, profitability and the extent of research and development (R&D) investments. There are also studies that evaluate clauses of the Patents Act in terms of benefits or gains for Indian industry or public interest in India. However these studies do not assess whether a specific provision or clause in the Patents Act is an efficient way of satisfying these interests. For example, Chaudhuri (2005: 90) criticises the adoption of a strict compulsory licensing clause in India (including the publication of the compulsory licensing application in the Official Gazette, the right to the patentee to oppose the application and the opportunity given to the patentee to be heard, the right of the patentee to submit an appeal against the decision of the administration to order compulsory licensing, etc.), on the ground that such a clause can delay the process of granting compulsory licensing. However, no attempt is made in this regard to see whether there 1 There exist also a number of writings by legal scholars on successive patent legislations in India, and even these, starting from the ‘Report on the Revision of the Patents Law’ (1959) by Justice N. Rajagopala Ayyangar, have been concerned with what India would gain by following a strict patent regime. Thus, issues of economy were of importance in this literature too.

20  V. Santhakumar

are some long-term disincentives in granting compulsory licences without giving an adequate opportunity to the patentee to oppose this decision. An economic analysis of the Patents Act, 1970 (hereafter Patents Act) carried out in Bagchi et al. (1984), evaluates the legislation as ‘progressive’ due to the perceived benefits to (sections of) Indian industry or certain sectors, etc., but does not analyse whether the statute facilitates and protects intellectual property in the country in the long run, and the economic implications, if it does not do so.

The Patents Act in Brief The Patents Act (as amended in 2005) defines an ‘inventive step’ as a ‘feature of invention that involves technical advance to the existing knowledge or having economic significance or both and that makes invention not obvious to a person skilled in the art’. It defines new invention as that which has not been anticipated by publication in any document or used in the country or elsewhere in the world before the date of filing of patent application with complete specification, i.e., the subject matter has not fallen in public domain or that it does not form part of the state of the art.

In defining inventions excluded from being patented, the amendment of 2005 includes the following: [T]he mere discovery of the new form of a known substance, which does not result in the enhancement of the known efficacy of that substance or the mere discovery of any new property or new use of a known substance or of the mere use of a known process, machine or apparatus, unless such known process results in a new product or employs at least a new reactant.

Other inventions excluded from being patented include inventions which would be contrary to law or morality or injurious to public health, mere discovery of scientific principle or abstract theory, inventions relating to atomic energy and others, as listed in Box 2.1. The clause in the Patents Act in 1970, which allowed patenting of only methods or processes of manufacture, which facilitated reverse-engineeringbased development, mainly in Indian pharmaceutical industry (Chaudhuri 2005), was deleted in 2005. The Patents Act also details the process of applying for patents, the requirements of submitting (and the nature of) specifications of inventions (which are provisional initially but completed



‘Law and Economics’ of Indian Patent Law  21 Box 2.1.  Inventions That Cannot Be Patented in India

Inventions excluded from patenting in India include: 1. a substance obtained by a mere admixture, resulting only in the aggregation of the properties of the components thereof or a process for producing such substance; 2. the mere arrangement or re-arrangement or duplication of known devices, each functioning independently of one another in a known way; 3. a method or process of testing applicable during the process of manufacture for rendering the machine, apparatus or other equipment more efficient or for the improvement or restoration of the existing machine, apparatus or other equipment or for the improvement or control of manufacture; 4. a method of agriculture or horticulture; 5. any process for the medicinal, surgical, curative, prophylactic or other treatment of human beings or any process for a similar treatment of animals or plants to render them free of disease or to increase their economic value or that of their products. Source: Compiled by author.

within a stipulated time) and other information (for example, regarding filing of patent application(s) in other countries), the process of examining the patent application by which a search is carried out to check whether the invention (for which the patent is sought) has been anticipated in any previous publication or previous patent application, the consideration of examiner’s report by the Controller of patents, the opportunity given to the applicant to respond to objections, if any, raised by the examiner, the provision to submit a further application if necessary, the powers of the Controller in the case of potential infringement on other patents, the provision to advertise the acceptance of complete specification which will make the information in the specification available in the public domain, the provision for others to oppose (with specific reasons as mentioned in the statute) the granting of patent to the applicant, the provision to enable others to file complaints after the grant of patent, the procedure to be followed by the Controller including the constitution of the Opposition Board to deal with such complaints and so on. The Patents Act provides for multiple durations for different patents; five years (from the date of patenting) for substances used as food or drugs and 14 years for others, but 20 years in case of international applications filed under the Patent Cooperation Treaty (from the international filing date). There is also a provision for seeking and granting patents of addition (for a term equal to that of the patent) for improvements in the invention to the holder

22  V. Santhakumar

of the original patent. The patentee can also apply for restoring a patent if it lapses due to the non-payment of the annual fee. The Patents Act mentions that compensation (though the extent of or the method of fixing it is not specified) must be given to other persons, if any, who may have taken steps to use that invention after the lapse of the patent and before its restoration. The patentee can surrender the patent too. Courts can revoke patents on finding that the specification given along with the application was contained in the specification of another patent granted in the country, or on finding that the patent should not have been granted to the one who received it, or for some other illegality in the application or in granting of the patent. There exists a provision for compulsory licensing in the Patents Act. After the expiration of three years from the date of granting patent, any person interested may make an application to the Controller alleging that the reasonable requirements of the public with respect to the patented invention have not been satisfied or that the patented invention is not available to the public at a reasonable price and praying for the grant of a compulsory licence to work the patented invention. If there is merit in this application, then the Controller can exercise his power and make an order to grant a licence upon such terms as he may deem fit. The Controller needs to take into account the efforts made by the patentee to commercially use the invention and the ability of the complainant to use it if provided a licence. For inventions after a period of three years, and inventions related to substances used as food, drugs or medicines, which will be endorsed as licences of right immediately after the granting of patents, any person can request the patentee to grant her a licence for mutually agreed upon conditions. If they cannot reach an agreement, the Controller can fix the terms after hearing both the parties, and the royalty and other remuneration reserved to the patentee under a licence granted to any person shall in no case exceed four per cent of the net ex-factory sale price in bulk of the patented article (exclusive of taxes levied under any law for the time being in force and any commissions payable) determined in such manner as may be prescribed. Two years after granting the first compulsory licence based on a patented invention or endorsing licences of right, any person or government can approach the Controller to revoke the patent, if the benefits of the invention are still not available to the public. Thus, the maximum period one can hold the patented invention without commercializing it (without valid reasons) is five years. There are also clear provisions explaining what is meant by failing to satisfy the requirements of the public, warranting compulsory licensing or revocation of the patent, which have implications for the development of the domestic economy. They are listed in Box 2.2.



‘Law and Economics’ of Indian Patent Law  23 Box 2.2.  When Reasonable Requirements of the Public Are Deemed Not Satisfied

For the purposes of Sections 84, 86 and 89, the reasonable requirements of the public shall be deemed not to have been satisfied: 1. If, by reason of the default of the patentee to manufacture in India to an adequate extent and supply on reasonable terms the patented article or a part of the patented article which is necessary for its efficient working or if, by reason of the refusal of the patentee to grant a licence or licences on reasonable terms:



(a) an existing trade or industry or the development thereof or the establishment of any new trade or industry in India or the trade or industry of any person or classes of persons trading or manufacturing in India is prejudiced; or (b) the demand for the patented article is not being met to an adequate extent or on reasonable terms from manufacture in India; or (c) a market for the export of the patented article manufactured in India is not being supplied or developed; or (d) the establishment or development of commercial activities in India is prejudiced;

2. if, by reason of conditions imposed by the patentee (whether before or after the commencement of the Patents Act) upon the grant of licences under the patent, or upon the purchase, hire or use of the patented article or process, the manufacture, use or sale of materials not protected by the patent, or the establishment or development of any trade or industry in India, is prejudiced; or 3. if the patented invention is not being worked in India on a commercial scale to an adequate extent or is not being so worked to the fullest extent that is reasonably practicable; or 4. if the demand for the patented article in India is being met to a substantial extent by importation from abroad by:

(a) the patentee or persons claiming under him; or (b) persons directly or indirectly purchasing from him; or (c) other persons against whom the patentee is not taking or has not taken proceedings for infringement; or (d) if the working of the patented invention in India on a commercial scale is being prevented or hindered by the importation from abroad of the patented article by the patentee or the other persons referred to in the preceding clause.

Source: Compiled by author.

24  V. Santhakumar

The amendment of 2005 allows granting of compulsory licences not only for commercial use within the country, but also for export of pharmaceutical products to countries where the capacity of production of such products is limited and where from the permit to import is obtained by those seeking the compulsory licence. The government can compulsorily ‘use’ a patented invention for purposes within India. In addition, governments can ‘acquire’ inventions. The relevant clauses show that goods or services produced from government-acquired patents can be exchanged commercially as the right to make, use, exercise and vend an invention for the purposes of Government…shall include the right to sell the goods which have been made in exercise of that right, and a purchaser of goods so sold, and a person claiming through him, shall have the power to deal with the goods as if the Central Government or the person authorised…were the patentee of the invention.

Such acquisition is meant for ‘public purposes’ as in the following clause: The Central Government may, if satisfied that it is necessary that an invention which is the subject of an application for a patent or a patent should be acquired from the applicant or the patentee for a public purpose, publish a notification to that effect in the Official Gazette, and thereupon the invention or patent and all rights in respect of the invention or patent shall, by force of this section, stand transferred to and be vested in the Central Government.

There is also a provision to pay compensation: The Central Government shall pay to the applicant, or, as the case may be, the patentee…such compensation as may be agreed upon between the Central Government and the applicant, or the patentee…or, as may, in default of agreement, be determined by the High Court on a reference under section 103 to be just having regard to the expenditure incurred in connection with the invention and, in the case of a patent, the term thereof, the period during which and the manner in which it has already been worked (including the profits made during such period by the patentee or by his licensee whether exclusive or otherwise) and other relevant factors.

Experience with Patenting in India There are a number of studies that look into what they see as favourable or negative outcomes of instituting, changing and enforcing the Patents Act



‘Law and Economics’ of Indian Patent Law  25

in India. These outcomes include, among others, the number of patents or extent of patenting before and after certain changes were made to the Patents Act. The situation before 1970 in India was that most patents were held by foreign firms, and they did not show much interest in producing (or commercializing their inventions through production) in India, and this gave the impression that the Patents Act was used as a tool by foreign firms to extend their monopoly power in the Indian market (Bagchi et al. 1984). After the Patents Act was passed in 1970, with provisions considered more biased against patentees than those in the pre-existing act of 1911, there was a decline in the number of patent applications filed by foreigners and the number of patents granted in their favour (Bagchi et al. 1984), whereas those filed by, and granted in favour of, Indians increased over a period of time. This can be at least partly attributed to the clauses biased against patentees, especially those insisting on commercial use of inventions through production within India, the likelihood of granting compulsory licenses if patentees did not commercialize (and produce) within the country, providing for government acquisition of patents if needed, etc., and these might have had discouraged foreign firms. The number of patent applications in the mid1990s rose sharply in India (Ramanna 2002). This may indicate that Indian companies too were becoming keener on protecting intellectual property, facilitated by the Patents Act. From the 1990s, Indian companies showed increased interest in filing patent applications in other countries including the United States. However, 10 developed countries still dominate in terms of global patents, or those obtained in United States. There are studies on whether the institution of and the nature of the Patents Act have an impact on what are broadly considered indicators of economic development.2 There are many problems identified in such studies. First of all, the relationship between patent protection and economic development can be mutual and not unidirectional. There seems to be a nonlinear relationship between patent protection and economic development (with the protection declining as development catches up initially since there is more interest in imitating technologies). The other problem of linking patents with outcomes of economic development is the difficulty in isolating the influence of other variables. There is an argument that economic growth achieved in a number of Asian countries in the 20th century was facilitated by weak rather than strong patent rights. Though not about economic development in general, there was an argument in the case of India that the patent system that prevailed in the 1970s and 1980s in the country 2

For a review of empirical literature on developing countries, see Kumar (2003).

26  V. Santhakumar

had facilitated the growth of, and technological development in, the Indian pharmaceutical industry. Another outcome of patenting, which attracted scholarly attention with regard to India, is the prices of products, especially that of drugs. It has been noted that after the adoption of the Patents Act in 1970, which granted patents only to the processes (and not to the products), the country achieved self-sufficiency in the production of bulk drugs (through reverse engineering and imitation that enabled Indian companies to market internally patented drugs in India within a relatively shorter period, without getting licences from the patentees) and the prices of most drugs in the country are much cheaper than in other parts of the world (Alam 1996; Lanjouw 1998). Probably due to the facilitation of process patents, the Indian pharmaceutical industry has grown substantially in terms of number and capability, and is currently able not only to meet domestic requirements but also to capture a substantial share of market for bulk drugs in international trade. Indian pharmaceutical companies are dominant exporters to many other developing countries including those in Africa (Chaudhuri 2005). However, a sizable section of this pharmaceutical industry that benefited from the previously existing process-patent regime have become increasingly comfortable with India moving towards a product-patent regime to comply with TRIPS (Trade-Related Aspects of Intellectual Property Rights).3 This comprises those that became multinational companies and started investing in R&D for new drug recoveries as well as filing patent applications in the United States. These Indian companies felt that they would also benefit from product-patent protection in India since they are ready to launch new drug discoveries and they can be competitive globally in this regard. However, the other section which still benefits from reverse engineering and process-patents actively lobby against strengthening the patents and giving more protection to patentees.4 The scientists in state-supported research laboratories also started demanding greater protection for intellectual property and that too facilitated the passing of the amendment effectuating product patents in 2005. Though the provision of compulsory licensing existed in the Patents Act of 1970 (incorporated in the 1911 act itself through an amendment prior 3 In brief, this is an international agreement administered by the World Trade Organization (WTO) that sets the minimum standards of intellectual property regulations that are to be followed by its members. India is also a signatory to this agreement. The changes in the Patents Act made in India can be viewed as part of the steps taken by the country to meet the standards set by TRIPS. 4 This is evident from the submissions made to the Expert Committee to determine the TRIPS compliance of the Patents Act amended in 2005.



‘Law and Economics’ of Indian Patent Law  27

to 1970), it was seen in the mid-1980s that the number of applications for such licences had been very small. Part of this could have been due to the internalization of restrictive clauses by patentees and the reflection of their cautious behaviour (which may even include a reduction in the number of patent applications).5 As noted in Bagchi et al. (1984), the information provided in the specification given to the patent office may not be sufficient for a third party to develop and commercialize that invention using the route of compulsory licensing. The delays and procedural hurdles in the administrative process (and in legal process, if used) of compulsory licensing might have also constrained this process.

Provisions That Facilitate Economic Efficiency in the Patents Act The fact that India adopted an amendment to the Patents Act, and made changes recently to move towards one complying with the requirement of TRIPS (though not all clauses), with a product-based patent is a positive development. It may be possible that a restrictive patent statute may be more discouraging than the absence of one. For example, the use of strategies other than patent applications as a way of ensuring secrecy over inventions or new knowledge developed by firms and individuals. The fact that if one applies for a patent, the details of the intellectual property become part of the public domain and that this can be acquired by the government or adopted by other private firms through reverse engineering might be more discouraging to inventions than the situation marked by no patent legislation. The definitions of inventive step and new invention used in the Indian statute after the amendments in 2005 are closer to international definitions. It is better to have international commonality and compliance with international agreements like TRIPS for economic efficiency, even if the efficiency implications of the specific clauses are disregarded. This is so since divergent definitions used in different countries increase transaction costs in acquiring patents, and this can have a negative impact on global inventions or the use of such inventions in countries where such definitions are very 5 Bagchi et al. (1984) argue that even if somebody sees the viability of commercializing an invention, which has not been done by the patentee of that invention, the market power of the patentee may deter the person from seeking a compulsory licence.

28  V. Santhakumar

different from those in mainstream countries (those with a larger market for patented products and services). There are provisions in the Indian act that facilitate both incentives for inventions and the use of such inventions for the society. These include multiple durations, compulsory licensing, elaborate opportunities to contesting parties to be heard and file appeals before courts, etc. The multiple-duration structure with a short-duration for food and drugs (and longer one for others) can encourage higher ‘turn-out’ of inventions in certain cases. However the broad categorisation of certain substances or products warranting shorter-term patents can be problematic since some specific items within this category (for example, a cure for AIDS) may require much higher investments requiring longer-duration protection, but such a broad definition is unavoidable to some extent in any law (since specifying a duration to each and every potential invention can be very costly). The logic of short-duration patents adopted in the Indian act may be similar to ‘petty patents’ prevailing in a few other countries. Compulsory licensing is a provision, which if used properly through a transparent and effective process with adequate safeguards for protecting the interests of well-meaning patentees (and not those who use patents as a blocking strategy), would ensure that the benefits of inventions are derived by the larger society at the earliest. One need not worry too much about the actual low-level use of the provision of compulsory licensing, which seems to be the case in India, since the provision of such licensing would encourage (a) the patentee to start operations within a reasonable time or to exchange the patented invention voluntarily with a licensee for a royalty higher than that prescribed as the maximum in the Patents Act, without becoming the target of compulsory license application, (b) discourage patentees to seek patents if they do not plan to commercially use it, and (c) encourage only those patentees who are almost sure that what is revealed in the specification provided as part of the patent application is inadequate for a licensee to start commercializing the invention. The Indian statute, like those existing in many other countries, excludes diagnostic, therapeutic and surgical methods (for the treatment of humans and animals). This exclusion is absent for products or instruments of diagnosis, therapy or surgery (DTS). There is an economic logic here, which is that even if a process of DTS (not instrument) is protected by a patent, the cost of enforcing payment of royalty each time it is used by a service provider can be very high. On the other hand, for a patented product, the licensed company can generate royalty somewhat automatically through monopoly pricing. Thus, the high cost of enforcement of royalty payment for using the processes of DTS may make it socially desirable to treat them



‘Law and Economics’ of Indian Patent Law  29

as ‘public goods’. This may be an economically efficient strategy. Using this logic, one can consider the extension of exclusion of patenting to not only DTS processes but also curative and similar treatments, again not only for humans and animals but also for plants in the Indian act. Such extension may not have any serious negative implications for economic efficiency, if the products and instruments used for all these procedures are strictly protected by patents. However, much of the legal discussion on this issue seems to be around whether the exemption granted for humans and animals needs to be extended to plants (Watal 2001), in neglect of economic considerations. Since Section 3(j) of the Patents Act excludes ‘plants and animals in whole or any part thereof other than micro-organisms but including seeds, varieties and species and essentially biological processes for production or propagation of plants and animals’, micro-organisms created through biotechnological research (and those not available in nature) can seek patent protection (as noted by the Mashelkar Committee). This facilitates the growth of the biotechnology industry and associated economic opportunities. However, other clauses such as ‘the need to protect public order, morality, or human, animal and plant life’ [Section 3(b)] can be used to prevent the patenting of any bio-technically engineered micro-organism if it is a threat to society (such as those which can be used as biological weapons, or that can endanger life when it is beyond the control of the inventor). Thus, the provision with regard to micro-organisms6 is adequate to encourage beneficial research on biotechnology and genetic engineering on the one hand and to curb their misuse on the other.

Limitations of the Patents Act Though the Patents Act of 1970 has been changed to make it TRIPS compliant through amendments in the 1990s and in the decade that followed, there are certain provisions which not only violate the spirit of TRIPS but are also against the interest of economic efficiency. For example, there is the provision in the amended act that says that the following is excluded from patenting: [T]he mere discovery of a new form of a known substance which does not result in the enhancement of the known efficacy of that substance or the mere 6 It may be noted that there exist many opponents to the clause that facilitates patenting of micro-organisms in India.

30  V. Santhakumar discovery of any new property or new use for a known substance or of the mere use of a known process, machine or apparatus unless such known process results in a new product or employs at least one new reactant.7

The committee constituted by India to look into the TRIPS compliance of its amended Patents Act (Mashelkar Committee) has noted that ‘it would not be TRIPS compliant to limit granting of patents for pharmaceutical substance to New Chemical Entities (NCE) only’. If patenting is practised in a country, then it is desirable to have a provision, such as the one noted in Article 27 of the TRIPS Agreement, ‘for any inventions, whether products or processes, in all fields of technology, provided that they are new, involve an inventive step and are capable of industrial application’. As Article 27 goes on to declare, the exclusion from patenting needs to be only for those: …inventions, the prevention within their territory of the commercial exploitation of which is necessary to protect ordre public or morality, including to protect human, animal or plant life or health or to avoid serious prejudice to the environment, provided that such exclusion is not made merely because the exploitation is prohibited by their law.

If the concern were that granting patents to new uses would prolong the monopoly power of the inventor, a little introspection would show that this need not be correct. While granting an original patent, the information on the substance is available in the public domain. Moreover, the material can also be used or imported for experimental purposes. Thus, any new firm can explore the new properties or the new use of this substance, and not only the patent-holding firm. Hence, patent protection for new uses or new properties can be sought by not only the original patentee but also by other firms. Moreover, excluding new uses or new properties of known substances from patenting seems to be in conflict with the economic principle that if the social value of investments in fundamental research is less than the social value of developing applications, then patents should be narrower, since the social value of applications is likely to be much higher in developing countries like India. It is also important to note the last part of the exclusion principle in Article 27 of the TRIPS Agreement. If something is excluded just because of the prevailing law, the prevailing institutional inefficiencies can limit the economic application of new inventions. On the other hand, only those inventions which society collectively thinks are harmful for social interest need to be excluded from patenting. Thus, the limiting of patenting 7

Section 3(d) as per the Indian Patents (Amendment) Act, 2005.



‘Law and Economics’ of Indian Patent Law  31

in India to NCE or the denial of patenting to the new use or new property of a known substance can have negative implications in terms of the expected desired effects of patenting to inventions and economic growth. It may be noted that many countries including the United States, Japan, Australia, Israel and a number of them in Europe have provision for granting patents to new uses of known substances (Watal 2001). As noted by the Mashelkar Committee, if the denial of patents to new property or new uses of known substances is to avoid evergreening (the strategies used by patentee to extend its monopoly control), the existing provisions of the Patents Act to distinguish obvious from non-obvious can be applied. But there should be a provision for protecting incremental innovations. Since India does not have the provision of petty patents, one should see whether the provision of ‘patent of addition’ could be made adequate to provide support for incremental inventions. The powers of the state in acquiring patents or using patented inventions with or without adequate compensation are wider in the Indian act, and may have negative economic implications. The provisions in the Patents Act seem to be beyond what TRIPS considers as cases where the patented product or process could be used without the authorization of the patentee under specified conditions for limited purposes which do not unreasonably prejudice the legitimate interests of the right holder, and which do not unreasonably conflict with a normal exploitation of the patent.8 There is an economic logic in limited exclusions which allow governments or agencies authorised by them to use patents for non-commercial uses (such as public epidemics or natural disasters affecting the public) or experimental uses where the benefits outweigh the costs (mainly those imposed on the patentee, and the associated disincentives, if any, on future inventions). However, the provisions applicable in India are not merely for these non-commercial and experimental uses. This is especially so since government’s use and acquisition of patented inventions can be done (theoretically, within the provisions of the Patents Act) for not only government departments but also for corporations owned by them. It may also be noted that the government’s use is not tied down to public goods, or a clearly articulated definition of public interest, for which government intervention is necessary and may be the only way of enhancing social welfare (for example, like the mass production of a patented medicine in the context of an epidemic). It is theoretically possible for the Indian state to acquire a patent for the benefit of a private goods-producing government-owned corporation against the interests of competing private 8

These are highlighted in Articles 30 and 31 of the TRIPS Agreement.

32  V. Santhakumar

firms. However, it is a reality that the state uses this provision only very rarely, and that may be a redeeming feature.

Political Economy Shaping the Patents Act It is noted that political economy considerations, including lobbying by certain interest groups shape the contents of the law on patents, not only in developing countries but also in developed ones (Cooter and Ulen 2004). However there are some additional reasons why political economy considerations would play a dominant role in developing countries in determining the nature of patent protection. This section reviews some of these issues. In the early stages of development of an economy, it is very unlikely that many domestic firms or persons will produce industrially or commercially usable inventions. Thus, the demand for patenting in such societies is likely to come from foreign firms, which sell their products or services there, but fear that the absence of patenting may encourage the budding domestic firms to copy their inventions. Therefore, it is likely that patenting in a less developed economy would mean that inventions are made abroad and monopoly rights to sell or license is made in the host country (Bagchi et al. 1984). Thus, there will be a conflict of interest between domestic and foreign firms. Moreover, if the domestic firms can copy the technology and produce the goods or services locally, the prices are likely to be cheaper due to the cost advantages in such developing economies. Thus, there may be a strong public pressure too (in addition to the interest of domestic firms) for weak patenting in such contexts. However domestic governments may try to strike a balance between this strong domestic interest in weak patenting and the pressure from foreign firms and foreign governments for strong patents. The stick or carrot for foreigners is likely to be related to foreign direct investment in the developing country or the export (of mostly raw materials) from the less developed economy. Moreover, as noted by Bagchi et al. (1984), the marginal incentive for inventions by the global firms when a poor developing country adopts patenting can be very low, and thus they may not insist on strong patent protection when the country is very poor or its market share for the global firms is lower. However, such an attitude may change when the economy and markets of these poorer countries grow. Thus, the loose IP regime prevalent in China (which manifests in the production of copyright materials without licensing) may become a major concern once the Chinese economy and market grow. Thus, it is not surprising to see the adoption of



‘Law and Economics’ of Indian Patent Law  33

a less protective patenting regime in many developing countries during their early stages of growth. However, India had a strong patent regime in the first half of the 20th century, which was inherited from its colonial rulers. But, under pressure from domestic interests, India moved towards a weak patenting regime (which allowed patenting only on processes and not on products) in the 1970s. However, on achieving significant levels of economic development, a weak patent regime can be harmful for domestic firms of the developing country themselves. This is realized when such firms, those that have grown out of using imitated inventions or others, increase their exports (especially due to the cheaper cost of inputs, mainly labour). Then, foreign countries would have a greater leverage to demand strengthening the patent regime in the developing country through the threats of trade sanctions. This may gradually build a constituency for strong patents within the developing economy. The interest of domestic firms to favour strong patenting increases not only because of their ability to export. Gradually some of these firms may acquire technological capacity and develop inventions, which they would like to protect from copying by other domestic firms. Thus, technologically advanced domestic firms would start lobbying for strong patents. There is also another reason for such a demand. Due to the globalization of research markets, there are many opportunities for domestic technology firms to participate in global R&D efforts. This participation is more likely in those activities that have cost advantages for developing countries (for example, testing newly developed drugs or insecticides, analysis of experimental data, etc.). A strong patenting regime is necessary for research firms in the developing country to participate in such global R&D initiatives. This is so since with weak patenting, there is a high likelihood of dispossessing valuable information, and this may discourage developed countries or multinational firms from collaborating with firms from the developing country. Thus, increase in exports from the developing country in general and of technological products in particular, enhanced ability of a section of domestic firms to produce inventions and the interest in participating in global R&D business would encourage a section of domestic firms to argue for strong patenting (though they were arguing for a weak patenting earlier.) Thus, the amendments to the Patents Act in the late 1990s (and later) to make it stronger in India is not only due to the need for TRIPS compliance but also due to the increasing domestic pressure of competitive and technologically capable domestic firms (Ramanna 2002). However, this turnaround of technologically advanced firms and their positive attitude towards strong patenting need not be shared by the whole society. There may be a section of domestic industry thriving on imitation. Moreover, lower prices in the past, aided by weak patenting, might also

34  V. Santhakumar

encourage significant sections of public to demand the continuation of a weak patent regime. Here, the government of the developing country is forced to strike a balance between its interests in the global market (enhancing the share of exports, need for imports, demand by inventing firms and those participating in global R&D, pressure of foreign governments) and its local interest (the public looking for cheap prices aided by imitation, survival and profitability of imitating firms) in deciding the nature of its patenting regime. This shows that if the economy in a developing country is growing fast, then there will be more and more interest in stronger patenting (especially due to its global opportunities due to cost advantages). Since it may take time (and effort) to change the patent law once it is adopted, the adoption of a weak patent regime as India did in the 1970s might create more inefficiency and this cannot be changed easily to meet the increase in demand for stronger patenting within the economy. Due to the likely impact of patents on increase in product prices and lesser benefits likely to be derived by domestic firms in developing countries, the debate on patenting may also be distorted by sectors where prices of innovative products have a greater social impact. That is why the pharmaceutical sector and drug prices have driven the public debates on patents in India. This sometimes neglects that fact there are other sectors where patenting by domestic firms or inventors are equally important, that prices are not so important for the domestic economy (if the products or services are exported), that in sectors where reverse engineering or imitation may not be that easy, and hence a patent law that deals with inventions in all sectors and technology should not be tailor-made to meet the requirements on one sector or technology. There is a tendency in India to address all social objectives and distributional considerations in each and every institutional mechanism (including law), rather than separating economic laws and addressing social and distributional concerns through other laws or policies. For example, the question of whether patent law is the appropriate mechanism to see that poor people get access to life-saving drugs is relevant in this context. There is no doubt that access to drugs is a laudable social concern, but addressing it through an economic law such as the Patents Act may have many unforeseen consequences, and sometimes work against domestic inventors and firms, or against the availability of drugs itself. Thus, one lesson that needs to be learnt from law and economics is the need to separate distributional concerns and develop direct means to address them rather than adding clauses in economic laws with consequent negative implications for economic activities and efficiency.



‘Law and Economics’ of Indian Patent Law  35

What Can Be Done to Improve the Patents Act in Terms of Its Economic Outcomes? Based on the discussion of the limitations of the Patents Act and the political economy pressures in this regard, one can identify a few areas or clauses that can be changed in the law for more beneficial economic outcomes. These including the following: 1. In the definition of what sort of inventions can be granted patents, change is needed to grant patents to new uses or new properties of a known substance, since such incremental inventions are globally important, but more so for developing countries like India in which many firms or inventors may not be able to invest in a generation of radically new inventions such as new chemical entities. However, if it is perceived that such marginal inventions do not qualify for the full patent duration, these have to be addressed in the multiple duration structure. Or, the provision for ‘patent addition’ may have to be redefined to incorporate such marginal inventions. On the other hand, if the concern were that the granting of such patents to new uses or properties would prolong socially harmful monopoly rights without commensurate benefits from additional inventions, these can be addressed using existing provisions such as compulsory licensing. Whatever be the mechanism, incremental inventions such as the identification of new properties or new uses that are not obvious to ‘a person skilled in the art’ may have to be granted patent protection. 2. The provisions of government use need to be clearly defined in Indian law. Currently such compulsory acquisition can be made not only for non-commercial or emergency purposes, but also for commercial uses. Such acquisition can be made even for government-owned companies involved in private goods production, which operate in markets competing with other private firms. Though in reality, one does not see much use of this provision against the spirit of patenting, the relevant clauses in Patents Act seem to bestow wide powers on the state, and they can be a disincentive to inventors. The government’s use may have to be limited to those ‘public goods’ and to those circumstances in which the failure to acquisition would lead to gross loss in social welfare. 3. There is a need to address distributional or social concerns with regard to particular sectors of industry (such as the drug industry) through

36  V. Santhakumar

means other than patent legislation. Special taxation or subsidy instruments that would make firms supplying products such as drugs with social objectives or subsidy or resource transfer schemes directly to consumers are more appropriate rather than tailoring patent law to meet the social requirements in this regard.

References Alam, G. 1996. ‘Impact of Proposed Changes in IPR in India’s Pharmaceutical Industry’, Prepared for Indian Council for Research on International Economic Relations (ICRIER) and UNDP, New Delhi. Ayyangar, N. R. 1959. ‘Report on the Revision of the Patents Law’. Government of India. Bagchi, A. K., P. Banerjee and U. K. Bhattachraya. 1984. ‘Indian Patents Act and Its Relation to Technological Development in India: A Preliminary Investigation’, Economic and Political Weekly, 19 (7): 287–304. Baker, S. and C. Mezzetti. 2005. ‘Disclosure as a Strategy in Patent Race’, The Journal of Law and Economics, 173–194. Chaudhuri, S. 2005. The WTO and India’s Pharmaceuticals Industry: Patent Protection, TRIPS, and Developing Countries. New Delhi: Oxford University Press. Cooter, R. and T. Ulen. 2004. Law and Economics. New Jersey: Pearson Addison Wesley. Grabowski, H. G. and J. M. Vernon. 1992. ‘Brand Loyalty, Entry, and Price Competition in Pharmaceuticals after 1984 Drug Act’, The Journal of Law and Economics, 35 (2): 331–350. Kitch, E. W. 1977. ‘The Nature and Function of the Patent System’, The Journal of Law and Economics, 20 (2): 265–290. Kumar, N. 2003. ‘Intellectual Property Rights, Technology, and Development: The Experience of Asian Countries’, Economic and Political Weekly, 38 (3): 209–226. Lanjouw, J. 1998. ‘The Introduction of Pharmaceutical Patent Product Patents in India: Heartless Exploitation of the Poor and Suffering?’ NBER Working Paper No. 6366, Cambridge. Lichtenberg, F. R. and T. J. Philipson. 2002. ‘The Dual Effects of Intellectual Property Rights, Within- and Between-Patent Competition in the US Pharmaceuticals Industry’, The Journal of Law and Economics, 45 (2): 643–672. Ramanna, A. 2002. ‘Policy Implications of India’s Patent Reforms: Patent Applications in the Post-1995 Era’, Economic and Political Weekly, 37 (21): 2065–2075. Sedjo, R. A. 1992. ‘Property Rights, Genetic Resources, and Biotechnological Change’, The Journal of Law and Economics, 35 (1): 199–213. Shavell, S. and T. Y. Ypersele. 2001. ‘Rewards verses Intellectual Property Rights’, 44 (2): 511–524. Watal, J. 2001. Intellectual Property Rights: In the WTO and Developing Countries. New Delhi: Oxford University Press.

3

Two Cases of IPR Issues in India: Pharmaceutical Patents and Film Industry Copyrights Sushma Kindo

Introduction Information is generally costly to produce and cheap to transmit. It has the characteristics of a public good, i.e., it is non-rival and non-excludable. Hence, the private sector tends to under-supply information. One way to solve this problem of under-supply of information is through intellectual property rights (IPR). There are three types of IPR: patents, copyrights and trademarks. This chapter attempts to provide an economic analysis of two cases pertaining to patents and copyrights respectively in the context of the Patents Act, 1970 and the Copyright Act, 1957 of India. The chapter consists of two broad sections, each devoted to the economic analysis of patents and copyrights using the Novartis case and the Bradford case respectively.

Patents: The Novartis Case Summary of the Case1 Novartis AG, a Swiss pharmaceutical multinational corporation, applied for grant of patent for its blood cancer drug ‘Glivec’ (imatinib mesylate) in 1 See the order passed in Novartis AG v. Union of India (et al.) on 6 August 2007 (WP Nos. 24759 of 2006 and 24760 of 2006).

38  Sushma Kindo

India (the Indian Patent Office in Chennai) in 1997. It was granted exclusive marketing rights (EMR) in 2003, for a period of five years in expectation of the product patent regime that was due to be enacted in India by 1 January 2005 (Mueller 2007). The EMR (this provision was incorporated in India in the Patents (Amendment) Act, 1999), which is the first such granted in the country, gave Novartis the right to be the only company that could produce and market the drug in India. Novartis began enforcing EMR for Glivec by asking for an injunction against generic manufacturers of the drug in the Madras High Court. In January 2004, the court granted Novartis an injunction, restraining companies such as Cipla, Ranbaxy and Sun from manufacturing, selling, distributing or exporting the drug. Once the generic manufacturers stopped producing Glivec, the price of the drug jumped from approximately `10,000 for a month’s requirement to around `120,000. Indian drug companies went in appeal, which was heard by a Division Bench of the Madras High Court. They contended that Novartis had obtained a patent in the United States and Canada in respect of ‘pyrimidine derivatives and processes for preparation thereof’. The EMR was fiercely challenged in courts by generic producers of the drug on the grounds that the compound, being a derivative of a molecule known prior to 1995, did not satisfy the novelty criterion in the Patents Act. In 2005, the Patents Act was amended. The Patents (Amendment) Act, 2005, which allows granting of product patents, provides that EMRs would either be replaced by patents (if granted) or cancelled (if patents were rejected). By way of opposition, Cipla Limited, along with other generic producers, filed their representations under the Patents Act, 1970 (Section 25(1) as amended by Patents (Amendment) Act, 2005) and the Patents Rules, 2003 (Rule 55 as amended by Patents (Amendment) Rules, 2005). The following two issues were argued. • Whether the product seeking a patent qualified as an invention, as the product was anticipated by prior publication and obviousness. • Whether the Patent Specification brought out any improvement in the efficacy of the beta crystals over the known substance, as required by Section 3(d) of the Patents Act, 1970 as amended by the Patents (Amendment) Act, 2005. The Assistant Controller held that imatinib mesylate is already known from prior publications because the claims numbered 6–23 of the United States patent claim a pharmaceutically acceptable salt of the base compound, and the patent term extension certificate specifically mentioned imatinib



Two Cases of IPR Issues in India  39

mesylate as the product. Further the United States patent disclosed ‘methane sulphonic’ acid as one of the salt-forming groups and the patent specification clearly stated that the required acid addition salts are obtained in a customary manner. It also said that imatinib mesylate normally exists in the beta crystals form, which is thermodynamically the most stable product, and thus the invention is obvious and anticipated by prior publication, and hence not an invention under the Patents Act. The Assistant Controller agreed with the contention of the opponent that a difference of 30 per cent upon comparing the relative bioavailability of the freebase with that of the beta crystal form of imatinib mesylate, which could be due to difference in their solubility in water, did not bring out any improvement in the efficacy of the beta crystals over the known substance and thus was not patentable under Section 3(d) of the Patents Act. Thus, in January 2006, the Indian Patent Office in Chennai rejected Novartis’ patent application on the grounds that the application claimed ‘only a new form of a known substance’.2 Aggrieved by the decision of the Indian Patent Office in Chennai, Novartis filed a writ petition before the Madras High Court in January 2006 challenging the constitutional validity of Section 3(d) of the Patents Act and also for quashing of the order of the Indian Patent Office refusing to grant a product patent. It asked the court to declare Section 3(d) as noncompliant with Trade-Related Aspects of Intellectual Property Rights (TRIPS)3 and 2 Judgement of V. Rengasamy, Asst. Controller of Patents and Designs on Novartis Ag vs Cancer Patients Aid Association on 25 January 2006. Available at http://indiankanoon.org/ doc/994049/(accessed on 23 October 2012). 3 The World Trade Organization’s (WTO’s) Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), negotiated in the 1986–1994 Uruguay Round, introduced intellectual property rules into the multilateral trading system for the first time. The WTO’s TRIPS Agreement is an attempt to narrow the gaps in the way these rights are protected around the world, and to bring them under common international rules. It establishes minimum levels of protection that each government has to give to the intellectual property of fellow WTO members. In doing so, it strikes a balance between the long term benefits and possible short term costs to society. Society benefits in the long term when intellectual property protection encourages creation and invention, especially when the period of protection expires and the creations and inventions enter the public domain. And, when there are trade disputes over intellectual property rights, the WTO’s dispute settlement system is now available. The agreement covers five broad issues:

• How basic principles of the trading system i.e., national treatment, Most Favoured Nation and balanced protection and other international intellectual property agreements should be applied. • How to give adequate protection to intellectual property rights.

40  Sushma Kindo

arbitrary and in violation of Article 14 of the Indian Constitution. On 6 August 2007 the Madras High Court dismissed the petition filed by the Swiss company. The court upheld the decision of the Indian Patent Office and held that Section 3(d) of the Patents Act, as amended in 2005, along with its explanation is valid. The above case can be better understood by keeping the theoretical and legal background in mind.

Relevant Theoretical Issues A patent provides the creator of an invention or intellectual property exclusive rights to get much of its social value so as to get an incentive to innovate. This social value appropriated by the inventor is similar to the rent captured by a monopolist. A patent enables an inventor to exceed the ordinary rate of return on investment (Cooter and Ulen 2004). However, a payoff of creativity entails social costs for dissemination. The restriction of patents for a limited time by law ensures that the inventor does not hamper the dissemination forever. A patent law has to strike a balance between incentives for innovation and dissemination. A way of achieving this is by adjusting the scope or breadth and duration of a patent.

• How countries should enforce those rights adequately in their own territories. • How to settle disputes on intellectual property between members of the WTO. • Special transitional arrangements during the period when the new system is being introduced. (http://www.wto.org/english/thewto_e/whatis_e/tif_e/agrm7_e.htm [accessed on 23 October 2012]) India joined the WTO and the TRIPS Agreement in 1995. The Doha Ministerial Conference declaration in 2001 on the TRIPS Agreement and public health recognized the gravity of public health problems afflicting many less developed countries. The declaration stressed the need for the TRIPS Agreement to be part of wider international action to address these problems. It acknowledged the concerns about its effects on prices. The Ministerial Conference agreed that the TRIPS Agreement should not prevent members from taking measures to protect public health. WTO members were under obligation to implement TRIPS provision by 2000, 2005, or 2016, depending on their level of development. India was given an extended period of time to make its patent regime complaint to TRIPS. Consequently India passed the Patents (Amendment) Act, 2005, which came into force on 1 January 2005. Earlier India had allowed for the manufacture of generic versions of many drugs. Through this amendment it has now implemented a product patent regime and product patents in the pharmaceutical sector (http://www.rajdeepandjoyeeta.com/trips-a-india.html [accessed on 25 October 2012]).



Two Cases of IPR Issues in India  41

The breadth of a patent refers to ‘how similar another invention can be without infringing on patent for the original invention’ (Cooter and Ulen 2004: 124). A broad patent would encompass a pioneering discovery and its applications. Thus, it would provide the rights to develop and market applications to the firm that developed the more basic invention. A narrow patent regime would give a separate property right (patent) to each application developed. It would allow firms focusing on applications (of more basic innovation carried out by other firms) to reap commercial benefits.4 Broad patents encourage more basic, pioneering research whereas narrower ones encourage the development of applications. Hence, whether to have a broad or narrow patent depends on the socioeconomic context. The principle is that ‘if the social value of investments in fundamental research exceeds the social value of investment in developing applications, then patents should be broadened’ (Cooter and Ulen 2004). This is less likely to be the case in developing countries like India, where the social value of developing applications is likely to be high, as the cost of basic, pioneering research is very high. Thus there can be a case for narrower patents for countries like India. However, the more basic invention and applications based on them together are joint products, and in a developing country like India, very few pharmaceutical companies have the capacity to engage in both research and development. Mostly, firms either undertake basic research activity or development of applications. Here the role of the patent law regime would be to facilitate sharing of benefits (of exchange) by lowering transaction costs.

Relevant Clauses from the Patents Act, 1970 (Amended in 2005) The patent regime in India is based on the Patents Act, 1970 (as amended in 2005), which defines an ‘inventive step’ as a ‘feature of an invention that involves technical advance as compared to the existing knowledge or having economic significance or both and that makes the invention not obvious to a person skilled in the art’. It defines ‘new invention’ as that invention or technology which has not been anticipated by publication in any document or used in the country or elsewhere in the world before the date of filing of patent application with complete specification, i.e. the subject matter has not fallen in public domain or that it does not form part of the state of the art. 4

See the chapter on the Patents Act by V. Santhakumar in this volume.

42  Sushma Kindo

The Patents Act, 1970 could be said to promote a narrow patent regime as it allowed for patenting of only processes and methods. However, after joining the World Trade Organization, India had to amend the Patents Act, 1970 in 2005 to make it complaint with TRIPS. The new patent regime allows for product patents, providing patent rights to basic inventions and their applications, hence broadening its scope. However to avoid ‘evergreening’ (extension of patent life by making trivial changes to the original product) and other misuse of the Patents Act, Section 3(d) defined inventions excludable from patenting as follows: The mere discovery of a new form of a known substance which does not result in the enhancement of the known efficacy of that substance or the mere discovery of any new property or new use for a known substance or of the mere use of a known process, machine or apparatus unless such known process results in a new product or employs at least one new reactant.

Prior to becoming TRIPS compliant, the patent regime in India allowed only process patents, so once a drug was discovered, a second producer could also produce it, provided it found a different way of doing so. The rationale behind this was that the same chemical product could be obtained by different processes and methods and even starting from initially different materials and components. Hence, there is social value in patenting a new process, as it rewards the innovator without preventing further innovation. This, along with other policy aspects such as strict price control, import restrictions, high tariffs, equity ceilings on foreign participation, etc., encouraged the growth of the domestic pharmaceutical industry. One of the most important impacts of the Patents Act, prior to its recent TRIPS-related amendments, and the resulting development of a generic pharmaceutical industry, has been significantly lower prices for drugs compared to other countries (Lanjouw 1998). A shift from a process patent regime to a product regime has important economic implications for the pharmaceutical sector. A brief analysis of the social costs and benefits involved in broadening a patent regime would aid us in understanding the issue. Social Benefits Patents assume importance in the pharmaceutical sector because of the unique characteristics of molecules and compounds and the fact that they are extremely easy to copy once discovered (Nogués 1993). A product patent regime increases the patent protection standard significantly compared to



Two Cases of IPR Issues in India  43

a process patent regime, generating incentive for fundamental, pioneering research. Social Costs Patenting a specific product entails a negative social value as this would exclude all others from producing it, even through different processes (Boldrin and Levine 2008). There is a loss of social welfare due to the monopoly regime created by a product patent. Since the pharmaceutical industry is characterized by the ‘cumulative innovation’ paradigm (Basheer 2005), a product patent could obstruct competition. A drug is created through broadly three processes—drug discovery, drug development and commercialization. A product patent, being an upstream patent, has the potential to block downstream research and consequently have a negative impact on drug development (Li 2008). The Novartis case is centred on whether the drug sought to be patented qualifies to be patented at all. This issue pertains to the interpretation of the provisions of the changed Indian patent law. As per Section 3(d), a ‘mere discovery of new form of a known substance’ cannot be patented unless its ‘efficacy’ is proven to be enhanced. The interpretation of this section has significant bearing on the changed settings of the pharmaceutical sector in India post TRIPS. It could have two kinds of implications. On one hand, it could have a beneficial impact by preventing frivolous patenting and ‘evergreening’. On the other hand, it also includes the issue of granting patents to ‘incremental’ innovations, which are crucial to the pharmaceutical sector. The law is not clear in defining what constitutes ‘evergreening’ and ‘incremental innovations’. This ambiguity was an issue that rose in the case and could become a subject matter for more litigation (Dhar and Gopakumar 2003). The need for more clarity in the law in this matter is articulated by the Mashelkar Committee, which was assigned with the task of determining terms and definitions for the new patentability criteria and the scope of patenting micro-organisms in accordance with the TRIPS Agreement. It pointed out, ‘Entirely new chemical structures with new mechanisms of action are a rarity rather than a rule. Therefore, “incremental innovations” involving new forms, analogs, etc. but which have significantly better safety and efficacy standards, need to be encouraged’ (Mashelkar Committee 2009). Thus, the Madras High Court’s dismissal of Novartis’ petition, which challenged the rejection of the application for patenting ‘Glivec’, had an important implication in setting a precedent for the examination of crucial

44  Sushma Kindo

drug applications. Its interpretation of the patent law as excluding ‘Glivec’ as it is a ‘new form of known substance’, would serve, one hopes, as a deterrent to other potential non-meritorious patent applications. Another implication is that the law in its present form is good enough to prevent ‘evergreening’. However, the case also brings to light the need for provisions in the patent law to encourage incremental innovations, which are very important to the pharmaceutical sector.

Copyrights: Barbara Bradford versus Sahara Media Entertainment Summary of the Case5 Romance novelist Barbara Taylor Bradford filed a suit in the Calcutta High Court against Sahara Media Entertainment Ltd. for the unauthorized use of her work. She claimed that her novel, A Woman of Substance, was being made into a 260-part television series, Karishma—The Miracles of Destiny, by Sahara, with the same content as the novel. Bradford had not received any payment, nor had she authorized the television series (BBC News 2004). A Woman of Substance is a rags-to-riches story chronicling a woman’s rise from being a street sweeper to the head of an international corporation. Karishma—The Miracles of Destiny is also a rags-to-riches story about a woman who rises from being a street sweeper to becoming the head of an international corporation. In both, the story is from the perspective of the main character, the old woman looking back on her life. Karishma is the most expensive series produced for Indian television, costing nearly `600 million (US$13 million). After Bradford filed the suit, the Calcutta High Court issued an interim injunction against Sahara which was to broadcast Karishma. Sahara appealed the decision, citing the incredible amount of money already invested in the project and the fact that Bradford had earlier filed a similar suit in the Mumbai High Court. On 12 May 2003, the very day on which the series was set to premiere, the Calcutta High Court vacated the stay. Bradford immediately filed a special leave petition in the Indian Supreme Court to block the airing 5 Barbara Taylor Bradford v. Sahara Media Entertainment Ltd. (G.A.No. 2310/03 decided on 16 July 2003).



Two Cases of IPR Issues in India  45

of the programme. The Supreme Court reinstated the injunction, but not before Sahara had already aired the first episode. On 21 July 2003, the case came before the Calcutta High Court. The court ruled in favour of Sahara and found no proof of plagiarism as the Indian copyright law protects expressions, not ideas. After reading a summary of Bradford’s novel and listening to the evidence presented, the judgment read, ‘In [the court’s] opinion, this is just an idea. The plaintiff cannot have a monopoly on a woman making it from rags to riches’ (Desai 2005: 266). The court fined Bradford about US$30,000 ($3,000 for every week that the show was delayed) and ordered her to pay Sahara’s court costs. Bradford again appealed to the Supreme Court. On 4 August, the Supreme Court upheld the lower court’s substantive decision but reversed its damage award. The court based its decision on the fact that, while the rags-to-riches idea may have been copied, the expression had not. Therefore, although the director even admitted in an interview that Karishma was based on A Woman of Substance, the Supreme Court refused to order an injunction against the network. Although Bradford was unsuccessful in her copyright infringement claim, the case is momentous because it represents one of the few times the Indian entertainment industry has been sued for its alleged plagiarism (Desai 2005). However, Indian courts have held that a work ‘inspired’ by another copyrighted work is not necessarily a copyright infringement. Copyright infringement hinges on whether a substantial portion of the original work has been copied; as long as the theme of the ‘inspired’ work is treated differently from its inspiration, there is no violation.

Relevant Theoretical Issues The main economic motivation behind instituting copyright is to prevent free riding on the author’s expression. A copyright grants writers, composers and other artists a property right to their creation on demonstration that their work is original expression (Cooter and Ulen 2004). The cost of creating a work that can be copyrighted (like novels, movies, songs, lithographs or computer software programmes) is often higher than the cost of reproducing the work. Once copies are made available, users of these copies often find it inexpensive to make more copies. If the copies made by the creator of the work are priced at, or close to, marginal cost, others may be discouraged from making copies, but the creator’s total revenues may not be sufficient to cover the cost of creating the work. Thus, by giving the creator

46  Sushma Kindo

the right to prevent others from making copies, copyright protection trades off the costs of limiting access to a work against the benefits of providing incentives to create the work in the first place. According to Landes and Posner (1989), the main problem of copyright law is to strike a balance between access and incentives. For copyright law to promote economic efficiency, its principal legal doctrines must, at least approximately, maximize the benefits from creating additional works minus both the losses from limiting access and the costs of administering copyright protection. There are two broad components of the cost of producing new work. One is the cost of expression (time and effort, editing manuscript, setting it to type, etc.) and the second is the cost of making copies (printing, binding and distribution of individual copies). The second component of the cost of producing a work increases with the number of copies produced. The cost of expression does not enter into the making of copies because, once the work is created, the author’s efforts can be incorporated into another copy virtually without cost (Landes and Posner 1989). Creation of new work requires that the expected return from the sale of copies must exceed expected cost. Assuming a downward sloping demand curve for a given work, say a novel (as there may be no perfect substitutes for a novel), the creator will make copies up to the point where the marginal cost of one more copy equals its expected marginal revenue. The resulting difference between price and marginal cost, summed over the number of copies sold, will generate revenues to offset the cost of expression. Since the decision to write the novel must be made before the demand for copies is known, the novel will be written only if the difference between expected revenues and the cost of making copies equals or exceeds the cost of expression. If we assume that the cost of creating (equivalent) works differs among authors, the number of works created will increase until the returns from the last work created just covers the (increasing) cost of expression. In the absence of copyrights, anyone can buy, make and sell copies, eventually bidding down price to marginal cost of copying. Hence, the author will not be able to cover the cost of expression, and this might result in no works being produced at all. Like patents, breadth is an issue in copyrights. A broad copyright prevents all unauthorized copying from a copyrighted work. This raises the cost of expression for other authors as they would have to incur high search costs (the time and effort spent in going through copyrighted material) to avoid copying protected works or incur licensing and other transaction costs to obtain permission for copying. This would eventually lead to fewer works.



Two Cases of IPR Issues in India  47

A narrow copyright, on the other hand, allows the other creator to borrow from previous works without infringing copyright. This leads to lower costs of creating new works and increases the number of works (Cooter and Ulen 2004).

The Copyright Act, 1957 (Amended in 1994) and the Merit of This Case The Copyright Act, 1957 of India covered original literary, dramatic, musical and artistic works, cinematography, films and sound recordings. In 1994, the Copyright Act, 1957 was amended to include satellite broadcasting, computer software, and digital technology to the areas traditionally protected by copyright. According to the law, in order to obtain a copyright on a film, the work must be ‘original’. Originality is defined as ‘[Originating] from the producer and not a copy of some other copyrighted work’ (Desai 2005: 264). Copyright generally protects two classes of rights: exploitable rights and moral rights (Desai 2005). Exploitable rights (also referred to as economic rights) are those that the owner of the work may commercially develop. The copyright owner has the exclusive right to make copies, adaptations or photographs of the copyrighted material and the right to license these rights to others. Moral rights are those that the author of the work will always possess. Included in moral rights are the right to decide when and if to publish the work, the right of authorship and the right to prevent any alteration that may harm an author’s honour or reputation. Ideas, concepts and facts, however, cannot be copyrighted. Only the ‘form, manner and arrangement, and expression of the idea’ are copyrightable. Thus, different authors are not prevented from independently developing the same idea, even if their products have some similarities (Desai 2005). The core issue is that copyright law protects expressions and not ideas and this pertains to the breadth of the copyright. Indian copyright law can be said to be narrow in that it allows for copyrighting only ‘form, manner and arrangement, and expression of the idea’. In other words it allows for ‘independent recreation’ of the original copyrighted work (Landes and Posner 1989). This does not amount to free riding as the author has incurred the full cost of expression (Landes and Posner 1989). Sahara’s Karishma—The Miracles of Destiny borrowed the idea of the story of a woman making it from rags to riches from Bradford’s A Woman of Substance, but the expression was its own. We can make a brief analysis of the social costs and benefits from a legal regime that provides copyright protection to ideas to understand the process of arriving at an economically efficient outcome.

48  Sushma Kindo

The social benefits of copyrighting ideas are providing incentives for new work and discouraging free riding. On the other hand, there are different types of social costs. First, it greatly increases the cost of expressing ideas for other authors as it raises search costs to a great extent, eventually leading to fall in the number of works and welfare per work, thus causing loss in social welfare (Cooter and Ulen 2004). Second, it would have a negligible effect on the copier’s cost of producing copies. Copiers are copying expression either unlawfully, in which case the marginal deterrence from protecting ideas is likely to be small, or lawfully, for example, because their copying is deemed a fair use. In either case, copyright protection for ideas would have little effect on the copier’s cost of copying (Landes and Posner 1989). Third, monopoly of an idea would lead to an increase in the price of copies due to lack of competing ideas. Fourth, it would encourage rent seeking. Since the costs of developing a new idea are likely to be low in most cases relative to the potential reward from licensing the idea to others, there would be a great incentive to develop and copyright ideas. This might lead to acceleration of development of ideas without much dissemination in the hope of future gains from later authors who use the ideas. Finally, the administrative cost involved in defining ideas is very high. Courts would have to define each idea, set its boundaries, determine its overlap with other ideas, and, most difficult of all, identify the idea in the work of the alleged infringer. Thus, looking at the high social costs involved we can say that it is economically efficient that copyright law does not cover ‘idea’. If Bradford’s idea of a rags-to-riches story of a woman was granted copyright protection, it would entail greater social losses than benefits as this would increase the cost of expression for others with similar storylines without offsetting benefits. A study of the judgment in the case provides instances substantiating the rationale motivating the law to protect expressions and not ideas. As per the judgment, the theme or plot by itself, which in this case is ‘ragsto-riches’, or identical characters is not a matter of copyright because of two reasons.6 The first reason is that the plots and characters are in themselves ‘not original’. This can be said to be so as the first writer or novelist creates the novel by combining stock characters and situations (many of which go back to the earliest writings that have survived from antiquity) with his particular choice of words, incidents and dramatis personae. He does not create the stock characters and situations, or buy them. Unlike the ideas for which patents can be obtained, they are not new and the novelist acquires 6

Refer judgment mentioned in Note 5.



Two Cases of IPR Issues in India  49

them at zero cost, either from observation of the world around him or from works long in the public domain (Landes and Posner 1989). The second reason for not giving protection is that if plots and ordinary prototype characters were to be protected by the copyright law, then ‘soon would come a time in the literary world, when no author would be able to write anything at all without infringing copyright’. As mentioned earlier, this would occur as the search costs for other authors would be greatly increased. This is succinctly expressed in the Calcutta High Court judgment: Fathers, mothers, revenge, lust, sudden coming into fortune, Count of Monte Cristo themes, beggar girl marrying rich boy, rich girl marrying poor boy, and one thousand such themes, characters and plots would become the subjectmatter of copyright and an intending author, instead of concentrating upon the literary merit of his expression, would be spending his life first determining whether he is infringing the copyright of the other authors who have written on this topic or that. The law of copyright was intended at granting protection and not intended for stopping all literary works altogether by its application.

The court further states: It is always open to any person to choose an idea as a subject-matter and develop it in his own manner and give expression to the idea by treating it differently from others…Care, however, must be taken to see whether the defendant has merely disguised piracy or has actually reproduced the original in a different form, different tone, different tenor so as to infuse a new life into the idea of the copyrighted work adapted by him. In the latter case there is no violation of the copyright…In other words, in order to be actionable the copy must be a substantial and material one which at once leads to the conclusion that the defendant is guilty of an act of piracy.

Hence, we see that the court’s verdict on Sahara’s Karishma not infringing Bradford’s copyright on A Woman of Substance is on basis of Karishma being an independent recreation of A Woman of Substance. As mentioned earlier, an independent recreation does not involve free riding as the author bears the full cost of expression, and hence cannot be said to be an infringement of copyright. Thus, the court’s verdict that Bradford ‘cannot have a monopoly to the idea of a woman making it from rags to riches’ can be said to promote economic efficiency. It is based on the legal principle of providing copyright protection to expression and not ideas. This principle aims at maximizing the benefits of creating additional works against the losses from limiting access and the costs of administering copyright protection.

50  Sushma Kindo

References Basheer, S. 2005. ‘Block Me Not: Are Patented Genes Essential Facilities?’, Bepress Legal Series, Working Paper No. 577. Available at http://law.bepress.com/expresso/eps/577. BBC News. 2004. ‘Author Loses India Plagiarism Case’, BBC News, 21 July. Available online at http://news.bbc.co.uk/2/hi/entertainment/3084401.stm (accessed on 31 July 2010). Boldrin, M. and D. K. Levine. 2008. Against Intellectual Monopoly. Cambridge, UK: Cambridge University Press. Cooter and Ulen. 2004. Law and Economics, 4th edition. USA: Pearson Addison Wesley. Desai, R. 2005. Copyright Infringement in the Indian Film Industry’, Vanderbilt Journal of Entertainment Law & Practice, 7 (2, Spring): 259–278. Dhar, B. and K. M. Gopakumar. 2003. ‘Effect of Product Patents on the Indian Pharmaceutical Industry’. Available online at http://wtocentre.iift.ac.in/Papers/3.pdf (accessed on 31 March 2010). Landes, W. M. and R. A. Posner. 1989. ‘An Economic Analysis of Copyright Law’, The Journal of Legal Studies, 18 (2): 325–363. Lanjouw, J. O. 1998. ‘The Introduction of Pharmaceutical Product Patents in India: Heartless Exploitation of the Poor and Suffering?’, NBER Working Paper No. 6366, NBER, Cambridge, MA. Li, Xuan. 2008. ‘The Impact of Higher Standards in Patent Protection for Pharmaceutical Industries under the TRIPS Agreement—A Comparative Study of China and India’, World Economy, 31 (10): 1367–1382. Available at SSRN: http://ssrn.com/abstract=1276776 or http://dx.doi.org/10.1111/j.1467-9701.2008.01133.x. Mashelkar Committee. 2009. ‘Report of the Technical Expert Group on Patent Law Issues (Revised)’. Submitted to Government of India. Mueller, Janice M., J.D. 2007. ‘Taking TRIPS to India—Novartis, Patent Law, and Access to Medicines’, New England Journal of Medicine, 356: 541–543, 8 February 2007, doi: 10.1056/NEJMp068245. Available at nejm.org (accessed on 25 October 2012). Nogués, J. 1993. ‘Social Costs and Benefits of Introducing Patent Protection for Pharmaceutical Drugs in Developing Countries,’ The Developing Economies, 31 (1): 24–52.

Section Two: Contracts

4

Economics of Contract Law in India* Indervir Singh

Introduction A contract is a promise exchanged among parties to do something in the future; hence there is a time lag between the exchange of promises and the performance. A mutually agreed contract creates wealth or brings Pareto– improvement, since a party will enter into a contract only if he expects to gain more than the cost of performing his part of promise. However, there is always an uncertainty regarding the future, and a party (any time between making contract and his performance) may find that his net gains from performance, opposite to his expectations, are less than its cost. In addition, the party may receive a better offer for the same performance. So, the party may find breach more beneficial than the performance of the promise. Further, when there is a time lag between the performances of two parties, a party who has already received the performance from other party can gain by not performing as per his promise. The relationship-specific investment presents another issue in contract enforcement. When a party has made asset-specific investment based on a contract with the other party in period one, the bargaining power of second party increases in period two. Renegotiation of contract terms in period two leaves a lesser share of surplus (generated by the exchange between to parties) for the first party than what he would have got if the exchange had happened * This paper is a part of my ongoing PhD thesis. I would like to thank my supervisors, Professor V. Santhakumar and Dr N. Vijayamohanan Pillai, for their constructive criticism and continuous encouragement. I also thank Professor George Geis for his helpful comments. The usual caveat applies.

52  Indervir Singh

as per the contract in period one. Therefore, the relationship-specific investment requires long-term credible commitment to ensure investment, and lack of it results in underinvestment (Joskow 1987; Williamson 1983). As a result, when there is no way to ensure performance, there will always be an uncertainty regarding fulfilment of the promise, and the investment that requires contract will not happen. Further, a party will not enter into a contract until there is a mechanism to ensure performance. A party, which finds breach more beneficial than performance, will keep his promise only if there is mechanism to punish him in a way that the breach no longer remains beneficial to him. Despite the uncertainty regarding performance, not all contracts need external enforcement. There are many contracts which are self-enforcing. A contract is self-enforcing when the interaction is repeated for infinite rounds and a party’s payoff from cheating is less than his discounted long-run payoffs of future contracts (Telser 1980). However, a contract to be self-enforceable has to fulfil following two conditions. First, no party should know when their relationship would end. In case parties know about the number of rounds they are going to play, one party may find it beneficial to cheat in the last round, whereas expecting the cheating, the other party will not perform in the last round. Since the defaulting party will also expect this, he will cheat one round earlier, and the other party will again expect this. Extending this logic will show that the parties will expect cheating even in the first round, therefore the contract will not be performed even in the first round. Second, the party should not discount the future payoffs at very high rate, because a high discount rate may make the aggregate present value of the expected gains from the repeated future contracts less than the gains from cheating. In addition, a party may also not breach if the breach could lead to loss of his reputation in the market which, in turn, results in financial loss because he would lose the trust of other potential contracting parties (see Klein and Leffler 1981), and hence would not be able to make profits by entering into contracts with them. However, for reputation to be effective in contract enforcement, the potential future contracting parties should know the reasons for the breach, which is often difficult in large economies.1 As a result, the role of a third party, particularly law, as contract enforcer becomes important when the contract is not self-enforcing and loss of reputation may not pose a credible threat to stop breach. 1 Reputation may play a role in contract enforcement in case of trade associations, where trade union records can provide the credible information about the history of a firm (see Bernstein 2001).



Economics of Contract Law in India  53

Law, by enforcing the contract, can remove the uncertainty regarding the gains from the contract. An efficient law requires courts to enforce the terms of the contract in the original form. This is because parties will agree to a term only if it benefits each one of them, which signifies Pareto–improvement. In other words, a contract provision is efficient only if it has been agreed upon by all parties. Thus, courts by enforcing the contractual terms ensure efficiency, which requires writing a complete contract that provides a solution for all possible contingencies. However, bargaining over a remote situation may increase the cost of the contract more than the expected loss due to that contingency (Cooter and Ulen 2004). Therefore, it will be efficient to leave gaps in the contracts in certain cases. Moreover, parties may not posses the information required to bargain on each possibility. Shavell (1980) argues that damage remedy provides the substitute for the complete contract. Law prevents breach by requiring the defaulting party to pay damages to the innocent party. However, compelling the party to perform when the breach results in greater total wealth of all parties is not efficient. A breach is efficient when a party is better off by defaulting even after paying the innocent party the profits he expected from the contract, that is, no one is worse off and at least one is better off. Hence, an efficient law prevents all inefficient breaches and allows the efficient breach. Though, law, by requiring the defaulting party to compensate for the loss of plaintiff’s profit, can discourage inefficient breach, the calculation of damage in itself poses a problem. In addition, the amount paid by the parties as damages and other provisions that h have an indirect impact on the amount of damages or post-breach bargaining also have efficiency implications. There are a number of studies that have analysed contract law in the economic framework in the context of common and civil law countries (see, for example, Hatzis 2003 and Posner 2003). There are also some studies in the law and economics framework in the Indian context. These studies are mainly in the area of property rights (Morris and Pandey 2007 and Sarkar 2007), tort (Babu 2010 and Jain 2010), financial regulations (Somashekar 2010) and weak enforcement of the law (Santhakumar 2003). However, the issues related to Indian contract law have been remained unexplored. In this context, this chapter is an attempt to understand Indian contract law in light of the existing literature. Since efficient enforcement of the contract largely depends on the remedies provided for breach under different situations, it is important to examine the legal remedies available under contract law. Therefore, the focus of the present chapter will be limited to the economic analysis of the breach remedies provided by Indian courts.

54  Indervir Singh

The Indian legal system is based on the English common law tradition, and has similarities with other common law countries. The Indian Contract Act (ICA) was first instituted in 1872 and amended from time to time.2 Though ICA is not applicable to situations or places where any statute, act or regulation exist to deal with the issue, it is applicable to large number of cases and lays down the general principles of legal solutions for breaches of contract. Sections 73 and 74 of ICA state the law for award of damages, where the former discusses the provisions regarding the measurement of damages by courts and the latter deals with the rules applied in case the damages are liquidated. Further, the Specific Relief Act, 1963 (SRA) discusses situations under which the remedy of specific performance is available for breach of a contract. Though Indian contact law is similar to English contract law in most aspects, it shows significant difference with respect to a few rules as well as their explanations. The arguments by many law and economics scholars who consider contract rules in common law countries efficient by and large favour ICA too. However, all scholars do not consider contract law of common law countries efficient, and they significantly differ in their views on many issues. One issue, which is highly debated in literature, is the relative efficiency of damage remedy and specific performance (see Ulen 1984). Another debated issue is the rule regarding penalties, where the common and civil law countries differ significantly (see Hatzis 2003). As already mentioned, ICA does not follow all rules of English contract law, and also differs from the American Uniform Commercial Code. Three ways, in which the Indian law departs from English law, are allowing exclusion of the right to claim damages by express contract, not allowing court-calculated damages when the sum to be paid in case of a breach is named in the contract and the right of the promisee to accept less consideration than that mentioned in the contract. The studies have argued in favour of the first provision, especially when the contracting parties are sophisticated enough to take a rational decision (see Schwartz and Scott 2003). The second departure may also be efficient, when loss to the plaintiff, which is often difficult to assess, is less than the damages calculated by available methods. For example, the damages calculated, by considering the difference between the contract price and the market price at the time of breach may be higher than the loss of plaintiff’s profit, which is often unobservable. However, the last provision may provide the promisor an opportunity to compel the promisee to accept less consideration, if the 2 The discussion on Indian contracts in this chapter is based on Pollock and Mulla (2006), unless mentioned otherwise.



Economics of Contract Law in India  55

promisee expects to bear significant indirect losses, which cannot be recovered under contract law. This chapter is divided into six sections. The section following the introduction discusses the remedies provided by the courts for breach of contract, which includes general principles of damage measurement, rules regarding liquidated damages and specific performance. The third section analyses the rights of the contracting party to exclude contract damages by an explicit provision in the contract. The rules regarding burden of proof are examined in the next section. The fifth section points out the efficiency implication of the right of the promisee to remit the performance of the promise made by promisor. The last section concludes the study.

Breach Remedies in India Indian courts provide relief to the innocent party for breach of contract under ICA and SRA. ICA, under Sections 73 and 74, covers the general principles of contract damage and liquidated damages respectively. SRA deals with the remedy of specific performance. These statutes establish the basis of damage measurement and have implications on the decision of parties to enter into a contract. For, a remedy, which does not secure the rights of any of the parties or gives one party a chance to profit at the cost of the other, has an adverse impact on economic activity. The present section discusses the remedies provided under the two statutes and analyses them, taking into account the insights from law and economics literature. The section first analyses the general principles of awarding damages, which is followed by the discussion on the issue of liquidated damages and penalties. Finally, the importance of specific performance in relation to the remedy of damages is examined.

General Principles of Contract Damages When the contract is broken, the readily available remedy in India under the ICA is the award of money damages. The ICA (like other common law countries) is based on the principle that the promisor has the choice to either perform as per the contract or default and pay for the plaintiff’s loss. In this regard, Section 73 of ICA discusses the provisions of awarding damages to the plaintiff. In case a contact has been broken, Section 73 entitles the party who has suffered the breach to receive compensation from the defaulting

56  Indervir Singh

party for any loss or damage arising in the usual course of things from such breach (that is, what the defendant can expect to happen in the event of breach from the nature of contract even when plaintiff has not explicitly mentioned it) or that the parties knew while making contract was likely to occur from such breach. Any remote or indirect loss resulting from the breach is not compensable under ICA. Thus, the basis of calculation of damage is the loss to the plaintiff and not the gain to the defendant. ICA tends to cover the losses suffered by the innocent party so that he can achieve the same welfare level as if the contract were performed. However, the plaintiff can claim only those losses that are the direct result of breach (that is, indirect losses are not allowed) and can be reasonably foreseen. Further, the plaintiff is also required to take necessary steps to mitigate his loss due to the breach, and cannot claim damage for any loss that is the result of not taking these steps. In principle, the courts try to compensate the innocent party for his loss due to breach. Nonetheless, the courts also use other damage measures. The courts use three types of damage measures, namely, restitution damages, reliance damages and expectation damages. Expectation damages put the plaintiff in as good a position as he would have been if the defendant had not defaulted. Reliance damages cover the expenditure that an innocent party has undertaken in reliance of the defaulting party. It seeks to put the plaintiff in a position as good as if the contract was never made. Restitution damages are awarded to prevent the defendant from benefiting from the plaintiff’s loss. This damage measure compels the party at fault to surrender those gains from the breach, which are reflected in the plaintiff ’s loss. Though the primary objective of the contract is to protect expectation interest (by awarding expectation damages), reliance damages are paid to avoid wastage when it is difficult for the plaintiff to prove the benefit that he expected from the contract. Restitution damages normally are not preferred. Courts generally do not ask the defendant to surrender his gains from the breach, unless they are reflected in the loss to the plaintiff. However, the restitution measure is used in cases such as when the defendant fails to provide the goods or services after receiving payment for it, or the defendant has made a profit by doing the same thing that he agreed not to do, or the defendant wrongfully used a trade secret or other confidential information in the breach of contract for his gain. A dominant strand in law and economics literature shows that the award of expectation damages leads to an efficient outcome (see Posner 2003). To illustrate, suppose A contracted with B to provide him q quantity of a good G at a price ps per unit, which he intends to use as input to produce other goods. A fails to deliver the good on the due date, and B suffers a loss as a result



Economics of Contract Law in India  57

of it. Suppose, at the time of breach B has an option to purchase the good from the market at a price pm (the case of non-market goods will be discussed later), where pm > ps. Therefore, B’s loss due to A’s non-performance is q*pm – q*ps. Let g represents the A’s gain from breach. In case A is required to compensate B for his loss, he will choose not to perform only if his gain from the breach is more than the B’s loss, that is, g > q*pm – q*ps. On the other hand, any compensation, which compels the defaulting party to perform, results into a net loss of g – (q*pm – q*ps). This is because the loss for the defaulting party from performance will be g, whereas the gain for the innocent party will be q*pm – q*ps. Since gains from breach are higher than the loss, there will be a net loss of g – (q*pm – q*ps) due to performance. Hence, the aggregate welfare of both the parties will be higher from the breach than the performance when the gains for defaulting party are higher than the loss to the innocent party. However, the economic analysis of reliance and restitution damages does not support the efficiency criterion (Ulen 1984). To understand the implication of restitution damages, suppose B, in the previous example, has given an amount ‘a’ as an advance for the good. In this case, the restitution damages will require A to return the advance paid to him. Since the advance was the part of total amount that A was supposed to get on completion of the contract, returning the advance will not have any impact on his decision. A will have an incentive to breach for any amount greater than the contract amount. Thus, restitution damages will be inefficient, since the breach will be efficient only if A is ready to pay a sum larger than B’s loss (which is required to ensure that aggregate wealth is higher because of breach), whereas the restitution does not require A to compensate for B’s loss. On similar lines, Kull (1995), criticizing the present proposition of restitution, argues that restitution should not be considered as remedy for breach, since it creates a liability independent of the breach. Nonetheless, it may be efficient to award restitution damages in cases like mistaken transfer of wealth (Mather 1982). Similarly, reliance damages are also not based on the loss to the plaintiff, and hence may lead to inefficiency. Taking the earlier example, suppose that B in reliance of the performance of the contract has spent an amount ‘t’ for the transportation arrangement of the good. This amount cannot be recovered even if the transportation facility is not used, which results into B’s loss due to A’s breach. Under reliance damages, A is required to cover B’s loss. However, the expenditure ‘t’, which B has undertaken in reliance of the contract, has nothing to do with his expected gain from the contract. While reliance damages, which are less than B’s gain (that is, t < q*pm – q*ps)

58  Indervir Singh

may lead to inefficient breach, reliance damages may prevent an efficient breach, when the plaintiff has spent more than his loss from the breach (that is, t > g – q*pm – q*ps). Rogerson (1984) also shows that the reliance decision under expectation damages Pareto–dominates the reliance damages. The study analyses reliance decisions under different contract remedies in a situation where the buyer or the seller must spend on specific capital before the exchange and post-breach bargaining is possible. The most common way to assess damages (whenever the market price is available for the contracted good) is the difference between the contract price and the market price of the good. This method allows expectation damages to the plaintiff, which is the most efficient and also the most preferred damage measure among the three. For example, if ps is the contract price of the good and pm is the market price of the good on the date of delivery, the damages are generally assessed by taking the difference of the two prices, that is, the extra amount which a buyer has to pay to purchase the same quantity of the good from the market, or the amount lost by the seller due to breach by the buyer. However, in case of delivery of defective goods, which were accepted, the difference between contract price and market price of the good is not used to calculate damages. In this case, damages are assessed using the difference between the actual price of the good and the price that the good would have had if the good were of right quality. The reason for this is that it is easy for the party who has the good to sell it or put it to other use (Posner 2003). That is why the seller is responsible to put the good to its best use if he occupies the good, and the responsibility of putting the good into right use lies on the buyer once the good is delivered to him. The market price of the good is generally measured at the place of breach. When the market price for the good is not available at the place of breach, the price in the nearby market is used to assess the damage. If the good is not marketable, such as, goods made to order, the price of the good is the measure of damage. In case the good cannot be purchased from the market, the price at which the good is sold to a third party is used as the measure of damage. The reason for this rule is that the value of the non-market good for the plaintiff cannot be calculated. Therefore, it is efficient to compel the defendant to perform (the case of the non-market good is discussed in detail under the section titled ‘Specific Performance’ later). The courts, by asking the defendant to pay the difference between the contract price and the price at which the good was sold to a third party when the good is non-market, compels him to surrender his benefit from the breach, thereby ensuring the performance since the defendant will not have any incentive to breach.



Economics of Contract Law in India  59

Another important issue is that contract law requires the plaintiff to take necessary steps to mitigate the losses, that is, he is expected to minimize his losses due to breach. This rule prevents unjust enrichment of the plaintiff at the cost of the defaulting party, and ensures efficiency. Taking the previous example, the innocent party can purchase the same good from the market at price pm. If he purchases the good from the market, his losses from breach will be pm – ps. However, in case the plaintiff fails to take the necessary steps to minimize the loss, his loss will be higher than pm – ps. Compensating the plaintiff for such loss, which occurred as a result of his failure to take necessary steps, will lead to higher losses from the breach. On the other hand, if the loss due to failure to mitigate is deducted while calculating the damages, the innocent party will have an incentive to act in a way that will minimize the loss and maximize the aggregate benefits. Further, putting the responsibility to mitigate on plaintiff will be efficient, because he is in a better position to understand his needs and search for an alternative at lesser cost. In fact, if the promisor is in a better position to find the alternative at lesser cost than the promisee, then it would be in his own best interest to purchase the good from the market and supply it to the promisee (see Posner 2003).3 Furthermore, under ICA only those damages for losses are allowed that may reasonably be considered as arising in the usual course of events from such breach, or have been in the contemplation of the parties while making the contract as the likely result of the breach. ICA does not allow damages for any event that is remote. The logic behind this provision is not compensating for any loss that was not known to the parties and could not have influenced the decision of the parties about entering into a contract. This is because the parties will not include any remote damage into their calculation while taking a decision about contract formation and default. Therefore, providing such damages will not result in increase in efficiency. On the contrary, allowing compensation for remote damages will lead to inefficiency, since the contracting parties will have no incentive to reveal the information about the possible effect of breach (Posner 2000). By allowing only those damages, which are in contemplation of the contracting parties at the time of making the contract, the law gives the parties an incentive to reveal information regarding the possible impact of the breach. As a result, the party while defaulting will take the likely damage into consideration. In addition, the literature also argues that not compensating for indirect loss is efficient and puts the responsibility to avoid the risk on the party that 3 Some studies have expressed disagreement over this point. Schwartz (1979) argues that the seller, being a supplier of the good, is more likely to know about the market than the buyer.

60  Indervir Singh

has superior knowledge of the risk. The rule is based on the famous contract case, Hadley v. Baxendale.4 In this case, the court refused to award damages for unforeseeable consequential loss. Posner (2003) explains the economic implication of the rule with an example of a case between a commercial photographer and a company, which makes and sells as well as develops films. The commercial photographer purchased a roll of film from the company. The price of the film also included the cost of developing it. The film was used to take the pictures of the Himalayas for a magazine, and huge expenses were incurred to complete the assignment. He later sent the film to the manufacturer for development. However, it was lost in the developing room and never found. Considering the facts of the case, Posner argues that allowing the photographer to recover damages will not have much effect on the manufacturer’s efficiency, since it will be difficult for him to separate out defective films from others, or know which films are important. At the same time, the photographer will not have any incentive to take precaution. On the other hand, not allowing the damages will give the photographer an incentive to take more precautions, like using more than one role of the film or requesting developer to take extra care. Here, it will be less costly for the photographer to take precaution than the manufacturer; hence it will be efficient not to allow the damages. However, Geis (2005) points out that not allowing unforeseeable indirect loss is efficient only when the distribution of buyers is such that most of the buyers value the performance low. If many buyers place a high value on the performance, the efficiency analysis does not favour the Hadley rule. The study does an empirical analysis of three goods and finds that the Hadley rule is efficient in all three cases. However, the results may vary for other goods based on the buyer’s valuation as well as the seller’s market power. The author emphasizes the empirical work to find situations under which the rule would be efficient.

Liquidated Damages and Penalties The previous section discussed the efficiency implications of the principles that are used by Indian courts to measure damages. However, there are situations when the contracting parties name a sum that has to be paid by the defaulting party in case of breach, that is, damages are liquidated. English law differentiates liquidated damages, which are the pre-estimates of the loss, from penalties, which put a much higher burden on the defaulting party than 4

Hadley v. Baxendale, 1.9 Ex. 341, 156 Eng. Rep. 145 (1854).



Economics of Contract Law in India  61

the loss of the innocent party, and damages are not allowed above what the court considers a reasonable pre-estimate of the plaintiff’s loss. Nonetheless, English law allows the promisee to disagree with the penal clause and sue for damages at large. In India, Section 74 of ICA states the law on liquidated damages and penalties. ICA also does not allow damages that are of the nature of penalty. However, ICA does not give the promisee the freedom to choose between the penalty clause (that is, claim under Section 74) and court-measured damages. ICA provides the general principle for contracts in which a sum is named in the contract as damages to be paid in case of breach. As per Section 74 of ICA, the party complaining of the breach is entitled to receive from the defaulting party a reasonable compensation not exceeding the amount mentioned in the contract or stipulated by the court as penalty, that is, the sum named is treated as the ceiling in Indian contract law. The present section first discusses the issue of penalty and then examines the impact of the departure of Indian law from English law on efficiency. The issue of penalty is highly debated in the law and economics literature on contract (see Hatzis 2003). Common law judges are generally reluctant to award liquidated damages when the stipulated sum named in the contract is higher than the reasonable pre-estimate of the loss of the promisee. On the contrary, courts in civil law countries are quite liberal in enforcing liquidated damages even when they are in the way of penalties, unless the amount mentioned is extravagant or purely gambling. The law and economics scholars who favour the common law approach consider penalties inefficient since they do not permit efficient breach (as discussed earlier, the efficiency requires the damages to be equal to the loss of innocent party) and there are chances of exploitation of the promisor when there are significant differences in the bargaining powers of both parties and there is information asymmetry. In addition, if the promisee is to receive a higher sum in case of breach than by fulfilment of the promise, allowing the penalty clause may give the promisee an incentive to create the situation leading to the default. On the other hand, proponents of the penalty clause argue that awarding penalties will compensate for the idiosyncratic value of the promisee and help the risk-averse to better manage the risk. It also helps avoid the uncertainty regarding contract and reduce transaction cost of litigation. Sometimes, the promisee may also like the promisor to act as the insurer, and the premium of the insurance is reflected in the higher price for the fulfilment of the promise (Mattei 1995). In these cases, not awarding the agreed sum would be inefficient because it would prevent the parties from entering into a contract. The penalty clause also works as a guarantee for new firms that do not have a strong reputation in the market, and other firms are reluctant

62  Indervir Singh

to do business with them. The proponents of the penalty clause, without completely rejecting the arguments against the penalty clause, point out that the contracting parties are the best judge of their interests and the exploitation of one party by using the penalty does not happen in commercial transactions among sophisticated parties (see Schwartz and Scott 2003). Goetz and Scott (1977) argue that initially the restriction on penalties came as a rule to prevent fraud and duress. The rule was further supported by the fact that a rational assessment of nature and extent of risk by parties was often prevented by information barriers. However, the rule was applied to all contract cases due to the high cost of finding fraud or information barriers in individual cases. Therefore, the restriction should only be applied to unfair bargaining. The restrictions on penalties will be efficient in standard form consumer contracts where the bargaining power of the consumer is much less than a corporate firm, but inefficient when applied on sophisticated players (Hatzis 2003). Further, when there is a risk that one party can make the performance difficult, the contracting parties can write provision to restrain the other party from such activities. Thus, researchers who are in favour of the penalty clause argue that the penalties should be restricted only when there is a large difference in the bargaining power of the two parties, such as in standard form consumer contracts, or when the penalties are unreasonably high. Polinsky and Shavell (1998), however, argue that penalties should be allowed only when the defendant has a significant chance of escaping liability for the damage he has caused. They also point out that the award of punitive damages against corporations may not serve the purpose because any penalty on corporations often does not affect the employees responsible for the breach, instead it will have an effect on shareholders and consumers. The law and economics literature seems to agree on the point that penalties should not be allowed in contracts where parties have unequal bargaining power, such as standard form consumer contracts. The dominant strand in law and economics literature is also of the view that party-provided remedy is efficient if the parties are sophisticated enough to take informed decisions, and neither party enjoys the dominant position in bargaining. Nonetheless, limiting the penalty clause to some issues, such as standard form consumer contracts, rather than making it a general rule is highly debatable, and remains more or less inconclusive. De Geest and Wuyts (2000) are of the view that a dominant strand of the literature considers penalties inefficient, whereas Hatzis (2003) argues that the widely accepted view in law and economics literature is that penalties should be allowed. One of the reasons that the debate on the penalty clause remains inconclusive is the lack of empirical studies. The rules regarding penalty clauses have important implications



Economics of Contract Law in India  63

for Indian economy, and empirical studies may provide useful insight into the issue. Though there are many problems with allowing penalties, the argument that sophisticated parties are in a better position than the courts to provide a remedy and therefore rejecting party-agreed remedy may lead to inefficiency cannot be rejected. In this regard, Indian contract law also needs to provide more flexibility in dealing with contracts involving wellinformed sophisticated parties. The calculation of the amount above which the liquidated damages are considered as penalty is another important issue. In India, the penalty is calculated on the basis of promisor’s liability in original contract, that is, the damage that an innocent party has to bear due to the default. For example, an illustration of the section provides that if A, who owes money to B, a moneylender, undertakes to repay him by delivering to him 10 maunds of grain on a certain date, and in the event of his not delivering the stipulated amount by the stipulated date, he shall be liable to deliver 20 maunds (see Pollock and Mulla 2006: 1658). ICA considers this stipulation a penalty and the plaintiff can only recover reasonable amount as damage. Here, law, while stipulating 20 maunds of grain as penalty, has considered the original deal, which was delivery of 10 maunds of grains. As a result, the same amount of damages, depending on the original liability, can be liquidated damages in one situation and penalty in the other. The same principle is applied in case of interest rates. Though the interest rate in India is controlled under the Interest Act, 1978, the provisions regarding increase in interest in case of default are set out under Section 74. For example, ICA does allow change of simple interest either into compound interest or to increase rate up to a certain point in case the money is not repaid on the due date. However, both increasing the rate of interest as well as making it simple to compound cannot be done in the event of breach. For instance, assume that B is supposed to pay A simple interest at the rate of 10 per cent per annum on a debt of `10,000. A, in case of not paying debt on time, can either demand from B an interest higher than 10 per cent or charge compound interest at the same rate (that is, 10 per cent) instead of simple interest, but cannot do both, that is, the promisee cannot increase the rate of interest and change simple interest to compound. A person who is already charging compound interest of 10 per cent per annum on the same amount of debt can demand a higher rate of compound interest in case of default. The rules regarding interest rates are based on the assumption that the opportunity cost of money for the lender remains almost the same, and the defendant has agreed on a large increase in the interest rate in case of breach due to unequal bargaining power. However, these assumptions need not be

64  Indervir Singh

true. For a person, the cost of parting with money may be different at different points of the time, which depends on external factors like interest rate in the economy, availability of opportunities and the use to which he is thinking to put that money. At any point of time, a person may reasonably expect much higher returns from the money at a future date. It is also unlikely that a lender, who has given the money at a rate lesser than what is allowed under the Interest Act, had unequal bargaining power at the time of contract to include the penalty, especially when the after-breach rate is still below the maximum allowed under that statute. Hence, there is no basis in the case of interest rate not to allow increase in the interest rate, which is much higher than the original rate but is still below the rate that can be considered as the result of unequal bargaining power (for example, the rate allowed under the statute cannot be considered the result of unequal bargaining power). Moreover, the penalty may be just an instrument to ensure the timely payment of debt, which may be crucial for the lender to avoid direct and indirect losses (for example, he may need money for an important business or personal need). In fact, the large difference in the original interest rate and interest rate after breach may be the result of a concession that the lender had agreed to give on the original rate in return for the promise of a penalty in case of default. Here, lower original interest rate is like an insurance premium for securing the performance. There are also certain prohibitions regarding the penalty clause that can be avoided merely by changing the wording of the contract. For instance, in case of default, the innocent party can charge a higher rate from the date of breach (that is, the date on which debt was payable), but not from the beginning. For example, if A lent B money on the compound interest rate of 10 per cent per annum, and stipulated that in case of late payment, the interest payable will be 12 per cent from the date of contract, this stipulation is by the way of penalty and the increased interest is allowed only from the date of breach. However, charging an interest of 12 per cent from the date of contract is not by the way of penalty, if A mentions in the contract that the interest rate on the loan is 12 per cent per annum but that B has to pay only 10 per cent in case he pays the debt within specified time limit. Even though the result of both the contracts is identical, the former is a penalty, whereas the latter is not. Rules that can be avoided by changing the wording of the contract do not serve any purpose, instead they can be misused by the more informed party; hence they lead to inefficiency. However, these aspects need to be studied further in detail. Though Indian and English contract laws are similar in their treatment of penalties, they diverge on the issue of the right of the promisee to sue for



Economics of Contract Law in India  65

court-measured damages. ICA does not allow the innocent party to choose to sue for the penal clause or damages at large. In Indian contract law, the sum named in the contract acts as the upper limit against which the damages can be awarded to the plaintiff, even when the damages assessed by the courts would have been higher. This provision may increase efficiency when courtcalculated damages are more than the actual loss. To illustrate, consider the example in earlier section, where A contracted with B to provide him q quantity of a good G at a price ps per unit, and A agreed to pay damages of d per unit in case of default. The market price of the good, at the time of delivery, is pm. The damages calculated by the court will be q*pm – q*ps, whereas the damages calculated as per the remedy provided by the parties is q*d. If the market price of the good, at the time of delivery, is higher than the price at which damages are calculated by the parties (that is, q*d < q*pm – q*ps), then it will be beneficial for the plaintiff to sue for court-provided damages. Suppose the sum named in the contract is the profit per unit that the plaintiff was expecting to make from the completion of the contract. Since the parties are in a better position to know the expected profit, but may not have enough evidence to prove it, the actual loss to the plaintiff will be less than the difference between the market price and the contract price. In this case, awarding court-calculated damages would result in an inefficient outcome, since it will not allow an efficient breach.5

Specific Performance In previous sections, the chapter discussed the rules regarding award of damages in the event of breach. Though money damages is the most frequently used solution by courts to compensate the innocent party, it is not the only measure available. An additional remedy provided by courts is specific performance under which the defendant is required to perform as per the contract. The remedy of specific performance is at the courts’ discretion and is not normally awarded in India (as well as in other common law countries). Specific performance, like liquidated damages, is a highly debated topic in law and economics literature. Some studies argue that the present provision of specific performance in common law countries (which are the same in India) 5 Going with the argument of researchers who stressed on payment of damages equal to actual loss to ensure efficient default, making the sum mentioned in the contract as the upper limit may create some inefficiencies, since it allows the promisor to default even when there may not be any net increase in surplus. Further analysis of the provision is needed to understand its efficiency implications.

66  Indervir Singh

leads to efficient outcomes (Kronman 1978), whereas others show that it will be efficient if specific performance is readily enforced (Ulen 1984). In India, the Specific Relief Act, 1963 (SRA) provides an alternative to the remedy of damages in ICA, by way of specific performance. Under this provision, the plaintiff can file a suit to compel the breaching party to perform as per the contract. SRA is an improved version of the Specific Relief Act, 1877, which was originally drafted on the lines of the Draft New York Civil Code, 1862. The main provisions of the older act were based on the doctrines developed by the equity court of England. Since SRA is based on English law, the substantive law and practices related to different sections can be interpreted in light of decisions by English courts, unless the provision in any section explicitly conflicts with the principle prescribed by English courts. The remedy of specific performance is not generally available to the innocent party, and the relief is at the discretion of the courts. This also means that the contracting parties cannot agree on the specific enforceability of the contract. Specific performance is available only in those cases when it is difficult to calculate the actual damage suffered by the innocent party, or money damages cannot compensate the plaintiff adequately. Hence the plaintiff, for enforceability of specific relief, has to prove that money damages are not adequate in that particular case. Law presumes that damages are not adequate relief in case of an agreement to purchase land, or an article that has high subjective value for the plaintiff, or an article that is not an ordinary commerce item or is not easily available in the market. The court generally does not allow specific enforcement in contracts, which need continuous supervision, such as, contracts to build, repair or provide personal services. However, courts may enforce contracts requiring supervision, if the contract clearly specifies (expressly or by implication) the work to be done, so that courts can enforce the contract in a way that the defendant knows what precisely he has to do to comply with the courts’ orders. The rule is provided because it is difficult for courts to verify the defendant’s compliance with the contract, in case the performance of the contract cannot be ascertained without incurring huge supervision costs. Further, the contract can also be specifically enforced when liquidated damages are provided in the contract and the party at fault is willing to pay the sum. The courts, in this case, determine whether the intention of the parties to provide liquidated sum is an alternative contract in case of default or to secure performance as per the contract. If courts finds that liquidated damages were to secure performance, specific performance can be awarded. Specific performance also ensures efficient breach by allowing post-breach negotiation. Suppose B contracts with A to provide him a service at a price w. A expects to gain a profit m from the service, where v is the cost to B to



Economics of Contract Law in India  67

perform the task, and he earns an amount w – v by providing the service. Now, B’s cost of providing the service has increased to z. Therefore, B will lose z – v by providing the service compared to his earlier situation. In this situation, efficiency considerations require him to complete the project if z – v < m, and pay damages to A in case z – v > m. Awarding specific performance gives the same results. Here, A will be better off if he gets any amount more than m. On the other hand, B will better off by paying any amount less than z – v to purchase the right of specific performance from A. If B’s loss is more than A’s expected profit from the service (that is, z – v > m), B will renegotiate with A and buy A’s right to claim specific performance. As a result, the service will not be provided, and the damages, which will be something between m and z – v, will be paid to A. However, the post-breach negotiation will not happen if A’s expected gain from the project is higher than B’s loss from completing the project (that is, m > z – v), because B will find it less costly to complete the project than to pay A his expected profits. These results are the same as required by the efficient breach theory, that is, breach will happen only when the promisor’s gain from breach is higher than the promisee’s loss. Another important aspect of awarding specific performance is that it protects the subjective valuation of the promisee. For example, in case A, in addition to the profits m, attaches a subjective value, k, to the service, the breach will happen only if z – v > m + k. Therefore, awarding specific performance will ensure efficient breach. However, making specific performance a routine remedy is a debated issue. The main focus of the debate is cost of pre- and post-breach negotiation. Where one strand of literature argues that pre- and post-breach negotiation cost is lower under the present provision, the other strand of literature advocates making specific performance a routine remedy. Kronman (1978) favours the above provision of specific performance. He argues that the provision is in line with what the contracting parties would have wanted as the remedy, and therefore has lower pre-breach negotiation cost. The rules in this case are based on the difference between unique and fungible goods. While fungible goods can be easily purchased from the market, the unique goods have either very limited market or no market. Due to the limited market in case of unique goods, it is highly unlikely that seller will get a better offer. As a result, he would like the contract to be specifically enforced. The buyer would also like to have the contract specifically enforced since the damages in case of unique goods are under-compensatory due to the high subjective value he attributes to them. On the other hand, both the seller and buyer would prefer money damages when the contract is for fungible goods. This is because the seller will expect better offers due to the developed

68  Indervir Singh

market, whereas the buyer will have less risk of under-compensation due to low subjective value. Therefore, the present provision regarding specific performance will reduce the pre-negotiation cost of the parties. Posner (2003) also favours specific performance only in those cases where the good is not available in the market and it is difficult to ascertain the value of the good for the promisee. He argues that providing specific performance for fungible goods will lead to high cost of post-breach negotiation, where the promisee will try to get a share of the promisor’s surplus. For example, A contracts with B to supply 100 quintals of rice for `10,000. Before the fulfilment of the contract C offers A `12,500 for the same quantity of rice. Suppose the market value for the contracted quantity of the rice increases to `11,000. In case of money damages, A will pay B `1,000 and sell the rice to C and gain a surplus of `1,500. However, when the remedy is specific performance, B has to negotiate with A to get a share of surplus. Posner argues that post-breach negotiation cost will be less in case of money damages, because the cost of determining the difference between the contract price and market price is easier than the negotiating for the division of surplus. Schwartz (1979), contradicting the above arguments, supports specific performance as the routine remedy for contract breaches. The study points out that the argument that the present provision regarding specific performance leads to lower pre- and post-breach negotiation cost is not valid. The claim that contracting parties would prefer specific performance in case of unique goods and money damages when the contract is for fungible goods does not hold, because the unique goods do not lack a developed market, and the seller can reasonably expect a better offer. On the contrary, the seller may like to have the contract specifically performed in the case of fungible goods when the competition is very high, so that he can secure the performance. Further, the view that the post-breach cost of negotiation will be higher in case of specific performance depends on the assertion that the buyer, in case of default, has a lower cover cost then the seller. However, it need not be so. It is argued that the seller already knows the exact specification of the goods and often has superior knowledge of the market for the goods than the buyer, since each producer has to be fully aware of the market for his products and his competitors. In addition, the argument that the buyer, by posing a threat of legal suit, may compel the seller to perform even when his own cover cost is less is valid only if the difference between the cover cost of the buyer and seller is larger than the cost of preparing a legal suit to pose the threat. Another argument in favour of the remedy of damages is that specific performance in cases of high inflation may lead to dead-weight loss if the



Economics of Contract Law in India  69

promisee stresses on performance even when the breach is more efficient than the performance. For example, A contracts with B to complete a construction project for `10,000 (this example is taken from Schwartz 1979). A expects to earn a profit of `3,000 from the project. The cost of production for B is `8,000 and he earns `2,000. Suppose the cost of production increases from `8,000 to `15,000. In this situation, enforcing specific performance will impose a burden of `7,000 on B, whereas A will gain only `3,000 from the completion of the project. The remaining `4,000 will be a dead-weight loss. In this case, allowing the breach and paying expectation damages of `3,000 will increase the efficiency. Schwartz (1979), however, notes that the dead-weight loss may also happen under the damages remedy, as the promisee, in the example, can claim `7,000 as the difference between the cost of contract (`10,000) and the market price (`15,000 as production cost and `2,000 as profit). On similar lines, Ulen (1984) argues that specific performance is a better remedy than money damages, and strongly recommends specific performance as a routine remedy. Most importantly, specific performance is most efficient in protecting the idiosyncratic values of the promisee. The cost of determining the subjective value for the promisee in the courts is very high, as a result the damages remedy leads to under-compensation. Specific performance, on the other hand, provides the innocent party at least the value he had contracted for, thus allows the contract breach only when it is efficient, that is, someone is better off and no one is worse off. Under specific performance, the parties determine which among them values the good the highest. Since the contracting parties are in a better position to determine the highest-value user, the post-breach cost of moving the good to that party will be lower under this remedy. Further, if all valid contracts are specifically enforced, the parties will have more incentive to exchange reciprocal promises efficiently by better risk allocation at the time of contract formation, rather than depending on the courts or post-breach negotiations under the threat of legal action. Ulen (1984) accepts that the cost of contracting for the sale of a fungible good will increase under specific performance. Nonetheless, he points out that Kronman’s (1978) logic says that the cost of contracting around specific performance will be less when the contract is for a unique good, therefore the cost of contracting around specific performance will not be large when the good is not purely homogenous, that is, there is a certain amount of product differentiation. On the other hand, the cost of contracting will increase with increase in uniqueness of the good when money damages is the routine remedy. Since the share of contracts for purely homogenous products is likely to be small, the net increase will not be significant. His

70  Indervir Singh

study also rejects the conclusion that the post-breach negotiation cost will be higher under specific performance. He argues that the post-breach cost may be high only in the beginning, and will come down in later periods due to lesser number of court cases and lower litigation costs. Ulen (1984) also asserts that not enforcing the contract specifically, even when the supervision cost is high, may not be efficient if the promisor fears that the bad performance will hamper his reputation, which, in turn, has an impact on his future returns. Brooks (2006) also argues that the efficiency criterion does not favour the present law which gives the promisor power to choose to perform or pay damages (efficient breach hypothesis) over its alternative that gives the same power to choose between performance and damages (efficient performance hypothesis) to the promisee, which is also considered better on moral grounds. Rogerson (1984) analyses reliance decisions under three contract remedies, namely specific performance, expectation damages and reliance damages in a situation where the buyer or the seller must spend specific capital before the exchange and post-breach bargaining is possible. The study shows that reliance decisions under specific performance Pareto–dominates the other two remedies. Further, specific performance may be awarded even if the contract specifies liquidated damages and the defaulting party is ready to pay that. The liquidated damages are ignored when courts, on the basis of evidence, are satisfied that the sum named in the contract was to secure the performance, and was never meant to be the remedy for breach of the contract. This provision is to compensate the innocent party for his subjective losses, because law does not allow penalty clauses in contracts and the parties cannot contract for specific performance as the breach remedy. Therefore, law allows specific performance even when a sum is named in the contract to be provided in case of breach, if courts are satisfied that it was to secure performance. This aspect of specific performance has got little attention. This is because the rule will not be required if law allows the parties to agree on specific performance or starts awarding penalties. Therefore, studies mainly analyse those situations under which awarding specific performance would be efficient. The literature shows disagreement on the likely effect of making specific performance a regular remedy. The answer to the question mainly lies in pre- and post-breach negotiation cost under money damages and specific performance. Though the lack of empirical evidence is the main reason that the debate remains inconclusive, the view that specific performance should be made a regular remedy seems to dominate the other. Nonetheless, there is an immense need to have more studies (especially empirical) to



Economics of Contract Law in India  71

understand the effect of the present rules regarding money damages and specific performance.

Express Provisions in the Contract for Relief The Indian Contract Act, in certain cases, allows remedies that are agreed on by the contracting parties. One of these kinds of provisions is the right of the parties to exclude the right to claim damages. ICA gives the parties a right to exclude the right to claim damages by writing such a provision in plain and unambiguous language in the contract. In this case, ICA is different from English law, which restricts any provision to exclude or limit the right to claim by statute. Nonetheless, ICA does not allow a party at fault to benefit under this provision. In addition, the remedy of suit for damages in case of breach is not excluded, even if the contract has provided other remedies. Schwartz and Scott (2003) show that it is efficient to provide a remedy according to the terms of the contract, when the contract is of a commercial nature and the parties in a contract are sophisticated enough to take a decision in their best interest. The choice of exclusion of the right to claim damages to the parties by express contract under ICA provides more flexibility to the parties to contract around the terms that they think are best suited to them and maximizes the aggregate wealth of the parties. Nonetheless, with respect to those contracts where the contracting parties do not have equal bargaining power and information, as in the case of consumer contracts, should be protected by different laws.

Burden of Proof Besides the provisions of contract remedies, the other important issue is burden of proof. The issue of providing proof is crucial for proving the breach, choice of breach remedy and right assessment of the damage. Before measuring and awarding damages, the law requires the plaintiff to prove the breach. Further, the plaintiff is supposed to provide proof of the loss that directly resulted from the breach, and prove that the loss is such that it was expected to arise naturally from such breach or it was in the contemplation of the parties at the time of formation of the contract that loss would arise from the breach. Nonetheless, the loss need not be proved

72  Indervir Singh

with certainty, and no proof of loss is not a reason not to award damages. Further, the burden of proof to reduce the amount of damages from the maximum possible (based on the proof provided by the plaintiff) is on the defendant. Studies argue that the burden of the proof should fall on the person who can provide the evidence at lower cost, and also compel them to disclose the information (see Hay and Spier 1997 and Posner 1999). In light of the economic analysis of law of evidence, the provisions in ICA can be considered efficient. The plaintiff is in a better position to prove the breach and provide evidence regarding his loss due to the breach than the defendant. This is because there may be many ways that a breach can occur, and the defendant has to prove that breach did not occur in any of those possible ways, whereas the plaintiff has to provide evidence only on the kind of breach that has occurred. In addition, the plaintiff is more likely to have information regarding the breach, for instance, it is easy for the plaintiff to prove that the goods delivered were defective, because he will have the possession over the goods and can point out the exact defect in them. Similarly, it is easier (or less costly) to prove that the defendant had the information regarding the likely impact of the breach on the plaintiff than proving that information was not provided. Therefore, it is efficient to require the plaintiff to prove that his loss is the direct result of the breach and that it was in the contemplation of the defaulting party at the time of making the contract that loss would arise from such breach. However, proving the exact amount of loss is difficult in most contracts; hence, requiring the plaintiff to prove the loss with certainty may lead to under-compensation when loss is hard to measure, which will result in an inefficient outcome by giving the defaulting party an incentive to opt for inefficient breach. It will also be costly for the plaintiff to prove that he had taken all the necessary measures to mitigate the loss, since he has to prove that no other option was available. However, the defendant has to prove just the availability of the solution that could have been used to mitigate the loss. Thus, the cost of proving that other alternative/s was available would be lower for defendant.

Dispensing With or Remitting the Performance The earlier sections have analysed the remedies for breach remedies the provisions regarding burden of proof in light of the law and economics literature. The rules regarding dispensing with the performance of the promise do not come under the breach remedy. Nonetheless, breach remedies cannot



Economics of Contract Law in India  73

be fully separated out from other provisions. These provisions may result in inefficiency when combined with breach remedies, especially when they give an advantageous position to one party after the breach. It is important to discuss rules regarding dispensing with the performance, because these rules have important implications for post-breach negotiation between the parties. Indian contract law on this provision significantly differs from the English law. The English law does not allow any contract by which a party agrees to receive consideration lesser than in the original contract, unless the benefited party has provided or agreed to provide something as compensation for the other party’s loss. On the contrary, Section 63 of ICA allows the promisee to grant a concession or waive wholly or in part the performance of the promise made by the promisor without the requirement of any consideration for it. This rule has important implications for the contracting parties, and may create inefficiencies. The promisee does not have any incentive to dispense with or remit the performance of the promise without any consideration. If the promisee has waived wholly or in part his right to performance without any compulsion, there are few chances that he would insist upon it later. Therefore, the acceptance of less consideration, unless promisee is compensated for his loss, may be the result of fraud, duress or any other compulsion. There are situations when the promisee may benefit from the acceptance of less consideration. For instance, the courts do not award damages for indirect losses resulting from the breach. If the partial performance of the contract at that time can save promisee from large losses, he may find it beneficial to remit the performance of promise partially. However, if the promisor knows about the promisee’s situation, he may compel the promisee to accept less consideration by threatening to default. For example, consider a situation where A owes B `1 million. B needs the money for his business. B has to suffer a loss of `500,000, if A does not pay back the money in time. Suppose B can save his business if A can pay `700,000. In this situation, it will be in B’s benefit to accept any amount greater than `700,000, because his maximum loss will be `300,000 and he will be able to prevent the loss of `500,000. However, allowing the promisee, B to remit the performance will also give the promisor, A a chance to compel B to give him a concession by taking advantage of his need. Therefore, giving the promisee the right to claim the remaining amount will prevent the promisor from seeking a concession by taking advantage of the promisee’s situation, and allow only efficient defaults. Moreover, the acceptance of less consideration, where a promisee has willingly abandoned his right to performance of the promise, will not have any problem, since the promisee, most probably, would not insist upon it in future.

74  Indervir Singh

Conclusion The present chapter analyses the efficiency implications of damage remedies provided under ICA and SRA. In addition, the implications of their provisions, the exclusion of the right to claim damages by explicit contract and allowing a promisee to accept a consideration lesser than in the original contract, are also discussed. The study finds that Indian contract law has many similarities with English law, and is based on the principle of compensating the innocent party for his foreseeable loss. A large number of studies in the context of common law countries show that Indian contract law is efficient. However, there is a significant amount of literature, especially in the context of penalty clauses and specific performance, which criticize the common law approach on efficiency grounds, and emphasize that courts, to ensure efficiency, should enforce contract terms related to penalties and also adopt a more liberal approach towards specific performance. Despite the fact that Indian law is based on English law, they differ significantly on some points. One of the points where Indian law is different from English law is that Indian contract law does not allow a promisee to sue for court-calculated damages if liquidated damages are provided for in the contract. It is argued that Indian law is efficient in this case, because the parties know more about their expected loss from the breach, and the court’s assessment of damages (which are often calculated as the difference between the market price and the contract price) may be higher than the actual loss due to lack of proof on profits. Another point of difference is that the exclusion of the right to claim damages by explicit contract is allowed under Indian law. This rule is in accordance with the argument by many scholars that parties, especially sophisticated ones, are in a better position than the court to decide about such provisions. Hence, the exclusion should be allowed unless there is a chance that the provisions are the result of unequal bargaining power or asymmetric information among parties. The third provision where Indian law departs from English law is allowing a promisee to accept a consideration lesser than in the original contract. Though this provision may be useful in avoiding indirect losses, the fact that the plaintiff may suffer indirect losses may provide the defendant a chance to compel the plaintiff to accept lesser consideration. Moreover, any concession by the plaintiff, if voluntary, is less likely to lead to litigation. The litigation is more likely to happen if the plaintiff has to accept it against his will.



Economics of Contract Law in India  75

References Babu, P. G. 2010. ‘Suit and Settlement under Asymmetric Information: The Case of Bhopal Gas Disaster’, in P. G. Babu, T. Eger, A. V. Raja, H. B. Schfer and T. S. Somashekar (eds), Economic Analysis of Law in India: Theory and Application, pp. 94–112. New Delhi: Oxford University Press. Bernstein, L. 2001. ‘Private Commercial Law in the Cotton Industry: Creating Cooperation through Rules, Norms, and Institutions’, Michigan Law Review, 99 (7): 1724–1790. Brooks, R. R. 2006. ‘The Efficient Performance Hypothesis’, Yale Law Journal, 116 (3): 568–596. Cooter, R. and T. Ulen. 2004. Law and Economics, 4th edition. New Delhi: Pearson Education. De Geest, G. and F. Wuyts. 2000. ‘Penalty Clauses and Liquidated Damages’, in B. Bouckaert and G. De Geest (eds), Encyclopedia of Law and Economics, Vol. III, The Regulation of Contracts. Cheltenham: Edward Elgar. Geis, G. S. 2005. ‘Empirically Assessing Hadley v. Baxendale’, Florida State University Law Review, 32 (3): 1–60. Goetz, C. J. and R. E. Scott. 1977. ‘Liquidated Damages, Penalties and the Just Compensation Principle: Some Notes on an Enforcement Model and a Theory of Efficient Breach’, Columbia Law Review, 77 (4): 554–594. Hatzis, A. N. 2003. ‘Having the Cake and Eating It Too: Efficient Penalty Clauses in Common and Civil Contract Law’, International Review of Law and Economics, 22 (4): 381–406. Hay, B. L. and K. E. Spier. 1997. ‘Burdens of Proof in Civil Litigation: An Economic Perspective’, The Journal of Legal Studies, 26 (2): 413–431. Jain, S. K. 2010. ‘Negligence Rule: Some Strategic Aspects’, in P. G. Babu, T. Eger, A. V. Raja, H. B. Schfer and T. S. Somashekar (eds), Economic Analysis of Law in India: Theory and Application, pp. 77–93. New Delhi: Oxford University Press. Joskow, P. L. 1987. ‘Contract Duration and Relationship-Specific Investment: Empirical Evidence from Coal Market’, American Economic Review, 77 (1): 168–185. Klein, B. and K. B. Leffler. 1981. ‘The Role of Market Forces in Assuring Contractual Performance’, Journal of Political Economy, 89 (4): 615–641. Kronman, A. 1978. ‘Specific Performance’, University of Chicago Law Review, 45 (2): 351–382. Kull, A. 1995. ‘Rationalizing Restitution’. California Law Review, 83 (5): 1191–1242. Mather, H. 1982. ‘Restitution as a Remedy for Breach of Contract: The Case of the Partially Performing Seller’, Yale Law Journal, 92 (1): 14–48. Mattei, U. 1995. ‘The Comparative Law and Economics of Penalty Clauses in Contracts’, American Journal of Comparative Law, 43 (3): 427–444. Morris, S. and A. Pandey. 2007. ‘Towards Reform of Land Acquisition Framework in India’, Economic and Political Weekly, 42 (22): 2083–2090. Polinsky, A. M. and S. Shavell. 1998. ‘Punitive Damages: An Economic Analysis’, Harvard Law Review, 111 (4): 869–962. Pollock and Mulla. 2006. Indian Contract and Specific Relief Acts. 2 vols. New Delhi: LexisNexis Butterworths. Posner, E. A. 2000. ‘Contract Remedies: Foreseeability, Precaution, Causation and Mitigation’, in B. Bouckaert and G. De Geest (eds), Encyclopedia of Law and Economics, Vol. III, The Regulation of Contracts, pp. 162–178. Cheltenham: Edward Elgar.

76  Indervir Singh Posner, R. A. 1999. ‘An Economic Approach to the Law of Evidence’, Stanford Law Review, 51 (6): 1477–1546. ———. 2003. Economic Analysis of Law. New York: Aspen. Rogerson, W. P. 1984. ‘Efficient Reliance and Damage Measures for Breach of Contract’, RAND Journal of Economics, 15 (1): 39–55. Santhakumar, V. 2003. ‘Citizens’ Actions for Protecting the Environment in Developing Countries: An Economic Analysis of the Outcome with Empirical Cases in India’, Environment and Development Economics, 8 (3): 505–528. Sarkar, A. 2007. ‘Development and Displacement: Land Acquisition in West Bengal’, Economic and Political Weekly, 42 (16): 1435–1442. Schwartz, A. 1979. ‘The Case for Specific Performance’, Yale Law Journal, 89 (2): 271–306. Schwartz, A. and R. E. Scott. 2003. ‘Contract Theory and the Limits of Contract Law’, Yale Law Journal, 113 (3): 541–619. Shavell, S. 1980. ‘Damage Measures for Breach of Contract’, Bell Journal of Economics, 11 (2): 466–490. Somashekar, T. S. 2010. ‘Unit Linked Insurance Policies: Do regulatory Gaps Pose a Moral Hazard?’, in P. G. Babu, T. Eger, A. V. Raja, H. B. Schfer and T. S. Somashekar (eds), Economic Analysis of Law in India: Theory and Application, pp. 28–39. New Delhi: Oxford University Press. Telser, L. G. 1980. ‘A Theory of Self-Enforcing Agreements’, Journal of Business, 53 (1): 27–44. Ulen, T. S. 1984. ‘The Efficiency of Specific Performance: Toward a Unified Theory of Contract Remedies’, Michigan Law Review, 83 (2): 341–403. Williamson, O. E. 1983. ‘Credible Commitments: Using Hostages to Support Exchange’, American Economic Review, 73 (4): 519–540. Yorio, E. 1982. ‘In Defense of Money Damages for Breach of Contract’, Columbia Law Review, 82 (7): 1365–1424.

5

The Economics of Contract Law: A Business Outsourcing Application George S. Geis

Introduction Business outsourcing has become a common strategy for firms seeking to cut costs, renovate their value chains or concentrate on narrower areas of activity. Outsourcing can be defined as a transaction where a company decides to stop performing a given function internally by hiring a vendor to take over production of the input.1 The assets and human capital used to generate this good or service may move to the vendor, or they may simply be replaced with the vendor’s own resources. In other words, the decision to outsource replaces the centralized, and often undocumented, control that arises via corporate ownership with an explicit governance structure defined by the contractual arrangements between client and vendor. The strategic and economic factors driving an outsourcing transaction are varied—and quite similar to the considerations underlying an initial ‘make or buy’ production decision. Indeed, an outsourcing move should simply be viewed as the result of a firm changing its mind, for whatever reason, on this question and deciding to ‘buy’ what it used to ‘make’. Accordingly, the structure of business outsourcing transactions raises important issues 1 See Gilley and Rasheed (2000), defining outsourcing as ‘a discontinuation of internal production (whether it be production of goods or services) and an initiation of procurement from outside suppliers’. A distinction sometimes arises between domestic outsourcing (to a nearby firm) and offshore outsourcing (to a firm based in a more distant location). The definition of outsourcing is also complicated by hybrid sourcing strategies, such as when a firm elects to ‘cosource’ an activity by moving some production to a vendor while continuing to perform the same activity, albeit on a smaller scale, within the firm. See Parmigiani (2007).

78  George S. Geis

underlying both contract law and the theory of the firm. Why does a firm decide to adopt an outsourcing contract? How are these transactions structured? What do these arrangements tell us about the goals and motives of contract law in facilitating efficient exchange and production? This chapter examines the structure of business outsourcing transactions from a law and economics perspective. It begins by describing the typical contractual framework that is used for outsourcing. It then turns to a discussion of the links between explicit contractual terms and general theories about the role of contract law in supporting the exchange of scarce resources. Much of what we observe in outsourcing contracts can be explained by the conventional economic theory that underpins the normative goals of contract law. Yet, at the same time, additional explanation is needed for some of the features that can arise in outsourcing relationships. This chapter discusses these various insights and concludes with some closing observations about the economic purposes of outsourcing in a system of production and exchange.

The Structure of Business Outsourcing Contracts We are only starting to understand how business outsourcing transactions are conducted—that is, what governance terms and norms are adopted in the formal contracts and informal relationships of the parties.2 Nevertheless, many outsourcing projects seem to share a common legal framework where the relationship is defined through a complicated array of sequential, overlapping contracts. This plurality of contracts may simply reflect the transactions costs of negotiating and drafting detailed contract terms.3 Outsourcing relationships are known to take some twists and turns, and this uncertain path progression makes it difficult to spell out the entire scope of commitment upfront. Parties may be better off waiting until important contingencies play out before fully documenting their agreement. Further, as described in more detail below, postponing some contractual commitments until a relationship ripens may help to better align incentives. 2 See Blair and O’Hara (2011), analysing seven outsourcing transactions in the context of a push towards business process modularity; Gilson et al. (2009), analysing outsourcing transactions in the context of inter-organizational innovation; and Geis (2010). 3 See Posner (2005), describing a tiered cost model for contract drafting, and Schwartz and Scott (2003), discussing how all contracts are incomplete because it is costly to specify every potential contingency.



The Economics of Contract Law  79

In any event, parties will commonly draft four distinct contracts to establish an outsourcing relationship—although, importantly, some of these documents and terms may be bundled into a single agreement. The typical contracts include: (a) a confidentiality agreement; (b) a master agreement; (c) multiple statements of work; and (d) a service level agreement. Consider each agreement in turn. First, during the initial negotiations the parties will often sign a confidentiality agreement to protect the business information of both client and vendor. This is typically structured as a stand-alone contract, signed in advance of the other deal documents. This is true because it can take a long time to write an outsourcing contract and because the very process of negotiating the deal will often reveal sensitive information. Furthermore, an outsourcing client may bargain simultaneously with multiple vendors, and everyone will want proprietary information protected if a contemplated relationship fails to materialize. These confidentiality agreements do not differ significantly from those used in other business transactions, and they will not be discussed further here. The second deal document is called the master agreement and sometimes referred to as the framework agreement. This contract provides an overview of the anticipated relationship and outlines the business goals in broad strokes. Typically, it will also set a timeline for moving the project forward and provide a governance structure—including the establishment of joint decision-making committees and meetings between high-level and operational-level personnel. It may also include other important terms, such as a mechanism for resolving disputes, termination rights and post-termination cooperation obligations. But while the master agreement can be quite lengthy, it rarely defines the exact work that will be performed by the vendor. Accordingly, a third collection of contracts, the statements of work, are subsequently negotiated to flesh out the project details. These documents are numerous and often fairly short—sometimes just a few pages. They will include detailed work orders and project functionality requests. For example, a master agreement may stipulate that a firm’s human resources activities— including recruiting, training and other administrative duties—will be outsourced to a vendor specializing in this area. The follow-up statements of work will then specify the exact recruiting duties, training sessions and other responsibilities that the vendor will assume. The usual approach is to have mid-level managers from both parties, those closer to the details of the project, draft statements of work that follow the general guidelines and timeline set by the master agreement. Because the scope of an outsourcing project may change frequently over time, it is also common to have numerous amendments and modifications to the statements of work.

80  George S. Geis

Finally, a fourth contract, the service level agreement (SLA), is usually signed to govern the ongoing quality of project execution. The SLA provides concrete performance metrics that the outsourcing vendor must maintain during the term of the contract. It may also establish measurable penalties for failing to meet these performance requirements. For example, the SLA may define the number of calls to be managed each person-hour (in a call centre contract), measures of bandwidth to be provided (in an IT hosting contract) or other metrics for assessing suitable performance levels. In other words, while the statements of work govern what will be done, the SLA governs how well this work will be conducted. As with the other deal documents, the SLA is sometimes bundled into the master agreement. But over time, as the project’s scope changes with the statements of work, SLAs may evolve to support these changes. Furthermore, the parties will sometimes agree to renegotiate service levels as the vendor moves down a learning curve or as future technology impacts the business activity that will be conducted. A brief example helps illustrate how these multiple contracts fit together. Roughly a decade ago, the energy giant BP Amoco decided to outsource much of its human resources department to a firm called Exult.4 The SEC filings surrounding this transaction contain a particularly large amount of information.5 The master agreement provides a general description of the project, which includes ambitious, though nebulous, goals such as ‘automating relevant transactional processes and employee access through the implementation of web-enabled human resources support’, ‘consolidation and integration of human resources transactional processing support into client service centres’, and ‘rationalization and integration of third party service providers’.6 The master agreement also sets out a timeline for Exult to submit detailed plans to provide this human resources support and for both parties to conduct due diligence of these plans.7 It then goes into great detail on how Exult will be paid, how the project will be governed, how 4 For an excellent overview of human resources outsourcing and further background on the BP Amoco deal, see Adler (2002). Exult was subsequently acquired by Hewitt Associates in 2004. 5 See Framework Agreement between BP Amoco PLC and Exult, Inc., dated 7 December 1999, available at http://contracts.onecle.com/exult/bpamoco.svc.1999.12.07.shtml (hereafter ‘Framework Agreement [1999]’). Very similar master service agreements were also executed on the same day between Exult and subsidiaries of BP Amoco in the United States and the United Kingdom. See US Country Agreement between BP America Inc. and Exult, Inc., dated 7 December 1999, available at http://contracts.onecle.com/exult/bpamerica.svc.1999.12.07. shtml; UK Country Agreement between BP International Ltd. and Exult, Ltd., dated 7 December 1999, available at http://contracts.onecle.com/exult/bpintl.svc.1999.12.07.shtml. 6 Framework Agreement (1999: i). 7 Framework Agreement (1999), at §2.



The Economics of Contract Law  81

disputes will be resolved, which employees will be transferred to Exult and other general terms.8 But the master agreement is exceptionally vague on the exact activities that Exult will perform for BP Amoco. One appendix lists about 20 different services—including training, HR strategy, labour relations, managing employee records, payroll, recruiting, severance and so on—and provides a two or three paragraph description of how each activity will be divided between Exult and BP Amoco.9 These descriptions are remarkably ambiguous,10 and the true scope of the project will only come to light as the detailed statements of work and the SLAs are subsequently negotiated by project managers at BP Amoco and Exult. This general framework only scratches the surface of any arrangement, however, as the specific terms and provisions in each of these different contracts—along with the extra-contractual relationships that develop between parties—are what really matter from a governance perspective. How should we understand these features of an outsourcing transaction? What do they tell us about the goals of contract law in promoting economic exchange?

The Goals of Contract Law Broadly stated, the goal of contract law is to facilitate the voluntary transfer of resources to their highest value users with a minimum of transaction costs (Posner 2005). Said differently, it should seek to maximize the joint gains 8 Framework Agreement (1999), at §§ 2, 9 16, 25. The master agreement also contains confidentiality provisions, Framework Agreement (1999), at § 14. I do not know whether the parties signed a separate confidentiality agreement before negotiating this master agreement. 9 Framework Agreement (1999: 62), Schedule A. 10 For example, under the heading for the first service, training, the Framework Agreement (1999: 63–64) runs as follows:

Training as a process includes training needs assessment, course/materials development, logistics co-ordination, conduct of training and training leader selection, training effectiveness assessment and post training follow-up. Delivery of training materials includes traditional classroom, self-study, computer-aided training and third party training delivery mechanisms. BP [BP Amoco] shall develop training strategies and policies, develop and deliver training programs based on needs analyses and assess the cost/benefit of training programs. Exult shall administer course schedules, registration, confirmations and training materials. Exult shall also administer attendee evaluations of programs and tuition reimbursement.

82  George S. Geis

from trade (Schwartz and Scott 2003). Importantly, however, lawmakers need to understand that private parties will incur two distinct categories of transaction costs: (a) the upfront costs that arise with the negotiation and documentation of a relationship between parties and (b) the back-end costs of adjudicating a contractual dispute (through the court system or otherwise) if a breakdown occurs in the economic relationship. This means that a court should not require an upfront contract to fully document how every contingency in the relationship will be treated before bestowing the protection of law. Indeed, despite the formidable planning capabilities of some managers, absolute documentation of all paths that a relationship might take is futile. Parties will rationally choose to leave contracts incomplete. For this reason, contract law should simply insist that the parties define a set of key terms—thus evidencing a real intent to be bound—and draw upon courts to interpret the ambiguities in a contract that can subsequently become important. A secondary goal of contract law should be to provide parties with legal assurance that binding agreements will be honoured—or, in the alternative, that a breached-against party can recover adequate damages. This will allow each side to make outside investments, in reliance on a contract that further increase the economic surplus from trade (Schwartz and Scott 2003). For example, a seller, having formed a deal to provide a large quantity of unique goods, may cut costs and decrease the chances of breach by purchasing special machinery to make the goods cheaper. Or a buyer may similarly gain by investing in reliance on the contract—perhaps by building facilities next to the seller’s factory in order to reap inventory benefits or transportation cost savings. Yet both investments will only pay off if the contract is honoured; in economic terms, the investments are relation-specific (Cooter and Ulen 2004: 205–217).11 Without the ability to make binding commitments, these specialized investments will not occur because each side will expose itself to hold-up renegotiation demands by the other (Cooter and Ulen 2004: 205–210). Both parties will suffer a net welfare loss because they cannot invest to increase the gains from trade. In short, contract law should strive to accomplish the twin goals of efficient trade and efficient investment by providing a mechanism for parties

11 Using an example of buyer investment, Richard Craswell (1996: 490) offers this helpful definition of relation-specific investment: ‘It is any choice, be it action or inaction, which will (1) make [seller’s] performance more valuable to [buyer] if [seller] does in fact perform, but (2) make [buyer] worse off than if he had not relied if [seller] fails to perform.’



The Economics of Contract Law  83

to inexorably bind themselves. The exact legal rules for accomplishing these ends can be difficult to determine,12 but the high-level mandate of maximizing the gains from trade in this manner is compelling. This, however, only outlines the macroeconomic goals of contract law. Perhaps as importantly, the law should also strive at a micro level to understand and preserve (or at least not distort) the careful balance of economic incentives in a given transaction.13 Sophisticated parties will often craft customized terms and clauses to trade risks, balance incentives or accomplish other objectives. Overeager judicial tinkering can upset these carefully negotiated agreements ex post, thereby undermining the law’s ability to provide the needed legal certainty described above. Accordingly, there is often good reason to follow the textual prescriptions of the contract precisely. The bigger challenge arises when a court needs to interpret contractual ambiguities or adjudicate disputes that arise under a legal standard (as opposed to an explicit legal rule). To this end, it is helpful to look into the governance tensions underlying outsourcing contracts in order to better understand and preserve the likely aims of the parties.

Why Outsource? Relocating economic production is nothing new, and companies have continually sought to move business activity to areas with cheaper supply markets as transportation costs drop. It is possible, therefore, that an outsourcing decision is simply part of a never-ending quest for lower production costs. This cannot be the entire story however because firms have another option for taking advantage of less expensive inputs: establish a captive offshore facility that remains within the corporate fold. For example, a firm looking to obtain cheap microchips in China might outsource chip production to a Chinese firm or, alternatively, set up its own, wholly owned manufacturing facility in China. Why do some firms take the former course, thereby jettisoning the activity completely? The answer takes us back to trade-offs underlying the theory of the firm. It is important to recognize that in a world with perfect markets and costless contracting there would be no need to organize any economic activity within 12 13

See, for example, Posner (2003). See, for example, Goldberg (2006).

84  George S. Geis

a firm. Instead, firms would prefer to outsource everything by writing a series of detailed contracts on the open market for the inputs and activities needed to create new goods. This would allow a producer to tap into the decentralized and disciplined price system of the market. Essentially, an entrepreneur could simply stitch together a ‘virtual’ business venture by grabbing inputs and services through complete contracts covering every possible state of the world. We do not live in a world of perfect markets or costless contracting, of course, and one of Ronald Coase’s great insights was that performing some activity within a firm might reduce the transaction costs (broadly defined to include the ‘hold-up’ problem described earlier) of using external market pricing mechanisms (Coase 1937). Production costs are likely to be higher when business is conducted within the firm—because the activity is walled off from the relentless pricing pressure that comes with well-functioning markets.14 Yet these greater production costs might nevertheless be worth paying if a firm can save even more by avoiding other costs related to arm’slength transactions. The alternative approach, of course, is simply to vertically integrate complimentary assets into one legal organization. There is no need to worry about a contractual hold-up problem when everything rests under one roof. Firms can simply reserve decisions about how these assets will be deployed over time and retain the managerial hierarchy needed to eke out all future benefits from complementarity. More robust theories have been built upon this theoretical cornerstone—such as the ‘property rights’ theory of economic organization (Grossman and Hart 1986; Hart and Moore 1990), or frameworks exploring the use of specialized human capital within a firm (Rajan and Zingales 1998). Yet all of this work shares the common insight that aggregating production into one legal entity can protect against the hold-up problem inherent with relationspecific assets. This implies that the optimal location of economic activity is the result of a careful balance between the production cost savings from using contracts and the transaction cost savings (again, broadly defined) from using firms. In other words, we would expect a firm to optimize the cost of producing 14 In other words, a firm that does everything itself will probably pay more for most of its economic inputs—after all it is unlikely to be the lowest cost producer of everything. Nevertheless, the firm may still choose to keep control of many activities to guarantee a source of supply (thus protecting against a form of market failure) or to cut the transaction costs (broadly defined) of securing the input.



The Economics of Contract Law  85

each input against the transaction costs of establishing a market contract in order to decide exactly where to erect the firm’s borders. In a static world, this balance should hold, and the division of activity between firm and contract would be roughly constant. However, because we live in a dynamic world, both production costs and transaction costs for various activities can increase or decrease as new technologies or suppliers come online. Accordingly, the borders of a firm may continue to change as the underlying tension between production costs and transaction costs oscillates. This provides us, then, with at least two plausible explanations for a business outsourcing decision. The first possibility is simply that production costs have dropped somewhere outside the firm in a manner that is difficult for the firm to replicate within its corporate borders. For example, a vendor might develop a new process for making a needed input that enjoys trade secret or patent protection. Or, there may be some other production breakthrough that can only be realized with greater economies of scale or scope. These savings should lead to a recalculation of the trade-offs described earlier and possibly to an outsourcing decision—even if there has been no change to the transaction costs incurred by making such a move. The second plausible explanation for an outsourcing move is the exact opposite: even assuming that there has been no change to the production economics (or that cheaper inputs can be sourced equally well both inside and outside the firm), a firm may be interested in outsourcing if the transaction cost barriers to securing economic inputs via market contracts drop—perhaps due to standardization, technological change, regulation or some other factor. These reduced transaction costs (assuming, again, they disproportionately affect trade outside the firm versus the handoff of economic inputs inside the firm) would rebalance the economic trade-offs and push some firms to outsource activity beyond their corporate borders. Both explanations are plausible—indeed, they may operate simultaneously, in some contexts, as dual engines driving an outsourcing decision. The second explanation, however, relating to reductions in external transaction costs, is more interesting to contract law scholars. A significant portion of transaction costs arises during governance breakdowns, where a counter party fails to honour its promise or attempts to perform the agreement in a slipshod manner. In other words, one fundamental objective of an outsourcing contract is to invest in governance mechanisms that mitigate the risk that one side will take advantage of information asymmetries to shirk on a deal. Said differently, both parties to an outsourcing contract face an agency cost problem.

86  George S. Geis

The Agency Costs of Outsourcing The defining feature of an outsourcing transaction is that a firm contracts with another entity to take over activity that was previously produced inside the firm. Outsourcing deals thus generate agency risk under a familiar logic: the entity that controls a business activity does not ultimately ‘own’ the economic result (Jensen and Meckling 1976). Just like a corporate executive controls the property of shareholders, or like a borrower manages the money of a lender, an outsourcing vendor manages the business activity of an outsourcing firm. Because the vendor can be seen as an agent, it faces continual incentives to cut corners, take excessive risks or engage in other forms of self-dealing. For example, in a typical call centre outsourcing project, the vendor decides who to hire and how it will train these employees. Similarly, it decides when to replace aging capital with more efficient technology. Further, the vendor takes charge of quality control to ensure that employees are polite on the phones and adept at solving callers’ problems. But the outsourcing client takes the fallout from many of these choices. If the vendor hires rude callers who chase away loyal customers, then the client loses business. If the caller misses obvious sales opportunities, then the client foregoes the revenue. Of course, the outsourcing vendor might ultimately be accountable for shoddy work if the client decides not to renew a contract or if word gets out to other potential clients that this vendor shirks.15 But punishment will only be meted out if clients become aware that the vendor is engaged in selfish behaviour. Most outsourcing relationships are rife with asymmetrical information, and the client is unlikely to guard against—or even know about—every potential abuse. Unless resources are spent on monitoring, bonding or other contractual protections, business outsourcing can breed a den of distorted incentives. Lawmakers should therefore understand that many features of an outsourcing contract are designed to deal with this agency cost problem. In other words, the open-ended governance that directs most activity inside a corporation (where managers do not typically follow written protocols or service level agreements) is replaced by a set of explicit governance rules 15 The corporate agency literature considers similar questions by asking whether a CEO’s desire to renew her contract, or to take a CEO position at another firm, might serve as a reputational check on opportunistic behaviour.



The Economics of Contract Law  87

defined by contract.16 In short, it becomes possible to understand outsourcing contracts as a governance tool.

Outsourcing Contracts as a Governance Tool Firms may take many steps to mitigate the agency costs of outsourcing, ranging from explicit performance mandates in an SLA (which should be understood as a form of monitoring) to less formal structural protections. To illustrate the range of possibilities here, a partial list of contractual features that are sometimes used will be discussed: (a) Staged commitment through the interlocking, multi-contractual framework; (b) the use of redundant agents or the retention of duplicate activity within the firm; (c) incentive compatible compensation; (d) explicit monitoring and control rights; and (e) ‘for cause’ and ‘for convenience’ exit rights. Recall that outsourcing contracts rarely specify the exact work to be performed at the outset of a project, but rather rely on a series of sequential work orders to define the jobs. As stated above, one reason for this delay is simply that it can be difficult to know exactly what business activities are conducted throughout a firm—and which ones the vendor will assume—at the outset of a relationship. But, importantly, there is a second reason why a client may wish to proceed in this manner. Multiple, asynchronous contracts allow firms to stage their commitment, freeing them to reduce the scope of a project if hints of vendor opportunism arise. In this way, the use of sequential commitment parallels a technique used by venture capital firms to mitigate agency risk by staging their investment in target companies (Gilson 2003: 1078). Just as a VC investor commits slowly via multiple funding rounds—each round contingent upon the achievement of business milestones (Gilson 2003: 1078)—an outsourcing firm can stage its contractual commitment by delaying detailed specification of scope and performance requirements. The master agreement may set out the contemplated scope of a project, but the devil is always in the details. A principal detecting any sign of poor performance, or losing trust in the vendor, will usually have sufficient flexibility to scale back commitment when drafting additional statements of work. 16 Of course, even with outsourcing contracts the parties cannot specify every governance rule with precision and will rely on their reputation, long-term contracting incentives or some other means to work out unexpected or low-probability contingencies that arise.

88  George S. Geis

A second way to manage agency risk involves the use of multiple agents and co-sourcing—that is, keeping some of the outsourced activity within the firm. By dividing a project into pieces, and awarding each piece to different agents, principals can, in theory, introduce an element of competition that will help them monitor the agents’ performance. Further, the agents will often be aware of this scrutiny, and this knowledge may also prevent some misbehaviour. For example, a firm outsourcing the hosting and management of its networking infrastructure might divide the company in half and award part of the project to one vendor and the rest to another. This tactic gives the principal an automatic way to benchmark the performance of both agents: it can directly compare packet transmission rates, bit-loss frequency, service call response rates and other relevant metrics. This might help the principal identify hidden risks or uncover agency distortions. Related to this strategy of using multiple agents, a principal will sometimes benchmark performance by co-sourcing a project. Under this approach, the principal simply outsources part of the work, keeping a share of the same activity within the firm. Cosourcing also allows the principal to pace the performance of agents, and it may have added strategic benefits.17 Consider the approach taken the Indian telecom company, Bharti Airtel. As part of a fascinating recent transition, Bharti has outsourced much of its business activities—including technology infrastructure, IT services, billing, provisioning and more—to other firms.18 It also decided to outsource its customer service centres via a multiple agent strategy. After dividing operations into five regions, Bharti hired a different vendor to handle the inbound service calls of customers in each region (Aggarwal 2005).19 This strategy gives Bharti a basis for comparing the efficiency of each vendor. In some cases, Bharti also assigned two vendors to a geographic region—to counter the potential excuse that an agent’s poor performance should be blamed on local market conditions (Aggarwal 2005). To be sure, the use of multiple agents or co-sourcing may come with a price. A firm can lose economies of scale by splitting an outsourcing project See, for example, Kaka (2006), discussing Dell Computer’s co-sourcing strategy. Bharti’s strategy is particularly interesting, because many of its deals have involved ‘reverse offshore outsourcing’, the movement of business activity from an Indian principal to agents in the United States. See Buckman (2005: A1) and Marcelo and Taylor (2004: 1). 19 See also Final Transcript of Bharti Tele-Ventures Limited Earnings Conference Call, Thomson StreetEvents, 27 October 2005, discussing Bharti’s outsourcing strategy; available online at http://www.alacrastore.com/research/thomson-streetevents-Q2_2005_Bharti_Tele_ Ventures_Limited_Earnings_Conference_Call-T1143681. 17 18



The Economics of Contract Law  89

into several smaller chunks. Additionally, the principal will need to incur extra transaction costs to stitch together the outsourcing project. Instead of dealing with just one big partner, it must manage a collection of smaller relationships and divide project responsibilities among the vendors. Thus, like all other monitoring investments, there is a fundamental tension underlying the use of this strategy: adding more vendors increases competitive pressure and mitigates agency risk, but it also raises transaction and coordination costs. Furthermore, it may be impossible to break up some outsourcing projects into meaningful pieces. Some assignments may not have parallel processes—or may not be divisible and measurable in a sensible way. A third strategy for mitigating outsourcing agency risk is to set a compensation scheme that seeks to align vendors’ economic interests with those of the principal. Just as a corporation issues options to top managers to focus their efforts on boosting stock prices (Jensen and Murphy 1990: 261), an outsourcing principal might negotiate incentive compatible compensation that narrows the agency gap. While these tactics may help at the margins—and are thus worth understanding—it is important to note that they will never fully solve the agency problem. Anything short of transferring a complete ownership interest to the agent leaves room for economic distortions. Consider first the problems with conventional outsourcing fee arrangements. There are two extreme compensation paradigms: a time and materials contract and a fixed price contract. Time and materials compensation, where the agent simply adds a mark-up to the project’s input costs, presents obvious agency problems. The vendor has no incentive to tackle a project in a cost-effective manner because she will be paid for shirking or other inefficient behaviour. A fixed price contract, by contrast, transfers the pain of excessive input costs to the vendor. But this paradigm raises other incentives to perform shoddy work; the vendor might, for example, take excessive risks or use inferior materials and still get paid in full. Sometimes, an outsourcing principal will try a compromise approach—by imposing a time and materials contract with a maximum price cap—but this strategy just presents the concerns of both payment schemes. In response, some outsourcing contracts substitute incentive-compatible compensation terms. For example, they might award half of all cost savings below a specific target to agent vendors or impose other ‘earn-out’ requirements (Craig and Willmott 2005). Other contracts require annual negotiation of compensation (a strategy similar to staged commitment) or impose fines if service levels miss contractual requirements. Still other contracts pay vendors with the principal’s stock to directly align agent incentives—though this

90  George S. Geis

appears to happen infrequently.20 Other compensation arrangements might plausibly serve similar purposes; the key is simply to focus the attention of both parties on the same goals. At the end of the day, however, these strategies will never eliminate agency risk completely—they can only influence vendor behaviour at the margins. The fourth governance strategy uses explicit monitoring and control rights to mitigate the agency costs. For example, the contract may require vendors to participate in extensive business audits that allow a client—or his designated third party—to come on site and inspect financial records and operating procedures. In theory, these monitoring provisions should reduce an agent’s temptation to make self-interested decisions because the principal will have a better chance at uncovering the bad behaviour. Explicit control rights work in a similar, but ex ante, manner by giving a client power over decisions that are typically delegated to the vendor. The scope of these rights will differ from transaction to transaction, but a major outsourcing client may hold sway over important decisions such as employee hiring and training, equipment upgrades, managerial ratios, the selection of subcontractors and other operating activities. Sometimes this control comes through contractual carve-outs allowing principals to veto particularly important decisions rather than make direct decisions.21 The final governance strategy that is discussed involves exit rights: the legal power to terminate an outsourcing contract before the contemplated term expires. Exit rights raise an interesting tension because of two competing concerns. On the one hand, allowing an outsourcing principal to exit at will provides a check on vendor opportunism. If the principal detects poor quality work, excessive costs or any other problem, it can simply end the relationship. In this sense, liberal exit rights serve as the ultimate check on agency cost problems, and even the threat of early termination may keep vendors in line. On the other hand, an outsourcing vendor will often need to incur relationspecific investments to perform new work, and it may worry about writing an open-ended put option on the project. Easy exit rights thus raise another potential problem: outsourcing principals might use, or threaten to use, these exit rights opportunistically to extort a better deal ex post. 20 Equity alignment can also go the other way when a client purchases stock in the vendor. For example, in the BP Amoco—Exult deal described earlier, BP bought eight per cent of Exult’s stock as a ‘sign of good faith’ (Adler 2002:16). 21 This can be seen as analogous to veto provision in debt contracts. See Amihud et al. (1999: 453–456), discussing the agency costs that arise with the use of debt and various bond covenants and Fischel (1989: 133–140).



The Economics of Contract Law  91

Owing to these competing tensions, exit rights can be one of the most heavily negotiated provisions in an outsourcing contract. The parties may take great pains to define when and how a firm may walk away from the project—and to set the financial consequences of an early exit. The typical compromise involves dual exit rights: ‘termination for cause’ and ‘termination for convenience’. Under the ‘for cause’ track, the client is entitled to exit the relationship if the vendor does not live up to service level requirements or if other problems arise. Under the ‘for convenience’ track, the client is entitled to cancel the project without proof of bad service, but it may need to reimburse the vendor for upfront asset purchases or pay other, sometimes significant, financial penalties.22 Under both tracks, the vendor is usually required to support transition to another vendor, although the obligations may differ between the two types of exit. An outsourcing partnership between Priceline.com, the online travel firm, and Calltech, a call centre vendor, provides a nice example of a typical exit arrangement. In 1998, Priceline decided to outsource all of its inbound call centre work to Calltech.23 The deal would last for a year with automatic annual renewal unless notice was provided prior to the end of a term.24 Priceline then secured ‘for cause’ termination rights, allowing it to cancel the agreement if Calltech failed to meet performance obligations—as defined in the statements of work and the SLA—or if Calltech suffered financial problems.25 Priceline also negotiated ‘for convenience’ exit rights, allowing termination with just 90 days’ notice. However, if Priceline triggered these rights, it would incur an early termination fee designed to ‘compensate Calltech for all costs and expenses actually and reasonably incurred by Calltech for personnel and equipment engaged in providing services to Priceline’.26 The agreement goes on, however, to require Calltech to take good faith efforts to discharge this fee by redeploying these assets—and, interestingly, that the termination fee will be capped at the total bill charged

22 Without financial penalty, liberal exit rights of this sort might raise mutuality of obligation concerns in contract law—although a good faith requirement would most likely be read into the exit terms to skirt the problem. See Farnsworth (2004); Murray (2001); Perillo (2003). 23 For general background on this deal, see Pledger (2000: 1C). 24 Master Agreement for Outsourcing Call Center Support Between Priceline.com LLC and Calltech Communication Inc. (hereafter ‘Master Agreement’), available at http://contracts. onecle.com/priceline/calltech.svc.1998.shtml, at §5.1. 25 Master Agreement, at §5.2. The financial problems triggering ‘for cause’ termination include bankruptcy, liabilities in excess of assets and other indicators of financial distress. 26 Master Agreement, at § 5.3.

92  George S. Geis

by Calltech during the month prior to termination.27 The exit structure of this deal thus stages a complicated dance to mitigate dual opportunism: Priceline can threaten to leave if it thinks Calltech is behaving badly, it must, however, pay a termination fee to do so, and the termination fee is limited by agent redeployment requirements. Of course, the complexity of a large outsourcing project can sometime make it hard to determine whether termination is ‘for cause’ or ‘for convenience’. A client might unjustifiably claim for cause termination in order to avoid the financial penalties of a convenience exit. Conversely, a vendor may refuse to accept evidence that bad behaviour amounts to cause.28 Since language is ambiguous and context dependent, the parties will never be able to set upfront exit rights to govern every possible distortion. Although, the overall framework of dual track termination does make economic sense when it is viewed as a technique for mitigating agency risk. In short, firms have a large menu of contractual governance strategies. Only a few are mentioned in this chapter: detailed service level requirements, staged contractual commitment, redundant agents, structural control provisions and liberal exit rights. Many other provisions might also be employed towards the goal of aligning incentives. Collectively, these terms should be understood as establishing a contractual governance framework that seeks to replace the open-ended, discretionary governance of a firm with a more explicit and rules-based (though never fully complete) governance system grounded in contract law.

Concluding Observations We are only starting to understand why firms rethink a make-or-buy decision to shift economic activity outside their legal borders. Yet these business outsourcing transactions clearly have much to contribute to our 27 Master Agreement, at § 5.3. This clause illustrates, perhaps, Priceline’s clout in negotiating the deal. It is likely that the financial pain to Calltech from termination would exceed one month’s billings. 28 This problem surfaced several years ago in litigation between Sears and Computer Sciences Corp. (CSC), a large outsourcing vendor. When Sears sought to cancel the project for cause, CSC countered that the exit was really for convenience and that Sears was attempting to dodge large financial penalties. See Perone (2005: 24) and Weisman (2005: E3). Unusually, CSC also tried (unsuccessfully) to seal the appellate court record of this lawsuit as a trade secret. See Silwa (2005).



The Economics of Contract Law  93

understanding of contract law and production economics. It may very well be that these decisions are often explained by the theories of production cost savings or transaction cost economics discussed in this chapter. As business processes grow more modular—becoming easier to negotiate, monitor and replace—we may see increased outsourcing activity. Related to this, economic production may develop new cross-over points, where it becomes easier to move a partially finished end product towards the next stage of development. It is important to note, however, that these two accounts are not the only plausible explanations for a business outsourcing decision. Recent work has focused on the role of outsourcing in spurring innovation.29 For example, parties may learn significantly from working with another business entity—perhaps in a way that contributes to new economic breakthroughs that are unobtainable by either party in isolation. If this is true, then firms may wish to engage in outsourcing even when the production and transaction costs of doing so are identical to those incurred by performing the activity internally. Nevertheless, a careful examination of business outsourcing contracts reveals many insights related to the governance of economic production. Business outsourcing may have thrived in recent years not only because globalization has unlocked inexpensive production markets, but also because firms are finding it easier to monitor and prevent the agency costs of outsourcing. The interesting question here is what exactly is accomplished by moving from discretionary governance systems—that are the hallmark of firm production—to the more rules-based governance systems of individually tailored contracts.

References Adler, Paul S. 2002. ‘The Human Resources Business Process Outsourcing Industry: The BP-Exult Partnership’. Available online at http://ssrn.com/abstract=317502 (accessed on 22 October 2012). Aggarwal, Balaka Baruah. 2005. ‘Bharti’s Outsourcing Innovation’, Dataquest, 27 September. Available online at http://www.dqindia.com/content/industrymarket/2005/105092702. asp (accessed on 22 October 2012). Amihud, Yakov, Kenneth Garbade and Marcel Kahan. 1999. ‘A New Governance Structure for Corporate Bonds’, Stanford Law Review, 51: 447–492.

29

See, for example, Blair and O’Hara (2011) and Gilson et al. (2009).

94  George S. Geis Blair, Margaret M. and Erin O’Hara. 2011. ‘Outsourcing, Modularity, and the Theory of the Firm’, Brigham Young University Law Review, 263–314. Working Paper. Available online at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1443357. Buckman, Rebecca. 2005. ‘Outsourcing with a Twist: Indian Phone Giant Bharti Sends Jobs to Western Firms in a Multinational Role Switch’, Wall Street Journal, 18 January, New York. Coase, R. H. 1937. ‘The Nature of the Firm’, Economica, 4 (16): 389–405. Cooter, Robert and Thomas Ulen. 2004. Law and Economics. Boston: Addison-Wesley. Craig, David and Paul Willmott. 2005. ‘Outsourcing Grows Up’, McKinsey Quarterly, Special edition: 1–6. Craswell, Richard. 1996. ‘Offer, Acceptance, and Efficient Reliance’, Stanford Law Review, 48: 481–553. Farnsworth, E. Allan. 2004. Contracts. Austin, TX: Wolters Kluwer, Aspen Publishers. Fischel, Daniel R. 1989. ‘The Economics of Lender Liability’, Yale Law Journal, 99: 131– 154. Geis, George S. 2010. ‘An Empirical Examination of Business Outsourcing Transactions’, Virginia Law Review, 96: 241–300. Gilley, K. and A. Rasheed. 2000. ‘Making More by Doing Less: An Analysis of Outsourcing and its Effects on Firm Performance’, Journal of Management 26 (4): 763–790. Gilson, Ronald J. 2003. ‘Engineering a Venture Capital Market: Lessons From the American Experience’, Stanford Law Review, 55: 1067–1103. Gilson, Ronald J., Charles F. Sabel and Robert E. Scott. 2009. ‘Contracting for Innovation: Vertical Disintegration and Interfirm Collaboration’, Working Paper, Columbia Law Review, 109: 431–502. Goldberg, Victor. 2006. Framing Contract Law: An Economic Perspective. Cambridge, MA: Harvard University Press. Grossman, Sanford J. and Oliver D. Hart. 1986. ‘The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration’, Journal of Political Economy, 94 (4): 691–719. Hart, Oliver D. and John Moore. 1990. ‘Property Rights and the Nature of the Firm’, Journal of Political Economy, 98 (6): 1119–1158. Jensen, Michael C. and Kevin J. Murphy. 1990. Performance Pay and Top-Management Incentives’, Journal of Political Economy, 98: 225–264. Jensen, Michael C. and William H. Meckling. 1976. ‘Theory of the Firm: Managerial Behaviour, Agency Costs and Ownership Structure’, Journal of Financial Economics, 3 (2): 305–360. Kaka, Noshir F. 2006. ‘Running a Customer Service Center in India: An Interview with the Head of Operations for Dell India’, McKinsey Quarterly, May: 22–29. Available online at http://www.mckinseyquarterly.com/article_abstract.aspx?ar=1779&L2=13&L3=13 (accessed on 22 October 2012). Marcelo, Ray and Paul Taylor. 2004. ‘IBM Turns Tables on Indian Outsourcing’, Financial Times, 27 March, UK edition, London. Murray, John Edward. 2001. Murray on Contracts. Lexis Publishing. Parmigiani, Anne. 2007. ‘Why Do Firms Both Make and Buy? An Investigation of Concurrent Sourcing’, Strategic Management Journal, 28: 285–311. Perillo, Joseph M. 2003. Calamari and Perillo on Contracts. Saint Paul MN: West Group. Perone, Joseph R. 2005. ‘Possible Fort Closing and Lawsuit Dog Computer Sciences’, Star Ledger (Newark N. J.), 23 May. Pledger, Marcia. 2000. ‘Netting New Business Ohio Companies are Finding that Fulfillment, Delivery are Entry Points into the Hot World of E-Commerce’, Cleveland Plain Dealer, 17 February.



The Economics of Contract Law  95

Posner, Eric A. 2003. ‘Economic Analysis of Contract Law After Three Decades: Success or Failure?’, Yale Law Journal, 112: 829–880. Posner, Richard A. 2005. ‘The Law and Economics of Contract Interpretation’, Texas Law Review, 83: 1581–1614. Rajan, Raghuram G. and Luigi Zingales. 1998. ‘Power in a Theory of the Firm’, The Quarterly Journal of Economics, 113 (2): 387–432. Schwartz, Alan and Robert E. Scott. 2003. ‘Contract Theory and the Limits of Contract Law’, Yale Law Journal, 113: 541–619. Silwa, Carol. 2005. ‘Sidebar: CSC Tried, Failed to Seal Court Records on Appeal’, Computerworld, 23 May. Available online at http://www.computerworld.com/action/article.do?command =viewArticleBasic&articleId=101909 (accessed on 22 October 2012). Weisman, Robert. 2005. ‘Technology Outsourcing Comes Home’, Boston Globe, 29 May.

Section Three: Tort Law

6

Economic Analysis of Tort Law: Some Conceptual and Interpretative Issues Satish K. Jain

Introduction In the sub-discipline of law and economics, laws are analysed, using the economic method, from the perspective of economic efficiency. Economic method consists of systematically working out the actions that would be undertaken by purposive individuals, given the rules of the game and the outcomes that are arrived at in consequence of these actions. In the economic analysis of law, the outcomes that materialize through the interaction of rational individuals are almost invariably analysed from the perspective of economic efficiency. Thus there are two distinguishing features of economic analysis of law: it uses economic method to analyse law, and the analysis is done from the perspective of economic efficiency. These two features are conceptually distinct. On the basis of the results that have been obtained in the area of law and economics, it has generally been concluded that by and large common law is efficient.1 Some scholars in view of this have advanced the thesis that efficiency provides a unifying explanation for law as it exists. The purpose of this chapter is to take a close look at these claims in the context of tort law. For this purpose, in this chapter, we will be discussing in detail the results that have been obtained regarding the efficiency of liability rules2, some of 1 For main results of the law and economics literature, see, among others, Cooter and Ulen (2003) and Posner (2007). 2 On the efficiency of liability rules, see Brown (1973), Calabresi (1970), Coase (1960), Grady (1983, 1984, 1989), Jain (2006, 2010), Jain and Singh (2002), Landes and Posner (1987), Posner (1972) and Shavell (1987), among others.



Economic Analysis of Tort Law  97

which are among the earliest and most important results of the law and economics literature. The second section discusses the main results that have been obtained on the efficiency of liability rules. These results, establishing the efficiency of most of the liability rules used in practice, depend crucially on the way the notion of negligence has been formalized in the literature. In the law and economics literature an alternative formulation of the idea of negligence has also been discussed. The third section is concerned with the implications of this alternative notion of negligence for the efficiency of liability rules. The importance of this alternative notion of negligence arises from the fact that there are cogent reasons to believe that it is probably this notion of negligence that is generally used by courts for determining negligence or otherwise of parties. The fourth section looks at some strategic aspects of liability rules having a bearing on their efficiency. In the light of the discussion in the third and fourth sections, the concluding section of the chapter looks at some of the conceptual and interpretative issues that arise because of claims pertaining to the efficiency thesis of common law and because of according primacy to the normative criterion of economic efficiency.

Harmful Interactions and Tort Law Consider a simple example of a harmful interaction between two individuals A and B. Suppose A is undertaking an activity that results in a gain to him of 100 and causes harm of 10 to B. Suppose further that A can eliminate harm to B altogether by taking measures that will cost 1. We further assume that transaction costs of negotiating an agreement by A and B between themselves are prohibitively high. If A takes care then the social costs of interaction are 1; and if he does not take care then they are 10. Thus, from a social point of view the optimal course of action by A is to take care. Whether A will take care or not, however, depends on the liability law in force. If the law is such that A is not liable for the harm that his action is inflicting on B, then he will not take care; as his net gain if he takes care is 99 and if he does not then it is 100. On the other hand, if he is liable, then he will take care; as his net gain is 99 if he takes care and 90 if he does not. Thus, making the injurer (A) liable for the harm to the victim (B) results in obtaining the socially optimal outcome. The liability law under which the injurer is liable for the entire harm that he causes to the victim is called the rule of strict liability. Thus, we saw that the rule of strict liability in the context of our example gives rise

98  Satish K. Jain

to an efficient outcome, an outcome under which the total social costs of interaction are minimized. On the other hand, the rule of no liability, the rule under which injurers are not liable for harm that they cause to victims, does not lead to an efficient outcome in the case of our example. Before proceeding further with the discussion regarding the relationship between liability law and efficiency in the context of harmful interactions, the rationale for the assumption made above regarding transaction costs needs to be clarified. Suppose in the context of the example considered above, the transaction costs of negotiating an agreement between the injurer and the victim are zero. If the injurer is liable for the harm to the victim, then as we saw above he will take care and thus the efficient outcome will be obtained. When transaction costs are prohibitively high, if the injurer is not liable, then an inefficient outcome results. If the transaction costs are zero, then this conclusion does not hold. In our example expenditure of 1 on care by A can eliminate harm of 10 to B. Under the rule of no liability, the victim will be willing to offer any amount less than or equal to 10 in exchange for A’s agreement to take care. A will be willing to take care provided an amount greater than or equal to 1 is offered to him. Thus, given that the transaction costs are zero, an agreement will be reached between A and B under which B will pay an amount belonging to the closed interval [1,10] to A in exchange for his agreeing to take care for eliminating harm to B. Thus the outcome that is reached is efficient, and the same as that which is reached under strict liability. Thus, when transaction costs are zero, an efficient outcome is reached regardless of whether the liability law makes injurers pay for harm that they cause or not. The liability law, however, makes a difference if transaction costs are prohibitively high as we saw above. Therefore, from here onwards throughout this chapter we will be assuming that transaction costs are high enough to preclude private bargaining among parties involved in interactions.3 Let us call a rule efficient if it invariably gives rise to efficient outcomes. Although the rule of strict liability gave rise to an efficient outcome in the context of our example, it does not possess the property of invariably yielding efficient outcomes. Consider for instance the following, somewhat more complicated, case of interaction between two individuals A and B. Suppose, as before, undertaking of an activity by A results in a gain of 100 3 This is application of the Coase Theorem to the example. The Coase Theorem states that if transaction costs are zero, then the allocation of resources will be efficient regardless of liability assignments. In other words, if transaction costs are zero, then liability law is irrelevant from the perspective of efficiency. It is when transaction costs are non-zero that liability law becomes relevant for efficiency. See Coase (1960). Demsetz (1972) contains a lucid exposition of the Coase Theorem.



Economic Analysis of Tort Law  99

to him and harm of 10 to B. Furthermore, let there be two courses of action open to A; to take care costing 1 (d = 1) and to take no care costing nothing (d = 0). Similarly, let there be two options for B; to take care, which costs 1 (c = 1) and to take no care, which costs nothing (c = 0). Assume that if neither party takes care loss to B will be 10; if only one of the two parties takes care loss to B will be 5; and if both parties take care then loss to B will be 0. Schematically, we can represent the situation as in the following array, where entry in a cell is the magnitude of harm to B corresponding to the costs of care of that cell: d 0 0 10

1 5

1

0

c 5

It is immediate that the social costs are minimized when both A and B take care. If the liability law in force is the rule of strict liability, then is easy to see that this efficient outcome is not obtainable. First we note that given that by spending 1 on care the injurer can reduce harm to the victim by 5 regardless of what the victim does, and under strict liability he has to pay for the victim’s loss whatever it might be, he will take care. Now, if B does not take care then the harm will be 5, and he will receive liability payment of 5 from A. Thus net costs of B, net of liability payment, will be 0. On the other hand, if B does take care then there will be no loss and consequently no liability payment from A. His net costs will, however, be 1 as this is the amount he will be spending on care, and for which he will get no compensation. Clearly, being rational, he will not take care. Consequently, the outcome that will result will be inefficient. One of the most commonly used liability rules is the rule of negligence. Under the rule of negligence the injurer is liable for the entire harm to the victim iff he is negligent; and is not at all liable iff he is non-negligent. One way to define the notion of negligence is to specify a due care level; and consider conduct to be negligent if the care taken is less than the due care, and non-negligent if the care taken is greater than or equal to the due care. We now reconsider the example in the context of which we found strict liability to be inefficient by assuming that the liability law in force is the rule of negligence, and that the legally specified due care for the injurer is the care level that costs 1. If the injurer takes care, then regardless of what the victim does he will not be liable for harm to the victim; and therefore his costs will

100  Satish K. Jain

be 1, the cost of taking care. On the other hand, if he does not take care, then he will be liable for harm to the victim, which will be 5 in case the victim is taking care and 10 in case the victim is not taking care. Thus the injurer’s costs will be at least 5 in case he does not take care. Therefore, it follows that a rational injurer will take care regardless of what the victim does. Next, we consider the costs of adopting alternative courses of action by the victim. The victim, arguing as above, will come to the conclusion that the injurer will be taking care. Therefore, if the victim does not take care then his costs will be 5, equal to the harm that he will have to bear in view of the injurer being non-negligent; and if he takes care then his costs will be 1, equal to the cost of taking care as there will be no harm in view of both parties taking care. For the victim, taking care is clearly a superior alternative. Thus we see that under the negligence rule, in the context of our example, both parties will be taking care and thus the efficient outcome will result. As long as the due care for the injurer is specified appropriately from the perspective of minimization of total social costs, the outcome that will result under the negligence rule will be efficient, regardless of whether the efficient outcome involves care-taking by one party or both parties, that is to say, whether care is unilateral or bilateral. The property of the negligence rule of yielding efficient outcomes is shared by two variants of it, which are also widely used, namely, comparative negligence, and the rule of negligence with the defence of contributory negligence. Comparative negligence differs from the negligence rule only in one respect: under comparative negligence, when both parties are negligent, the accident loss is apportioned between the two parties according to some pre-specified provision. Under the negligence with the defence of contributory negligence, the injurer is liable for the entire harm if he is negligent and the victim is non-negligent; otherwise he is not at all liable. If the context is such that total social costs are minimized at a configuration of care levels where the victim’s care level is zero, then strict liability leads to an efficient outcome. However, when efficiency requires caretaking by both parties, strict liability will result in an inefficient outcome, as the victim will have no incentive to take care. If the defence of contributory negligence is added to strict liability, then it results in efficient outcomes in both unilateral and bilateral cases. Under strict liability with the defence of contributory negligence, the injurer is fully liable iff the victim is nonnegligent; and not all liable iff the victim is negligent.4 4 Efficiency of rules of negligence, negligence with the defence of contributory negligence, and strict liability with the defence of contributory negligence was first established by Brown (1973).



Economic Analysis of Tort Law  101

When the notion of negligence is defined as shortfall from due care, and due care is specified appropriately from the perspective of minimizing total social costs, then the following general result holds regarding efficiency of liability rules: a liability rule is efficient iff it satisfies the condition of negligence liability. The condition of negligence liability requires that: (a) If the victim is negligent and the injurer non-negligent, then the entire loss due to harm to the victim must be borne by the victim himself and (b) if the victim is non-negligent and the injurer negligent, then the entire loss due to harm to the victim must be borne by the injurer (Jain and Singh 2002). Efficiency of the rules of negligence, comparative negligence, negligence with the defence of contributory negligence, and strict liability with the defence of contributory negligence follows as a corollary of this general result as all these rules satisfy the condition of negligence liability. The rules of negligence, comparative negligence, negligence with the defence of contributory negligence, and strict liability with the defence of contributory negligence are some of the most commonly used rules. Thus, on the basis of the results that have been obtained regarding liability rules with respect to their efficiency characteristic, one can assert that much of tort law as it exists is efficient. Common law is regarded by and large efficient as similar results have been obtained in other branches of common law. It is on the basis of the efficiency of much of judge-made law that the thesis of economic efficiency providing a unified explanation for common law has emerged.

Efficiency of Liability Rules under an Alternative Notion of Negligence The notion of negligence on the basis of which the results on the efficiency characteristic of liability rules have been derived, as mentioned above, is defined as shortfall from a due care level chosen appropriately from the perspective of minimizing total social costs. This way of defining negligence can be questioned on the ground that courts in fact do not determine negligence or otherwise of parties in this way. Rather, courts regard conduct to be negligent iff it can be shown that the party in question could have taken some precaution that would have resulted in averting loss of an amount greater than the cost of precaution. In other words, a party’s conduct is negligent iff there exists a cost-justified untaken precaution.5 5 This viewpoint has been cogently and consistently articulated by Grady (1983, 1984, 1989).

102  Satish K. Jain

If negligence is defined in terms of existence of cost-justified untaken precautions, rather than as shortfall from due care, then the results on the efficiency of liability rules change radically. Consider for instance the negligence rule which invariably gives rise to efficient outcomes when negligence is defined as shortfall from due care and due care is chosen appropriately from the perspective of total social costs minimization. If the notion of negligence is defined in terms of existence of cost-justified untaken precautions, then it is no longer true that the negligence rule will invariably induce both the parties to take optimal levels of care. Consider for instance the following simple example: (a) The injurer has two courses of action open to him: taking care which will cost him 2, and not taking care which will cost 0. (b) Similarly, the victim has two courses of action open to him; taking care which will cost him 1, and not taking care which will cost 0. Let loss to the victim for various configurations of care costs be as shown in the following array: d 0 0 10

2 1

c 1

1

0.5

It is immediate that total social costs are minimized when the victim takes care (c = 1) and the injurer does not (d = 0). But when the notion of negligence is defined as existence of a cost-justified untaken precaution, this outcome is not obtainable. To see this, first we note that given the fact that there are only two courses of action open to the injurer, namely, to choose d = 0 or d = 2, it follows that when the injurer is choosing d = 2, the question of there existing a cost-justified untaken precaution does not arise. Consequently, when the injurer is taking care then he will be judged as non-negligent. Also, when the victim is taking care (c = 1) and the injurer is not taking care (d = 0), the injurer can bring about a reduction in harm to the victim of only 0.5 if he increases his care from d = 0 to d = 2. In this case, the cost of care being greater than the reduction in harm, it follows that given (c = 1 ∧ d = 0), d = 2 does not constitute a cost-justified untaken precaution; and consequently when (c = 1 ∧ d = 0) the injurer will be judged as non-negligent. When neither the victim nor the injurer is taking care, i.e., (c = 0 ∧ d = 0), if the injurer increases his care from d = 0 to d = 2 then a reduction in harm of an amount equal to 9 can be brought about; consequently at (c = 0 ∧ d = 0),



Economic Analysis of Tort Law  103

d = 2 constitutes a cost-justified untaken precaution. Thus at (c = 0 ∧ d = 0), the injurer will be judged negligent. Therefore, it follows that at various configurations of care costs (c, d ), the parties’ costs will be as shown in the following array, where the entry in each cell is (victim’s costs, injurer’s costs) corresponding to the care costs of that cell:

d 0 0 (0,10)

2 ( 1 , 2)

1

(1.5,2)

c (2, 0)

Thus the only Nash equilibrium is (c = 0, d = 2), which is not total social costs minimizing. The above example shows that when the notion of negligence is defined in terms of existence of cost-justified untaken precautions, the negligence rule does not possess the property of invariably yielding efficient outcomes. In fact, it is not merely the negligence rule that lacks the property of always resulting in an efficient outcome when the notion of negligence is defined as existence of a cost-justified untaken precaution. It can be shown that with this alternative notion of negligence there is no liability rule that possesses the property of always yielding an efficient outcome.6

Strategic Aspects There is one rather neglected aspect of negligence determination that needs to be discussed. The basic rationale behind negligence determination, whether negligence is defined as shortfall from due care appropriately chosen from the perspective of total social costs minimization, or as existence of a cost-justified untaken precaution, is that persons who are better at bringing about reduction of harm should be induced to make greater efforts for that purpose compared to those with less competence for harm reduction. But, For a formal statement and proof of the theorem that there is no liability rule that is efficient when negligence is defined in terms of existence of cost-justified untaken precautions, see Jain (2006). 6

104  Satish K. Jain

this means that those with greater diligence and industriousness may end up with a greater burden. Consider, for instance, the following application of the negligence rule with negligence defined as shortfall from due care, and due care being chosen appropriately for minimizing total social costs. Let both the victim and the injurer have three alternative courses of action open to them: To take no care; to take moderate care; to take high care. Let the quantum of harm for various configurations of care levels be as given in the following array:

Victim’s Care Level

No Care Moderate Care High Care

No Care 20 15 13

Injurer’s Care Level Moderate Care 15 10 8

High Care 13 8 6

If we assume that to each of the two parties, no care costs 0, moderate care costs 2, and high care costs 5, then for various configurations of care levels total social costs will be as given in the following array:

Victim’s Care Level

No Care Moderate Care High Care

No Care 20 17 18

Injurer’s Care Level Moderate Care 17 14 15

High Care 18 15 16

Thus, total social costs are minimized when both parties take moderate care. If the due care for the injurer is set at the moderate care, then under the negligence rule the efficient outcome of each party taking moderate care will be obtained. Under this efficient outcome, the injurer’s costs will be 2 and the victim’s costs will be 12. Now suppose the injurer, by expending some resources, can bring about reduction in the costs of care. Suppose after the investment, his costs of taking no care are 0, moderate care 1, and high care 2.5. Assume that the costs of taking various levels of care by the victim are unchanged. Then, total social costs at various configurations of care levels will be as given in the following array:



Economic Analysis of Tort Law  105

Victim’s Care Level

No Care Moderate Care High Care

No Care 20 17 18

Injurer’s Care Level Moderate Care 16 13 14

High Care 15.5 12.5 13.5

Thus, total social costs are uniquely minimized when the victim takes moderate care and the injurer takes high care. If the due care for the injurer is set at a high level, then the negligence rule will give rise to this efficient outcome. Under the efficient outcome, the injurer’s costs are 2.5 and the victim’s costs 10. Comparing these costs with the pre-improvement situation we find that the injurer’s costs have gone up by 0.5 and the victim’s costs have come down by 2. If the injurer makes the investment which results in decrease of costs of taking care by him, then total social costs come down by 1.5. Therefore, from a social point of view, as long as the cost of investment is less than 1.5, it is socially worthwhile. However, it is clear that such an investment will not be made even if it is costless, because the injurer, instead of getting rewarded for superior performance, ends up paying more than before. When negligence is defined as shortfall from due care, and due care is chosen appropriately from the perspective of minimizing total social costs, in general it does not pay to be dextrous. On the contrary, if one is dextrous, then advantages can flow from suppressing information regarding one’s dexterity. Similar perverse incentives are present if the notion of negligence that is used is based on the existence of cost-justified untaken precautions. We illustrate it with the same example that we discussed just now. First we consider the pre-improvement situation. If the injurer is taking high care then there being no greater care the question of there being a costjustified untaken precaution does not arise; and consequently the injurer will be judged non-negligent. If the injurer is taking moderate care, then by increasing care to the level of high care a reduction in harm of an amount 2 can be brought about regardless of what the victim is doing. However, the increase in cost if the injurer shifts from moderate care to high care will be 3. Consequently, when the injurer is taking moderate care there does not exist any cost-justified untaken precaution; consequently the injurer will be judged non-negligent when his care level is moderate. If the injurer is taking no care, then he can bring about a reduction in harm of 5 if he increases his care to the moderate level at a cost of 2 regardless of what the victim is doing. Therefore, the injurer will be judged negligent if he takes no

106  Satish K. Jain

care. Thus, the situation with regard to negligence or non-negligence of the injurer is identical to what it was when the negligence was defined as shortfall from due care, and due care was chosen to be the moderate level of care. Thus, there will be no change in configurations of costs of the two parties and therefore the Nash equilibrium will continue to be the configuration of moderate care by each party. Next we consider the post-improvement situation. If the injurer is taking high care then he will be judged non-negligent, as there is no care level higher than that. If the injurer is taking moderate level of care, then he can bring about a reduction in harm of 2 by increasing his care to the high care level at an additional cost of 1.5 regardless of what the victim does; consequently, if he is taking moderate care he will be judged to be negligent. If the injurer increases care from no care level to moderate level, he can bring about a reduction in harm of 5 at a cost of 1 regardless of what the victim does; and therefore will be judged to be negligent when he is taking no care. In the post-improvement situation as well we find that the injurer will be judged negligent precisely in those cases where he will be judged to be negligent when negligence is shortfall from due care and due care is set at the level of high care. Thus, in the post-improvement situation all configurations of costs of the two parties will be the same as before and therefore the Nash equilibrium will remain the configuration of moderate care by the victim and the high care by the injurer. In view of the fact that the Nash equilibria in this case are identical to what they were when negligence was defined as shortfall from due care and due care was appropriately chosen from the efficiency perspective, all the remarks that were made regarding the perverse incentives earlier remain applicable.

Efficiency of Liability Rules: Conceptual and Interpretative Issues As mentioned earlier, one reason why tort law is considered by and large efficient is because of the results establishing efficiency of most of the liability rules used in practice. As we saw these results crucially depend on the way the idea of negligence is formalized. These results hold if negligence is defined as shortfall from due care, and due care is chosen from the perspective of minimizing total social costs. But these results do



Economic Analysis of Tort Law  107

not hold if negligence is defined as failure to take a cost-justified precaution in view of the fact that with this notion of negligence there is no liability rule that is efficient. If negligence is to be construed as shortfall from due care, with due care being chosen from the efficiency perspective, then the courts will have to acquire relevant information to enable them to determine the appropriate level of due care. The idea of courts acquiring information with a view to determine the due care level appropriately, however, is not in harmony with common law being an adversarial legal system. In an adversarial system, the information that the court has is what the parties to the dispute find in their interest to present before the court. As a rule, the court on its own does not acquire information. Now, it is certainly possible that in specific instances the parties to the dispute behaving rationally may end up supplying enough information to the court to enable it to calculate the appropriate level of due care. But there is no reason why such should be the case generally, in view of the fact that calculating the due care level correctly in most cases will require rather detailed information on cost and loss functions. In the absence of sufficient information it is unlikely that the courts will be able to specify the due care level correctly. Determination of negligence on the basis of failure to take a cost-justified precaution is more in tune with the adversarial nature of common law. If the court is going to adjudge the negligence or otherwise of the conduct on the basis of failure to take a cost-justified precaution, then it is in the plaintiff’s interest to show that there exists a cost-justified untaken precaution and in the defendant’s interest to make a refutation of such a showing. The court has to merely decide on the reasonableness of these claims and counter-claims in order to make a determination of negligence, instead of requiring and processing of a large amount of information. According to Grady (1983, 1984, 1989), reading of a number of important tort cases is more in tune with negligence as failure to take a cost-justified untaken precaution rather than with negligence as failure to take at least the due care. In any case, as we saw in the previous section, whether negligence is defined as failure to take a cost-justified untaken precaution or failure to take at least the appropriately chosen due care, it tends to reward ineptitude and penalize skill. Consequently, perverse incentives are generated which discourage skillformation with consequent implications for efficiency. When negligence is defined as failure to take a cost-justified untaken precaution there are two sources of inefficiency: possibility of there existing a cost-justified untaken precaution for a party who is taking socially optimal level of care; and perverse

108  Satish K. Jain

disincentives on account of this way of defining negligence for socially useful skill-formation. If negligence is defined as shortfall from appropriately chosen due care then there is only the latter source of inefficiency. It is thus clear that one can assert that tort law as it exists is by and large efficient only by abstracting from important sources of built-in inefficiency. The thesis of common law being efficient does not seem to hold, at least for the tort law part of it, in as unambiguous terms as has been made out by the proponents of the thesis. In this connection, there is another important factor which needs to be pointed out. Suppose we ignore the factors that seem to cast doubt on the assertion of most of the liability rules being efficient; and accept the standard framework under which most of the liability rules used in practice are efficient. As mentioned earlier, under the standard framework efficient liability rules are characterized by the condition of negligence liability. There is an infinite number of liability rules satisfying the condition of negligence liability; hence an infinite number of efficient liability rules. Consequently, even if every rule used in practice were efficient, efficiency alone would not constitute an explanation for tort law. If there had been a single liability rule that was efficient and this rule alone was used, one would have been entirely justified in claiming that efficiency constituted a complete explanation of tort law. However, when all rules which are used in practice are efficient, but together they constitute only a proper subset of all efficient liability rules, one also has to explain why some efficient rules are used and others are not. Suppose for instance that all liability rules which are used in practice are efficient, all liability rules which are used in practice satisfy a particular condition other than efficiency, and every efficient liability rule which is not used in practice violates this condition, then the correct position is that efficiency and this condition together constitute an explanation for tort law. Thus, given that only a proper subset of efficient liability rules are used in practice, at most what can be asserted is that efficiency constitutes a partial explanation for tort law. In addition to providing an explanation for existing law, one may also be interested in designing laws in such a way that they satisfy those normative criteria which are considered important in the institutional context of law. The law and economics literature is not only concerned with examining efficiency or otherwise of existing laws, but is also concerned with designing rules that will lead to efficient outcomes. The implications of exclusive preoccupation with efficiency are rarely, if ever, discussed. It is not always clear as to whether those who are engaged in the enterprise of designing efficient rules, regulations and procedures regard efficiency



Economic Analysis of Tort Law  109

as the pre-eminent value, a value to be protected against every other value in cases of conflict; or consider efficiency a benign value which is nonconflictive with other values that might be considered important in the context of legal institutions; or consider all values which are important in the context of law like justice as derivatives of economic efficiency. Although, some prominent law and economics scholars seem to hold the position mentioned last in the preceding sentence, most people will find it difficult to accept the position that justice is nothing but efficiency or some consequence thereof.7 Once the position of values like justice being derivatives of economic efficiency is rejected, and the position that efficiency on the one hand and values like justice on the other are independent values is accepted, the thesis of efficiency being non-conflictive with values like justice becomes immediately suspect. Given the fact that it is difficult to think of even two similar but independent values being such that they will never conflict, the case of efficiency and justice being non-conflictive does not seem to lie within the realm of possibility. Thus, it seems that when one is engaged in the enterprise of designing efficient rules, regulations and procedures, one must be according to economic efficiency the preeminent position, at least implicitly. But if laws are such that they satisfy the normative criterion of efficiency, then it must be the case that at least in those instances when efficiency and justice conflict they could not be serving the cause of justice. To conclude, in this chapter the efficiency thesis of common law has been questioned regarding one component of common law, namely, tort law. It has been argued that a close scrutiny of the claims regarding efficiency of tort law puts a question mark on them. Also it was argued that, notwithstanding the correctness or otherwise of these claims, to say that efficiency constitutes a unified explanation for common law is not justified. Finally, the acceptance of the thesis of common law being efficient has the implication that in cases of conflict between efficiency and justice, the common law could not be serving the ends of justice.

7

Foremost among these scholars is Richard Posner. According to him (1981):

A second meaning of ‘justice’, and the most common I would argue, is simply ‘efficiency’. When we describe as ‘unjust’ convicting a person without a trial, taking property without just compensation, or failing to require a negligent automobile driver to answer in damages to the victim of his carelessness, we can be interpreted as meaning simply that the conduct or practice in question wastes resources. It is no surprise that in a world of scarce resources, waste is regarded as immoral.

110  Satish K. Jain

References Brown, J. P. 1973. ‘Toward an Economic Theory of Liability’, Journal of Legal Studies, 2 (2): 323–350. Calabresi, G. 1970. The Costs of Accidents. New Haven: Yale University Press. Coase, R. H. 1960. ‘The Problem of Social Cost’, The Journal of Law and Economics, 3 (October): 1–44. Cooter, R. D. and T. S. Ulen. 2003. Law and Economics, 4th edition., New York: AddisonWesley. Demsetz, Harold. 1972. ‘When Does the Rule of Liability Matter?’, Journal of Legal Studies, 1 (1): 13–28. Grady, M. F. 1983. ‘A New Positive Theory of Negligence’, Yale Law Journal, 92 (5): 799–829. ———. 1984. ‘Proximate Cause and the Law of Negligence’, Iowa Law Review, 69 (2): 363–449. ———. 1989. ‘Untaken Precautions’, Journal of Legal Studies, 18 (1): 139–156. Jain, S. K. 2006. ‘Efficiency of Liability Rules: A Reconsideration’, The Journal of International Trade and Economic Development, 15 (3): 359–373. ———. 2010. ‘Negligence Rule: Some Strategic Aspects’, in P. G. Babu, Thomas Eger, A. V. Raja, Hans-Bernd Schafer and T. S. Somashekar (eds), Economic Analysis of Law in India: Theory and Application. New Delhi: Oxford University Press. Jain, S. K. and R. Singh. 2002. ‘Efficient Liability Rules: Complete Characterization’, Journal of Economics (Zeitschrift für Nationalökonomie), 75 (2): 105–124. Landes, W. M. and R. A. Posner. 1987. The Economic Structure of Tort Law. Cambridge, MA: Harvard University Press. Posner, R. A. 1972. ‘A Theory of Negligence’, Journal of Legal Studies, 1 (1): 28–96. ———. 1981. The Economics of Justice, Cambridge, MA. Harvard University Press. ———. 2007. Economic Analysis of Law. New York: Wolters Kluwer Law & Business. Shavell, S. 1987. Economic Analysis of Accident Law. Cambridge, MA: Harvard University Press.

7

Efficiency of Liability Rules: An Experimental Analysis in India Sanmitra Ghosh Rajendra P. Kundu

Introduction This chapter presents the results of an experiment that was conducted in India to analyse the question of efficiency of liability rules in the context of interactions between two parties in which only the injurers, by taking care, can affect the expected loss from accident. Analysis of the data obtained from the experiment shows that the behaviour of the subjects was not in conformity with the predictions of the theoretical model. There is extensive literature on the efficiency of liability rules. Pioneering contributions came from Calabresi (1961, 1965, 1970), who dealt with the effect of liability rules on the behaviour of parties. Posner (1972) analysed the efficiency of the rule of negligence. A formal analysis of some of the most important liability rules was first put forward by Brown (1973). He demonstrated the efficiency of both the rule of negligence and the rule of strict liability with the defence of contributory negligence in the context of interactions between two parties in which both parties, by taking care, can affect the expected loss of accident. Efficiency of the negligence rule and the rule of strict liability in the context of unilateral care models was demonstrated by Shavell (1980). Systematic and detailed treatment of liability rules is contained in Shavell (1987), Landes and Posner (1987) and Miceli (1997). Jain and Singh (2002) obtained a complete characterization of efficient liability rules in the context of interactions involving bilateral care.

112  Sanmitra Ghosh and Rajendra P. Kundu

Experimental verification of claims made in the law and economics literature about the efficiency properties of liability rules was first done by Kornhauser and Schotter (1990).1 They tested the relative efficiencies of the rules of strict liability and that of negligence in the context of unilateral care models. They found that under the negligence rule the subjects’ choice of care levels corresponded to the predictions of the theory. However the predicted behavioural equivalence of strict liability and negligence rules, when the care standard for the latter rule was set at the social cost minimizing level, was not borne out by data. The authors concluded that the rule of negligence dominates the strict liability rule in terms of accident costs and also performs robustly in cases where the care standard is set far above the socially optimal level. We focus on the rules of no liability and strict liability in the context of two-party interactions in which only the injurer’s choice of care affects the expected loss from the accident. The theory predicts that under the no liability rule the injurer will not take any care and under the rule of strict liability the injurer will always take the efficient level of care. Analysis of data obtained from the experiment leads to rejection of the theoretical predictions mentioned above. We also investigate whether differences in the levels of wealth of the two parties lead to any change in the behaviour of the injurer. The data does not reveal any systematic pattern in behavioural change arising out of asymmetric wealth levels. The injurer’s choice of care level, however, seems to be greatly affected by subjective information about the nature of the interaction. The unilateral decision problem described here in the context of liability rules is structurally similar to that of dictator games. There is extensive experimental literature on such games, which demonstrates that theoretical predictions do not hold good in the laboratory. In fact, Roth (1995) found that in laboratory implementation of dictator games, egalitarian distribution was the second most frequent outcome. Forsythe et al. (1994) argued that ‘fairness’, as an utilitarian ethic per se could not explain this generosity. Hoffman et al. (1994, 1996) hypothesized that this generosity arose from unconscious expectations of reciprocity. They attributed other-regarding behaviour to a minimal social distance between the proposer and the receiver. They defined social distance as the degree of reciprocity that people believe is inherent within a social interaction. Conferring property rights or entitlements to the proposer (by allowing the proposer, for example, to 1 Till date, this is the only published paper that has applied experimental methods to study the efficiency properties of liability rules.



Efficiency of Liability Rules  113

‘earn’ her role as a proposer) or increasing her social isolation (by ensuring anonymity, for instance) helps enhance the social distance and lead the proposers systematically towards self-regarding behaviour. In the context of the extant designs on dictator games, our design can be situated near the lower end of the social distance spectrum.2 The rest of the chapter is organized as follows. The second section presents a general model of the liability rules with unilateral care by the injurer. The third section contains some illustrative examples and states the theoretical predications. In the fourth section we present the experiment in detail. Finally, the fifth section discusses the significance of the experimental results and concludes.

The Model3 We consider interactions between two parties who are strangers to each other. The interaction can result in an accident in which, to begin with, the loss can fall on any one of the parties, called the victim (plaintiff). The other party is referred to as the injurer (defendant). We denote by a ≥ 0 the index of the level of care taken by the injurer. Let A = {a|a ≥ 0, where a is the index of some feasible level of care taken by the injurer}. We assume: 0 ∈ A.

(A1)

We denote by c(a) the cost to the injurer of care level a. Let C = {c(a)|a ∈ A}. We assume: c(0) = 0.

(A2)

We also assume that c is a strictly increasing function.

(A3)

In view of (A2) and (A3) it follows that: (∀c ∈ C )(c ≥ 0). A consequence of (A3) is that c can be regarded as the index of care level adopted by the injurer.

2 We assigned the roles of victims and injurers randomly, and except for the last three treatments we did not disseminate subjective information about the roles or the nature of the interaction. We maintained anonymity conditions similar to the single blind treatments of Hoffman et al. (1996). Details are in the section titled ‘The Experiment’. 3 The model is based on Jain and Singh (2002).

114  Sanmitra Ghosh and Rajendra P. Kundu

Let p denote the probability of occurrence of an accident and H denote the harm in case an accident occurs. p and H will be assumed to be functions of c. Let L = pH. L is thus the expected loss from accident. We assume: (∀c, c' ∈ C)[c > c' → L(c) ≤ L(c' )]

(A4)

In other words, it is assumed that greater care by the injurer results in lesser or equal expected accident loss. Total social costs (TSC ) are defined as the sum of cost of care by the injurer and the expected loss due to accident; TSC = c + L(c). Let M = {c' ∈ C | (∀c ∈ C ) (c' + L(c' ) ≤ c + L(c))}. Thus M is the set of all cost of care c which are total social cost minimizing. We further assume that C and L are such that M is non-empty.

(A5)

Let I denote the closed unit interval [0,1].4 Given C, p, H and c*∈ M, we define function p: C → I as follows:    1   if  c ≥ c* p(c) =    c/c*  if  c < c*; p would be interpreted as the proportion of non-negligence of the injurer. (1 – p) consequently would denote the proportion of negligence of the injurer. In case there is a legally binding due care level for the defendant, it would be taken to be identical with c* figuring in the definition of function p. A liability rule is a rule that specifies the proportion in which the two parties are to bear the loss in case of occurrence of an accident, as a function of the proportion of the injurer’s non-negligence. Formally, a liability rule is a function f : I → I 2, such that: f (p) = (x, y) and x + y = 1. Let C, p, H and c*∈ M be given. If an accident takes place and harm of H(c) materializes, then xH(c) will be borne by the victim and yH(c) by the injurer. Since, to begin with, in case of occurrence of an accident, the entire loss falls upon the victim, yH(c) represents the liability payment by the injurer to the victim. The expected costs of the victim and the injurer Throughout the chapter we use the following standard notation to denote by (0,1) the set {x|0 ≤ x ≤ 1}, by (0,1) the set {x|0 ≤ x < 1}, by (0,1) the set {x|0 < x ≤ 1}, and by (0,1) the set {x|0 < x < 1}. 4



Efficiency of Liability Rules  115

are xL(c) and c + yL(c) respectively. Both parties are assumed to prefer smaller expected costs to larger expected costs and be indifferent between alternatives with equal expected costs. This chapter is concerned with the rules of no liability and strict liability. The rule of no liability is defined as [∀p∈[0,1]][f (p) = (1,0)] whereas the rule of strict liability is defined as [∀p∈[0,1]][f (p) = (0,1)]. Thus, under no liability the expected costs of the victim and the injurer are L(c) and c respectively and under strict liability the expected costs of the victim and the injurer are 0 and c + L(c) respectively. Let f be a liability rule. An application of f consists of specifications of C, p, H and c*∈ M. A liability rule is efficient for a given application if and only if every care level that minimizes the expected cost of the injurer is total social cost minimizing, and there exists at least one care level which minimizes the expected cost of the injurer. A liability rule is efficient if and only if it is efficient for every application. A liability rule f satisfies the condition of negligence liability (NL) if and only if [∀p∈(0,1)][f (p) = (0,1)]. In other words, a liability rule satisfies the condition of negligence liability if and only if its structure is such that whenever the injurer is negligent the entire loss in case of an accident is borne by the injurer. Note that while the rule of strict liability satisfies the above condition the rule of no liability does not. Theorem 1: f is efficient for every application satisfying (A1) to (A5) if and only if it satisfies the condition of negligence liability. Proof: See Appendix A. Corollary: The rule of strict liability is efficient and the rule of no liability is inefficient.

Illustrative Examples In this section, we illustrate the generalized model of the previous section with some numeric examples and generate numeric predictions for the same. The experimental design, discussed in the next section, is based on these examples.5 Thus, the examples serve two purposes. First, they help the reader grasp the theoretical set-up underlying the efficiency analysis of tort laws. Second, they provide us with the parameter configuration for the 5

For details, see ‘Instructions’ in Appendix B.

116  Sanmitra Ghosh and Rajendra P. Kundu

experimental design and offer precise numeric predictions, which can be put to test in the laboratory.

Examples Example 1 (Application 1): Consider an application of the rule of no liability in which there are five possible levels of care for the injurer and the loss from accident decreases moderately as care level increases, as shown in Table 7.1. Table 7.1. Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Let p = 1. Note that the total social costs are minimized at medium care level. Note also that the expected costs of the injurer are minimized at no care. Thus, the no liability rule is inefficient for the given application. Example 2 (Application 1): Consider the same example, but now, as an application of the rule of strict liability, as shown in Table 7.2. Table 7.2. Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Let p = 1. Note that the total social costs are minimized at medium care level. Note also that the expected costs of the injurer are minimized at medium care. Thus, the strict liability rule is efficient for the given application. Example 3 (Application 2): Consider another example in which the care is relatively less effective in reducing the loss from the accident, as shown in Table 7.3. Table 7.3. Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8



Efficiency of Liability Rules  117

Let p = 1. Note that the total social costs are minimized at no care level. Note also that the expected costs of the injurer are minimized at no care under both rules. Thus, both rules are efficient for the given application. Example 4 (Application 3): Now consider the following example in which care is highly effective in reducing the loss, as shown in Table 7.4. Table 7.4. Care Levels Cost of care Loss

No Care

Low Care

Medium Care

High Care

Very High Care

0

1

2

3

4

10

6

1

0.5

0

Let p = 1. Note that the total social costs are minimized at medium care level. Under the no liability rule the expected costs of the injurer are minimized at no care. Under the strict liability the expected costs of the injurer are minimized at medium care. Thus, the rule of no liability is inefficient and the strict liability rule is efficient for the given application.

Predictions of the Model According to the theoretical model, a rational injurer should choose: 1. No care under the no liability rule. 2. Efficient (medium care for Applications 1 and 3 and no care for Application 2) care under the strict liability rule.

The Experiment Hypotheses 1. Under the no liability rule, an injurer will always choose no care. 2. Other things remaining unchanged, a change in the effectiveness of care does not affect the care level chosen by the injurer under the rule of no liability. 3. Relative differences in the wealth of the injurer and the victim do not affect the care level chosen by the injurer under the rule of no liability.

118  Sanmitra Ghosh and Rajendra P. Kundu

4. Subjective information does not affect the care level chosen by the injurer. 5. An injurer will always choose efficient care under the rule of strict liability.

Design In each of the experimental sessions we recruited even numbers of subjects who were randomly split into two groups of equal sizes. Subjects in the first group were assigned the role of injurers whereas subjects in the other group were given the role of victims. However, except for ‘subjective information’ treatments, the words ‘injurer’ or ‘victim’ were not mentioned explicitly.6 Next, each subject in the first group was matched randomly with a subject in the other group. However, the identities of these matches were not disclosed to the subjects. In each session we played each treatment twice. In the first, subjects in the first group acted as injurers and in the second, they acted as victims and vice versa. Note however that before swapping their roles we randomly re-matched the subjects of the two groups so as to rule out the possibility of retaliation. In order to simulate the environment of interaction between strangers, as warranted by the theory, we allowed injurers to play each treatment only once so as to avoid the effect of learning on the subject’s behaviour. In the entire experiment victims have a passive role: they are not asked to make a choice but only receive the payoff resulting from the care level chosen by her respective partner (injurer) in the other group. The subjects are then given their endowments; they are informed about the costs of care that may be incurred by the injurer and the corresponding amount of loss resulting from the interaction.7 All of this is common knowledge. We conducted three hand-run sessions with a total of 42 subjects.8 Each session had 15 treatments as shown in Table 7.5. For details of the treatment-specific instructions, see ‘Instructions’ in Appendix B. In all treatments a certain amount of loss (depending on the care level chosen by the injurer) occurs with probability one. In other words, we set p = 1 in all treatments in order to control for potential differences in the subjects’ preferences towards risk. 8 The subjects were recruited from the population of graduate and undergraduate students of Jadavpur University, Kolkata, India. A typical session lasted about one hour and fifteen minutes. An experimental rupee was taken to be equal to one Indian rupee. The average payment per subject was `300. 6 7



Efficiency of Liability Rules  119

Table 7.5.  Treatments Treatment

Rule

Endowment

Subjective Information

Effectiveness

1

NL

(15, 15)

No

Moderate

2

NL

(15, 15)

No

Low

3

NL

(15, 15)

No

High

4

SL

(15, 15)

No

Moderate

5

SL

(15, 15)

No

Low

6

SL

(15, 15)

No

High

7

NL

(20, 10)

No

Moderate

8

NL

(20, 10)

No

Low

9

NL

(20, 10)

No

High

10

NL

(10, 20)

No

Moderate

11

NL

(10, 20)

No

Low

12

NL

(10, 20)

No

High

13

NL

(15, 15)

Yes

Moderate

14

NL

(15, 15)

Yes

Low

15

NL

(15, 15)

Yes

High

In Table 7.5, ‘NL’ refers to the rule of no liability and ‘SL’ refers to strict liability. The entries in the endowment vector refer to the endowments of injurers and victims respectively. We made three changes in the informational content of instructions in the ‘Subjective information’ treatment. Specifically, we made explicit use of the words ‘injurer’, ‘victim’ and ‘accident’ in these rounds. Finally, the entries in the ‘Effectiveness’ column refer to the extent to which an additional unit of care by the injurer is able to reduce the loss resulting from the accident.

Results We present the experimental results in two parts. The first section looks at the patterns in individual behaviour. The second section presents the salient features of the aggregate data. Individual Behaviour We focus mainly on treatments 1–6 of Table 7.5 in this section. These are treatments pertaining to the benchmark cases of no liability (1–3) and strict liability (4–6) rules.

120  Sanmitra Ghosh and Rajendra P. Kundu

With moderate care effectiveness, about 52 per cent of the subjects chose the predicted care level of 0 under the no liability rule (treatment 1 of Table 7.5). Among the substantial 48 per cent, who had deviated from the theoretical prediction, almost half the subjects (21 per cent of the entire sample) chose a care level equal to 2, which happens to be the efficient (total social cost minimizing) care level. Under the strict liability rule and moderate care effectiveness (treatment 4 of Table 7.5), about 74 per cent of the subjects chose the predicted (and also the efficient) care level of 2. However, among the 26 per cent deviations exactly half the people (14 per cent of the entire sample) chose a very high care level of 4. Recall that when the effectiveness of care is low, the predicted as well as the efficient care level is 0 under both no and strict liability rules (treatments 2 and 5 in Table 7.5). About 55 per cent and 69 per cent of subjects adopted 0 care under these rules respectively. With high care effectiveness, the predicted care levels under the no and strict liability rules are 0 and 2 respectively, whereas the efficient care level for both cases is 2 (treatments 3 and 6 in Table 7.5). About 57 per cent of the people chose 0 care under the no liability rule and 86 per cent chose care level 2 under the strict liability rule. In case of no liability, among the 43 per cent deviations, almost 10 per cent chose the efficient care level of 2. Next, we investigated whether there is persistence in subject behaviour across rules. Specifically, we wanted to know whether a subject who chooses the predicted (efficient) care level under one rule also chooses the predicted (efficient) care level under the other rule. If a person chooses the same predicted care level under both rules, she is acting in an individually rational way as predicted by the theory. On the other hand, if she always chooses an efficient care level she probably has an innate preference towards efficiency or has a tendency to over invest in care. Table 7.6 shows the joint frequency distribution of subjects under the two rules with moderate care effectiveness. Table 7.6.  Joint Distribution of Care Choice (Moderate Effectiveness) Strict Liability No Liability

0

1

2

3

4

Total

0

0

1

18

1

2

22

1

0

0

4

0

2

6

2

0

1

7

1

0

9

3

0

0

1

0

2

3

(Table 7.6 Continued)



Efficiency of Liability Rules  121

(Table 7.6 Continued) Strict Liability No Liability

0

1

2

3

4

Total

4

1

0

1

0

0

2

Total

1

2

31

2

6

42

Notes: Pearson chi-square (16) = 32.9198; Pr = 0.008.

Note that the conditional frequency of care choice 2 under the strict liability rule (the predicted choice under strict liability), when care choice under no liability is held constant at 0 (the predicted choice under no liability), is about 82 per cent. It implies that most subjects who choose the predicted care under no liability also opt for the predicted care level under the strict liability rule. However, the reverse is not true. Only 58 per cent of the subjects who had chosen the predicted care under the strict liability rule (care level 2) opted for the individually rational predicted care level under the no liability rule (care level 0). About 23 per cent of the subjects in this group chose care level 2 under the no liability rule (the efficient care level). Since the efficient care level under both these rules is 2, it could well be the case that this small group of people (about 17 per cent of the entire sample) always prefer the efficient outcome. The reported value of the Pearson chisquare indicates that with moderate care effectiveness, care choices under the two rules are strongly associated. Consider, next, the joint frequency distribution of subjects under the two rules with low care effectiveness, presented in Table 7.7. Table 7.7.  Joint Distribution of Care Choice (Low Effectiveness) Strict Liability No Liability

0

1

2

3

4

Total

0

21

1

1

0

0

23

1

3

3

0

1

0

7

2

3

1

0

1

0

5

3

1

0

1

1

1

4

4

1

0

0

1

1

3

Total

29

5

2

4

2

42

Notes: Pearson chi-square (16) = 32.0662; Pr = 0.010.

The predicted and efficient care choice under both rules is 0 when effectiveness of care is low. Table 7.7 shows that 50 per cent of all subjects chose 0 care under both rules. There is also a fair amount of association

122  Sanmitra Ghosh and Rajendra P. Kundu

between care choices under the two rules in this case as evidenced by the Pearson chi-square. Finally, the joint frequency distribution with high care effectiveness is given in Table 7.8. Under high care effectiveness, the efficient care choice under both rules is 2. However, the predicted care choice under the no liability rule is 0, whereas that for the strict liability rule is 2. About 55 per cent subjects made the predicted choice under both the rules. It appears, however, that a small proportion of subjects always tend to over invest in care. Specifically, there were a few subjects who took more than efficient care even when the marginal efficiency of care was the highest. Table 7.8.  Joint Distribution of Care Choice (High Effectiveness) Strict Liability No Liability

2

3

4

0

23

1

0

24

1

8

1

8

11

2

3

0

1

4

3

2

0

1

3

36

2

4

42

Total

Total

Notes: Pearson chi-square (16) = 7.4110; Pr = 0.285.

Aggregate Behaviour Table 7.9 shows the predicted choice and the average actual choice of the injurer for each of the 15 treatments; the average being calculated over the actual choices of the subjects in all the sessions taken together. The last column of Table 7.9 indicates the t-values of the one sample t-test (one tailed) against the predicted choice. They are always significant, showing that actual choices differ significantly from theoretical predictions. Table 7.9.  Predicted and Actual Choice Treatments

Predicted Choice

Average Actual Choice

t-value

1

0

0.98

5.189***

2

0

0.98

4.808***

3

0

0.67

4.654***

4

2

2.24

1.815**

5

0

0.69

3.669***

6

2

2.24

2.5*** (Table 7.9 Continued)



Efficiency of Liability Rules  123

(Table 7.9 Continued) Treatments

Predicted Choice

Average Actual Choice

t-value

7

0

0.6

3.489***

8

0

0.79

4.242***

9

0

0.6

3.358***

10

0

0.38

2.891***

11

0

0.40

3.42***

12

0

0.36

3.048***

13

0

1.52

6.308***

14

0

1.38

5.733***

15

0

1.55

6.909***

Notes: *** denotes 99 per cent; ** denotes 95 per cent.

Next, we compare the efficient choice and the average actual choice under the no liability rule with and without subjective information. The results are summarized in Table 7.10. As before, the average actual choices differ significantly from the corresponding efficient care levels. However, when subjective information is used (treatments 13, 14 and 15 of Table 7.10) the divergence between the actual choice and the efficient care level under the no liability rule shrinks considerably, especially when care is moderately or highly effective. It is also interesting to note that the divergence between the actual and efficient care levels under the strict liability rule (treatments 4–6 in Table 7.9) is much smaller compared to that under the no liability rule.9 Dissemination of subjective information greatly reduces this difference and causes the actual choices under the two rules to converge from the perspective of efficiency. Table 7.10.  Efficient and Actual Choice Treatments

Predicted Choice

Average Actual Choice

1

2

0.98

t-value

2

0

0.98

4.808***

3

2

0.67

–9.308***

13

2

1.52

–1.9714**

14

0

1.38

5.733***

15

2

1.55

–2.0195**

–5.44***

Notes: *** denotes 99 per cent; ** denotes 95 per cent. 9 Note that under the strict liability rule, predicted choice always coincides with the efficient choice.

124  Sanmitra Ghosh and Rajendra P. Kundu

The behaviour of the subjects, however, changed in important ways when we looked at the choices across treatments. The main points of comparison are the following. We looked at the difference in care levels chosen by injurers between no liability and strict liability rules, for an asymmetric increment in injurers’ endowments, in victims’ endowments and finally with additional subjective information about the environment. We proceeded as follows. Treatments 4, 5 and 6 in Table 7.5 differed from treatments 1, 2 and 3 in the same table, respectively, only in the application of the liability rule: the former represented the case of strict liability whereas the latter referred to no liability rule. In treatments 7, 8 and 9 the injurers’ endowments were increased from `15 to `20, whereas the victims’ endowments were reduced to `10—compared to treatments 1, 2 and 3, other things remaining unchanged. Treatments 10, 11 and 12 did just the opposite: injurers’ endowments were reduced to `10 and victims’ endowments were raised to `20. Finally, in treatments 13, 14 and 15 we gave subjective information to the players: specifically we used words like ‘victim’, ‘injurer’ and ‘accident’ to describe the environment under which care level was to be chosen. We conducted two sample t-tests with unequal variances to compare the mean care levels across treatments. The results are presented in Table 7.11. Table 7.11.  Mean Comparison Tests Mean Care Level Treatments

Former

Latter

t-value

1 versus 4

0.98

2.24

–5.502***

2 versus 5

0.98

0.69

3 versus 6

0.67

2.24

–9.135***

1 versus 7

0.98

0.6

1.4998*

2 versus 8

0.98

0.79

0.693

3 versus 9

0.67

0.6

0.313

1 versus 10

0.98

0.38

2.591***

2 versus 11

0.98

0.4

2.432***

3 versus 12

0.67

0.36

1.673**

1 versus 13

0.98

1.52

–1.788**

2 versus 14

0.98

1.38

–1.285

3 versus 15

0.67

1.55

–3.313***

Notes: *** denotes 99 per cent; ** denotes 95 per cent; * denotes 90 per cent.

1.032



Efficiency of Liability Rules  125

The main findings are summarized as follows. 1. The average care chosen under no liability is equal to 1, approximately (row 1 in Table 7.9). 2. The average care under the no liability rule is not 0 even when taking care is hardly effective in reducing the loss (row 2 in Table 7.9). However, the average care chosen declines when care becomes highly effective (row 3 of Table 7.9). 3. An enhanced difference in the levels of endowment (wealth) of the two parties reduces the choice of care (rows 4 to 9 in Table 7.11), with reduction being more prominent when the injurer is relatively impoverished compared to the victim (rows 7, 8 and 9 of Table 7.11). 4. Subject behaviour is somewhat enigmatic, especially in treatments 7, 8 and 9. Even though the injurers’ endowment levels were raised in these treatments, whereas those of the victims were reduced compared to treatments 1, 2 and 3, we find that injurers revise their choice of care downwards (rows 4, 5 and 6 in Table 7.11). This behaviour cannot be explained by underlying ‘fairness’ considerations. We pose it as a puzzle. However, two out of these three reductions are not statistically significant. 5. Subjective information increases the average care level substantially (rows 10, 11 and 12 of Table 7.11). 6. Under the strict liability rule, average care is always higher than the efficient level (rows 4, 5 and 6 in Table 7.9). In particular, even when taking no care is efficient, subjects on average spend 0.69 on care (row 5 of Table 7.9).

Concluding Remarks Economic analysis of liability rules suggests that in unilateral care tort contexts the rule of strict liability leads to efficient outcomes while that of no liability does not. However, the subjects in the role of injurers in our experiment, on average, chose inefficient levels of care under both rules. It is particularly interesting to note that under the rule of no liability, the average care of injurers is significantly higher than the theoretical prediction of no care. Furthermore, subjective information greatly reduces the difference between the deviations of average care levels under the two rules from the efficient care

126  Sanmitra Ghosh and Rajendra P. Kundu

level. Thus, the results of the experiment do not support the efficiency claims about the rules of no liability and strict liability in unilateral care models.

Appendix A: Proof of Theorem 1 Proposition A.1 If a liability rule f satisfies NL then it is efficient. Proof. This proposition follows from (A5) and the fact that if f satisfies NL then the expected cost of an injurer is less than or equal to the minimum social cost when he is taking due care, is equal to total social cost when he is negligent and is more than or equal to his expected cost at the due care level when he takes more than due care.  Proposition A.2 If a liability rule f is efficient then it satisfies NL. Proof. Suppose liability rule f violates NL. Then (∃p ∈ [0,1))(y(p) < 1). Let t be a positive number. y(p)t < t. Choose a positive number r such that y(p) r t < r < t. Let c0 = . (1 p ) Let C and L be specified as follows: C = {0, pc0, c0}, L(0) = t + pc0, L(pc0) = t, L(c0) = 0. Note that here M = {c0}. Let c* = c0. EC(c0) – EC(pc0) = c0(1 – p) – y( p)t = r – y(p)t > 0. Therefore f is not efficient.  Theorem 1 follows immediately from Proposition A.1 and Proposition A.2.

Appendix B: Instructions Welcome You are about to participate in a decision-making experiment. The experiment will last for several periods. At the beginning of every period you will be given a fixed amount of experimental rupees which can be treated as your endowment. You will have to meet all your costs (cost of care and share of loss) from this fixed amount. In each period you will be matched with another person at random. You will not know the identity of the person with whom you will be matched and vice versa.



Efficiency of Liability Rules  127

In each round you will be asked to take a decision (choosing a care level). The interaction will result in a loss falling on the other person. The amount of the loss depends on the care level you choose. In different rounds, after the choice is made, you may or may not have to bear the loss falling on the other person. Information specific to different rounds will be provided during the experiment. Your total payoff in a round = your endowment – your costs. Your total payoff is the sum of your payoffs in all the rounds. You are free to leave now if you do not wish to participate in the experiment. In that case please collect your turn-up fee. Before we proceed please write down your name here. ___________ For each of the first three rounds your endowment is 15 experimental rupees. The other party also has the same endowment. You are not liable for the loss falling on the other party. Your payoff for each round = your endowment – your cost of care. The other party’s payoff = endowment – the loss. Round 1 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Your choice of care: _______________ Round 2 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8

Your choice of care: _______________ Round 3 Care Levels Cost of care Loss

No Care

Low Care

Medium Care

0

1

2

10

6

1

High Care 3

Very High Care 4

.5

Your choice of care: _______________

0

128  Sanmitra Ghosh and Rajendra P. Kundu

For each of the next three rounds your endowment is 15 experimental rupees. The other party also has the same endowment. You are strictly liable for the loss falling on the other party. Your payoff for each round = your endowment – your cost of care – the loss falling on the other party. The other party’s payoff = endowment. Round 4 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Your choice of care: _______________ Round 5 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8

Your choice of care: _______________ Round 6 Care Levels Cost of care Loss

No Care

Low Care

Medium Care

0

1

2

10

6

1

High Care 3

Very High Care 4

.5

0

Your choice of care: _______________ For each of the next three rounds your endowment is 20 experimental rupees. The other party has an endowment of 10 experimental rupees. You are not liable for the loss falling on the other party. Your payoff for each round = your endowment – your cost of care. The other party’s payoff = endowment – the loss. Round 7 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Your choice of care: _______________



Efficiency of Liability Rules  129

Round 8 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8

Your choice of care: _______________ Round 9 Care Levels Cost of care Loss

No Care

Low Care

Medium Care

High Care

Very High Care

0

1

2

3

4

10

6

1

0.5

6

Your choice of care: _______________ For each of the next three rounds your endowment is 10 experimental rupees. The other party has an endowment of 20 experimental rupees. You are not liable for the loss falling on the other party. Your payoff for each round = your endowment – your cost of care. The other party’s payoff = endowment – the loss. Round 10 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Your choice of care: _______________ Round 11 Care Levels Cost of care Loss

No Care 0 10

Low Care

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8

Your choice of care: _______________ Round 12 Care Levels Cost of care Loss

No Care

Low Care

Medium Care

0

1

2

10

6

1

High Care 3

Very High Care 4

.5

Your choice of care: _______________

0

130  Sanmitra Ghosh and Rajendra P. Kundu

For each of the next three rounds the loss falling on the other party can be non-pecuniary (for example, accidental damage, injury, etc.) The injurer’s (your) endowment is 15 experimental rupees. The victim also has the same endowment. You are not liable for the loss falling on the other party. The injurer’s payoff for each round = your endowment – your cost of care – the loss falling on the other party. The victim’s payoff = endowment – the loss. Round 13 Care Levels Cost of care Loss

No Care

Low Care

0 10

Medium Care

High Care

Very High Care

1

2

3

4

8.5

7

6.5

6

Your choice of care: _______________ Round 14 Care Levels Cost of care Loss

No Care

Low Care

0 10

Medium Care

High Care

Very High Care

1

2

3

4

9.5

8.75

8.25

8

Your choice of care: _______________ Round 15 Care Levels Cost of care Loss

No Care

Low Care

Medium Care

0

1

2

10

6

1

High Care 3

Very High Care 4

.5

0

Your choice of care: _______________

Acknowledgements The authors would like to thank Sugato Dasgupto, Satish K. Jain and Shubhro Sarkar for helpful comments and suggestions. Thanks are also due to participants of the International Conference on Experimental Economics and Other Social Sciences organized by the Centre for Experiments in Social



Efficiency of Liability Rules  131

& Behavioural Sciences, Jadavpur University, Kolkata in December 2008 for helpful suggestions and comments. The Centre for Advanced Studies, Department of Economics, Jadavpur University funded the experiment. We would also like to thank Abhishek Das, Sarat Pathak and Abhirup Roy for helping us in conducting the sessions.

References Brown, John Prather. 1973. ‘Toward an Economic Theory of Liability’, Journal of Legal Studies, 2 (2): 323–350. Calabresi, Guido. 1961. ‘Some Thoughts on Risk Distribution and The Law of Torts’, Yale Law Journal, 70 (4): 499–553. ———. 1965. ‘The Decision for Accidents: An Approach to Non-Fault Allocation of Costs’, Harvard Law Review, 78 (4): 713–745. ———. 1970. The Costs of Accidents: A Legal and Economic Analysis. New Haven: Yale University Press. Forsythe, Robert, Joel Horowitz, N. E. Savin and Martin Sefton. 1994. ‘Replicability, Fairness and Pay in Experiments with Simple Bargaining Games’, Games and Economic Behavior, 6 (3): 347–369. Hoffman, Elizabeth, Kevin McCabe, Keith Shachat and Vernon Smith. 1994. ‘Preferences, Property Rights and Anonymity in Bargaining Games’, Games and Economic Behavior, 7 (3): 346–380. Hoffman, Elizabeth, Kevin McCabe and Vernon Smith. 1996. ‘Social Distance and Other Regarding Behavior in Dictator Games’, American Economic Review, 86 (3): 653–660. Jain, Satish K. and Ram Singh. 2002. ‘Efficient Liability Rules: Complete Characterization’, Journal of Economics (Zeitschrift fur Nationalokonomie), 75 (2): 105–124. Kornhauser, Lewis and Andrew Schotter. 1990. ‘An Experimental Study of Single Actor Accidents’, Journal of Legal Studies, 19 (1): 203–233. Landes, William M. and Richard A. Posner. 1987. The Economic Structure of Tort Law. Cambridge, MA: Harvard University Press. Miceli, Thomas J. 1997. Economics of the Law: Torts, Contracts, Property, Litigation. USA: Oxford University Press. Posner, Richard A. 1972. ‘A Theory of Negligence’, Journal of Legal Studies, 1 (1): 28–96. Roth, Alvin E. 1995. ‘Bargaining Experiments’, in J. Kagel and A. E. Roth (eds), Handbook of Experimental Economics, pp. 253–348. Princeton, New Jersey: Princeton University Press. Shavell, Steven. 1980. ‘Strict Liability Versus Negligence’, Journal of Legal Studies, 9 (1): 1–25. ———. 1987. Economic Analysis of Accident Law. Cambridge, MA: Harvard University Press.

Section Four: Regulation

8

Regulation in the Case of a Natural Monopoly: The Electricity Act in India V. Santhakumar

Introduction The Indian Parliament passed a new Electricity Act in 2003. This Act intends to facilitate regulation and competition in the electricity sector in the country. This act is analysed here as a case of legislation facilitating government intervention in the provision of goods and services that may have features of a natural monopoly, i.e., the decrease in the average cost of provision as more is provided due to the economy of scale and hence the desirability to have one firm providing service in a geographical area or for a set of consumers, rather than a number of firms competing to provide the service. The issue of natural monopoly is about how to derive the benefits of the economy of scale with having one firm operating, without incurring the losses due to the monopoly behaviour of that single firm.1 Multiple strategies have been adopted in different parts of the world to sort out the issue of natural monopoly. These strategies include: governments owning monopoly companies, regulating private companies operating as natural monopolies, franchise bidding or auctioning the right to serve as natural monopolies and allowing competition in the set of activities where competition is feasible within the overall provision of a service having natural monopoly. The extent 1 A monopoly firm does not supply its good or service as much as that supplied by a competing firm, and hence the former can charge a higher price (than that can be charged by a competing firm), subject to the demand for that good or service.



Regulation in the Case of a Natural Monopoly  133

of natural monopoly for a given service can also vary depending on the socioeconomic context. For example, in a place where electricity consumption is not high, there may be economy of scale in its generation, whereas it may cease to exist in the same place when electricity consumption becomes much higher. Given these possibilities and multiple strategies, what should be the role of law in addressing the issue of natural monopoly? Is it in prescribing that it is the right of government-owned companies to provide all such goods and services? Is it in insisting on a particular mode of regulation? This is the issue addressed here in the specific context of the Electricity Act in India. The remaining part of the chapter is organized as follows: The historical experience as well as the theoretical debates on the institutional mechanisms to address the issue of natural monopoly in the international context is summarized briefly in the following section. The provisions of the Electricity Act adopted in India in 2003 are discussed in the third section. This is followed by an evaluation of the Electricity Act regarding its appropriateness as a legislation to address natural monopoly.

Institutional Mechanisms to Address Natural Monopoly Societies in different parts of the world have had experimented with different institutional mechanisms to get services such as electricity, water supply, telecommunications and so on during the last 150 years—the period during which technologies to provide such services on a large scale became available. Most of these services were initially available in the towns and cities, and expanded to the countryside only in later periods. In the initial stages, one or more private companies supplied services such as electricity with (formal or informal) consent or licence of the city governments. One reason for this consent was the need to use public spaces (such as streets) for certain activities (such as constructing lamp posts or digging pipes) for the provision of these services (as electricity or water supply).2 Such licensed 2 In certain cases there was only implied consent of the people living in area. Even today one can see such small private suppliers operating in countries such as Cambodia where an owner of a small generator develops a local area wire network through the streets and provides electricity to paying consumers. This is visible in some parts of India too, for example in the remote parts of Assam and Bihar, where people depend on local generator owners for minimal electricity supply due to the unreliability of centralized grid supply.

134  V. Santhakumar

private supply prevailed not only in countries such as the United States,3 but also in India until its independence. The 20th century witnessed changes in this institutional mechanism. One tendency was for governments to take over the ownership of these companies and expand their operation to other parts of the country. This has happened in post-independent India and postcolonial Latin American countries, for example in Brazil after the 1930s. In the United States, there was strengthening of government regulation even from the beginning of 20th century, though the ownership of some the companies remained in private sector.4 There has been an intense debate during the last three or four decades on the effectiveness of government ownership and regulation of utilities, providing services such as electricity supply. Government-owned companies failed to meet the growing demand in many countries like India. Thus, those countries that depended solely on government-owned companies for generating and distributing electricity started encouraging private firms to generate more electricity (as in China in the 1980s and India in the early 1990s). Thus, a mix of public and private firms came to exist in certain activities of the provision of electricity supply. This encouraged these countries to move towards regulation (of private and public companies), which depended until then solely on government ownership. On the other hand, regulation came under critical analysis from the 1960s onwards in the United States. The 1960s witnessed theoretical arguments that the social losses due to natural monopoly can be controlled through strategies such as franchise bidding—auctioning the right to serve as a monopoly service provider in an area among the competing firms. This period also saw an increase in the consumption or the development of technologies, which nullified or minimized the ‘natural monopoly features’ of at least certain goods and services. Electricity consumption has increased manifold so that many firms can compete viably in electricity generation, or 3 In fact, municipal governments in the United States awarded licences to more franchisees, approximating the situation of very high competition (Jarrell 1978). However, there was also consolidation of these competing franchisees, indicating that even when competition is enforced, the economy of scale in operations may encourage ‘market-based’ consolidation of firms. 4 Priest (1993) argues that the changes in the United States were not very radical in the sense that there were franchisees licensed through competition in the 19th century, and these licensed franchisees were operating in some form of regulated environment, and this gradually had evolved into a formal regulation. Thus, according to him, the argument by Demsetz (1968) and others that franchise bidding could be a more efficient alternative to regulation was not based on an understanding of the actual problems of dealing with a natural monopoly in the history of the United States.



Regulation in the Case of a Natural Monopoly  135

the development of mobile telephony has lessened the need for landlines—a feature that contributed to the natural monopoly in telecommunications. These changes also facilitated the use of competition amongst private firms as a strategy to provide these services efficiently without causing the many social losses associated with monopoly power. However, each of the different institutional mechanisms used to address natural monopoly has merits and demerits. Moreover, how appropriate each of these is in a specific socioeconomic realm depends on contextual factors, which may change over time. Thus, one mechanism suitable for a locality during a particular period need not be the optimal one in the same locality where the level of consumption, technology, etc., undergoes significant changes.

Benefits and Costs of Government Ownership Government ownership of public utilities such as those providing electricity has produced certain important outcomes in many developing countries. 1. The provision of public service was intended to have a demonstration effect, and most people who did not demand electricity initially did so after seeing its benefits in reality. It is unlikely that private firms would be willing to engage in such creation of supply-induced demand. 2. It might have been the case that private firms would not be able to mobilize and invest the large amount of resources required to reap the benefits of the economies of scale in the case of services such as electricity in countries such as India, and government’s direct involvement may have helped in this regard. 3. There was the fear that even if private firms were able to invest adequately, the provision of such services involved activities that lent themselves to being natural monopolies, and private monopolies would behave in a socially harmful manner. However, government ownership has several problems. For example, it has not performed well in assessing the demands of people in many parts of the world. This is facilitated by the fact that government-owned firms face soft budget constraints and hence they do not go bankrupt, even if revenues do not meet expenditure. It has been fairly slow in responding to social and economic changes. In many countries, the provision of electricity or water supply was carried out without charging the appropriate cost of

136  V. Santhakumar

supply to several consumers. This created either financial constraints for the government-owned utilities or fiscal problems for the governments. Often, such financial problems curtailed the ability of state-owned utilities to invest in new projects that would enhance the quantity or quality of supply and to provide the service to new consumers. This led to the deterioration of the quality of service in a context where more and more people demanded more of these services. Efficiency of government-owned organizations is also open to question; there is anecdotal and other evidence indicating large-scale inefficiencies, corruption and wastage of resources by the government-owned companies. One feature noted in many developing countries is the spending of a greater amount of financial resources to pay salaries of permanently employed staff, with little left for the operation and maintenance of assets providing public services. Thus, there have been inappropriate allocations of resources between capital investments and operating expenditures in the public sector. The rewards to the staff tend to be misaligned, with the indexing of salaries of employees without much concern for market wages and this leads to the ballooning of the overall budget. Thus, government-owned utilities do not have adequate checks and balances to be efficient and to minimize the cost of providing the services. This is compounded by the rigidities in the formal or public sector labour market operating in many countries. The incentives of individuals within these organizations are rarely in tune with the stated objectives of their organizations. Apart from these issues of allocative inefficiency, there can also be issues of simple or x-inefficiency in public organizations, manifested by wastage of resources, corruption and poor planning. Theoretically inefficiencies in government-owned companies can arise due to their lack of profit maximization objective, and hence the lack of incentives for increasing productivity (Stiglitz 2000). Due to the perception that managers of government-owned firms are likely to misuse their position, there are many procedures imposed on them while hiring or procuring inputs, including labour. These procedures can enhance administrative and transaction costs and create inflexibility in using the appropriate type and quantum of inputs determined by the economic or market conditions of the time. Owing to the fact that the managers of government-owned firms do not usually gain significantly by improved efficiency, they look forward to other benefits. Power and prestige associated with organizational size could be one such motivation. Increased size of the budget could be another. These managers may also absolve their responsibility for the failures and mistakes affecting the functioning of the organization. They may not be willing to take appropriate risks in decision-making.



Regulation in the Case of a Natural Monopoly  137

The sources of inefficiency in government-owned organizations have been discussed. However, some of these problems can be addressed without changing the ownership pattern. The organizations can be turned into ‘corporations’ where the role of government as owner is limited to that of shareholders in a corporate company, and the day-to-day affairs are handled by mangers, without political interference. There can also incentives and disincentives within the organizations against measured targets of outputs. The hiring and firing in these organizations can be separated from civil service rules and can be made more market-oriented and less rigid. There are public organizations in many parts of the world whose performance on the basis of a number of indicators is comparable to that of private firms engaged in similar activities. However, there are inherent difficulties in keeping these organizations out of the political web of decision-making, and hence many attempts to enhance efficiency in government-owned organizations have not been successful in countries like India.

Benefits and Costs of Privatization The basic idea that private firms are likely to be more efficient comes from the following argument. When somebody hires an agency to provide a service, one can imagine the possibility of a detailed contract about how to produce that service, how much effort should be put in incurring how much cost, and so on. However, in reality, such detailing is impossible in any contract. All contracts have to be incomplete. Thus, the person or the agency providing the service would have the incentive to provide the required service at the minimum possible cost, only if the provider has a right over the residual amount or the savings made through such cost minimization. Such a right exists when the service is provided by a private firm. While considering the service provided by a government department, even if one or a few people struggled hard to minimize costs, they cannot claim the residual amount or the savings, and thus, they are not assumed to have adequate incentives to minimize the costs (since all conditions to minimize costs cannot be put in the contract of hiring). Private firms may not facilitate the achievement of social objectives. For example, if a society wishes everybody in it to have electricity or water supply, irrespective of its affordability, then such an objective may not be achieved by a private organization alone (without the support of the government). There are also cases such as nuclear power plants, where it may be better to keep decision-making within the control of public bodies

138  V. Santhakumar

directly for safety and other considerations. It is noted that the need to exercise control over contingency periods has encouraged municipalities in the United States to have some permanent staff (for tree cutting during ice falls) even when they are allowed to outsource many of their activities (Levin and Tadelis 2010). The debates on whether privatization would lead to improvements in firms’ performance in the real world are unsettled.5 There are accounts which see improved performances by state-owned enterprises (SOEs) after reforms in China, acknowledging that in comparison with feasible private alternatives, SOEs often perform dismally in terms of productivity, cost control, technical development, customer satisfaction and in some cases even output growth. We can presume that there will be some minimization of actual costs in the private firm, even though stated costs may not come down if it operates in a regulatory environment, which is very common for example in electricity industry in many parts of the world. Or that part of the costs, which is simply wasted in a government-owned firm, may come down. There can also be institutional rigidities or inadequacies with which private firms have to operate (such as the restrictions on hiring and firing of workers, long delays in penalizing and recovering charges from the non-paying consumers, etc.) and these can also restrict the possibilities of efficiency improvements achievable by a private firm. Even if cost is minimised, how far the benefits get transferred to the consumers depends on the nature of the market. This is so because if only one private firm operates, it may not produce at socially desired levels (and produce less than what is desired by the society so that it can charge a price higher than the cost of producing it). There are two ways of solving this problem. One is to have more number of firms (and see legally that they do not collude to behave like a single firm), so that they would compete and in the process society (or consumers) would get the quantity of supply at a price that meets only the costs (including the managerial costs and opportunity cost of capital) of production. However, in the case of a natural monopoly, it is cheaper to have only one firm, since covering every consumer together by a single firm reduces the cost per consumer. This may be the case in water or electricity supply in a village. The stage of growth of many public services such as electricity supply in many parts of India may not be conducive for sustaining competition. All these reasons point to the difficulties faced in achieving real efficiency improvements through privatization. 5 For reviews, see Boycko et al. (1996), Megginson and Netter (2001), Shleifer (1998) and Vickers and Yarrow (1995).



Regulation in the Case of a Natural Monopoly  139

Regulation of Private Companies The other way to reduce the social losses due to monopoly power is to have regulation. Here, an independent regulator, mostly appointed by the government looks into the quantity produced and/or price charged by the monopolistic private firms and mandates certain restrictions, such as that they should produce a specified amount, or that they should not charge beyond a price for a specific type of consumer and so on. Usually the regulator insists that the firms charge in such a way that they get a particular rate of return on its capital (in addition to meeting all costs including managerial costs). The regulator would also see whether the costs (reported to be) spent by the firms are the really required one to efficiently produce or deliver the service. The government may provide subsidies to firms if they are asked to provide some services (to some clients) at tariffs lesser than the costs of production. The advantages of regulation include the possibility of avoiding government ownership of firms and the associated inefficiencies. Moreover, a more transparent and consistent policy can be pursued. For example, if the government wants to support supply of electricity to poor households, this can be achieved through regulation and subsidy without meddling with the incentives of the firms. Thus, firms that can achieve this social objective efficiently would gain. Further, the cost calculations would be transparent since firms would continue to operate with economic principles whereas social objectives are dealt with separately. Thus, the inherent efficiency of the private firms can be tapped without its associated social losses, through regulation. There are however possible problems with regulation too. Firms may invent loopholes to overcome the restraint of regulations. Rate-of-return regulations might provide incentives to over invest in capital assets. There are also significant administrative costs of regulation. Regulation can work well only if the regulator is in a position to acquire the required information on how much is demanded by the consumers, what is the minimum required cost to produce the service and so on. Usually the firms (producing the service) have superior information on such matters than the regulator. Then the regulator has to design incentives in such a way that the firms reveal actual information. This happens often when the firms perceive benefit from revealing or using the superior information to their advantage, and hence they may look for rents or profits beyond what is admissible under regulation. Then there may not be any benefit of regulation, as such rents would become similar to the social losses due to monopoly power. Thus, in the case of services where more than one firm cannot exist and hence

140  V. Santhakumar

the benefits of competition cannot be derived, the benefits of regulated privatization are uncertain, especially if the regulator has lesser information or lesser expertise or resources to impose its conditions effectively. There is also a possibility of ‘regulatory capture’ by the firms due to their superior information as well as their ability to collectively and coherently articulate their interest in regulatory affairs (Posner 1974 and Stigler 1971). There can also be regulatory capture by the consumers,6 by which regulators appointed by elected legislators are discouraged from charging cost-based tariffs, creating disincentives for the firms to make adequate investments. Thus, regulation is not free from political economy pressures which sometimes create outcomes that need not maximize social welfare (Levine and Forrence 1990, Peltzman 1976 and Stigler 1971).

Regulated Competition It is argued that competitive strategies can be used to avoid regulation, at the same time reducing the social losses associated with monopoly (Demsetz 1968). Franchise bidding is one such measure. This is to use competition to give the right to serve as a regulated monopoly, and this can sometimes lead to the capturing of as much benefits to consumers that can be derived through competition of firms in service provision. Here, the benefit of one firm (in having economy of scale of operations) is retained but incentive is provided to the firm to report and operate at least cost through competition at the time of bidding. However, if there are not many firms that have the resources and skills to operate in the given context, then there will not be effective competition at the bidding stage. This has happened when electricity distribution in some parts of India was opened for bidding, and in most cases there were no more than one or two willing firms. Such limited competition at the bidding stage will not lead to the capture of much benefit to the consumers. Once a firm has been awarded the licence, there is a need to adapt to the changing economic (demand and supply) conditions.7 Re-bidding can be costly and this may encourage the firm to demand renegotiation of conditions. There is also a likelihood of firms intentionally underestimating the cost at the bidding stage, and then asking for renegotiation after the award. Zupan (1989) notes (based on a reading of the literature) that the 6 As noted in the footnote on p. 196 in Stiglitz (2000), and as evident from the study of recently implemented electricity reforms in India by Santhakumar (2008). 7 This was seen as a major problem of franchise bidding by Tesler (1969).



Regulation in the Case of a Natural Monopoly  141

problems with franchise bidding include imperfect information at the time of initial bidding, producer capture of the regulatory process, the pursuit by franchisors of non-price concessions at the expense of lower prices for general service and difficulties in enforcing a franchise contract once it is awarded when the incumbent firm has advantages over potential rivals. While considering the problems associated with awarding and monitoring franchise agreements, Williamson (1976) notes that in many cases such arrangements may not be superior to the regulatory solution.8 However, some of the systematic empirical evidence shows that franchise bidding could avoid monopoly pricing, but was not very successful in promoting efficiency (as evident from supra-normal profits) (Zupan 1989). This may be partly due to the non-price concessions pursued by franchisors. As shown the case of cable television in the United States, this is owing to considerations of ‘crosscity and within’ contract reputation (which is necessary for the franchisee to get new contracts in other cities or within the same city) and the buyer’s monopsony power. If the licence given to the franchise is for a short period, it will not provide adequate incentives to make long-term investments for the provision of services, and such long-term investments are needed for services such as electricity or water supply. Though long-term contracts seem to be a solution, all possible contingencies cannot be outlined in a contract. For example, the currency crisis (the devaluation of the currency) in south-east Asian countries like Indonesia and Philippines led a to reduction in the economic activity and consumption of services such as electricity, and this created problems for electricity utilities that had gone into long-term power purchase contracts with independent power producers. What would happen in the event of a currency crisis could not be envisaged and were not outlined in the long-term contract for electricity purchase.

Competition The development of technology and increase in the consumption of services that have natural monopoly features enabled the use of competition. The emergence of mobile phones facilitated the functioning of more than one firm (due to the lack of need for landlines which accorded natural monopoly) and hence the entry of private firms and their competition have benefited telephone consumers all over the world including India (Li and Xu 2004). 8 There were also attempts to combine both systems (franchise bidding and regulation) (Loeb and Magat 1979).

142  V. Santhakumar

The increase in air traffic has facilitated the operation of a number of airlines and their competition is also turning out to be successful in India. The need for more electricity facilitates the functioning of a number of large-scale generators of electricity, and hence the economy of scale in generation need not restrain the number of firms operating. That is why privatization and competition have been allowed in generation in many parts of the developing world during the last two or three decades.

What Should Be the Role of Law? The discussion above should demonstrate that none of the institutional mechanisms described above might be considered superior in each and every context. Contextual factors play an important role in making one mechanism better than the other. Such factors include the actual level of economy of scale in each of the activities in the provision of these services, and also the performance of other institutional mechanisms. Given this situation, any law that insists on one mechanism and pre-empts others can be costly in the long run, if not immediately. Law plays an important role in the evolution of the organizational form in the provision of these services. For example, several states in the United States prevent the oil companies from operating gasoline retail stations (Blass and Carlton 2001); this study showed that it was efficiency considerations (facilitated through technological change) that encouraged the oil companies to directly operate retail stations, and hence they incurred significant costs due to the legislation banning such direct operation. Jayaratne and Strahan (1998: 240) found that the restrictions on the geographical (territorial) scope of banks imposed by states in the United States ‘retarded the natural process of selection whereby better managed, lower-cost banks expand at the expense of inefficient ones, and this has raised the costs associated with average bank asset’. On the other hand, promoting open access to natural gas transportation (separation of gas sales from its transportation, allowing open access to pipeline transportation for gas producers and consumers) in the United States helped the integration of regional markets to form a national competitive market for natural gas. Thus, while considering the need for societal intervention in a natural monopoly, and the possibilities and limitations of different institutional mechanisms in addressing this issue in specific contexts, the law in this regard needs to be analysed for its impact on the evolution of the organizational form that may facilitate dynamic efficiency. Such an analysis is attempted in the case of the Electricity Act in India, in the following sections.



Regulation in the Case of a Natural Monopoly  143

Analysing the Electricity Act, 2003 The Electricity Act, passed in 2003, is a culmination of a series of legal efforts taken in the 1990s to amend the then prevailing Electricity (Supply) Act, 1948, which strengthened government monopoly in the generation, transmission and distribution of electricity. (As a follow up to the 1948 statute, electricity generation and distribution then carried out by private companies in a number of urban or municipal areas were taken over by the respective state governments.) There was a realization in the mid-1980s that government-owned companies alone would not be able to provide adequate electricity supply to meet the growing demand in the country.9 This led to the allowing of private generators in the early 1990s.10 Then the attention was focused on tariff reforms, and there was a feeling that tariff fixation should be entrusted to independent regulators, which was made possible by amendments in Electricity (Supply) Act, 1948 in the mid-1990s (first in 1996 for establishing state regulators, and then another one in 1998 for the central electricity regulatory commission). Later there were discussions regarding the need for facilitating open access to transmission (or distribution) lines, as a step towards achieving competition in the sector. All these pointed towards the need for a comprehensive legislation in electricity sector in India, and this is the background to the passing of Electricity Act, 2003. One important objective of the Act passed in 2003 was to provide a national policy framework for electricity development and tariff fixation. This was necessary due to the important role accorded to state governments in matters related to electricity distribution in the Indian Constitution. There is a perception in India that state governments are less willing to take steps to reform the electricity sector, probably due to their closeness to interest groups, and hence the need for a national framework. This is one of the rationales for such a provision in the Electricity Act 2003, though it is worded to say that such national framework will be prepared in consultation with state government.11 There is a provision to seek the opinions of licensees For details see Santhakumar (2008). This was made possible through an amendment in 1991 of the then prevailing Electricity (Supply) Act, 1948. This step did not lead to much success due to continued financial difficulties of the government-owned distribution utilities, which prevented them from entering into competitive and reliable contracts with the generators. 11 To some extent, the ‘mandating’ of such a national framework in India is similar to the strengthening of the electricity regulatory framework in the United States in the early part of 20th century, when the municipalities were seen as incapable regulators in this regard. See Jarrell (1978). 9

10

144  V. Santhakumar

(allowed to distribute electricity in an area or a set of consumers), generating companies and consumers too in the preparation of the National Electricity Plan to be prepared by the Central Electricity Authority once in five years. There is also a provision for preparation of a national policy in consultation with state governments for rural electrification and bulk purchase of electricity and management of local distribution in rural areas through panchayat institutions, users’ associations, cooperative societies, non-governmental organizations (NGOs), and franchisees (Section 5). Universal service obligation is worded to say: ‘The respective governments shall endeavour to supply electricity to all areas including villages and hamlets’ (Section 6). The Act has de-licensed electricity generation12 in the country, allowing any company meeting technical standards of connectivity to the grid to generate electricity (Section 7). However, those proposing the hydropower plans need to submit more details and get permission from the relevant authority, to see that this proposal does not hamper the optimal generation of power from a river, or the optimal use of water for considering all uses. This is to take care of interconnectedness in the generation of power by various hydro schemes in a river basin, and the need to operate each hydro plant in such way to use the water resources of the whole basin in an optimal manner. Moreover, the Act contains a provision that insists on the generators’ responsibility to ensure the safety of the dam—to take care of the likely externalities in this regard. The Electricity Act (Section 9) has also made the operation of captive generation plants possible without requiring any permission or licence. This nullifies the prevailing provision in many Indian states to require licence for such captive generation, which does not have any economic logic for such control (and was meant to mainly protect the interests of the government-owned utilities). There is a provision that government can direct these generating companies and captive generators to generate electricity in extraordinary circumstances like war or natural calamities, on payment of costs in this regard to be determined by the appropriate regulatory commission. This is a facilitating clause to take care of exigency conditions (during which non-generation of electricity may cause severe social loss), as long as the provision is used only under such conditions. The owners of captive generation plants are to be provided open access to transmission facility for their own use subject to its availability, which is determined ultimately by the respective regulatory commissions. However, 12 There are however restrictions on hydroelectricity generation, for which the company has to submit a plan to the Central Electricity Authority that looks into the optimal development or use of water resources in the river, and regarding dam design and safety.



Regulation in the Case of a Natural Monopoly  145

here the availability of transmission facility is to be determined by the state transmission utility (STU), which is currently the state electricity board in many Indian states. Since these boards are also involved in distribution of electricity, they may not like industrial firms starting captive generation (and thereby losing such high paying consumers).13 This may encourage them to take a lukewarm attitude to the proposal to provide open access to captive generators. The Electricity Act allows generating companies to supply electricity to any consumer (subject to the rules made within the framework of the Act), but coordinate with the STU. However, here again the same conflict of interest of the STU may arise. Since STUs are involved in generation and distribution, they may not like to lose their consumers to other generators (since those who want to shift are those who see an opportunity to get electricity cheaply from other generators or those who pay the current supplier a tariff much higher than the required cost of supply). This conflict of interest for STUs is not directly addressed in the Electricity Act, though it is expected within the Act that the existing electricity boards will be unbundled into separate entities for generation, transmission and distribution within a reasonable time. This is yet to happen in a number of states. Though the central transmission utility is not expected to enter into generation and trading business (to avoid such conflict of interest), the law does not prohibit the STU from doing distribution business. Thus, the unfavourable attitude of STUs to the provision of open access may continue in future (though this can be restrained by the respective regulatory commissions). Section 12 of the Act insists on licensing for transmission, distribution and trading (buying electricity for resale). This may arise out of the recognition of the economies of scale and/or network economies in the activities of transmission and distribution. There are some exemptions in this regard. Since a captive generation plant is defined in Section 2(8) to include a power plant set up by a cooperative society or association for generating electricity primarily for the members of cooperative society or the association, distribution of electricity amongst the members of such a society/association does not seem to require licensing. Moreover, Section 13 authorizes the regulatory commission (to be exercised on the recommendations of the appropriate government and within the national policy) to exempt local authorities, panchayat institutions, users’ association, cooperative societies, NGOs or franchisees from licensing. In addition, any person can attempt distribution of electricity to a ‘rural area’ declared by the state. 13 This is since industrial consumers pay a tariff higher than the average cost of supply in a number of states in India.

146  V. Santhakumar

According to Section 14, existing firms transmitting or supplying electricity under previous acts are to be treated as licensees under the Electricity Act, without being affected by the provisions of the Act for some time (the effect of existing firms), and thereafter the regulatory commission may impose conditions, if any. The Act also considers STUs (which were until then operating as vertically integrated utility) as transmission licensees. There is a need for the regulatory commission to consider this transmission licensee’s recommendations in awarding licences to distributors. This can also be a source of trouble because of the negative attitude likely to be taken by STUs to new applicants for licences (due to their own interest in the ongoing business of distribution). The licence for distribution, transmission and trading is awarded for a period of 25 years, unless revoked prior to that. The application for licence may be rejected on specific grounds (capital adequacy, credit worthiness and so on), but not on the ground that another licensee already operates in an area. Regulators are expected to seek non-binding recommendations from the STU before issuing the licence. (This too can create some problem due to the prevailing conflict of interest of the STU as discussed earlier.) There are restrictions on acquisitions and mergers by the licensee without the approval of the regulatory commission (according to Section 17(1)), but it can acquire licensed firms in other states. There are restrictions also on the sale or exchange of licences. Such restrictions are imposed for avoiding vertical integration with an aim of the misuse of monopoly power, how far such restrictions can prevent the economies of scope and scale, if any, in the integrated operation of electricity supply is yet to be seen. The regulatory commission has the right to revoke a licence for violation of conditions and also for failing to carry out the activities for which licence is awarded. The commission also has the right to auction the utility of the revoked licensee. The purchaser of this utility will not be responsible for liabilities and this cannot be attached. Failure to comply with auction sale by the previous owner would lead to the removal of its assets in public spaces (like electric posts in streets) at its cost. There is also a provision for an administrator to operate the utility during the interim period. In order to ensure smooth functioning of electricity grid operations when multiple transmitters, distributors and traders function, national, regional or state load despatch centres (LDC) are created (as per Part V of the Act) either as a government company or corporation. However, such entities cannot engage in the business of trading electricity (to avoid conflict of interest). These despatch centres may give directions to achieve stability in



Regulation in the Case of a Natural Monopoly  147

grid operations, and to also to achieve efficiency in the operation of the power system, which are binding on every generating company, or licensee, or other persons connected with the system. This is to facilitate coordination of the functioning of the system, in the absence of which, there is incentive to free ride. The disagreements on such directions can be brought to the central regulatory commission, but compliance with LDC directions is needed, pending the decision of the commission. The state or regional LDC can levy or collect fees from generating companies or licensees as specified by the regulatory commission. Their operations are mediated through committees (i.e., regional power committees) having representatives from all stakeholders of electricity supply industry in the respective jurisdiction. Though this company or entity is likely to be a monopoly, this may be unavoidable due to the nature of its operation. However, there is a need to have safeguards and institutional options to see that the monopoly operation of LDCs does not increase their inefficiency. It is the duty of the transmission licensee to provide open access to its facilities. However, as mentioned earlier, though the central transmission utility is barred from conducting trading or distribution business, STUs are not prohibited from doing so. Thus, STUs having a stake in distribution can block the provision of open access through many means, either by taking a position that there is not adequate transmission facility, or quote an unrealistic wheeling charge, etc. In addition to wheeling charge, and a charge to meet the fixed costs involved in the provision of transmission facility, those seeking open access are expected, according the Act, to pay a surcharge to compensate for cross-subsidy (whereby industrial or commercial consumers pay a tariff higher than the average and marginal cost of supply to meet the burden to provide electricity to residential and agricultural consumers who get the supply at tariffs lesser than the cost of supply) currently prevailing in India. The STU, which currently distributes electricity, can also quote a much higher surcharge14 to make the open access-based supply unprofitable. Though the regulators are supposed to mediate and take a final decision on all these matters, and they are supposed to announce a time frame to avoid cross-subsidy and hence surcharge, the experience so far shows that there are formal and informal constraints to remove cross-subsidy. There is scope for political or electoral capture of the regulators in this regard. The experience in India so far clearly indicates this possibility. Thus, the surcharge, which can be decided by the STU, can be a real barrier against open access to the transmission system. 14 There is no surcharge for providing open access to captive generators for transmitting electricity for their own use.

148  V. Santhakumar

Open access is theoretically permitted in the distribution of electricity (Section 42(2)). The licensee with a distribution network in an area needs to provide access to its network for those within that area who wish to procure electricity supply from some other provider (other than the distribution licensee in an area), on payment of a wheeling charge to be determined by the regulatory commission. Here too, there is a provision to impose a surcharge in addition to wheeling charge (and an additional charge to meet the fixed costs of the distribution licensee) as long as cross-subsidy prevails, even though there is a vague presumption that such cross-subsidy will be eliminated within a time frame to be decided by the commission. Thus, the law allows open access, but its actual realization is limited by the persistence of cross-subsidy, and the law is less clear in terms of its removal. The distribution licensee is also required to supply any person deserving electricity within a specific time on payment of specified charges (Section 43). There is also a penalty to be paid by the licensee for any unjustifiable delays. This is to avoid any discrimination, and the consequent losses that may occur due to the monopolistic nature of operations of the distribution licensee in an area. There is a clear economic logic behind this section. Though there is a provision for the licensee to recover charges, fixed costs, rent and security deposits as specified by the regulatory commission, consumers using the provision of open access can enter into negotiated agreements with the service provider, and here the terms and conditions of electricity supply will be as determined by the consumers and provider. It is not clear why the negotiation option available to consumers seeking open access is not available to those getting supply from distribution licensees. This may be intended to avoid the licensees imposing more costly terms and conditions on some consumers through negotiated agreements. However, it is not clear why the licensee cannot arrive at less costly terms and conditions with some consumers, without imposing any cost higher than that specified by the regulatory commission on other consumers. There are economic opportunities arising out of the economies of scale, scope and network externalities leading to lower optimal charges for some consumers, and the insistence on commissionmediated terms and conditions for all the consumers for a distribution licensee may discourage the realization of these economic opportunities. When there are two distribution licensees in the same area, Section 62.d empowers the regulator to fix the maximum ceiling of tariff, allowing the licensees to fix anywhere below this ceiling to facilitate competition. In deciding the tariffs, no explicit rule or method is mentioned in the Act. However the need to ensure ‘commercial viability’, ‘recovery of costs in reasonable manner’, ‘promote efficiency and economic use of resources’, ‘good performance and optimal investment’, etc. are mentioned along with



Regulation in the Case of a Natural Monopoly  149

protecting consumer interests. This can be broadly interpreted as efficient performance without leaving scope for monopolistic extraction of rents. It would have been better to have an explicit interpretation of consumer interest so as not to interpret it as the need to provide electricity to any group of consumers at tariffs lesser than the (appropriate) cost of supply. Moreover, Section 62 requires that the regulator not show undue preference towards any consumer groups other than differentiating based on power factor, load factor, voltage, total consumption, geographical area, nature of supply or purpose for which supply is required. This indicates that income or economic status should not be considered while fixing the tariff, and we can presume that affordability or distributional considerations are not a concern in tariff setting. However, if the purpose of this elaborate set of factors for discrimination in tariff setting were to take into account the marginal cost of supply, then it would have been better to state that explicitly. Otherwise, there can be price discrimination even when there is no economic logic to do so. The powers of the regulatory commissions are wide ranging.15 These are assigned with the objectives of maintaining efficient supply, equitable distribution of electricity, promoting competition, specifying the standards of performance and so on. The power to determine tariff based on an examination of the costs of supply and reasonable rate of return on capital investments between the generator and distribution licensee, on wheeling, and also of the retail supply is vested with the regulators. However, in the case of consumers provided with open access only the wheeling charge (and related surcharge) need to be fixed by the regulator (since tariff is expected to be fixed through negotiations between the provider and user). The commissions may regulate the price at which distributors purchase or procure electricity for supply, as a way of having control over the cost of supply. The commissions issue licenses to transmit, distribute and trade electricity. The conduct of other business by electricity suppliers (such as facilitating the installation of cable television lines), and additional revenue in this regard too needs to be considered while determining the cost of electricity supply to be recovered through tariffs. However, there are provisions for competitive fixing of tariffs either by consumer who get into a contract with a supplier having open access, or when the tariffs in an area are decided through competitive bidding of suppliers (for example between generators and the distributors). Thus, the tariff determination role of the regulator cannot be an institutional restraint when the supply moves towards a more competitive phase. 15 There is also an appellate authority, namely Appellate Tribunal for Electricity, which considers the appeals against the decisions of regulatory commissions. Appeals against the orders of this tribunal can be made to the Supreme Court of India.

150  V. Santhakumar

Regarding the payment of subsidy, the Act has the following provision in Section 65. If the State Government requires the grant of any subsidy to any consumer or class of consumers in the tariff determined by the State Commission under section 62, the State Government shall, notwithstanding any direction which may be given under section 108, pay, within in advance in the manner as may be specified, by the State Commission the amount to compensate the person affected by the grant of subsidy in the manner the State Commission may direct, as a condition for the license or any other person concerned to implement the subsidy provided for by the State Government: Provided that no such direction of the State Government shall be operative if the payment is not made in accordance with the provisions contained in this section and the tariff fixed by State Commission shall be applicable from the date of issue of orders by the Commission in this regard.

However, when the government owns the utilities, the transactions between them need not be transparent even to the regulator, and hence there can be difficulties16 for the regulator to enforce this payment. This is already evident from the experience of a number of states in India. The law facilitates the reorganization of state electricity boards (SEBs). Using its provisions, the state government can take over the properties of SEBs and then hand over them to companies formed as part of the reorganization. However, the flexibility given to the state governments in terms of fixing the date of reorganization of the SEBs has resulted in a situation where many states continue to have SEBs in their conventional forms.

Potential Impact of the Electricity Act, 2003 There are people who are unhappy with the slow pace at which the Electricity Act is implemented in letter and spirit. The presence of cross-subsidy or continuing barriers against open access can be taken as indications of this slow pace. On the other hand, there are some who reckon that competition is not appropriate for electricity distribution in India. Thus, they object to the clauses facilitating competition, even if such clauses are yet to be used in any significant manner.

There can also be other difficulties arising out of the informal control or influence that the government wields over regulators. 16



Regulation in the Case of a Natural Monopoly  151

However, this analysis views the role of a piece of legislation like the Electricity Act in a different manner. The appropriate form of regulation or social intervention in a service such as electricity evolves over time in a specific context, and differs across different socioeconomic contexts (at the same point of time). The differences in this regard are to be determined by the level of electricity consumption (which in turn may be dependent on the levels of income and/or economic development), density of households (level of urbanization), demand from industrial and other users, potential sources of electricity, technological changes in the generation, transmission and distribution of electricity, etc. Thus, one form of intervention chosen for a context (like the city of Mumbai) may not be appropriate for the hinterlands of Assam or vice versa. Similarly, levels of consumption and technology change very fast these days, and hence there may be a need for changing the mode of social intervention accordingly. Thus, any law that mandates one form of intervention or regulation may not be suitable. For example, if the law vests electricity supply with governmentowned utilities, it blocks the use of other modes of regulation as and when they become feasible in a given context. On the other hand, if the law mandates competition, then it may be imposing a mode of regulation even in contexts where it is too early to have competing electricity suppliers. Moreover, it is difficult to have a central law suitable for each and every context. It is also costly to change the law now and then, according to the changes in consumption and supply of electricity services. This is especially since the effort required to create consensus and legislate the provisions of a law is costly. Hence, the most appropriate legislation is one that provides an umbrella framework that allows the electricity supply service to evolve as per the requirements of the time and space. Thus, it should neither mandate any one mode of regulation nor prohibit another one. It should facilitate smooth transition from one mode to another as and when the socioeconomic context evolves and poses a demand for newer modes of regulation. We need to see the Electricity Act, 2003 in such a perspective.

References Besanko, D., J. D’Souza and S. R. Thiagarajan. 2001. ‘The Effect of Wholesale Market Deregulation on Shareholder Wealth in the Electric Power Industry’, The Journal of Law and Economics, 44: 65–88. Blass, A. A. and D. W. Carlton. 2001. ‘The Choice or Organizational Form in Gasoline Retailing and the Cost of Laws that Limit that Choice’, The Journal of Law and Economics, 44: 511–524.

152  V. Santhakumar Boycko, M., A. Shleifer and R. W. Vishny. 1996. ‘A Theory of Privatisation’, Economic Journal, 106 (435): 309–319. Demsetz, H. 1968. ‘Why Regulate Utilities?’, The Journal of Law and Economics, 11 (1): 55–65. Doane, M. J. and D. F. Spulber. 1994. ‘Open Access and the Evolution of the US Spot Market for Natural Gas’, The Journal of Law and Economics, 37: 477–517. Eriksson, R. C., D. L. Kaserman and J. W. Mayo. 1998. ‘Targeted and Untargeted Subsidy Schemes: Evidence from Post-divestiture Efforts to Promote Universal Telephone Service’, The Journal of Law and Economics, 41: 477–502. Geddes, R. R. 1997. ‘Ownership, Regulation and Managerial Monitoring in the Electric Utility Industry’, The Journal of Law and Economics, 40: 261–288. Jarrell, G. A. 1978. ‘The Demand for State Regulation of the Electric Utility Industry’, The Journal of Law and Economics, 21 (2): 269–295. Jayaratne, J. and P. E. Strahan. 1998. ‘Entry Restrictions, Industry Evolution, and Dynamic Efficiency: Evidence from Commercial Banking’, The Journal of Law and Economics, 41: 239–273. Levine, M. E. and J. L. Forrence. 1990. ‘Regulatory Capture, Public Interest, and the Public Agenda: Towards a Synthesis’, Journal of Law, Economics and Organization, 6 (Special issue): 167–198. Levin, J. and S. Tadelis. 2010. ‘Contracting for Government Services: Theory and Evidence From US Cities’, The Journal of Industrial Economics, 58 (3): 507–541. Li, W. and L. C. Xu. 2004. ‘The Impact of Privatisation and Competition in the Telecommunications Sector Around the World’, The Journal of Law and Economics, 47 (2): 395–433. Loeb, M. and W. A. Magat. 1979. ‘A Decentralized Method for Utility Regulation’, The Journal of Law and Economics, 22: 399–404. Megginson, W. L. and J. Netter. 2001. ‘From State to Market: A Survey of Empirical Studies on Privatisation’, Journal of Economic Literature, 39 (2): 321–389. Peltzman, S. 1976. ‘Towards a More General Theory of Regulation’, The Journal of Law and Economics, 19 (2): 211–240. Posner, R. A. 1974. ‘Theories of Economic Regulation’, Bell Journal of Economics and Management Science, 5 (1): 335–373. Priest, G. L. 1993. ‘The Origins of Utility Regulation and the Theories of Regulation Debate’, The Journal of Law and Economics, 36 (2): 289–323. Santhakumar, V. 2008. Analysing Social Opposition to Reforms: Evidence from Electricity Sector in India. New Delhi: SAGE Publications. Shleifer, A. 1998. ‘State versus Private Ownership’, Journal of Economic Perspectives, 12 (4): 133–150. Stigler, G. J. 1971. ‘The Theory of Economic Regulation’, Bell Journal of Economics and Management Science, 2 (1): 3–21. Stiglitz, J. E. 2000. Economics of the Public Sector. New York: W. W. Norton & Company. Tesler, L. G. 1969. ‘On the Regulation of Industry: A Note’, Journal of Political Economy, 77: 937–952. Vickers, J. and G. Yarrow. 1995. Privatisation: An Economic Analysis. Cambridge, MA: MIT Press. Williamson, O. E. 1976. ‘Franchise Bidding for Natural Monopolies: In General and with respect to CATV’, Bell Journal of Economics and Management Science, 7 (1): 73–104. Zupan, M. A. 1989. ‘The Efficacy of Franchise Bidding Schemes in the Case of Cable Television: Some Systematic Evidence’, The Journal of Law and Economics, 32 (2): 401–456.

9

Initial Public Offerings: Institutions in India Padma Kadiyala

Introduction Privately owned enterprises can choose to convert to publicly listed firms through the process of issuing equity to the investing public. This process is called an initial public offering, or IPO. The IPO is an important watershed in the life of a company, securing for it access to public equity capital and greater publicity. At the same time, an IPO invites scrutiny in the form of greater transparency, disclosure and compliance with accounting standards. On balance, IPOs offer great benefits for young, growing firms. Since the first recorded IPO by the Dutch East India Company in the year 1602, firms have raised a total of US$488 billion in the 21 years between 1980 and 2001 through IPOs (Welch and Ritter 2002). IPOs have become an extremely popular vehicle for corporate financing in India when restrictions on new issues were eased in 1990. Indian companies raised a total of `1.716 trillion in the 17 years between 1990 and 2006. An astounding 5,833 firms went public during this time, with an average of 115 firms going public every month in just the two years between 1993 and 1994. Many of these firms in the early years were small issuers, with an average issue size of only `123 million. After rules for an IPO were tightened in 2003, the average issue size has increased almost thirty-fold to `4.17 billion (Agrawal n.d.). The popularity of IPOs in the Indian market can be traced to two main events: (a) economic liberalization undertaken by the government in 1991 which spurred industrial growth, and hence demand for capital, and (b)

154  Padma Kadiyala

financial reforms to protect small investors, primarily those undertaken by the Securities and Exchange Board of India (SEBI). The Indian economy was heavily regulated until it was hit by a balance-of-payments crisis in July 1991. The government was able to avert the crisis by adopting several IMFled macroeconomic and structural reforms. Notable among the reforms was the partial convertibility of the Indian Rupee (`), abolition of certain controls over industry, liberalized foreign investment norms and simplified import regulations. Underperforming government-owned companies were privatized to foster competition and to build a thriving domestic industry base. Entrepreneurship was encouraged to spur capital formation and economic growth (Teja 1992). Capital markets were also simultaneously deregulated. The Controller of Capital Issues (CCI), which enforced the Capital Issues (Control) Act, 1947, was abolished. The CCI, through its mandate from the government, was, until it was abolished, heavily involved in all aspects of the IPO process including the amount, type and price of a new equity issue. Firms were dissuaded by the involvement of the central government in the equity issuance process from seeking public equity. Consequently bank-financed capital became the primary source of financing. Bank financing led to inefficient allocation of capital, which led observers to blame the banking sector for the 1991 financial crisis. The CCI was replaced as a regulatory body in 1992 by SEBI. In June 1992, SEBI issued its Disclosure and Investor Protection (DIP) Guidelines, which allowed firms that had been in existence for several years prior to the IPO to set their own price at issuance. The new guidelines represented a significant departure from the earlier policy that required that firms issue only at par value. Under the new guidelines, only companies with fewer than 12 months of operating history were required to price their IPO at par. Older companies were allowed to set their own offer price. The requirement of a fixed price IPO was relaxed again in 1999 when book building was permitted to be used to set the offer price. The retail investor in India has embraced IPOs. IPOs collectively drew in about 15 million new investors into the stock market in just the four years between 1992 and 1996.1 Low minimum investment amounts, which has varied from `1500 to `10,000 most recently, makes IPOs very attractive to A report dated 6 June 2002 prepared by a Joint Parliamentary Committee appointed by the Indian government estimates that IPOs drew 15 million new investors into the stock market in the years between 1992 and 1996. Roughly 4,000 new IPOs were issued during this period, many of which have ceased trading (Ministry of Finance 2003). 1



Initial Public Offerings  155

retail investors. Firms are required by law to refund the entire subscription amount if at least 90 per cent of the issued amount remains unsubscribed. But companies hardly ever refund, as the average IPO is oversubscribed 16 times, with retail investors also being favoured in the share allotment phase. It is not uncommon for small investors in India to hold highly diversified portfolios, with some holding over 100 stocks (Nandha and Sawyer 2002). Indian capital markets have benefited from the active participation of retail investors. The financial industry in India has been growing rapidly to accommodate the surge in demand for IPOs. Several new brokerage houses and investment banks are competing for a share of this growing industry. Simultaneously, facilitators such as stock exchanges and depositories have had to adopt the most modern technology to allow the market to operate without any hitches. India is unique in that 100 per cent of trading has now moved to screen-based electronic trading. It is inevitable that excesses will arise when there is unfettered demand for a product. The Indian IPO market has suffered its share of frauds and unscrupulous activities. Regulators, primarily SEBI, have been performing valiantly in providing surveillance and monitoring of markets to proactively avert fraud in the markets. Examples of intervention by SEBI include tightening eligibility norms for IPOs to preclude fraudulent issuers, permitting the book building process to set the offer price and tightening restrictions on share allocations by underwriters. Some of these actions were in response to scams perpetrated by unscrupulous operators, and others, notably the book building rules, were put in place proactively to encourage capital formation. The rest of this chapter examines the institutional framework of Indian capital markets, and the IPO market, in particular. Differences between the process of going public in India and in the United States are highlighted. The impact of these differences on salient aspects of an IPO and its investment value are extensively discussed.

The Process of Issuing Equity In a typical IPO in the United States, the firm selects an investment bank or an underwriter to manage the offering. The investment bank helps the firm file a registration statement with the Securities and Exchange Commission (SEC). The SEC pores over the document to check the legitimacy of the claims made in it. After approval is received from the SEC, the investment

156  Padma Kadiyala

bank takes the company on a ‘road show’ to introduce the company to institutional investors and favoured clients. Feedback provided by investors during the road show is incorporated into the investment bank’s financial analysis to arrive at an offer price for the IPO. Shares are allotted to investors at the offer price and the proceeds, net of the investment bank’s fee, are paid to the issuing company. The company’s shares commence trading on the following day on the exchange on which its shares are listed. The early stages of the process for going public in India resemble those in the United States. Firms select an underwriter, who then files a draft red herring prospectus with SEBI. The underwriter mounts a publicity campaign to attract the attention of investors. This includes circulation of a report on the firm, which includes a preliminary price, among favoured clients, and a road show to elicit interest for the IPO from institutions. The underwriter uses the information obtained from the road show to set a price band. The price band is included in a red herring prospectus filed with the SEBI. The main difference between the process in the United States and that in India is the mechanism to set the offer price. In India, IPOs can be priced using a fixed price mechanism or through a book building process. Until September 1999, only fixed-price IPOs were permitted in India. New firms with less than 12 months of operations were restricted to issuing at par, but older, more established companies could choose their own offer price. The latter include firms with established brand names or companies with a strong history of profitability.

The Book Building Process In September 1999, on the recommendations of the Malegam Committee set up by SEBI in 1995, permission was granted for the book building process to price IPOs for firms issuing more than `250 million. After a slow start, book building now dominates the fixed-price offering mechanism, accounting for 80 per cent of the IPOs issued in the 2006–2007 period (SEBI 2000), the most recent period for which data is available. The actual process of book building in India differs in several respects from that in the United States. In India, the underwriter, who is known as the book running lead manager, sets a preliminary price range or a price band. The maximum price in the band is restricted to be no greater than 120 per cent of the minimum price. The price band is selected after the completion of road shows during which indications of interest are solicited from institutional clients. The price band on which the book is built is then filed with SEBI. Once the price band is



Initial Public Offerings  157

filed with the SEBI, it is almost never revised upward, although it is revised downward if the issue meets with poor demand. The underwriter forms a syndicate to manage the book. The syndicate consists of brokerage firms and other underwriters who collect bids submitted electronically by investors. Unlike in the United States, retail investors are allowed to bid directly for the IPO. The definition of retail investors has undergone several changes in the Indian market. Starting with an initial definition of those who bid up to 1,000 shares, the definition was changed in 2003 to those who bid up to `50,000, a threshold which was raised in 2005 to `100,000. Retail buyers are permitted to place market or limit orders with the requirement that their bid be accompanied by a deposit for the full amount of their bid. Non-institutional buyers, defined as retail buyers permitted to exceed an investment of `100,000, but limited to no more than `500,000 are also required to deposit the full amount at the time of bidding. Bids are legally binding contracts, with investors obligated to take up their allocation quotas. The third category of investors is qualified institutional buyers (QIBs). These investors are required to place limit bids, but are not required to deposit the full amount at the time of bidding; they are required to deposit only 10per cent of the amount bid. The final price for the IPO is set by the book runner in consultation with the issuer after the completion of the book building process which typically lasts for five days. The mandated minimum book-building period is three days and the maximum is 10 days. The entire book building process is conducted electronically and is carried live by the stock exchange. At the end of each day during the book building process, the stock exchange website displays cumulative bids at their respective prices, and bids consolidated by investor categories. Thus investors have full information about subscription levels and prices. All investor categories are permitted to use this information to revise their bids. Khurshed et al. (2010) argue that complete transparency in the book building process helps to reduce the winner’s curse. Thus the book building process in India is much more transparent than the process in the United States. The various regulations governing the book building process encourage participation of retail investors in the bidding process. The limitation of the Indian process is that the offer price is almost never revised upward even when there is strong demand.

Share Allocations SEBI has imposed regulations governing allocation of shares to the three categories of investors. In fixed price IPOs, at least 50 per cent of the shares

158  Padma Kadiyala

must be allotted to retail investors. In book-built IPOs, retail investors are mandated to receive up to 35 per cent of the issue, non-institutional investors up to 15 per cent, and QIBs 50 per cent of the issue. The Indian IPO market is characterized by oversubscription. The reluctance of underwriters and issuers in the Indian book building process to revise the offer price upward creates excess demand in the form of oversubscription for the IPO. Oversubscription is resolved through share rationing, while conforming to mandatory limits on allocations to the three investor categories. When the issue is oversubscribed, all categories of investors receive share allocations on a pro rata non-discriminatory basis. If the issue is under-subscribed in any investor category, then the firm can reallocate those shares to any oversubscribed category. The basis of allocation is made public by the registrar of the IPO. The actual allotment of shares to all categories, and their registration has to be completed within 15 days after the book building closes. Failure to do so requires the issuer to pay a penalty to the new shareholders. Underwriters and issuers were allowed to exercise discretion in allotting shares to QIBs until November 2005. There are three categories within QIBs, all of which were treated differently by underwriters: (a) Financial institutions (FIs), (b) domestic mutual funds and (c) foreign institutional investors (FIIs). Bubna and Prabhala (2010) find that financial institutions place the smallest bids and FIIs place the largest bids. Domestic mutual funds bid for moderate amounts, but are reported to be favoured in receiving higher allocations. FIIs are discriminated against receiving share allocations only when they place very large bids. For other groups, bid size did not affect share allocations. Underwriters thus appeared to use subjective information about the bidder in addition to bid size to determine allocations. The tendency of domestic mutual funds to receive more than proportional allocations received negative attention in the press. The popular press appeared to view the actions of underwriters as exhibiting favouritism towards large and powerful mutual funds. There were numerous media reports decrying the allocation policy on the grounds of favouritism. In response to this public outcry, SEBI changed the policy in November 2005 to a nondiscriminatory regime.2 Under the new regime, all investors, including QIBs receive proportional allocation. Underwriters no longer have discriminatory power in determining allocations. 2 Amendments were made to the Securities and Exchange Board of India (Disclosure and Investor Protection) Guidelines, 2000 on 19 September 2005, which were issued by SEBI’s Division of Issues and Listing.



Initial Public Offerings  159

The public outcry against the discriminatory allocation mechanism may have been short-sighted. It is well known in the United States that the book building process helps underwriters extract the true underlying demand curve for the IPO (Benveniste and Spindt 1989). Better information about demand helps the issuer receive the highest valuation and maximize the capital raised at the IPO. Domestic mutual funds may have been favoured by underwriters prior to November 2005 as a reward for truthfully revealing their demand for the IPO.

Direct Costs of IPOs Direct costs associated with a public offering are the underwriter fees, administrative fees, advertising and marketing fees, legal fees and, in the case of Indian IPOs, fees to be paid to a rating agency. The direct costs should depend on the type of mechanism chosen to price the IPO. Book-built issues should be more expensive as the underwriter is more heavily involved in setting the price. Consistent with this observation, Kumar (2008) reports a gross spread of 4.5 per cent for book-built IPOs, and a spread of 2.3 per cent for fixed price offers in the years between January 2003 and December 2007. Underwriter fees and hence direct costs of an IPO in India also depend on the issuer. In India, divestment of state-run enterprises through public offerings accounts for a large proportion of IPO volumes. Fierce competition among foreign banks, local Indian banks and Indian affiliates of foreign banks to take these large IPOs public has severely hurt underwriter fees. State-owned companies only paid 0.05 per cent of issue proceeds as underwriter fees in the first quarter of 2010. The average fee paid by private issuers is much higher, about 2.88 per cent according to a recent Bloomberg study (David 2010). In the United States, Chen and Ritter (2000) show that 90 per cent of the firms pay a fee equal to 7 per cent of the total capital raised at the IPO. Observers assert that the 7 per cent fee in the United States is too high since investment banking fees tend to be much lower, about half of that in the United States, for IPOs in Asia and Europe. Indeed, underwriter fees in Hong Kong and Singapore average 3 to 4 per cent. Investment banks in the United States have argued in their defence that the book building process in the United States justifies a higher fee. Their experience in India, with total direct costs for book-built IPOs of only 4.5 per cent, does not support their claim. On balance, Indian issuers appear to have been successful in lowering the costs associated with a public offering.

160  Padma Kadiyala

Period between Issue and Exchange Listing Listing on an exchange completes the IPO and enables secondary market trading in the shares of the issuer. In the United States, listing occurs on the very next day following the IPO. In India, IPOs are not immediately listed on an exchange. The average period between the IPO and listing is 72 days. SEBI requires this waiting period to be able to conduct its due diligence process. In the United States, the due diligence is undertaken by the SEC before the offer is made to the public. The underwriter submits a registration statement to the SEC, which contains detailed descriptions of the utilization of IPO proceeds and projections of future financial performance. The underwriter also files the registration statement with NASD, the stock exchange where the IPO will be listed, and Blue Sky regulators in the states where the IPO will be sold. While the SEC is conducting its review, the underwriter takes the IPO on a road show. It is only after SEC’s review is completed can the underwriter begin to sell the securities. The review process in India is slightly different. Initial buyers only receive brief summaries of firms that plan to issue an IPO. Most individual investors do not receive the prospectus. Further, it is all but impossible for the issuer to provide detailed information to the large number of retail investors who express interest in an IPO. Several consumer protection agencies enter the picture to perform due diligence on behalf of the retail investor. This process of vetting issues by proxy takes several forms, all of which are unique to the Indian market. First, all IPOs have to carry a rating issued by a registered credit rating agency, and the rating has to be disclosed in the final prospectus. SEBI has made IPO grading mandatory effective from 1 May 2007. IPOs are graded on a scale of one (lowest) to five (highest) based on fundamental factors. These factors are firms’ business prospects, financial prospects, accounting practices, management quality, potential future risks and the quality of corporate governance. Grading costs an issuer about `500,000 and takes between three and four weeks to complete. The process consists of extensive interviews with the firm’s top executive and on-site visits by representatives from the credit rating agency. Retail demand has been found to be higher for IPOs with higher grading. Demand from institutional investors appears to be less sensitive to IPO grading. Thus, IPO grading appears to satisfy the objective it was designed for, namely to provide third-party certification of an issuer. Second, major national newspapers carry ratings of all forthcoming new issues. About a month before an issue opens for public subscription and



Initial Public Offerings  161

continuing until the IPO is listed, financial newspapers publish premiums for a majority of new issues. The premium for each issue is calculated as a consensus estimate of premiums reported by investment clubs around the country. Third, many IPOs trade on an illegal, but thriving, when-issued market, also called a ‘grey market’. The price in the when-issued market is another piece of information that should reduce the information asymmetry. All these ‘pre-market’ activities enable ordinary investors to invest in several IPOs without having to incur time and effort conducting their own due diligence on individual issues. It is not surprising then that Indian investors have more than 100 stocks in their portfolio. A lively debate has ensued in Indian financial circles as to whether the excessive diversification does more harm than good since it is costly to monitor news developments in so many stocks. Ever since Markowitz’s path-breaking work, it is recognized that the benefits of diversification level off after about 30 stocks in the portfolio.

Exchange Listing The choice of stock exchange is an important consideration for many issuers. In the United States, the choice is between listing on the NYSE or on NASDAQ. Issuers with total market capitalization of more than US$40 million, with a public float of more than 1.1 million shares and with no fewer than 400 round lot holders, are eligible for listing on the NYSE. Listing requirements on NASDAQ are less stringent. Historically, large, stable firms chose to list on the NYSE, and the NASDAQ attracted technology stocks with high growth prospects. But, as trading has shifted even on the NYSE from the floor of the exchange to electronic trading, the distinction between the two markets has blurred in recent years. For Indian IPOs, the choice is between the Bombay Stock Exchange (BSE), the National Stock Exchange (NSE), and several regional exchanges. The BSE is India’s oldest exchange, established in 1875. The NSE is a newcomer, which was incorporated in 1992. The NSE is India’s first fully demutualized stock exchange. It is also the largest stock exchange in India in terms of volumes in the equity and derivatives segments. The emergence of NSE, whose trading systems and structure are specifically tailored to attract young firms, has displaced BSE as the top choice for IPOs. The BSE is competing to attract smaller companies by creating two categories of companies that are eligible for listing, large cap and small cap companies. Large cap companies are required to have a minimum market capitalization

162  Padma Kadiyala

of `250 million. Small cap companies have a lower requirement of only `50 million in market capitalization.3

Underpricing It is now a well-documented global phenomenon that the average IPO increases in price on the first day of trading. This increase is termed underpricing, and is calculated as the percentage difference between the offer price and the price at which the shares subsequently trade in the market. Since it was first discovered by Logue (1973) and Ibbotson (1975), the mean underpricing in the global market is 26 per cent. It is certainly noteworthy that there is no country which has escaped this phenomenon. Table 9.1 reports statistics for the magnitude of underpricing in each of 33 countries (Ritter and Welch 2002). Table 9.1. Average Initial Returns for 33 Countries Country Average Australia

Sample Size

Time Period

Initial Return

266

1976–1989

11.9 per cent

Austria

67

1964–1996

6.5 per cent

Belgium

28

1984–1990

10.1 per cent

Brazil

62

1979–1990

78.5 per cent

Canada

258

1971–1992

5.4 per cent

Chile

19

1982–1990

16.3 per cent

China

226

1990–1996

388.0 per cent

32

1989–1997

7.7 per cent

Finland

85

1984–1992

9.6 per cent

France

187

1983–1992

4.2 per cent

Germany

170

1978–1992

10.9 per cent

Denmark

Greece Hong Kong India

79

1987–1991

48.5 per cent

334

1980–1996

15.9 per cent

98

1992–1993

35.3 per cent (Table 9.1 Continued)

3 Information obtained from the respective exchange’s websites, http://www.bseindia.com/ static/about/listsec.aspx?expandable=2 and http://www.nse-india.com/corporates/content/ capital_market.htm.



Initial Public Offerings  163

(Table 9.1 Continued) Country Average

Sample Size

Time Period

Initial Return

Israel

28

1993–1994

4.5 per cent

Italy

75

1985–1991

27.1 per cent

Japan

975

1970–1996

24.0 per cent

Korea

347

1980–1990

78.1 per cent

Malaysia

132

1980–1991

80.3 per cent

Mexico

37

1987–1990

33.0 per cent

Netherlands

72

1982–1991

7.2 per cent

New Zealand

149

1979–1991

28.8 per cent

68

1984–1996

12.5 per cent

Portugal

62

1986–1987

54.4 per cent

Singapore

128

1973–1992

31.4 per cent

Norway

Spain Sweden Switzerland Taiwan Thailand

United States

1985–1990

35.0 per cent

1980–1994

34.1 per cent

42

1983–1989

35.8 per cent

168

1971–1990

45.0 per cent

32

1988–1989

58.1 per cent

138

1990–1995

13.6 per cent

2,133

1959–1990

12.0 per cent

13,308

1960–1996

15.8 per cent

Turkey United Kingdom

71 251

Source: Ritter and Welch (2002).

In India, the average underpricing has varied over the two decades since IPOs began to be issued in the early 1990s. One of the earliest studies of IPOs conducted by Madhusoodanan and Thiripalraju (1997) reports an average underpricing discount of 75.2 per cent during the period from 1992 to 1995. Shah (1995) reports a whopping underpricing discount of 105.6 per cent when IPOs issued in 1991 are also included in the analysis. Ghosh (2005) finds that underpricing is still high, at 97 per cent, even when IPOs issued during a longer time period from 1993 to 2001 are studied. All these studies are based on IPOs listed on the Bombay Stock Exchange (BSE). Pande and Vaidyanathan (2009) restrict their study to IPOs listed on the National Stock Exchange and find that the average underpricing has dropped to 26.2 per cent. Prabhala and Bubna in an updated study, which includes IPOs listed both on the BSE and on the NSE, report average underpricing of 22 per cent for IPOs issued between 2004 and 2006.

164  Padma Kadiyala

Issue size seems to be the most important factor in predicting underpricing of Indian IPOs. Larger issues are associated with smaller underpricing. Another significant predictor of underpricing is oversubscription. Although the average IPO in India is oversubscribed 21 times, higher the oversubscription, higher is the underpricing. The third predictive factor is the age of a firm at the time of an IPO. IPOs by younger firms are underpriced more. Finally, firms affiliated with large industrial groups and those that followed up the IPO with a seasoned offering are underpriced to a greater extent.

Explanations for Underpricing The underpricing phenomenon has invited a lot of scrutiny since it imposes costs on a firm’s owners. First, shares sold by owners from their personal account are sold at too low a price. Second, the value of shares they retain after the IPO is diluted by underpricing. So it is an enduring puzzle that firms are willing to leave billions of dollars ‘on the table’, which is the magnitude of the underpricing multiplied by the number of shares sold at the IPO. Firms would raise a greater amount of equity capital were there no underpricing. Several theories have been advanced to explain why this costly phenomenon endures. These theories can be classified into three groups: (a) asymmetric information, (b) institutional reasons and (c) control considerations. Behavioural explanations have also been offered, but these are in their infancy. Each of these three explanations is discussed in detail below. The participants in an IPO are the issuer, the investment bank and investors. Asymmetric information theories share the common feature that one of the participants in the IPO is assumed to have better information than others. The uninformed participants, according to these theories, protect themselves through underpricing. The earliest asymmetric information model was developed by Rock (1986), who assumes that some investors are better informed about the true value of an issuer than other investors. Informed investors will only bid for attractively priced IPOs, and the uninformed will bid without any prior knowledge. This uninformed bidding imposes a ‘winner’s curse’ with uninformed investors receiving a greater proportion of shares in overpriced IPOs, and a lower proportion of shares in an underpriced, and oversubscribed, IPO. These investors protect themselves against their information disadvantage by demanding underpricing in the average IPO. Indian IPOs provide a good context to test the winner’s curse hypothesis. The book building process is transparent, which allows a researcher to test whether underpricing is related to informed versus uninformed interest in the



Initial Public Offerings  165

IPO. One can test whether the underpricing discount is larger when there is greater participation by informed investors. The winner’s curse hypothesis argues that informed investors will only bid for attractively priced IPOs, and will stay away from overpriced IPOs. In the Indian market, bids placed by informed versus uninformed investors in the book building process is public information. Khurshed et al. (2010) find no support for the winner’s curse hypothesis as underpricing is high even in issues where there is no institutional participation. Even so, there is evidence that asymmetric information explains some of the underpricing. Deb and Marisetty (2010) study whether IPO grading affects the underpricing discount. As described in an earlier section, there is a unique process in India that requires IPOs to be graded by credit rating agencies. IPOs that receive high grading are underpriced less and enjoy greater liquidity in the secondary market. Signalling is another theory of underpricing that relies on asymmetric information between issuers and investors. According to the signalling theory, issuers are better informed about their true value than are investors. High-quality issuers have to distinguish themselves from low-quality issuers during the IPO if they plan to receive their full true value in subsequent rounds of financing. Underpricing is a tool used by high-quality issuers to signal to investors their true value. The usefulness of underpricing as a signal lies in the fact that it is costly for low quality issuers to mimic. A low-quality issuer suffers the cost of underpricing at the IPO and runs the risk of being discovered as a cheat before it can issue equity in a subsequent post-IPO financing stage. In India, signalling would appear to be unnecessary as higher quality issuers are allowed to issue shares at a premium, while lower quality issuers are restricted to par issues. We should expect this differentiation that occurs at the very time of issuance to reduce underpricing, but in fact, it has been observed that par issues are underpriced to a much higher extent than are premium issues. Clearly, the higher underpricing for the par issuers cannot be attributed to these issuers signalling that they are better than the premium issuers. There has to be another explanation for this cross-sectional difference between par and premium issuers observed in the Indian market. Signalling is supported by Ghosh (2005), who finds that firms that follow up an IPO with a subsequent equity offering underprice to a greater extent. He asserts that firms with high growth potential who expected to come back to the market to finance subsequent investment expenditures were willing to take a lower offer price at the IPO to signal their higher quality over small firms who were in the market to exploit investor optimism.

166  Padma Kadiyala

When it is investors who possess superior information, Benveniste and Spindt (1989) argue that underwriters offer underpricing as an incentive to induce investors to truthfully reveal that they want to purchase shares at a high price. Investors who bid conservatively during the book building process face the threat of being allotted too few shares; underwriters favour investors who bid aggressively. Underpricing is necessary to encourage aggressive bidding; otherwise truthful reporting is not attractive to investors. Aggressive bidders are rewarded with higher share allocations of the IPO. As described in earlier sections, the book building process in India differs from the process in the United States in some important respects, one of which is the inability of an underwriter to discriminate among different investors in the allocation phase. Bubna and Prabhala (2010) find that bookbuilt IPOs were underpriced by a smaller magnitude in the period prior to November 2005, when underwriters were allowed to discriminate among institutions in their share allotments. The negative relation between being book-built and underpricing disappears in the post-November 2005 period when underwriters lost the ability for discriminatory share allocation. It must be pointed out that another study by Gopalaswamy et al. (2008) finds that underpricing discounts are comparable for fixed price (29 per cent) and book-built offers (25 per cent) for IPOs issued between 1999 and 2004, a period when discriminatory allocation was permitted. Book building in India also differs from the United States’ process in that bids are binding in the Indian process. Investors are legally bound to take up their allocations in India. In the United States, where there is no formal bidding process in the book-building stage, demand is only indicative and is non-binding. Even so, Benveniste and Spindt (1989) argue that investors are reluctant to bid aggressively unless the issue is underpriced. This reluctance to bid should be greater in the Indian system where the bid is binding. So, one would expect higher underpricing in Indian IPOs. Unfortunately, to date, there is no study that formally compares the magnitudes of underpricing in the Indian and United States IPO markets. The second category of explanations for underpricing is institutional features of the IPO market. These include the threat of lawsuits, after-market trading and taxes. Tinic (1988) argues that investors are likely to sue if the price in the market after trading resumes falls below the IPO offer price. They also argue that the probability of a lawsuit increases with the offer price. That is, if the market price is lower than the offer price, then the probability of a lawsuit increases with the offer price. Underpricing is required to reduce the probability of a lawsuit. It is also required to reduce the probability of an adverse ruling were the lawsuit to be filed, and to reduce the damages awarded



Initial Public Offerings  167

in the event of an adverse ruling. Investors in IPOs in India are not allowed to sue the issuer or the underwriter. Hence, the threat of lawsuits cannot be a likely explanation for the underpricing phenomenon in India. A second institutional feature is after-market support offered by underwriters. Underwriters commit to supporting the price after trading begins if the price were to fall below the offer price. The expensive commitment encourages underwriters to underprice the offering, which should reduce the probability that the after-market price ends up falling below the IPO price. Other after-market activities have also been put forth as explanations. It has been argued that greater underpricing attracts greater trading volume once the issue commences trading. Greater revenue from trading compensates the underwriter for the underpricing. After-market support was allowed in the Indian market starting from 2003, when underwriters in India were permitted to hold the Green Shoe option, which gives underwriters the option to buy an additional 15 per cent of the issue. Price stabilization is allowed for up to 30 days after the issue is listed on an exchange. How this change affected underpricing is not clear since there are no studies that examine this issue. Control considerations have been put forth by Brennan and Franks (1997). Firms issuing IPOs are reluctant to cede control to outsiders. The threat of ceding control through a hostile takeover is lower when there is greater ownership dispersion. Further, when shareholding is dispersed, there is less incentive to monitor management. IPO issuers can strategically allocate shares to investors who are unlikely to threaten their control of their firm by underpricing their issue. Underpricing generates excess demand, which enables managers to ration share allotment such that investors end up holding smaller stakes in the business. A variety of mechanisms exist to allow transfer of control away from existing management in the United States. A well-functioning capital market and strong laws to protect shareholders’ rights allows the market for corporate control to thrive. As a case in point, the SEC now allows investors owning three per cent of a company to nominate directors on corporate ballots. The new law is meant to help shareholders evict board members who wilfully ignore the interests of shareholders. In India, the market for corporate control is still in its infancy. The enactment of the new Takeover Code, which went into effect in February 1997, based on the recommendations of the former Chief Justice of India, Justice P. N. Bhagwati, has created the right conditions for a vibrant takeover market to develop. Even as the takeover market grows in sophistication, there are realities that have to be acknowledged. Corporate ownership continues to

168  Padma Kadiyala

be concentrated within a few large family groups. These groups also exercise control over their subsidiaries through shareholdings and directorships. Family-owned businesses play a valuable role in an emerging market where capital markets are not fully developed, and legal institutions are not able to provide protections to uphold the sanctity of contracts. Familyowned groups do not have to rely on external capital markets, but can turn to internally generated funds to invest in new enterprises that they can then directly control. Placing family members in key managerial positions also reduces the risk of subjecting firms to fraud and to misuse of funds. Thus, family-owned firms in India benefit from stability and the commitment to keep a watchful eye on the management of the firm. Underpricing is especially useful in IPOs of subsidiaries controlled by large family-owned groups. Indeed, Ghosh (2005) finds this to be the case. Underpricing is higher when the IPO is a subsidiary of a large family group of companies. Although Ghosh (2005) interprets this evidence as signalling of future intent to issue shares, it may also be interpreted as a mechanism to control share allocations. Greater underpricing should lead to oversubscription, which allows the underwriter and the issuer to discriminate against large financial institutions that may present a threat to the family owned group. Indeed, Prabhala and Bubna find that rationing is more likely for equity investors such as mutual funds and foreign financial institutions. Submission of larger bids by foreign institutions results in greater rationing, which indicates a reluctance to cede control to foreign entities. Oversubscription of IPOs has led to unintended consequences in the Indian market. On 27 April 2006, SEBI indicted two depositories along with Karvy, a brokerage firm, as the main agents behind the demat scam in the IPOs of Yes Bank and IDFC, which were unearthed in 2005. The indicted entities had opened over 50,000 fictitious retail accounts and had used these accounts to make applications for IPOs in the retail category, with each application being of a low value so as to be eligible for allotment under the retail category. Subsequent to the receipt of IPO allotment these fictitious allottees transferred shares to their principals, who in turn transferred the shares to the financiers, directly or through a web of transactions. The financiers, in turn, sold most of these shares on the first day of listing thereby realising the windfall gain from the underpricing discount. Several changes were instituted by SEBI following the demat scam. One of the changes has already been described, namely elimination of discriminatory allocation of shares in the QIB category. Other changes dealt with the requirements to open demat accounts. Demat refers to a dematerialized account, through which investors can buy or sell shares without having to hold the share



Initial Public Offerings  169

certificates. Shares are held electronically with a depository institution, which is itself required to be registered with SEBI. Most trading in the stock market has now switched to the demat form, but for IPOs in excess of `100 million, SEBI requires that shares be issued only in demat form.4 This stipulation had the unintended consequence of fictitious accounts being opened by depository institutions to corner share allocations. To prevent such scams in the future, SEBI now mandates that all demat accounts should be associated with permanent account numbers (PANs). PAN is the sole identification number for participation in trading on the markets, which can be used to identify investors who open multiple demat accounts. SEBI hopes that this new requirement will discourage depositories from opening fake demat accounts.

Long-Run Performance of IPOs The underpricing phenomenon makes the IPO a very attractive short-term investment. But IPOs are not a very good long-term investment. Studies have found that in the United States, the average three-year post-IPO buy and hold return adjusted for the performance of a value-weighted index of NYSE/NASDAQ stocks is –23.4 per cent. It appears that investors are better off holding a well-diversified portfolio of non-IPO stocks. Long-run performance of Indian IPOs mirrors that of United States IPOs. Poor performance manifests in fact over a very short period of one month after listing. Pande and Vaidyanathan show that the one-month return of IPOs listed on the NSE, adjusted for the return to the S&P CNX Nifty 500 is –1.13 per cent with 58 per cent of IPOs underperforming the market. In addition to poor stock performance, IPOs also suffer from poor operating performance in the post-listing period. Mayur and Kumar (2007) show that profitability measured as the return on equity declines by 40.82 per cent from the year prior to the IPO to two years after the IPO is completed. They show declines of similar magnitudes in other measures of profitability as well. Ownership retained by promoters of the IPOs is shown by these authors to be a significant predictor of the decline in post-IPO operating performance. Firms in which promoters retained over 52.12 per cent of total shares outstanding had better operating performance than firms in which promoters retained less. 4 Section 68B of the Companies Act, 1956, inserted by the Companies (Amendment) Act, 2000.

170  Padma Kadiyala

There is controversy in the United States over the measurement of long-run abnormal performance. Long-run returns tend to be noisy which makes it difficult to draw statistical inference about abnormal performance, which is the return generated by a stock in excess of the return predicted by an equilibrium model such as the capital asset pricing model. IPOs, in particular, present another difficulty with measuring abnormal returns as their asset structure evolves as they go through their life-cycle phases. Typically, firms go public in the United States when they are confident of sustaining a period of high growth rate. Whether the expectation of high growth is met by actual performance depends not only on market conditions, but also on managerial decisions. If managers make poor investing decisions, the long-run performance of the firm may suffer, and this poor performance would be reflected in negative abnormal stock returns. Abnormal returns may not be negative however, if the poor performance arises from difficult market conditions. The evidence reported in the previous paragraph about the importance of ownership retained by promoters suggests that at least in India, poor managerial decision-making may be to blame for the poor long-run performance of IPOs. Agencies in the United States do not monitor managerial decisions in the post-IPO period, as long as the stock continues to meet exchange listing requirements, and meets mandatory disclosure requirements. In India, however, SEBI has decided to intervene in the operations of the firm after it goes through an IPO. SEBI has taken a proactive approach in monitoring utilization of funds raised through the IPO. Clause 43 inserted by SEBI in the listing agreement requires newly listed companies to provide a statement every quarter reconciling projected and actual utilization of funds. Raju and Conseicao (2007) find that of the 85 companies they studied, 50 firms were in non-compliance with Clause 43. The authors suggest that the language in the clause is ambiguous, permitting companies to avoid compliance by stating that variations between projections and actual utilization of funds are non-material. Listing agreements on the United States stock exchanges do not contain such a clause, so it is difficult to predict whether tightening the language in the clause will encourage compliance.

Conclusion IPOs are now in the mainstream in every major market in the world. Small companies aspiring to grow gain access to capital through an IPO, which



Initial Public Offerings  171

also brings greater visibility for their products. These benefits have not gone unnoticed as evidenced by a news article (Financial Times 2010), which reported that the Shenzhen Stock Exchange in China attracted the largest number of IPO listings in the world in the first half of 2010. Many of these IPOs were being completed by small growing firms in China. It is noteworthy that the Shenzhen Stock Exchange achieved this distinction even though it is not even the largest Chinese stock exchange. Stock exchanges are recognizing the benefits of IPO listings, and are competing with each other to become the top destination as a listing venue. The challenge for lawmakers in emerging countries is to protect the stakes of shareholders without stifling entrepreneurship by imposing too many rules and regulations on the operations of firms and markets. The United States experienced such a conflict when it enacted the SarbanesOxley Act, 2002 to stamp out accounting fraud by requiring corporate leaders to personally certify the firm’s financial statements. Section 404 of the Act went one step further in requiring auditors to certify the firm’s internal controls. The burden placed by this legislation on the board of directors and on top management has scared away firms that sought to be listed on United States exchanges. Foreign firms that would have listed in the United States chose instead to list in Europe and in other non-United States markets (Ascarelli 2004). Countries will have to devise legislation being mindful of the cultural, economic and political realities on the ground. Fortunately, lawmakers in India can draw on the rich experience of industrial economies with IPOs in dealing with issues that are specific to this security.

References Agrawal, D. n.d. ‘IPO Pricing—Book Building and Efficient Pricing Methodology’, Working Paper, Truba College of Engineering and Technology. Ascarelli, S. 2004. ‘Citing Sarbanes, Foreign Companies Flee U.S. Exchanges’, The Wall Street Journal, Eastern Edition, 20 September. Benveniste, L. M. and P. A. Spindt. 1989. ‘How Investment Bankers Determine the Offer Price and Allocation of New Issues’, Journal of Financial Economics, 24: 343–361. Brennan, M. J. and J. Franks. 1997. ‘Underpricing, Ownership and Control in Initial Public Offerings of Equity Securities in the UK’, Journal of Financial Economics, 45: 391–413. Bubna, A. and N. Prabhala. 2010. ‘When Bookbuilding Meets IPOs’, Working Paper, Indian School of Business. Chen, H. and J. R. Ritter. 2000. ‘The Seven Percent Solution’, Journal of Finance 55 (3): 1105–1131.

172  Padma Kadiyala David, Ruth. 2010. ‘Zero Fees from India Has Investment Bankers Relying on Private Share Sales’. Bloomberg, 5 April. Available at http://www.bloomberg.com/news/2010-04-05/ zero-fees-from-india-has-investment-bankers-relying-on-private-share-sales.html. Deb, S. and V. Marisetty. 2010. ‘Information Content of IPO Grading’, Journal of Banking and Finance, 34 (9): 2294–2305. Financial Times. 2010. ‘Shenzen Takes Over as China’s Listing Hub’, Financial Times, 18 October. Ghosh, S. 2005. ‘Underpricing of Initial Public Offerings’, Emerging Markets, Finance & Trade, 41 (6): 45–57. Gopalaswamy, A., K. Chaturvedi and N. Sriram. 2008. ‘Long Run Post-Issue Performance of Fixed Price and Book Built IPOs: An Empirical Study on Indian Markets’, Journal of Advances in Management Research, 5 (2): 64–76. Ibbotson, R. G. 1975. ‘Price Performance of Common Stock New Issues’, Journal of Financial Economics, 2: 235–272. Khurshed, A., A. Pande and A. Singh. 2010. ‘A Dissection of Bookbuilt IPOs: Subscriptions, Underpricing and Initial Returns’, Working Paper, Manchester Business School, UK. Kumar, S. S. S. 2008. ‘Is Bookbuilding an Efficient IPO Pricing Mechanism?—The Indian Evidence’, Working Paper, IIM Kozhikode. Logue, D. 1973. ‘Premia on Unseasoned Equity Issues’, Journal of Economics and Business, 25 (8): 133–141. Madhusoodanan, T. P. and M. Thiripalraju. 1997. ‘Underpricing in Initial Public Offerings: The Indian Evidence’, Vikalpa, 22 (4): 17–30. Mayur, M. and M. Kumar. 2007. ‘Ownership and Performance in an Emerging Market: Evidence from Indian IPOs’, The ICFAI Journal of Applied Finance, 13(2). Ministry of Finance. 2003. ‘Action Taken Report on the Report of the Joint Parliamentary Committee on Stock Market Scam and Matters Relating Thereto’. Available at finmin. nic.in/the_ministry/dept_eco_affairs/UTI&JPC/ATRMay2003.pdf. Nandha, M. S. and K. R. Sawyer. 2002. ‘Ex-ante Uncertainty in Initial Public Offerings: The Indian Market’, Finance India, 16 (3): 961–976. Pande, A. and R. Vaidyanathan. 2009. ‘Determinants of IPO Underpricing in the National Stock Exchange of India’, IUP Journal of Applied Finance, 15 (1): 14–30. Raju, G. and S. Conceicao. 2007. ‘Projections and Utilization of Funds by IPOs and Compliance with Clause 43 of Listing Agreement at Stock Exchanges in India: A Case Study’, The Journal of Wealth Management, 10 (3): 52–65. Ritter, J. R. 1998. ‘Initial Public Offerings’, in D. Logue and J. Seward (eds), Warren Gorham and Lamont Handbook of Modern Finance. Boston and New York: WGL/RIA. Ritter, J. R. and I. Welch. 2002. ‘A Review of IPO Activity, Pricing and Allocations’, Journal of Finance¸ 57 (4): 1795–1828. Rock, K. 1986. ‘Why New Issues are Underpriced’, Journal of Financial Economics 15: 187–212. SEBI. 2000. Annual Report 1999-2000, Part B. SEBI. Shah, A. 1995. ‘The Indian IPO Market: Empirical Facts’, Working Paper, Center for Monitoring Indian Economy. Teja, R. 1992. ‘Crisis, Recovery, and Transformation in India’ Finance & Development, 29 (4): 31–33. Tinic, S. 1988. ‘Anatomy of Initial Public Offerings of Common Stock’, Journal of Finance, 43: 789–822. Welch, I. and J. Ritter. 2002. ‘A Review of IPO Activity, Pricing and Allocations,’ Journal of Finance, 57 (5): 1795–1828.

10

Legal and Judicial Process in India: A Preliminary Economic Analysis of Some Issues V. Santhakumar

There are many well-known problems of Indian courts and judicial process. Long delays in decision-making and the very large number of pending cases fall in this category. This chapter discusses some of the problems of legal and judicial process.

What Are the Functions of the Court? Resolution of disputes is an important function of the courts, but that is not the only one, especially for courts in countries like India, where a major problem faced in the legal process is poor enforcement of the legal machinery. There can be several reasons for this state of affairs. There may not be adequate investment in the law enforcement machinery, in tune with the extent of law making. Or, there may not be adequate thinking about the investment required for law enforcement before making new laws. Laws are being made (their cost is relatively low) but law enforcement is left open. Thus, people who wish to see that a particular law is enforced well may approach the court. Then, the function of the court is to direct the enforcement machinery to act. There can also be several instances when courts are approached to enforce the judgments of the courts themselves. In one sense, when a law is made and not enforced, but enforced only when people approach court, it is like leaving law enforcement to ‘demand’ (as in the case of demand for a good or service in a market). It is expected

174  V. Santhakumar

that law making should also reflect the demand—not expressed through the market—but through the voice of people through their elected representatives in a democracy. However, such aggregation of social demand in the process of law making is very difficult. It may be possible that laws are made for certain ceremonial purposes or to meet the requirements of some section of society or to meet international requirements (and not to meet internal demands of the majority). Thus, one can expect that there may be some laws made, which are not demanded by people. (The opposite can also be true. There may be some laws demanded by people that are not made, and hence they may use non-state means to meet those objectives. They will be considered later.) When this is the case, not enforcing law all the time or enforcing it only when people demand them by approaching court may enhance social welfare. Thus, some are allowed to live without that law (since law enforcement is weak even though the law exists), if they do not approach the courts to enforce it. The other function of the courts, which is widespread everywhere, is the interpretation of statutes. Usually this interpretation is made with regard to the Constitution of India (to see whether any law made is in violation of the Constitution). Through this process, courts may end up making the law. Even otherwise, following the common law traditions, the precedents established by past decisions have a bearing on judicial decision-making. How does this affect the efficiency of judgments? The laws made at a point in time, including the Constitution, reflect the socioeconomic and ideological context of that particular time. Broadly, if law is considered part of the institutional framework, such institutions should be linked organically to the underlying economic and technological conditions. Thus, when there is a change in these economic and technological conditions, there should be changes in the institutional framework. For example, when telephony is based on landlines, there is an economic issue of natural monopoly here, and hence the legal provision in this regard should be the one facilitating and regulating this natural monopoly. However, in the era of mobile telephony, landline-induced natural monopoly does not exist. The new legal framework then should be one facilitating competition in telephony. If, however, the previous legislation exists without any change, it may make the evolution of mobile telephony costly and hence inefficient. Thus, one would expect the legal frameworks to change as and when underlying techno-economic conditions change. It is all the more important when it is noted that a law, once made, is likely to be in place for all time, despite changes in conditions that make the law obsolete. Debroy (2000) provides a detailed account of such laws. Altogether there may be between 25,000 and 35,000 laws in force in India, instituted by



Legal and Judicial Process in India  175

the central and state governments.1 All these laws are legally valid, since India has not accepted the legal principle of desuetude—if a law is not enforced for a long time, it is considered by the courts as having no legal effect. Thus, in order to invalidate the law it has to be ‘legally’ removed.

Some Factors That Determine the Number of Cases in the Courts Some factors that determine the number of cases in the courts are outside the legal or judicial system. For example, the level of economic development or certain social factors (that may lead to more or less conflicts) may have a bearing on the number of cases in which one of the parties may want to seek the assistance of the court. Such factors are not considered here. But there are certain factors that are endogenous to legal and judicial process that may influence the ‘demand’ for courts. Just like any other good or service, its price may be one among several factors that may influence the demand. Here, the price for a plaintiff is the cost of taking the service of the court. This will include the costs for filing the case, charges for hiring advocates, direct and indirect costs—opportunity cost, say by foregoing other income generating activities—of appearing before the court for a certain number of days (which may also depend on the distance to the court). Thus, all these put together constitute a significant expenditure that has a bearing on the number of cases. The more this expenditure is, the lesser the number of cases would be (with a given level of other potential factors). Thus, even when the filing cost is kept very low in India, with an objective to increase the accessibility of the lower income groups, if there is a very long delay in obtaining court decisions (as the case in India), the total cost to plaintiffs can be higher. This is so since plaintiffs may have to spend more time on the cases, and also may have to pay more to the advocates, if payment is made for each appearance. (Such a practice is also common in India.) When this total cost or the ‘price’ of getting court service is higher, plaintiffs will approach the courts only for those cases for which the expected benefit is greater than this cost. That would mean that certain cases (violations, breaches and injuries) might not reach the courts. There can be a social cost due to this. Either people will take excessive precaution in avoiding such cases (including avoiding mutually beneficial exchanges) or the possibility of 1 These include regulations/rules and those made by the British government before Indian independence.

176  V. Santhakumar

such uncontested violations may encourage people to divert their resources to ‘redistributive acts’ that are not socially beneficial. It is argued elsewhere that when one person infringes on another’s property, breaches an agreed contract without adequate compensation, and injures accidentally without giving damages, these are mere redistributive acts—zero sum games—and these do not increase welfare. Thus, the price of the court service may have to be decided on the basis of comparing the social cost when some conflicts do not come to the courts (when the price or cost is kept high) and the increased administrative costs when many or almost all conflicts reach the courts (when the price or cost is kept low). Or, a case can be encouraged to reach the courts (by suitably changing the cost of court service), only when the marginal social cost of not doing so is greater than the marginal administrative cost incurred by the courts in admitting such a case. The delay in court decisions can also be viewed in another way. Since court fees are not increased in consideration of accessibility to courts, governments do not have enough money (considering other public needs for which government money has to be spent) to increase the number of courts to the level demanded by society corresponding to the low court fees. (The demand for a service is to be seen for a particular price.) Thus, there is excess demand, considering the supply of services. In such a case, rationing is needed. One way of rationing is through allowing long queues. Thus, there is a long ‘queue’ before the courts. Hence, even when people do not pay directly through court fees, they pay indirectly by spending time due to the long delays in the courts.

Technology in Producing Court Services The court delays can also however be due to some ‘inefficiency’. Efficiency means the minimum possible input (in terms of cost) that is taken to produce a given output. This depends on the technology of production of a service. Court decisions take both labour (say, that of judges) and probably capital. For example, capital embodied in the form of, say, computers can be used to categorize cases, pool together evidence, provide easy referencing of previous cases for judges, etc. It is expected that an appropriate (cheapest) combination of capital and labour be used in the production of a service (of a given quality). Are Indian courts using such an appropriate combination of inputs? Although it is said that a large number of cases are pending before Indian courts, one is not sure whether the number of cases before the court is high



Legal and Judicial Process in India  177

or not. It is not known what would be the number of cases if there were less delay. This is because the delay may discourage some people from approaching the court (even when the formal fees in courts are kept lower for reasons of accessibility). Thus, when such delay is reduced there may be an increase in the number of cases. However, delay in courts and its differential impact on plaintiffs and defendants may create some incentive to use court cases (not judgments) as a means to punish some party, and this will be discussed later, along with what is called nuisance litigation.

Incentive to Settle Out of Court or Use Alternative Dispute Resolution (ADR) Methods In the discussion on theory (in Volume 1) it was seen that negotiated outcomes between parties are efficient since they are mutually beneficial (and may lead to the achievement of potential surplus from the exchange). But transaction costs reduce the scope of such negotiations. Formal conflictresolving mechanisms are likely to incur other costs for the society. Hence there will be a search for informal mechanisms, which have lesser direct costs, and which can also facilitate mutually beneficial negotiations better. Here, the discussion is limited to the impact of delay in court decisions on the incentive to use to alternative dispute resolution (ADR) forums. Two parties in conflict may decide to negotiate and settle out court without going through trial. In certain legal jurisdictions, such bargaining is very much part of the formal legal process (as in the case of settlement bargaining in the United States). But in India, though there may be many cases which are settled without going to court, it is seen as part of an informal conflict resolution process. These days there are efforts to promote ADR methods as part of legal process within India (Panchu 2007; Srikrishna 2007). Before discussing the details of such ADRs promoted in India, the question of why some parties in conflict may want to resolve their disputes out of court will be dealt with. Usually, trials in courts involve litigation costs for the parties. The costs are likely to be higher than the transaction costs of settlement even if they engage a mediator or neutral decision-maker or an arbitrator. There are also other costs of trial, which can be avoided through settlement. Bad publicity, loss of reputation and the possibility of damaging personal relationships between the parties who go through the adversarial trial process are some

178  V. Santhakumar

of these costs of trial. These can also be avoided through out-of-court settlement. However, the reduction of costs of trial by itself may not encourage one of the parties to opt for out-of-court settlement. This depends on what she expects to get from the trial process. What each of the conflicting parties get in court can be taken as their gains from not agreeing to settlement. Hence, these may become the threat values in their negotiation during the settlement. The plaintiff will not agree to anything less gainful for her and the defendant will not agree to anything more harmful to him at the time of settlement, than what they would expect to get at the court during the trial process. The role of ADR mechanisms is now seen. The ADR employs a mediator or a neutral decision-maker. The mediator is likely to give both the parties a reasonable picture of likely judgment. This may lead to similar expected probabilities (of getting a particular decision in court) for both the defendant and plaintiff. Hence, one important role for the ADR mechanism is to reduce what is called ‘relative optimism’. On the other hand, if the case is such that (or the contesting parties are such that) there exists relative optimism which cannot be corrected through the mediation or the neutral third party engaged, then ADR may not be effective in such cases.2 There may be other factors that influence the incentives to use ADR, and these can be interpreted within this framework. For example, the publicity produced by trials and the associated loss of reputation (say for a married couple that is seeking divorce) increase the cost of trials. That would mean that the gains (or the cost avoided) by opting for out-of-court settlement increases. This may enhance the incentive for such settlements. The delay in getting judgment may also have an impact on the incentive to settle out of court. All negotiations take place in an institutional context. The context determines the exit options for the negotiating parties. For example, if A is not willing to accept the offers made in a negotiated settlement, then what A gets (either as compensation or punishment depending on whether she is the plaintiff or defendant) is determined by the court. However, when the court decision is delayed (and there is no indexation of the court-determined award or punishment based on the duration of delay), the effective value of award or compensation is reduced. This is so since `100 received today is likely to be more valuable than `100 received after two years for the plaintiff. 2 The absence of such a theoretical understanding on why certain cases are more amenable to resolution through ADR is a limitation of the treatment of this subject in other writings, especially by lawyers like Panchu (2007).



Legal and Judicial Process in India  179

(Similarly `100 to be paid as damages today is likely to more costly for the defendant, than `100 to be paid after two years.) For an example of how delay affects the incentive to use ADR methods, assume that the plaintiff expects an award A with a probability p by filing a case. C is the cost of trial. The delay affects here in two ways. It reduces the discounted value of A (since `1 received today is more valuable than `1 after some delay). The delay also increases the cost of trial, even if there is no increase in court fee or even lawyer’s cost (if the lawyer is paid a fixed fee for the case). This is so due to the opportunity cost of time. Because of these two effects, the expected value of the judgment comes down if there is a delay. On the other hand, the impact of delay on the defendant is somewhat different. The cost of trial may go up due to an increase in opportunity costs (if not due to an increase in lawyers’ fees). On the other hand, the expected value of his loss may come down due to delay, since he needs to pay the compensation only at a later time (and the court is not taking such delay into account). If the compensation that he has to pay dominates the trial cost, the net effect on the delay is a reduction in his expected value of the loss. (It may be safe to assume that the delay is unlikely to influence the probability of a particular decision in the court, and hence the expected probabilities for both the plaintiff and defendant.) The reduction in expected gain of the plaintiff (due to delay in the court decision) may increase his incentive to settle out of court. But the reduction in the expected loss of the defendant (due to delay) may reduce his incentive to settle out of court. Hence the impact of delay on the likely use of ADR cannot be predicted ex ante. In the way that delay gives an incentive to use appeals to higher courts as a strategy—an incentive not to accept the decision by the lower court as final—there can be an incentive not to accept the verdict from ADR forums. This is known to different parties—thus, one party knows that the other has an incentive not to accept the verdict of ADR, and hence may not even take the effort to approach such informal mechanisms. This is especially so when the use of ADR is voluntary, and when ADR is not final and cannot impose its decision through an enforcement machinery.

Public Interest Litigations Citizen suits or what are called public interest litigations (PILs) are used widely in India. In a PIL, the person approaching the court need not be one among

180  V. Santhakumar

the affected persons. Thus, an industry polluting an area may be affecting a number of residents there but someone from outside (a concerned citizen or a non-governmental organization) may approach the court on this issue. The issue at hand need not be even one where the affected parties can be identified as easily as in a pollution case. For example, cutting of trees on a piece of land can be objected to in court by an unrelated individual, for the possible impact of such action on biodiversity. The real affected party of this loss to biodiversity could be future generations living all over the world. The PIL need not be only on issues of environmental conservation. It can be related to other social issues as well. The impact of a change in government policy on the poor (and the possible violation of rights in this regard on some constitutional grounds) can be challenged in court by a person who is not poor. The cost of filing a PIL petition is also very low, if not nil. Even if a letter sent to the court, it can be taken as a basis for starting this judicial process. There are many positive dimensions to PILs. First of all, there may be some cases where the affected party is not immediately available (like the biodiversity) case, and hence a PIL provides a window to sort out conflicts in this regard. Or, violations of the rights of those affected parties come to light through this process. There is also a perception that many people in countries like India are ignorant of their rights, do not have information on the implications of violation of certain rights, and even if they know, cannot afford to take such issues to the courts. The PIL is seen as a partial solution in this regard. In the absence of such efforts, it can be argued that one can expect to see more right violations against those who are unavailable (say future generations), or those who cannot afford the costs of the legal and judicial process. This need not be seen as a mere redistribution issue, but may have an impact on aggregate social welfare (when such welfare is considered as some form of aggregation of that of different generations). The PIL is expected to depend on the social concern or the altruism of the person taking the effort to file or proceed with the case. Though many people can be assumed to have an aspect of altruism, there can be some ‘rational’ patterns in this regard. For example, if the cost of PILs increase, people may be less willing to make such altruistic efforts (for a given level or distribution of altruistic preferences in society, and for a given set of other factors). Thus, keeping the cost of PILs low is to encourage many socially concerned people to take this step. There are however some negative aspects to PILs. First of all, the courts may find it difficult to limit the scope of PILs to those set of issues for which they can be socially beneficial—where the affected parties are not immediately



Legal and Judicial Process in India  181

available or identifiable, or when they cannot afford to go through legal and judicial process. Second, the parties filing the PIL may gain pecuniary benefits (not only that, they get the benefit of the lower cost of such litigation). For example, the need to become ‘noted’ or develop experience as a lawyer is not insignificant in a context like India where many junior lawyers suffer from lack of cases to develop expertise. Such pecuniary gains may lead to a situation where there can be excessive (beyond optimal, based on the social costs and gains) number of PILs. However, the greater problem is when a PIL can become a nuisance suit, or when it is deliberately used to inflict a cost on the party against whom it is filed. The fact that many impacts or effects against which a PIL is filed are intangible or indirect and the assessment of such impacts would require considerable expertise from outside the judicial system, and also that such expertise can be contested on ideological grounds, may cause inevitable delays in the delivery of court decisions on these matters. Even in normal cases, there is considerable delay in delivering justice by Indian courts. The possibility of delay, which itself can impose a significant cost on one of the parties to the case, can be a factor facilitating the misuse of PILs. This issue is taken up in the following section.

Nuisance Suits and the ‘Public Interest Litigation Mafia’ There are certain cases where the plaintiff is not a real affected party, or is one who has not incurred any real losses. Such a plaintiff cannot expect to win the case in courts, and thus may lose the expenditure incurred in approaching the court. However one does see such cases. What could be the rationality behind them? If one such plaintiff A files (or threatens to file) a case by spending `1,000, knowing that he cannot expect any benefit normally from the legal process, and can only expect to lose this amount. The defendant B in this case may have to spend some money to argue the case. Assume that B loses `10,000 due to the case. This includes the direct cost of defending the case and probably for delaying certain activities due to the case. (One can see B as a builder or one implementing a project.) Of course in the end, B will win, but he still loses `10,000 (the cost due to the case). Assume that both A and B know the costs, not only their own, but also the other’s. Thus, if B initiates an out-of-court settlement, he may be willing to pay some money

182  V. Santhakumar

to A. This is so since he can avoid the case, say by paying `5,000 to A. Even after this is paid to A, B is not that worse off compared to what he would have lost by defending the case. As for A, he could make `5,000, by showing a willingness to spend `1,000. Here the court case is intentionally used to get some money from the other party, even when there is no real (winning) case. The fact that the defendant loses money due to the case is used by some unscrupulous elements to make money. This is a called nuisance suit. It is apparent why this may not happen in a normal case. If A has to spend `1,000 on the case and B has to spend the same amount to defend it (and there are no other losses), what B would willing to pay would be some amount less than `1,000 (which is the maximum loss that he would incur by going through the case). Thus, the amount that A would receive would be less than what he has to spend to go ahead with the (losing) case. In the end, A will certainly lose some money. Thus, he has no incentive to go ahead with the case. This is known to B, and hence B is not willing to offer anything to A. (A’s threat to file the case is not credible). Thus, only when the defendant has more to lose than what the plaintiff has to spend by going through the case, is there a scope for nuisance suits. If the case is shown to have no basis, as evident from court procedure (and there is some indication of a nuisance suit), the plaintiff can be asked to pay compensation to the defendant. This may reduce the incentive to file such nuisance suits. Or, the filing of such nuisance suits can be made punishable—where such plaintiff not only bears the costs, but can also be made to pay fines. What about public interest litigations being used as ‘nuisance suits’ with the intention of making money or taxing the other party? The two features of public interest litigation that increase the possibility of such cases becoming nuisance suits are: (a) the person filing the case need not be a directly affected party and (b) usually the cost of filing the case is kept low; even a letter to the judge can be taken as adequate for registering the case. First of all, the plaintiff need not demonstrate any direct harm to file the case. Otherwise, this could have been used to weed out non-serious cases at the stage of preliminary hearing. Second, the fact that filing cost is very low, the loss to the plaintiff, even when the case is rejected can be very low. (There is a need for an advocate to argue the case, but the costs in this regard need not be very high, if a junior lawyer is used, or if the plaintiff himself is a lawyer. Many not-so-busy lawyers may use PILs as windows of opportunity in India to develop a name in the profession.) Usually such PILs are filed against parties involved in the implementation or operation of projects or industrial firms, and any stoppage or delay of work can be



Legal and Judicial Process in India  183

costly to these parties. Thus, these parties may have an incentive to think about an out-of-court settlement in such cases, by making some payments on the side to the plaintiff. This is known to the plaintiff, and can become an incentive to file many such cases, and this can develop into a mafia for making money. Within a PIL context, a nuisance case need not be just for taxing the defendant with money. There may be ideological positions that may prevent an out-of-court settlement by taking side payments. Thus, the sole objective could be to impose a cost on the other—the adversary or enemy. If it is less costly to oneself to impose a cost on the enemy, rational individuals may have a tendency to do so. Within a conventional notion of rationality, people may be called irrational, if they want to impose a cost on other, by bearing a significant cost on oneself. (This is not to imply that society need not have such ‘irrational’ individuals.) However, when the cost of imposing a cost on the enemy is low or minimal, then one can see an increased use of this option (when the level of rationality or the distribution of ‘rational’ people is unchanged in a society). Thus, the low cost of PIL may increase the frequency of nuisance cases, which aim at imposing a cost on another (and not to tax money as part of an out-of-court settlement). Hence the probability of misuse of PIL to cause a nuisance is high, if courts are not vigilant enough to discourage such litigants, either by disposing of the case early having seen the case as a ‘nuisance suit’ or by discouraging such litigants more explicitly.

Impact of Delay in Public Interest Litigations and Nuisance Suits It has been seen that even the normal time taken for delivering justice can be costly to some parties, and this can be misused by filing nuisance suits. Moreover, the possibility of PIL (where one does not have to demonstrate direct damage and the lower cost of filing case) may encourage the possibility of nuisance cases. Indian courts are known to exhibit considerable delay in delivering justice. The impact of this delay is that the cost imposed on clients involved in projects or firms for whom the work may have to be stopped during the judicial process can be very costly. Thus, the incentives to use court cases to cause a nuisance (either by taxing money or simply to instil a cost) to the other party) can be higher.

184  V. Santhakumar

The delay can create some other perverse incentives. For example, the party that is not happy with the judgment at one level of the court can use appeals to higher levels to delay the implementation of a court verdict. This is so since each appeal and the decision-making at this level can cause significant delay in implementing the original decision (even if appeal leads to the decision taken at the lower level being upheld). Such strategic use of court delay (using appeals at different levels) has been noted in India. Delay can impose costs asymmetrically on different parties. An example is presented to elaborate this case. Public interest litigations are being used widely for issues of pollution and environmental conservation. Think about a firm currently creating pollution. If people, agitated by this pollution, file a PIL, any delay in this regard in court is favourable to the polluter, since it can continue to pollute until a decision is arrived (if there is no immediate order to stop the functioning of the factory). This polluter may be even tempted to use appeals as a strategy to delay the final verdict of the court. On the other hand, assume that a new factory is planning to establish operations in a less industrialized area (or a place with no industry). Even if the factory plans to establish adequate pollution-control mechanism, the local people may have an interest in taking the firm to the court, if there is a delay. This can be analysed in more detail. The pollution control rule in most contexts is that, the level of pollution should not be more than a particular level—which is considered socially acceptable level. (This may be arrived at based on scientific information.) However, this level is unlikely to be ‘no pollution’. When pollution is at a very low level, the social harm can be small. Considering the benefits of the economic activity that produces the pollution, bringing it down to zero may not a desirable situation. That is why, it is rational to have a socially acceptable level of pollution, beyond which it is banned or restricted by the state. The problem is that such a socially acceptable level of pollution can be an increase in pollution (from no pollution to some pollution) for those people living in a non-industrialized area. Hence, they have an incentive to avoid it (but cannot do it in a well-functioning legal context, since if they complain, the law enforcement machinery will only see that the factory pollutes only within the socially acceptable level of pollution). The chance of delay in court gives them an opportunity. They may file a public interest case. The delay in getting court verdict can work against the firm—it can establish the operation only after getting the court verdict in its favour. The people may have an incentive to use appeals to delay the final verdict from being executed as part of the judicial process, since they can benefit from zero pollution until then.



Legal and Judicial Process in India  185

However this delay can be very costly for the firm even when it intends to start with legally mandated pollution control.3

Efforts to Reduce Delay There have been many attempts in India to reduce delay and make judicial process more efficient. The amendments in Code of Criminal Procedure Act in 2008 aim at providing some of these improvements. Section 309 has been amended to ensure that no adjournment is granted at the party’s request. Lawyers being in another court cannot be a reason for adjournment. The amendment also stipulates importantly that if a witness is present in court when one of the parties or his lawyer is absent, the court can examine the witness and pass orders without waiting for the other side to complete the cross-examination. It is necessary to understand that the lawyers do not have the incentive to avoid delay in many cases in India, since they are not paid per case. They are more likely to get paid for appearances before the court. Thus, they have an incentive not only to increase the number of appearances but also postpone the case, if they have some other case on the same day. Thus, seeking adjournments is a very common practice in Indian courts. Judges do not have the incentive to finish the case in time, and their payment structure only encourages them to minimize their time per day and they do not have an incentive to discourage adjournments. This not only delays the case, imposing significant costs on the defendant and the plaintiff. Much more than that, the practice of frequent adjournment enhances the cost for witnesses. It may be noted that the costs of prosecution witnesses to appear before the court in criminal cases is not reimbursed. (Though, buying out witnesses by the opposing side is not uncommon in India and the Indian judiciary is seriously considering how to penalize such hostile witnesses.) In such a situation, where witnesses’ costs are not reimbursed, their reluctance to be witnesses increases if such costs are likely to be increased due to frequent adjournments. Moreover, when such a witness appears before the court, none of the parties have an incentive not to adjourn the case (since they do not bear the cost of appearance of the witness). Thus, it is not surprising that in general people are very reluctant to be witnesses in criminal trials. 3

A detailed treatment of this issue can be seen in Santhakumar (2003).

186  V. Santhakumar

Thus, it is seen that though it is socially costly to have frequent adjournments, the lawyers have an incentive to do so. Thus, it was not surprising that lawyers across the country opposed the amendments to the criminal procedure code. There were national strikes by the lawyers. This is not to argue that all the amendments to minimize the cost of frequent adjournments are the best response to the current situation. However, it shows that a deeper analysis of the incentives of the lawyers to oppose reforms is needed and mere legal restrictions on unwanted adjournments may not be adequate.

Addressing Possible Corruption in the Judiciary Analysing the incentives for judges to indulge in corruption and the mechanisms to minimize them need to be part of law and economics in developing countries like India where corruption is not rare in all walks of public life. Incidents of corruption and its explicit forms like amassing wealth disproportionate to known assets, possibly through decisions as part of the job, are not very rare in the case of the Indian judiciary.4 The corruption needs to be addressed through methods of appointment and also methods of ensuring accountability while in service.5 The appointment is first considered. For judges in lower courts, merit-based appointment with considerations of social identity (as part of affirmative action) is likely to minimize the scope of biases, provided such process of appointment itself is free of corruption. For judges in the higher levels of courts, broadly, there are two ways of appointment: (a) by the executive and (b) peer selection by (senior) judges. There can also be consultation or joint role for both judiciary and executive in appointing judges. The usual problem attributed to executive appointment is that such decisions are likely to be influenced by the political machinery. Thus, the political biases (in terms of ideology or personal) may influence the selection of judges. Moreover people who seek jobs under such a situation and get favours from the political system may continue to reward political patrons while being in service. However, other accountability measures may be used to see that the political system does not appoint corrupt, ineffective 4 The events leading to the withdrawal of the nomination of Justice P. D. Dinakaran to the Supreme Court of India is one such recent case. For details, see Bhushan (2009). 5 Such measures of accountability may include dealing with post-service periods to see that inducements for post-service life do not influence the actions of judges during service.



Legal and Judicial Process in India  187

or incapable judges. Confirmation hearings as in the United States are one such strategy. Even otherwise, public discussions on such appointments initiated by the media may act as a restraint against the abuse of this power by the politically influenced executive. It is partly to avoid this problem that the judiciary has been given some role in the appointment of judges in India. It is difficult for the non-judicial world to assess the quality of judicial officers, and hence it is better to give the power to those who have that information advantage. It is not unusual to see such peer selection process in other occupations as in academic jobs, and it is the information advantage of insiders (and the disadvantage of outsiders) that justifies such an appointment process. However, history matters in this case. If in the past judges were not selected on the basis of ability, and there were certain biases in their selection, such features may influence the selection of newer judges. Moreover there may also be some equilibrium effect here. If the judges have a particular level of ability (integrity, legal knowledge, etc.), they are likely to select judges of that type. Thus, ‘lower’ quality peers cannot be expected to select ‘higher quality’ people. (Such a problem is visible in other occupations where peer selection is prevalent, as in academic jobs.) The advantages of peer selection cannot be avoided. However, in order to ensure that peers select qualified ones on the basis of certain principles, and not on the basis of nepotism (and for conforming to the practices of senior judges), the selection process as well as the merit and other qualities of judges have to be in the public domain. Thus, the assets of judges of different levels can be made public from time to time, so that those with disproportionate increase in assets can be identified and eliminated for selection to higher levels. The peer selection should also be based on information pooling within the judiciary, so that those peers having closer information on the performance of a particular judge should get an opportunity to share that information with those involved in the selection process. The criteria used by the peer group can be made objective (to the extent possible) and be made public. However, there is a suggestion for an independent mechanism for the section of judges to higher levels of courts. The benefits of such mechanism are not that obvious. It is not clear whether they would have the information advantages as those in the case of peer judges. Moreover, independent regulators may only be as good as how they are selected. The problems of political selection of such regulators cannot be wished away in countries like India. The crucial issue regarding corruption is how to ensure accountability when the judges are in service. Here too, peer pressure and punishment by peers when somebody errs can be an important strategy due to the informational advantages. However the incentives to retain status quo is

188  V. Santhakumar

likely to be much higher in this case, and if the status quo is one where there is a significant level of corruption, then corruption cannot be minimized through peer group pressure alone. There are some ‘externalities’ to peers due to the actions or decisions of a judge. This externality operates when one judge takes action against another judge for corruption, by affecting the reputation of the former (or other judges). If the tendency of judges is to hide the level of corruption and malpractices among the judiciary from the public, then the incentive for the peers is to take less stringent action against the corrupt judges. However, if one judge’s corruption is perceived to affect the good reputation of other judges, then peers may have adequate incentive to take appropriate action against erring members of the judiciary. These contrasting cases may depend on the extent of corruption (or proportion of corrupt versus non-corrupt judges) prevalent in a judicial system. When there is a significant level of corruption within the judiciary, the tendency among the peers may be to hide it, where as when corruption is very infrequent, the incentive may be to control it fully so as to avoid tarnishing the good reputation of other judges. The removal mechanism that exists in India is to impeach higher court judges requiring the support of two-thirds of the members of parliament. (The case needs the support of at least 100 members of parliament to bring it before the Parliament.) Such a difficult legislative procedure is envisaged to ensure independence, since such a rule will make the removal of judges at the whims and fancies of the executive or the political ruler very difficult. However this difficult rule can also make genuine action required against erring judges difficult to implement. This can also make action against judges a political issue—over which coalitions can be formed on the bases of issues and identity other than the merits of the case. There may be a need for taking action against erring judges without placing such issues in political domain. However, corruption in the judiciary depends to some extent on the quality of democracy and the functioning of other institutions.

Social Norms and Practices, and Legal and Judicial Process The role of social norms in the place of law is an issue that requires very detailed and longer treatment, and it is not attempted here. Only the role of social norms in the context of some problems of legal process like delay is considered.



Legal and Judicial Process in India  189

Social norms and community practices work in the following important ways: 1. There are certain rules of behaviour internalized by people as the right or wrong ways, and here the punishment for breaking the norm is shame or self-pity. 2. Another form of enforcement of norms is through community networks, where one member A decides not to interact with B for not following the norm, or C decides not to interact with A if she is interacting with B despite the latter’s violation of the norm. 3. Instruments of an informal state are used to impose correct behaviour through explicit punishments, which include taking away property, or even the life, of the person who breaks the social norm. These were the main forms of institutional enforcement that facilitated mutually beneficial and surplus-generating exchanges in all contexts where formal state and law enforcement machinery were not developed or fully functional. Thus, the residuals of such practices exist in all societies. There can be a dependence on such social norms when formal law enforcement becomes costly (due to delay). Some of the strategies used, like the punishment meted out by the informal state (armed men appointed by moneylenders to recover money and property and for thrashing out the people who do not repay the debt in time), are in fact breaking the law, which may invite punishment from the formal state. However, the delay in court decisions and its impact of reducing effective punishment may make such formal punishment of informal enforcers (like armed men) less costly. Thus, there can be two incentives to depend on informal punishment when the formal system is costly. The deal breakers cannot be punished effectively by the formal system, and hence the victims of such deal breakers have an incentive to use informal punishment. Moreover, informal enforcers are likely to get less effective punishment, and hence there are likely to be more active. There are also other features that may encourage people to follow social norms. People who have internalized norms (like feeling shame while breaking an agreement, or while not repaying debt, etc.) feel the pressure to follow such norms only when the deal and its enforcement are operationalized through the social norms. If a moneylender who has been traditionally using the social norm to get back his debt, starts using courts to enforce his lending contracts, those who are in debt may lose the pressure of ‘shame’, and may use courts to their advantage for not repaying money if that is advantageous. Thus, they may use court delays or errors to their advantage, since in the

190  V. Santhakumar

domain of formal law enforcement they are no longer affected by the shame involved in not repaying a debt on time. However, the enforcement of social norms requires certain community linkage. In terms of economics, norms are likely to be followed when there are repeated interactions (when people expect themselves or their future generations to have interactions within the community), and when they do not discount future rewards very heavily. If they are in a position that compels them to value immediate gains very highly compared to those received after some time, the rewards from repeated interactions in future may become less attractive. This may work against sustaining cooperation. The conditions that sustained cooperative behaviour between individuals may change when people move out of their communities (say outmigration to cities), and newer set of people come into a geographical space like a residential colony. Greater interaction with larger markets, urbanization, etc., may weaken community linkages. In such cases, where community norms are weakened, there may be lesser dependence on internalized social norms or exclusionary strategies used by the communities, even when the formal law enforcement machinery remains weak. There may be a dependence on informal state arms (like mafia) as tools for deal enforcement. This is so since it is more effective (due to the delay in courts), and also due to the increased supply of informal law enforcers since their cost—a major part of which is the punishment from formal courts for taking the law into their hands—is lower since there can be delays in these punishments by formal courts.

References Bhushan, P. 2009. ‘The Dinakaran Imbroglio: Appointments and Complaints Against Judges’, Economic and Political Weekly, 44 (42): 10–12. Debroy, B. 2000. In the Dock: Absurdities of Indian Law. New Delhi: Konark Publishers. Heaton, P. 2006. ‘Does Religion Really Reduce Crime?’, The Journal of Law and Economics, 49 (1): 147–172. Panchu, S. 2007. Settle for More: The Why, How and When of Mediation. Chennai: East West Books. Santhakumar, V. 2003. ‘Citizens’ Action for Protecting the Environment in Developing Countries: An Economic Analysis of the Outcome with Empirical Cases from India’, Environment and Development Economics, 8 (3): 505–528. Srikrishna, B. N. 2007. ‘A Superior Form of Dispute Resolution’, Economic and Political Weekly, 42 (30): 3099–3101.

About the Editors and Contributors Editors Shubhashis Gangopadhyay is currently the Research Director of India Development Foundation, Director of the School of Humanities and Social Sciences, Shiv Nadar University, and a Visiting Professor at the University of Gothenburg, Sweden. He obtained his PhD in Economics from Cornell University, USA, in 1983 and joined the Indian Statistical Institute where he became a Professor in 1991. In 2003, he became the founder-Director of India Development Foundation, an independent research institution. He was awarded a doctorate (honoris causa) by the University of Gothenburg, Sweden, in October 2006. In 2008, the year of the global financial crisis, he was appointed Advisor to the Finance Minister, Government of India. He is the Chief Editor of the Journal of Emerging Market Finance and has been on the editorial boards of Journal of Financial Stability and Review of Development Economics. He is a member of the Board of the Centre for Analytical Finance (Indian School of Business, Hyderabad) and International Development Enterprise India. He has been a member of the Bankruptcy Task Force of the Initiative for Policy Dialogue (IPD), Columbia University; the International Advisory Board of the Centre for Law and Economics of Financial Markets (Copenhagen Business School); the South Asia Chief Economist’s Advisory Council of the World Bank; and the board of Industrial Reconstruction Bank of India. Currently, he is also an advisor to the Competition Commission of India. He is also the founder-President of the Society for the Promotion of Game Theory and Its Applications. He has published widely in international journals and has a number of books to his credit on economics and finance. His published work (books) includes Waiting to Connect (co-authored), 2008; Economic Reforms for the Poor (co-edited), 2000; Counting the Poor: Where Are the Poor in India (co-authored), 1998; The Institutions Governing Financial Markets (edited), 1996; Enabling Financial Institutions (coauthored), 1997; Economic Theory and Development Policy (co-edited), 1992; and Economic Development and Policy (co-edited), 1991. V. Santhakumar is currently Professor at Azim Premji University, Bangalore. He had been on the faculty of the Centre for Development Studies, Trivandrum, for 15 years from 1996. He has held post-doctoral scholarships

192  Law and Economics

at the Wageningen Agricultural University (The Netherlands) and Vanderbilt University (USA). He has also carried out a number of consultant assignments for the Asian Development Bank, United Nations Environment Programme (UNEP), and United Nations Development Programme (UNDP) in different states of India, and in Tajikistan, Palestine, Laos, etc. He has received the research medal and the outstanding research award of the Global Development Network during its initial years. He specializes in analysing institutional issues in environment, natural resources, energy, infrastructure, and so on. In addition to publication in international journals, he has authored books titled Analysing Social Opposition to Reforms (SAGE, 2008) and Economic Analysis of Institutions: A Practical Guide (SAGE, 2011).

Contributors George S. Geis, John V. Ray Research Professor of Law, University of Virginia School of Law, Virginia, United States of America. Sanmitra Ghosh, Department of Economics, Jadavpur University, Kolkata, India. Satish K. Jain, Centre for Economic Studies and Planning, School of Social Sciences, Jawaharlal Nehru University, New Delhi, India. Padma Kadiyala, Associate Professor of Finance, Lubin Business School, Pace University, New York, United States of America. Sushma Kindo, Indian Economic Service, Government of India, New Delhi, India. Rajendra P. Kundu, Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi, India. Malabika Pal, Assistant Professor, Department of Economics, Miranda House, University of Delhi, Delhi, India. V. Santhakumar, Professor, Azim Premji University, Bangalore, India. Indervir Singh, Doctoral Scholar, Centre for Development Studies, Trivandrum, Kerala, India.

Index adjournment, costs of, 185–186 alternative dispute resolution (ADR) methods, 177–179 asymmetric information, and underpricing, 164–166 Barbara Taylor Bradford v. Sahara Media Entertainment Ltd., 44–45, 48–50, xiv–xv. See also copyright Bombay Stock Exchange (BSE), 161 business outsourcing, 77. See also outsourcing contracts capital markets, 154, 155 captive generation plant, 145 Cawasjee Nasarwanjee Dinshaw v. Special Land Acquisition Officer, 12 citizen suits. See public interest litigations (PILs) comparative negligence, 100 compensation, for land acquisition capitalization method, 10 comparable sales method, 10, 11 compensable vs. non-compensable takings, 4 Land Acquisition Act on, 9–13 market value formula, 7, 10 Michelman’s utilitarian framework and, xiii, 4–7 demoralizing costs, 4–7 efficiency gains, 4 settlement costs, 5, 6 test for requirement of compensation, 4 confidentiality agreement, in outsourcing contract, 79

contract burden of proof in breach of, 71–72 definition of, 51 dispensing with performance, rules regarding, 72–73 enforcement of, 51–52 relationship-specific investment and, 51–52 role of law in, 52–53 role of reputation in, 52 remedies for breach of, 55 general principles of awarding damages, 55–60. See also Indian Contract Act (ICA) liquidated damages and penalties, 60–65 specific performance, 65–71 self-enforcing, 52 contract law, goals of, 81–83. See also Indian Contract Act (ICA) contributory negligence, 100 Controller of Capital Issues (CCI), 154 copyright, 37 Bradford case, 44–45, 47–50 breadth of, 47 broad, 46 exploitable rights, protection of, 47 infringement, 45 moral rights, protection of, 47 narrow, 47 need of, 45–46 no protection to idea under, 48 protection of expression by, 47 search costs, 46 Copyright Act, 1957, 47–50

194  Law and Economics corruption, in judiciary, xx–xxi, 186–188 court fees, 175, 176 courts delays in, xx, 176–177 and alternative dispute resolution, 176. See also alternative dispute resolution (ADR) methods attempts for reducing of, 185–186 impact on PIL and nuisance cases, 183–185 factors determining number of cases in, 175–176 function of, 173–175 direction to enforcement machinery to act, 173–174 interpretation of statutes, 174 resolution of disputes, 173

reorganisation of state electricity boards, 150 revocation of licence, 146 tariff, determination of, 148–149 eminent domain, power of, 1–2. See also compensation, for land acquisition and just compensation, 3 evergreening, prevention of, 31, 43, 44 exclusive marketing rights (EMR), 37 exit rights, in outsourcing contracts, 90–92 expectation damages, 56–57

damage by severance, 11–12 demat, 168–169 demoralization costs, concept of, xiii, 4–7, 13–14 Disclosure and Investor Protection Guidelines (DIP), 154 domestic mutual funds, 158, 159

government-owned organizations, 134 benefits of, 135 efficiency of, 136 enhancing efficiency in, 136 problems of, 135–136 profit maximization objective in, lack of, 136

economic analysis of law, 96 economic method, 96 Electricity Act, 2003, xix, 132, 143 on coordination with state transmission utility, 145 impact of, 150–151 licence for distribution, transmission and trading under,