Is the good news about compliance good news about cooperation? (issue of compliance to international treaties and/or regulation)

In the past few years many social scientists interested in cooperation have turned their attention to the problem of compliance in international regulatory regimes. Much of the empirical research in this area has been conducted by a group composed mainly of qualitative political scientists and scholars interested in international law.(1) Its message is that (1) compliance is generally quite good; (2) this high level of compliance has been achieved with little attention to enforcement; (3) those compliance problems that do exist are best addressed as management rather than enforcement problems; and (4) the management rather than the enforcement approach holds the key to the evolution of future regulatory cooperation in the international system. As Oran Young notes, "A new understanding of the bases of compliance - one that treats compliance as a management problem rather than an enforcement problem and that has profound practical as well as theoretical implications - is making itself felt among students of international relations."(2) In short, not only are the dreary expectations born of factors such as relative gains concerns, collective action problems, anarchy, and fears of self-interested exploitation incorrect but also the enforcement limitations that always have appeared to sharply bound the contributions of international law and many international institutions now appear to have been exaggerated.

In this essay we will argue that the empirical findings of this group, which we refer to as the "managerial" school, are interesting and important but that its policy inferences are dangerously contaminated by selection problems. If we restrict our attention to those regulatory treaties that prescribe reductions in a collectively dysfunctional behavior (e.g., tariffs, arms increases), evidence suggests that the high level of compliance and the marginality of enforcement result from the fact that most treaties require states to make only modest departures from what they would have done in the absence of an agreement. This creates a situation where states often are presented with negligible benefits for even unpunished defections; hence the amount of enforcement needed to maintain cooperation is modest. Nothing is wrong with this situation in itself, but it is unlikely to provide the model for the future that the managerialists claim. Even if we assume that the absolute value of the benefits generated by this small amount of regulation is relatively large, further progress in international regulatory cooperation will almost certainly require the creation of agreements that present far greater incentives to defect than those currently in place (e.g., more demanding environmental standards, fewer nontariff barriers, steeper arms reductions). We have precious little evidence that such progress can be obtained in the absence of better enforcement.

After discussing the problems posed by endogeneity and selection, we present the theoretical argument for linking enforcement level to what we call "depth of cooperation" and examine the extent to which deep cooperation has been achieved without enforcement. We then present a number of prominent exceptions to the managerial school's unqualified generalizations about the causes and cures of noncompliance. Finally, we discuss the strategic implications of the evolution of increasingly cooperative regimes.

The managerial thesis

The bedrock of the managerial school is the finding that state compliance with international agreements is generally quite good and that enforcement has played little or no role in achieving and maintaining that record. In Abram Chayes and Antonia Chayes's words, what ensures compliance is not the threat of punishment but "a plastic process of interaction among the parties concerned in which the effort is to reestablish, in the microcontext of the particular dispute, the balance of advantage that brought the agreement into existence."(3) For the members of the managerial school, "noncompliance is not necessarily, perhaps not even usually, the result of deliberate defiance of the legal standard."(4) On those rare occasions when compliance problems do occur they should not be viewed as violations or self-interested attempts at exploitation, but as isolated administrative breakdowns. The causes of noncompliance are to be found in (1) the ambiguity and indeterminacy of treaties, (2) the capacity limitations of states, and (3) uncontrollable social or economic changes.(5)

Not surprisingly, the managerial school takes a dim view of formal and even informal enforcement measures. Punishment not only is inappropriate given the absence of any exploitative intent but it is too costly, too political, and too coercive. As Ronald Mitchell notes, "Retaliatory non-compliance often proves unlikely because the costs of any individual violation may not warrant a response and it cannot be specifically targeted, imposing costs on those that have consistently complied without hurting the targeted violator enough to change its behavior."(6) As a result, according to Young, "arrangements featuring enforcement as a means of eliciting compliance are not of much use in international society."(7) Since sanctions usually are more successful against economically vulnerable and politically weak countries and "unilateral sanctions can be imposed only by the major powers, their legitimacy as a device for treaty enforcement is deeply suspect," as Chayes and Chayes point out.(8) Moreover, retaliation for violating a treaty may risk the breakdown of current and future cooperation:

the actor considering retaliation must also think of the possible future costs. It may be dangerous to prejudice the possibility of support from the violator at some point in time in the future when it may be needed. . . . [T]he risk of setting off "a long echo of alternating retaliations" will often dwarf the consequences of overlooking what are arguably relatively minor or "technical" violations.(9)

Instances of apparent noncompliance are problems to be solved, rather than violations that have to be punished. According to Chayes and Chayes, "As in other managerial situations, the dominant atmosphere is that of actors engaged in a cooperative venture, in which performance that seems for some reason unsatisfactory represents a problem to be solved by mutual consultation and analysis, rather than an offense to be punished. Persuasion and argument are the principal motors of this process."(10) The strategies necessary to induce compliance and maintain cooperation involve: (1) improving dispute resolution procedures, (2) technical and financial assistance, and (3) increasing transparency. The last is especially important: "For a party deliberately contemplating violation, the high probability of discovery reduces the expected benefits rather than increasing the costs and would thus deter violation regardless of the prospect of sanctions."(11)

The endogeneity and selection problems

It is not difficult to appreciate why the findings of the managerial school suggest that both international institutions and even international law have a far brighter future than most international relations specialists have believed for the past fifty years. Apart from sharply contradicting the pessimistic expectations of many realists and neorealists about the inability of cooperation and self-regulation to flourish in an anarchic world, they also run counter to the claims of cooperation researchers in the rational-choice tradition. Such researchers emphasize the centrality of enforcement concerns in regulatory environments and characterize them as mixed-motive games, where the danger of self-interested exploitation is significant, as opposed to coordination games, where it is not.(12) Such findings certainly add credibility to the frequent speculation that the rational-choice tradition's affection for the repeated prisoners' dilemma has led it to overemphasize enforcement and underemphasize the potential for voluntary compliance and noncoercive dispute resolution.

In trying to understand the prescriptive significance of the managerialists' compliance findings, it is useful to consider the following hypothetical story. An article has recently appeared in an education journal criticizing the state of musical education in an age of funding cutbacks. The author, a longtime music teacher, argues that such cutbacks inevitably have dire consequences for the quality of school music programs. A member of the school board who has aggressively supported the elimination of frivolous expenditures is skeptical of what she believes to be characteristically self-interested reasoning. In an effort to get to the bottom of the issue, she attends fifteen concerts in her district and fifteen concerts in a rival district that has not reduced its support of music education or extracurricular activities. She finds that the quality of the two orchestras as measured by the number of mistakes they made to be pretty much the same and quite low in both cases. Noting that the orchestras in her district have achieved this high level of performance despite a 75 percent reduction in the number of rehearsals, she is delighted. Not only has she demonstrated that the cutbacks have had no effect on school orchestras but she believes that she has confirmed her long-held suspicion that rehearsals do not make school orchestras better, they simply line the pockets of music teachers eager to buy hot tubs and Steinway pianos.

These conclusions may, however, be invalid. It is likely that orchestras in her district may have adapted to the decrease in resources by playing less demanding pieces. No orchestra is eager to embarrass itself, and one of the most effective ways to avoid doing so is to play Haydn rather than Mahler or Stravinsky. Unless the school board member counting mistakes figures out a way to control for the difficulty of repertoire, we do not really know what her findings tell us about the impact of the budget cuts. A treaty, like the selection of an orchestra's repertoire, is also an endogenous strategy. States choose the treaties they make from an infinitely large set of possible treaties. If some treaties are more likely to be complied with than others or require more enforcement than others, this will almost certainly affect the choices states make. Just as orchestras will usually avoid music that they cannot play fairly well, states will rarely spend a great deal of time and effort negotiating agreements …