Coase—the Man and His Work
Richard A. Epstein*
Mr. Ning Wang has put together this historical collection of essays written by the late Ronald Coase over his long and distinguished academic career, first in England and then in the United States. It is especially fitting that this volume will be published by the Peking University Press, given that Coase, along with Wang, spent much of his last years engaged in research on the transition of China into a capitalist economy, which resulted in 2012 of the publication of their book, How China Became Capitalist. In addition to his extensive written work, Coase spoke and organized conferences devoted to this theme. It marks a profound transformation in the life of a man who started in England, moved to the United States and finished his career with a decided Chinese twist.
Wang has asked me to write a preface to this collection, for which he has supplied an introduction. My preface is intended to achieve two related purposes. First, to offer some personal reflections on my own relationship to Ronald Coase, and second, to speak about his influence, not so much on the economics profession, but on the legal profession, which has fully embraced his ideas. Indeed, lawyers continue to
use Coase’s work to explicate major areas of law that become intelligible only after taking into account the frictions that beset the legal system, which Ronald Coase first systematically identified with the beguilingly obvious term—obvious, that is, only after you figure it out—transactions costs. So much of my own intellectual development has been associated with Coase that it is hard to keep these two prefatory tasks in watertight compartments.
The Man
To start at the beginning, I had never heard of Ronald Coase when I graduated from Yale Law School in the late spring of 1968. His name was mentioned neither at Yale nor at Oxford Law School where I took my first law degree as a pseudo—Englishman in the spring of 1966. My first introduction to Coase came shortly after I arrived at the University of Southern California to begin my own teaching career. The late Michael Levine arrived that same summer, fresh from his one-year fellowship at the University of Chicago Law School, where he worked in depth with Ronald Coase. Our first substantive conversation concerned various theories of liability in the law of tort. It was only a few moments into the discussion before Mike piped up with some reference to “Coase,” and I did not know whether the word was a noun, verb, or name. But what followed was an outpouring of discussion about how the system of liability had to take into account transactions costs both between the parties and against the rest of the world. My initial response, like so many others, was that this supposed insight seemed to put the cart before the horse. Transactions costs were always some kind of added extra into a much larger equation, not very relevant to the traditional legal concerns of justice fairness, and equity in the particular case which had long dominated the landscape, just as they had throughout in my own legal education.
It was only a matter of time before the issues that piqued my attention at the University of Southern California came into sharper relief when I got to the University of Chicago in the fall of 1972, which is when I first met Coase. By that time he was already 61—a lot younger than I am today. It seems clear in retrospect that he had lost some of the fire that characterized his younger behavior. But Coase was always insistent about the soundness of his position, which he defended against one and all.
He was more than gracious to me, and, as editor of the Journal of Law and Economics, published my article “Unconscionability: A Critical Reappraisal” in 1975, which marked another step in my volution towards his transactions costs approach. Over the 40 years that I knew Coase he never wavered in his defense of the distinctive nature of his approach.
Several incidents during this period come quickly to mind. At some time in the 1980s Coase, James Buchanan, and myself were asked to attend a small meeting with members of the staff of the Liberty Fund. At that occasion Coase was quite upset at the way in which he thought the world, and most particularly Richard A. Posner, misunderstood and misapplied his work.[1] Coase wanted to use transactions costs to define the operation of the legal system. He did not see that his theory had powerful applications, often implicitly, for the way judges crafted their legal doctrine. Indeed, Coase often thought that lawyers had taken his ideas in directions of which he did not approve. He claimed that he had used the transactions costs language merely to explicate land disputes between neighbors under the law of nuisance.
Ironically, on this point Coase was surely wrong. His work had far greater reach than he supposed, and the legions of scholars that ported his work into different subject matter areas were right to do so given how far-reaching were the applications of the now famous “Coase Theorem” to all aspects of legal and social behavior. Indeed, it is instructive to ask why there is no second theorem by some other scholar with the same breadth as Coase’s. I have come to the conclusion that none has emerged in the near sixty years since the publication of his work because none is really needed. Restated modestly, that theorem posits that the correct way to maximize social welfare is to minimize transactions costs, for only this strategy can increase the volume of socially maximizing transactions. For readers who want to hear Coase talk about these issues in his own words, I had the privilege to conduct a longish interview with him in 2002, which has been been recorded for posterity as “A Conversation with Ronald H. Coase.”
Even toward the end of his life Coase was fiercely proud of his individuality. Here are two incidents illustrating this fact.
In 2010 Professor Thomas Hazlett and I planned a conference devoted to the works of Coase. Coase was not able to attend the conference but did record, with some evident difficulty, a few remarks that were played on the occasion. The main topic of those remarks was a provocative essay by Richard A. Posner that pointed out the asserted resemblances between Coase and John Maynard Keynes. Posner emphasized that both men had grand theories that eschewed the use of formal mathematics. Coase
would have none of it, so he gave his own account of the interconnections between himself and Keynes that went roughly as follows: “Mr. Keynes and I met on only one occasion shortly after the end of the war. It was at a tea party. Mr. Keynes was sitting down, and I was standing up behind him. He turned to me and asked if I could pour him a cup of tea, which I dutifully did. And that was the influence of Keynes on Coase.”
Indeed, Coase is right on this point. His every instinct was to break down complex aggregates into their constituent parts to see how they worked together. Keynes by contrast always moved in the opposite direction, ignoring individual transactions in order to postulate aggregates that can be studied for their macro-economic effects.
On this I think that Coase’s reductionism of macro to micro is much preferable to Keynesian aggregation from micro to macro, as the latter approach tends to ignore conflicts of interest between members within the aggregate group, which in turn leads one to overlook the public choice problems with government regulation, a dynamic which Coase correctly saw as a stumbling block to the more efficient operation of markets.
The last time I saw Coase was for lunch at his old-folks home. He was then over the age of 101 and obviously labored. Coase had always jealously protected both his reputation and the purity of his economic position in his final years. But on this occasion he was morose. He pronounced that his life had been an academic failure because the world had largely misconstrued his understanding of the role of transactions costs in the economic system, echoing points he had made earlier. Those scholars took the theory to places where it did not belong. I disagreed with him on this gloomy assessment, but got only a small smile when I said to him that he had a huge intellectual advantage—a last name with only one syllable—which he had exploited to the maximum, which happens whenever your name becomes an adjective ... like Coasean.
His Work
The adjectival use of his name is well-deserved because his simple insights were able to upend other conceptual frameworks. Raised as a typical English and American lawyer in the mid-1960s, I was in the midst of my own first legal project, which was to explain how some conception of individual autonomy provided the intellectual glue that held together the diverse common law fields of property, contract, tort, and restitution. Those models worked pretty well for simple two-party transactions, or at
least I thought. Those (relatively) simple cases involving disputes between two drivers over an automobile accident, or landlord and tenant over a rental agreement, did not bring the issue of transaction costs to the fore. But once the seed had been planted, it was difficult to erase the Coasean notions from my mind. Coase developed these ideas in “The Problem of Social Cost” initially with cases drawn from the English law of nuisance, chiefly during the late nineteenth century, with which I was familiar from my studies at Oxford. There was something odd about what Coase had to say about
them. The point here has less to do with the transactions costs model and more to do with the role of causation in tort law. Coase’s view of this subject was to treat the term “causation” as wholly reciprocal, so that it was impossible to say whether the fish died because they ingested pollution or because their lungs were in some way weak or defective.
Viewed in this light the term is largely useless as a way to decide particular cases, which should have been a warning sign to Coase that something was wrong because he generally loathed to disregard longstanding practice in business. Why then should he do so on linguistic matters? Ordinary language stood in stark opposition to the Coasean conception of causation, and the pervasive role of theories of causation in the law of tort offered a loud testament that these issues could not be ignored, which is not
to say that I could put together the law’s divergent ideas into any cohesive synthesis.
Part of the problem in this analysis stems from a challenge that Ning Wang addresses when he tries to conceive of how to think about a zero transactions costs world. The conceptual issues here are really more difficult than Coase let on for the simple reason that no one has any idea what a zero transactions costs world looks like, any more than they understand how the world looks once one accelerates beyond the speed of light. To put it more concretely, in a world of zero transactions costs there can
be no fraud because everyone knows everything. There can be no temporal barriers because all information is incorporated instantaneously in order to avoid the costs of delay. Nor can distances matter because whatever actions do take place also have to be instantaneous. The features of time and space that mark the world as we know it disappear in a world of zero transactions costs. What then do we make of the gap between contract formation and contract performance, or of particular acts and their consequences? What then can be the use of this construct?
The way to escape this puzzle lies in restating the question. We know that we all live in a world that has temporal and spatial dimensions. We also know that information is costly to acquire, which necessarily means that monitoring is imperfect and fraud or opportunism always remain serious threats given the fact that self-interested individuals do not always have the moral fiber Coase found so important in economic affairs—to steer the straight and narrow. At this point, the zero transactions costs model asks this question: suppose that we had all the time in the world at to figure out how to design and organize transactions that take place in a positive transactions costs world. How in our infinite leisure would we structure those transactions from formation to execution to avoid the strains and tension that take place when we leave this intellectual Nirvana—to use one of Harold Demsetz’s favorite phrases—and enter into the real world where transactions costs matter for all the obvious reasons?
The answers to this question are quite tractable, and they help explain why Coase thought that as transactions costs tended to zero the question of externalities disappeared. In my own writing about Coase I refer to this as the single-owner approach to thinking about social and legal relationships. Consider any kind of collective endeavor—the formation of a condominium association or the creation of a corporation or a club. All members at the outset are fully aware of the set of laudable and dangerous traits present in individual human beings, and thus the single owner.
This single owner knows that he will part with his property, often to multiple parties who then will be neighbors with relationships with each other in a decidedly positive transactions cost environment. That single owner also knows that all benefits and costs are correlative with each other, so that if he grants a set of rights to one potential buyer as against all others, he will get less from them; even as he gets more from the original purchaser, he will get less from those who follow. He therefore has to calculate whether the downstream losses, reduced to present value, exceed the initial set of gains. And that singular calculation has to be undertaken for each of these transactions with each of these parties. The effect of this is to impose an enormous discipline on the way in which these transactions are undertaken, because these concerns effect how to make the sales, what terms to include in contracts between the various parties, how to deal with subsequent sales and leases by the original parties, and so on down the line.
In this world there are by definition no externalities, just as Coase posited when he held that transactions costs were zero. But it was then instructive to see what kinds of obligations were imposed on various parties, which could serve as a template for how to deal with neighbors who did not derive their title from a common owner. The simplest approach is to ask which prohibitions on use were imposed as part of the deal by the single owner and then use those findings to develop the law of nuisance that Coase studies so intensely in “The Problem of Social Cost.” This point then ties neatly into one of Coase’s lifelong obsessions that is often derided by more theoretical economists and lawyers. If you want to understand, Coase always maintained, how people would contract with each other, then by all means look at how they do contract with each other. Far from starting with blackboard economics, as Coase liked to call it, it is much better to get into the nitty-gritty of agreements and use those actual transactions as the baseline to uncover how, for example, neighbors get along with each other.
And once you take Coase’s methodology in this light, it turns out that Coase was incorrect in thinking that causation should be understood as perfectly reciprocal when in ordinary language it is not. The simplest case of causation is the direct application of force by one person against the person or property of the other. The differentiation between the nominative and accusative in all languages shows how deeply the distinction is built into the universal architecture of the human mind. This notion of
causation carries forward in a somewhat attenuated form in the cases to which Coase devoted so much attention in the law of nuisance, which stresses various forms of physical invasion of noise, smells, and solids. These cases are recognized as needing special treatment in all these initial agreements promulgated by the single owner.
Emissions are sharply circumscribed in virtually all initial condominium agreements, for example, which are drafted and recorded in ways that make sure that the benefit and burden side of the prohibition descends to all takers, regardless of the order in which they take or the price they pay. The structure has this feature of permanence because the problem is one that is constant over time.
That fixity of vision is decidedly not true on all cases. Thus, these same master agreements contain terms on softer externalities having to do with styles, colors, decor and the like that vary much more with wealth and taste, and thus display a greater degree of diversity. So the common law of nuisance entrenches the first set of rights, and deals very cautiously with the second set of rights. Indeed, with public nuisance from factories and automobiles the concern with pollution is the same as it in the
private settings, but once again the notion of transactions costs comes in to explain the different responses to public nuisances, which has been built into the English common law since 1535.[5] The definition of what counts as a nuisance does not switch as the property damaged—we can now use that word—ceases to be a private pond and becomes a public river. But the cost of enforcing rights differs radically given the general rule that for common resources like rivers, open access was the initial rule— all could enter but none could exclude. In these settings private rights of action are very costly to administer relative to the lower amounts at stake to individuals.
Hence the common law distinction, which endures to this day, takes the position that those individuals who suffer special, e.g. large and unique, damages can maintain their traditional rights to sue for property damage. By contrast those individuals who suffer only general damages along with their fellow citizens have no private rights of action. But now the state can bring an administrative remedy against the offender, which allows it both to remove the obstacle or clean up the pollution while also fining that party in order to improve the deterrent effect. There are, of course, constant gradations between the two categories that cannot be ignored. But the basic point only shows the power of the Coasean concern with transactions costs, which now receives a much more central role in cases where the numbers of parties seeking redress increases. This simple point helps explain, as no other theory can, the switch from simple lawsuits based on principles of corrective justice to more complex class actions and administrative remedies. And so it is that the law of nuisance slowly turns into environmental law, with a larger role for direct public enforcement of prohibitions in the case of wide spread harms of diffuse or multiple origins. All too many people think of these as discrete when in fact they are not. The concern with keeping the entitlements that emerged from the early concerns of the single owner now gives way to greater understanding of the hard remedial choices that develop when the scope of problems increases.
The insistence on the role of transactions costs also influences Coase’s first classic article, “The Nature of the Firm” published in 1937 after his trip to study industrial organization in the United States. As ever, the basic observation Coase makes is perfectly obvious—at least after he makes it. As it turns out, the system of voluntary transactions is marked by a mixture of some discrete, or spot, transactions, where a price is paid for the provision of some good or service. But there are also many settings in which informal exchange or the formation of the firm takes hold and puts the sale of goods or services to one side. Coase asked the simple question: why the division between spot transactions and firm formation? His resulting simple answer has survived the test of time. It is not costless to organize an exchange system. There are transactions costs that have to be incurred during negotiation, contract formation, performance, and litigation. In some cases these costs are well worth bearing. But in other settings it turns out that these expenses are too high, so instead the firm arises such that employees are paid not for each discrete act but with a wage for overall services whose content and description is not fully specified in advance. The new strategy also has negotiation and enforcement costs, and it is often harder to monitor work than it is to inspect a good before it is purchased, or to repair work before it is paid for. But there is no general rule to push things in one direction or another, and Coase was right to insist that the notion of transactions costs lies at the differentiation of various forms of organization.
The world is, of course, even more complex than this simple dichotomy implies, so that it is often the case that firms have more complex structures.[7] Workers can receive base pay and a commission, or a bonus at the end of the year. It may be better for them to bear some residual risk along with the employer. Or the employer could become a partnership instead of an individual proprietorship. These permutations are not the work of an economist, but a product of lawyers who address these issues as they draft agreements for the provision of capital and labor. Lawyers also have to ask the hard
question of how different members of the firm fit together in order to take into account differences in tastes and competence, which militate against a simple firm design in which all workers have identical roles with respect to management. It is here that the genius of Coase lies. It is not just that he made observations about everyday events that many great minds had missed. It is that all of his work was capable of a principled extension to problems that Coase never addressed. And so it is that work which was always important has become more so over time, and will continue to do so. Coase’s preoccupation with transactions costs turns out to unlock many of the secrets of social organization such that familiar notions that were once afterthoughts in legal and social discussion have become the centerpiece of modern social science.
Thank you, Ronald Coase.