What Efficiency Demands: The Efficient Performance Hypothesis Defended
I appreciate the comments and thoughtful engagement by Professors Coleman, Kraus, and Posner on my essay “The Efficient Performance Hypothesis.” A few words of clarification may be helpful with regard to these comments, as well as the original essay. Let me begin with Professor Kraus’s comment—the most critical of the three in both tone and substance. Regrettably, Professor Kraus’ comment misses the point of the essay and his tangents contain a number of mistakes. I will attempt to clarify.
My essay did not advance a particular moral view of contract. It demonstrates, instead, that allocative efficiency can be made consistent with different moral views of contract. The competing views I employed to make this point were: (1) the promisor has the legal entitlement to choose nonperformance and payment of damages (i.e., the so-called Holmesian option) and (2) the promisee instead has the legal entitlement to choose nonperformance and payment of damages. Professor Kraus claims that these two views necessarily reduce to the promisee not having or having the right to specific performance. This reductionist claim is wrong, as I will demonstrate below, but even if it was correct it would not undermine the point of my essay, which again is simply to show that different and competing moral views can be made consistent with efficiency.
Professor Kraus obscures my thesis by focusing on specific performance. He says, “Brooks concludes that morality requires specific performance, and not expectation damages because ‘[b]reach of a legal contract without excuse is wrong conduct.’” In fact, I draw no such conclusion about specific performance and expectation damages. Clearly, either remedy could be consistent with viewing breach as wrong, and nothing I say in the essay denies this basic point. So why does Professor Kraus believe that I am invoking a moral argument in favor of the legal right to specific performance?
Because he conflates the economics of remedies with the morality of remedies. Here is his argument: he begins by rightly noting that promisees have options under the standard breach rule—that is, the option to get expectation damages or nothing at all. This option, he says, is not so different from the option they have under my proposal. Therefore, the standard rule and my proposal may be expressed as follows:
(i) Standard Rule: given promisor non-performance, promisee can elect damages or nothing; and
(ii) Proposed Rule: given promisor non-performance, promisee can elect damages or specific performance.
Since the only difference he can see between (i) and (ii), is “nothing” and “specific performance,” he argues that “Brooks’ moral thesis reduces to the claim that remedies that entitle the promisee to specific performance avoid the moral objection to expectation damages.” But, of course, there is another difference between (i) and (ii), which lies in the antecedent condition of the promisor’s non-performance. Under the efficient breach hypothesis, as structured in the essay, the promisor has the legal right to determine nonperformance, while under the efficient performance hypothesis that right belongs to the promisee. This is the key moral distinction between the two regimes, not the remedies. This is emphasized in the essay. Nothing for me here turns on the morality of the remedies (expectation damages, disgorgement or specific performance), which is not to say that there aren’t sound ethical reasons to prefer one remedy to another, but that was not the point of my analysis.
More generally, Professor Kraus fails to recognize that specific performance serves a distinct economic (as opposed to moral) function under the two regimes. Under efficient breach, specific performance is simply an economic instrument that allows the promisor to compare the cost of performance against the value of performance (measured by expectation damages), while under efficient performance, specific performance allows the promisee to compare the value of performance against the cost of performance (measured by disgorgement). Specific performance is no more morally required under one regime than the other. Specific performance, expectation damages and disgorgement serve economic functions here, encouraging the party with the option (in terms of rights or powers) to make an efficient allocation. And just as the efficient breach hypothesis works with right or power so does the efficient performance argument. Imagine that the two competing moral visions of contract enforcement were (1) the promisor has the right to perform or pay and (2) the promisor does not have the right to perform or pay. Note that the promisee does not necessarily have a right to performance under either vision, but assume that she is able nonetheless to secure performance through force, guile, or some fortunate circumstance. The promisee has the power to get performance, but she would only secure performance, if that meant giving up her claim to expectation damages, when performance is efficient.
Specific performance serves only as a proxy for the promisee’s value (with disgorgement proxing the promisor’s cost) under efficient performance, just as specific performance serves as a proxy for the promisor’s cost (with expectation proxing the promisee’s value) under efficient breach. To really make clear that the right to specific performance is not required consider the following: let’s say that the rule is that the promisee gets to choose between disgorgement and expectation damages. Now disgorgement proxies the promisor’s cost while expectation proxies the promisee’s value (specific performance is absent here: the promisee has no right to enjoy performance and will end up with money damages in any event). To get efficient allocation the court imposes the following decision rule: if the promisee chooses disgorgement there will be not performance and she will simply get the cash transfer. However, if the promisee chooses expectation damages then performance will be ordered and immediately liquidated for the promisee’s benefit (which is to say, performance will be sold on the market and the market price received will be handed over to the promisee). In this case, the decision rule (with the promisee effectively making the allocative choice) leads to efficiency, but the promisee never gets performance; nor does she have a right to it. This illustration is not offered as a practical alternative, it is merely offered to highlight the distinction between the economics of remedies and the morality of remedies, a distinction that Professor Kraus comment does not fully appreciate.
I have no quarrel with his general comments about the moral arguments for and against specific performance. They are beside the point, as I was neither explicitly nor implicitly asserting any moral distinction to specific performance. But if one must insist, despite the hazards of doing so, that the moral distinction uniquely present in the efficient performance hypothesis is the promisee’s right to specific performance then so be it. That insistence still preserves the thesis of the essay, namely that efficiency can be made consistent with competing moral views of contract enforcement.
Professor Kraus finally makes a number of challenges regarding the efficiency claims in my essay. These arguments too are somewhat unfocused, some concern specific performance as an efficient remedy and others are more focused on the efficient performance hypothesis. None are particularly compelling. I agree, of course, that the relative efficiency of a contractual regime depends on a host of empirical variables, including relevant market structures, judicial determination costs, litigation costs, search and negotiation costs, ex ante and ex post. The efficiency indeterminacy I describe is empirically contingent, and while Professor Kraus’ armchair empiricism doesn’t resolve the matter for me, I cannot say that his intuition is any worse than mine.
Another point of clarification concerns Professor Coleman’s claim that I mistakenly conflate the Holmesian notion of the promisor right (option) to breach with the efficient breach hypothesis. He urges instead that “efficient breach” entails a breach of a duty (not the exercise of a right or privilege). This a reasonable interpretation of “efficient breach,” but by no means the only reasonable one. There have certainly been plausible efficient breach arguments based on a right to breach (what Professor Schiffin called the strong version of the efficient breach argument in her recent article), not just the mere power or capacity to breach (the weak version of the argument). For instance, Professor Shavell advances an incomplete contracts framework to suggest that nonperformance and payment of expectation damages when performance is not cost justified (i.e., efficient breaching) may be considered default term in incomplete contracts because that is what the parties would have negotiated. While it is not clear to me that the parties would have negotiated necessarily for expectation damages, his argument of an implied term to efficiently breach suggests a form of efficient breach consistent with the Holmesian option to breach. Similarly, Judge Posner in Patton v. Mid-Contintent Systems, Inc., described “certain fundamentals of contractual liability” by claiming that “[t]he promisor promises in effect either to perform or compensate the promisee for the cost of non-performance.” So while Professor Coleman is correct to say that it would be a mistake to equate the right to breach necessarily with efficient breach (which I did not do), he is mistaken to argue that the right to breach is necessarily disjoin from efficient breach.
But this is really a trivial point of contention. I couldn’t agree more with Professor Coleman’s basic point. His prior work with Professor Kraus, calling for greater attention to the normative content of existing categories of legal entitlement is surely right and greatly under-appreciated in the literature. My essay, as Professor Coleman observes, is less ambitious than what they call for, but it is nonetheless sympathetic. That larger ambition was simply not the principal the objective of my essay, which was to show that efficiency embraces a much broader range of legal rights than is traditionally suggested by efficiency advocates.
My essay is also sympathetic to the prior work of Professor Posner and his comment. He is right to suggest that the demands of efficiency may become quite exclusive, ex post, once relevant facts are known. Professor Posner made a similar point several years ago, and Professor Craswell made a similar point before him, and Professor Rogerson before him, and Professor Shavell before him, and Professor Coase before them. Each identified key indeterminacies regarding efficiency claims about contract remedies. Professor Coase, for example, argued that if ex post bargaining costs are zero then all remedies lead to allocative efficiency. Without some assumptions of differing bargaining costs associated with one remedy or another, efficiency won’t tell you which remedy to pick. Of course, even with zero transaction costs the remedies create different distributive effects, which will have indirect consequences for resource allocation and investment efficiency. Hence the well-known argument that competing efficiency concerns (such as investment, search, precaution and bargaining) leave matters indeterminate until we know how the remedies specifically respond to these various concerns and contribute to overall welfare.
My point is that the indeterminacy is more profound: even along the sole dimension of allocative efficiency we are not in a position to say, analytically, which remedy is the most efficient. Efficient breach and efficient performance can be made analytically equivalent in terms of allocation of resources and distribution of wealth. Any tie-breaker must be an empirical one. I am deeply skeptical of our capacity and willingness to search out that empirical answer. And in any event, resolving fundamental economic indeterminacies won’t end the debate. It will merely signal a clean start of the debate over efficiency and morality in contracts: when we really know what efficiency demands. One can only hope, though I am equally skeptical here, that when that day comes we will also know what morality demands.
Richard R.W. Brooks is Professor of Law at Yale Law School. He thanks Jules Coleman, Jody Kraus, and Eric Posner for their contributions and insights. He also thanks Bruce Ackerman, Ian Ayres, Heather Gerken, Liam Murphy and Jed Rubenfeld for their colleagueship and comments.
Preferred Citation: Richard R.W. Brooks, What Efficiency Demands: The Efficient Performance Hypothesis Defended, 117 Yale L.J. Pocket Part 14 (2007), http://yalelawjournal.org/forum/what-efficiency-demands-the-efficient-performance-hypothesis-defended.