Ruling the World: Power Politics and the Rise of Supranational Institutions

By Lloyd Gruber

The last few decades have witnessed an extraordinary transfer of policy-making prerogatives from individual nation-states to supranational institutions. If you think this is cause for celebration, you are not alone. Within the academic community (and not only among students of international cooperation), the notion that political institutions are mutually beneficial--that they would never come into existence, much less grow in size and assertiveness, were they not "Pareto-improving"--is today's conventional wisdom. But is it true? In this richly detailed and strikingly original study, Lloyd Gruber suggests that this emphasis on cooperation's positive-sum consequences may be leading scholars of international relations down the wrong theoretical path.


The fact that membership in a cooperative arrangement is voluntary, Gruber argues, does not mean that it works to everyone's advantage. To the contrary, some cooperators may incur substantial losses relative to the original, non-cooperative status quo. So what, then, keeps these participants from withdrawing? Gruber's answer, in a word, is power--specifically the "go-it-alone power" exercised by the regime's beneficiaries, many of whom would continue to benefit even if their partners, the losers, were to opt out. To lend support to this thesis, Gruber takes a fresh look at the political origins and structures of European Monetary Unification and NAFTA. But the theoretical arguments elaborated in Ruling the World extend well beyond money and trade, touching upon issues of long-standing interest to students of security cooperation, environmental politics, nation-building--even political philosophy. Bold and compelling, this book will appeal to anyone interested in understanding how "power politics" really operates and why, for better or worse, it is fueling much of the supranational activity we see today.

• Language: English
• Publisher: Princeton University Press
• Release date: Mar 20, 2000
• ISBN: 9781400823710

Lloyd Gruber

Lloyd Gruber is an Assistant Professor in the Harris School of Public Policy Studies at the University of Chicago.

Book preview

CHAPTER 1

Introduction:

From Anarchy to Organization

Without cooperation, we will be lost. Without institutions there will be little cooperation. And without a knowledge of how institutions work—and what makes them work well—there are likely to be fewer, and worse, institutions than if such knowledge is widespread.

—Robert O. Keohane

A CALL FOR THEORETICAL REORIENTATION

If supranational structures like the World Trade Organization, the European Monetary Union, and NAFTA have recently been spreading like wildfire,¹ it is because they make it easier for states to cooperate—and in a world of rapidly increasing interdependence and complexity, the more interstate cooperation, the better. Inasmuch as there is a conventional explanation for the rise of supranational institutions, surely this is it.

It’s true that the last several years have seen questions about the role of power, distributional conflict, bargaining, and coercion rise to the top of the cooperation literature’s analytical agenda.² So far, however, few of the theorists responsible for this shift have thought to challenge the view, most closely associated with the neoliberal scholar Robert Keohane, that international cooperation—and hence international institutions—are fundamentally good things. Thus, while neorealists complain that Keohane’s theoretical orientation minimizes the importance of interstate disputes over the distribution of collective benefits (see esp. Grieco 1993 and Krasner 1991), they, too, assume that the cooperative equilibria that emerge once these distributional conflicts are resolved leave all participants, strong and weak alike, at least as well off as they would have been had the original, precooperation status quo remained in effect.

In many ways, this convergence in liberal and realist opinion is a sign of theoretical progress. The underlying explanatory logic of the new liberal-realist research agenda has the virtue of being firmly anchored in rational choice foundations, and there is little question that its emphasis on possibilities for (and impediments to) mutual gain captures a great deal of what is going on.³ As productive as it has been, however, the line of theory developed by Keohane and subsequently elaborated by Stephen Krasner, Joseph Grieco, and other scholars has led to an overemphasis on the mutual-gains side of international cooperation. For while the prospect of collective gain may well be driving some of the action, there is a winners-and-losers dimension to recent developments that may, in many cases, be even more important.

Might a number of the supranational arrangements we keep hearing about—even ones as successful as NAFTA and the newly reinvigorated European Monetary System (EMS)—end up working to the disadvantage of some of their participants? Although this conclusion runs directly counter to the positive-sum view of international cooperation that dominates today’s theoretical mainstream, the model of cooperation set forth in this book raises the possibility that, indeed, only some—perhaps a small fraction—of the supranational institutions that get created and endure actually benefit all of their signatories. To be clear, the issue here is not simply one of relative gains and losses. The NAFTA and EMS losers were indeed made worse off in relative terms—they did not benefit as much as the winners—but they were also worse off (or so I shall argue) in a more fundamental absolute sense. Had it been within their power to return the world to the original noncooperative status quo, they would have been only too delighted to do so.

As we will see, the task of adducing evidence for this winners-and-losers perspective is complicated by a number of factors, not least of which is the need to imagine a world in which the cooperative arrangements in question had never gotten off the ground. What would the utilities of each cooperation partner have been if, for some reason, these arrangements had never materialized? Might some of these partners have been better off in this noncooperative (albeit hypothetical) situation? Counterfactual questions of this sort are inherently problematic, making it difficult to offer a definitive answer. That said, these methodological difficulties do not explain—and they certainly do not justify—the literature’s enduring preoccupation with the positive-sum consequences of international cooperation. Insofar as previous scholars have misconstrued the driving forces behind such far-reaching cooperative arrangements as NAFTA and the EMS (as I believe they have), the problem has far more to do with deficiencies in our theoretical understandings of cooperation and institution building than it does with any empirical or purely methodological limitations. If we really want to understand why supranational institutions are ruling the world, we will first have to broaden our current theories of collective action, extending their logic in ways that acknowledge and incorporate the possibility of both winners and (absolute) losers. That is what I hope to do in this book.

POWER POLITICS AND THE INSTITUTIONALIZATION OF INTERNATIONAL RELATIONS

So what, then, is the logic of (international) collective action? Few questions in the study of international politics are more basic or, indeed, more relevant to the concerns of today’s heads of state. And yet even after years of research, and despite the glaring policy significance of the issues involved, the truth is that we are still some distance from a completely satisfactory answer.

It is tempting to think that this is because our empirical base is too limited, as if the deficiencies in our theoretical understanding of recent international developments could be overcome by simply piling up more facts or locating new sources of data. In fact, scholars of international relations and comparative politics have already accumulated a vast amount of information about interstate cooperation and institution building in Western Europe and the Americas, the two regions of the world in which current trends toward supranational governance have made greatest headway. Much is known, too, about the emergence of global arrangements like the General Agreement on Tariffs and Trade (GATT), whose member states recently established a new World Trade Organization (WTO) for the express purpose of mediating and resolving their future trade disputes. But while the literature is quite good at describing these momentous achievements,⁴ it has been markedly less successful at explaining why they occurred, and thus at telling us what sort of developments we ought to expect in the future. To this point, the newfound enthusiasm for supranational institutions—and for interstate collaboration more generally—has remained something of a mystery.

This book traces the problem to a widely overlooked conceptual bias in the academic literature. Until now, most scholars have taken the view that international institutions, whatever their distinctive purposes or politics, are fundamentally welfare improving. Underlying this view is a simple two-step logic wherein institutions facilitate international cooperation (step 1), and cooperation in turn facilitates utility gains for each participating state or government (step 2).

For well over a decade now, this positive-sum analytical orientation—most forcefully articulated in Robert Keohane’s groundbreaking book After Hegemony (1984)—is what has guided empirical inquiry in the study of international cooperation. Not that all has been sweetness and light, of course. Over the past few years, in particular, the disagreements among contributors to the cooperation under anarchy literature have been numerous, their internecine controversies heated and intense. Thus far, however, the debate inspired by After Hegemony has focused almost exclusively on the first leg of the institutions → cooperation → mutual-gains two-step: How, exactly, do institutions help their signatories overcome obstacles to collective action? Is it by weakening their incentives to free ride (as Keohane himself proposed), or do institutions work their magic in other ways—by providing focal points, for example, or by altering the identities, and hence the underlying preferences, of the state actors who participate in them?⁵

But what about the second link in Keohane’s causal chain? Must international cooperation always leave each actor better off than it was in the initial, pre-cooperation status quo? While most scholars see international institutions as efficient, Pareto-improving responses to collective-action problems, this is not, it turns out, the only logical possibility.

To demonstrate this point, the next chapter begins laying out a very simple theoretical model in which two states—the ultimate beneficiaries of cooperation—encounter an exogenous shock of some kind, be it a sudden shift in their relative structural positions or, more likely (since structural shifts tend to occur rather slowly), a sudden change in one of the state’s preferences, perhaps itself a consequence of a recently held election. In the wake of this shock, our two prime movers—neither of which had previously been inclined to coordinate its policies with the other—observe that they now have certain interests in common and thus could benefit, potentially a great deal, by structuring their future interactions in accordance with a common set of mutually agreed upon rules and procedures. In moving toward a more cooperative relationship, the two states (or, more accurately, their respective leaders) believe they will be able to pursue their joint interests more effectively than would otherwise be the case.⁶ What are their joint interests? Expanding the total pool of customers available to each other’s domestic producers is one possibility. Alternatively, the two states—let’s call them A and B—may be hoping to reduce the probability of their subsequently becoming embroiled in a costly military conflict. For present purposes, however, what matters is less the specific nature of A and B’s common interests than the fact that the cooperative arrangement they devise—their new multilateral regime—permits these interests to be realized.

Having identified the motivations of the regime’s prime movers, I then take a look at the loser’s side of the story: Why might other actors in the international system choose to take part in the new regime established by A and B, its enacting coalition? Note that the motivations of these other cooperators may be very different from those of the regime’s original sponsors: although the prime movers have an understandable interest in complying with what is, after all, their regime, that same regime need not further the interests of other states. In contrast with A and B, who comply because they anticipate utility gains relative to the status quo, these other states—call them C and D—may actually prefer the original, noncooperative state of the world. But why, then, would C and D ever agree to come onboard? The answer because A and B make them is too simple. What we would like to know, of course, is how A and B do this. How, precisely, do our two enactors succeed in getting other states or governments—some of whom would be perfectly content with the status quo—to agree to cooperate? The plot thickens.

One possibility is that A and B resort to force, threatening these other actors (i.e., C and D) with economic sanctions or other forms of direct coercive pressure. These days, however, one rarely finds states being forced to take part in new multilateral ventures. To the contrary, states often line up to be admitted into cooperative institutions whose current members do not even want them. Are we therefore to conclude that international cooperation and institution building really are mutually beneficial affairs, that (absent blatant acts of coercion) everyone involved really does expect to derive utility gains relative to the status quo ante?

Not necessarily. The fact that cooperation and institution building are matters of voluntary choice does not imply that the institutions in question raise (or even that their signatories expect them to raise) their baseline utilities.

Given the losers’ preference for a world in which these arrangements did not exist, theoretical inquiry properly begins with the disruption of the initial, pre-cooperation status quo. How might the winners succeed in removing this status quo from the choice sets of the losers? While engaging in coercive diplomacy is one way, there is often, I will argue, an easier alternative: the new regime’s beneficiaries can simply "go it alone."

By pursuing this line of argument, this book shows how institutionalized cooperation by one group of actors (the winners) can have the effect of restricting the options available to another group of actors (the losers), altering the rules of the game such that members of the latter group are better off playing by the new rules despite their strong preference for the original, precooperation status quo. Once the winners seem likely to establish their new cooperative structure—something they would do even if they were destined to be its only members—the losers conclude that being left out would be even worse than joining. Their entry decisions are, in this sense, purely voluntary; indeed, the losers may well plead to be admitted into the new arrangement. And yet, despite their unabashed enthusiasm for inclusion, the losers would actually be much happier if this arrangement had never been created.

While everyone who studies international cooperation is aware that power matters, most of the time it is either bargaining power (e.g., Fearon 1998; Krasner 1991; Oye 1992) or coercive power (e.g., Martin 1992a, O. Young 1983) that they have in mind. What I am talking about here, by contrast, is an altogether different kind of power, one that, unlike bargaining power, yields absolute—not just relative—losers and yet does not, like coercive power, require the winners to force their opponents into submission. At no point does anyone in the power-politics model I put forward coerce, or even attempt to coerce, anyone else. Nor, for that matter, do the beneficiaries of cooperation engage in linkage politics.⁷ The winners do not get the losers to acquiesce by compensating them (via side-payments) in some other sphere of interaction. The losers acquiesce because they know that the winners are in a position to proceed without them.

In a myopic sense, of course, the losers really are better off cooperating. Yet—and this is the important point—that is only because the winners have presented them with a fait accompli. The resulting cooperative regime is not, then, the Pareto-improving outcome it outwardly appears. On the contrary, even when all cross-issue deals and side-payments are taken into account, the losers would still be happier if the original, noncooperative status quo had never been disrupted. Faced with a choice between joining the winners’ new cooperative system or being completely shut out, the losers enthusiastically submit their applications for membership. They do so, however, only because the winners’ actions have had the effect of removing the status quo from their choice sets, leaving them with what they view as a bad option (cooperating with the winners) and an even worse alternative (incurring the costs of exclusion).

While this perspective offers a new way of thinking about why nations cooperate, it also provides new insight into how they cooperate. Among other things, for example, it suggests that the institutional preferences of certain cooperators (the governing officials who make up the enacting coalition) are likely to count for a great deal—and those of other cooperators, and of the collectivity as a whole, for practically nothing. Given the enacting coalition’s ability to proceed unilaterally, its members do not really need to take anyone else’s preferences into account at all. Whichever forms of cooperation they prefer will be the ones adopted in the end, whether or not these forms are conducive to the interests of other regime participants. But of course this simply moves the analysis back a step: Which kinds of institutional structures would the first group of participants—the enactors—tend to find most attractive?

The obvious answer is the kinds that are most likely to endure. But this begs the question: endure what? What future uncertainties or threats might the enactors be particularly concerned about, and therefore try especially hard to mitigate, as they go about the task of designing their institutions?

One might think that their greatest concern would be the possibility that one (or more) of the governments that made up the enacting coalition might refuse to hold up its end of the deal. Yet—as I am by no means the first to point out⁸—the international relations literature has tended to exaggerate the vulnerability of cooperative arrangements to free riding and other forms of opportunistic behavior. If the rules of the game are clear, if compliance is easily monitored, and, most importantly, if the future stream of cooperative benefits or payoffs is expected to be substantial, members of the enacting coalition would presumably have strong incentives to honor one another’s commitments from the outset.

In fact, when it comes to the issue of regime durability, the real questions have less to do with the incentives of the enacting coalition than with the incentives of its domestic successors. Although the aforementioned benefit stream may be large for the political elites who make up the enacting coalition, it may be small, or even nonexistent, for the officeholders who succeed them. Consequently, the latter might wish not just to defect from the cooperative regime, but to destroy it altogether. While this would deny the beneficiaries of cooperation the utility gains they might otherwise have enjoyed, in the end cooperation is like beauty: its value is in the eye of the beholder.

Insofar as their domestic opponents have their own conception of what it means to be better off, the governments that comprise the enacting coalition have a stake in finding a way to discourage one another’s internal successors from abandoning the international arrangement that they, the prime movers, are about to create. Why, it might be asked, wouldn’t the enacting coalition’s go-it-alone capabilities afford sufficient protection against this possibility? Wouldn’t the fear of exclusion suffice to keep the newly empowered beholders from sabotaging international regimes they dislike and whose initial creation they opposed? The answer is no—at least not necessarily—for unlike the enacting coalition’s external enemies (the losers in other countries), the coalition’s internal enemies (its domestic opponents) would be in a position to single-handedly bring down its new cooperative structures. Should they wish to do this, their withdrawal from these structures is all that would be necessary, since without their ongoing participation, the regimes in question would effectively cease to exist. And because these regimes would no longer exist, the fear of exclusion could hardly be expected to operate as a reliable deterrent.

This is not to say that the losers would incur no costs in opting out of (and thereby destroying) the regimes they inherited upon taking office, particularly if these arrangements had already been up and running for some time. Indeed, the prime movers could be counted upon to try to make these costs—the price that their (potentially) hostile successors would pay for exercising their exit options⁹—as high as possible. The flip side is at least as important, however, for one would also expect the enactors to try to reduce the costs to their regime’s would-be destroyers of staying in the regime. In short, the go-it-alone coalition’s interest in co-opting their domestic opponents might well influence the types of cooperative institutions they devise, inclining them toward thicker arrangements whose terms are not fully specified at the outset and thus leave the losers some (albeit limited) scope for ex post changes in the initial rules of the game.

This, it seems to me, is exactly what logic would suggest. The problem with fully specifying the terms of cooperation ex ante is not that it would be more time-consuming, and hence less efficient, than allowing for subsequent revisions or clarifications over time. The problem is that it would deny the enacting coalition’s domestic successors any opportunity to moderate these terms, reformulating—or simply reinterpreting—them in ways intended to make their continued participation in the regime somewhat less burdensome than it would otherwise be. It is for this reason, I will argue, rather than out of the (narrowly construed) efficiency considerations emphasized by previous scholars, that the contractual terms embodied in many of today’s international treaties take the supranational forms they do. Delimited though it is, the flexibility afforded by such arrangements works to co-opt the enacting coalition’s domestic opponents, reducing their propensity to mount a serious challenge to the new regime if, in future years, they should ever get the opportunity.

All of this presumes, of course, that the utility consequences of interstate cooperation may differ radically depending on which parties, or even which individuals, happen to control domestic public authority. Allowing that internal leadership changes and government turnover can bring about abrupt shifts in state preferences certainly complicates matters, requiring as it does that the analyst peer inside the black box of domestic politics. In this case, however, the analytical payoff is well worth the effort. For if I am right, the realist view of states as internally undifferentiated units¹⁰ threatens to obscure what may well be the central motivating force behind the ongoing supranationalization of international cooperation: the vicissitudes of domestic politics inside the pivotal enacting states.

In sum, power may permeate and infuse international cooperation just as it permeates and infuses international conflict. In the case of conflict, however, relations among nations are not strictly voluntary, and so the means by which the strong exercise power over the weak—outright bullying, brute force, and the like—are more transparent. Make no mistake: there are important differences between go-it-alone power and what might be termed structural or institutional power. Whereas the victims of go-it-alone power are outsiders—public officials in countries located outside the enacting coalition—the losers in the case of institutional power are insiders—the enacting coalition’s domestic successors. Other differences have to do with the powerful actors’ intentions. In one case (go-it-alone power) the pressure that is brought to bear on the losers is inadvertent; in the other case (institutional power) the enacting coalition goes out of its way to alter the losers’ decision-making calculus, deliberately moderating the costs of cooperation on actors who, were they to exit the cooperative arrangement in question, would effectively bring about its demise. As important as these distinctions are, however, my main concern is with the more fundamental point that power’s influence on international affairs may be far more pervasive than many scholars, including even many hard-nosed realists, currently appreciate. Whether one is talking about go-it-alone power or institutional power, the end result is the same: Without in any way being bullied or coerced, governments choose to participate in cooperative international arrangements they genuinely, and intensely, dislike.

THE PLAN OF THIS BOOK

Are all of the participants in today’s international institutions truly better off than they were in the pre-cooperation status quo, or can cooperative institutional arrangements spread like wildfire even when their most ardent supporters fully expect to get burned in the process? How much of the current trend toward collective decision-making, institutional flexibility, and delegated authority is being driven by efficiency considerations, and how much by power politics?

The first four chapters of the book provide the theoretical groundwork necessary for answering these questions. Chapters 2 and 3 offer a new perspective on the question why do states cooperate? After sketching out the conventional explanation for the recent proliferation of cooperative initiatives, I introduce a simple model to demonstrate how an apparently clearcut case of voluntary cooperation may in fact represent an instance of go-it-alone power. In this model, the weak cooperate only because the strong present them with a fait accompli. This basic framework is further developed in chapters 4 and 5, where the focus shifts from the question of why nations cooperate to that of how they cooperate: Is there a power politics of institutional design and, if so, how might its internal logic differ from that of the efficiency-centered perspectives we have grown accustomed to seeing in the literature?

The next four chapters supplement this theoretical discussion with analyses of two of the most important developments in the history of regional trade and monetary relations: the inaugurations of NAFTA (chapters 6 and 7) and the European Monetary System (chapters 8 and 9) during the mid-1990s and late-1970s, respectively. In offering these case studies, I hope to show more concretely how theories of international cooperation that fail to appropriately integrate power into their larger explanatory edifice can end up obscuring more than they reveal. At the same time, however, these case studies also serve an additional purpose: They allow me to explore some of the many interesting nuances and subtleties in the book’s larger theoretical argument.

Before launching directly into this argument, I should alert the reader to some of the empirical difficulties that it presents. One of these has already been noted: the outside observer must determine how well off a particular member government would have been if, counterfactually, that government had not joined the cooperative arrangement in question. To perform this sort of welfare comparison, the analyst needs to know (among other things) what that government’s leaders were originally seeking to accomplish during their tenure in office. At some point, therefore, the analyst is forced to specify the underlying preferences of the relevant government actors. This is admittedly a complicated task. On the other side of the coin, however, those scholars who embrace the conventional wisdom on international cooperation—that, as Keohane puts it in After Hegemony (1984, 88), [international] regimes are developed . . . because actors in world politics believe that with such arrangements they will be able to make mutually beneficial agreements that would otherwise be difficult or impossible to attain—seem to have no difficulty assuming (i.e., without even attempting to specify the underlying state or government preferences) that collective action, when it occurs, is in fact mutually beneficial.

In any event, the fact that the book’s core hypotheses are difficult to test does not mean they are impossible to test. To the contrary, parts III and IV offer a whole series of tests, albeit ones that are more suggestive than definitive. The NAFTA and EMS stories I flesh out below may not qualify as crucial cases for my theoretical argument. At the same time, however, it would certainly have been possible for my investigations of these cases to turn up no evidence whatsoever either that defensive considerations—rather than any expectation of positive-sum gain or fear of coercive sanctions—were what motivated some of the participants in NAFTA and the EMS to take part, or that these signatories would actually have been better off, in the sense that their governments would have been less disadvantaged politically, had the pre-cooperation (i.e., pre-FTA or pre-EMS) states of the world remained in effect. The fact that I do turn up such evidence, while it may not prove the empirical validity of my theoretical claims beyond any doubt, does at least lend them plausibility—which, given the counterfactual and preference-specification problems noted earlier, may be all one can reasonably expect at this stage of the game.

Let me note, finally, that in mapping out the logic of go-it-alone power, I am not seeking to account for international cooperation in any particular issue-area or geographic region. To be sure, I believe the theoretical approach developed here offers important new insights into the origins and structural foundations of NAFTA and the EMS. In singling out these two cases, however, I do not mean to suggest that this approach is applicable only to North America and Western Europe or only to cases of economic cooperation. In fact, though I dwell here upon the NAFTA and EMS experiences, there is every reason to think that the book’s larger analytical framework can also be used to shed light on other (e.g., non-Western, noneconomic) kinds of supranational institutions, and even other kinds of domestic institutions.¹¹ In much of what follows, however, I deliberately set aside these generalizability concerns so as to keep the analysis as clear and simple as possible.

¹ See, e.g., Whalley 1993.

² See, among many others, Grieco 1990; Krasner 1991; Garrett 1992; Knight 1992; Oye 1992; Morrow 1994; and Fearon 1998.

³ Cf. Olson 1965.

⁴ See, e.g., Kahler 1995.

⁵ The preference-altering transformational impact of EU membership is explored in Risse-Kappen 1996, and Sandholtz 1993. On the logic of focal points, see Garrett and Weingast 1993.

⁶ Although conceptualizing the beneficiaries of cooperation as states or nations is fine for present purposes, I will argue later on that it is better to think of the prime movers not as states but as governments; cf. Moravcsik 1998.

⁷ Cf. Frieden 1994; Garrett 1994; Martin 1992a and 1994; and Sebenius 1983.

⁸ See, e.g., Grieco 1988.

⁹ The phrase is Hirschman’s (1970).

¹⁰ See esp. Waltz 1979.

¹¹ The latter strikes me as a particularly promising avenue for future research (see chapter 10). Indeed, much of what I will have to say on the subjects of power, cooperation, and institutions was foreshadowed in the work of an Americanist, the great E. E. Schattschneider. See esp. Schattschneider 1960. That the concept I term go-it-alone power bears a strong family resemblance to Schattschneider’s notion of agenda power will become apparent as we go along. Nevertheless—and for reasons I discuss at some length in chapter 3—these are not the same concepts. Quite the contrary, the differences between go-it-alone power and Schattschneider’s earlier conceptualization are as noteworthy, it seems to me, as their similarities.

PART I

Why Do Nations Cooperate?

CHAPTER 2

Institutions, Collective Action, and the Prospect of Mutual Gain

THE NOTION that international institutions are worthy of close examination first gained broad acceptance during the early postwar years. Over the course of the 1960s and 1970s, however, the consensus among scholars shifted 180 degrees. As structural realism came to supplant the largely descriptive institutionalist traditions of the 1940s and 1950s, the supranational organizations that had once been the major focus of attention—principally, the United Nations—were deemed irrelevant or epiphenomenal. For realists, power and interests were the predominant forces driving world politics. Institutions were seen as mere facilitators, translating the interests of powerful states directly and unproblematically into international outcomes.¹

It would be natural to suppose that this movement toward an institutionfree, balance-of-power theory of international relations arose as a reaction to theoretical deficiencies in the institutionalist literature of the 1940s and 1950s. And, indeed, much of this literature was deficient, with scholars analyzing organizational rules and procedures indiscriminately, almost never pausing to ask where they came from or how, if at all, they influenced or constrained the behavior of member states.² Only recently, however, have realists felt the need to justify their neglect of institutionalist scholarship.³ From an empirical standpoint, supranational institutions just did not seem to matter very much, as states often violated their institutional obligations with impunity. Nor were realist scholars alone in thinking that institutions could be safely omitted from the analysis of politics. Proponents of pluralist theory (e.g., Truman 1951) also tended to leave institutions out of their larger explanatory framework.

By the mid-1980s, though, institutions had once again begun to emerge as a subject of interest among students of international affairs. And this time things were different: Unlike their intellectual forebears, proponents of the new institutionalism recognized the importance of grounding their analysis in an overarching theoretical framework.

Robert Keohane stands as the leading figure in this effort to go beyond the legalistic, institution-by-institution approach characteristic of earlier institutionalist thinkers. Initially dubbed regime theory, the voluminous literature spawned by Keohane’s work⁴ took direct aim at Waltz and other realists who, in their preoccupation with international conflict and the balance of power, had failed to conceive of interstate cooperation as anything more than the spontaneous formation of status-quo-preserving, marriage-of-convenience alliances. Since rechristened neoliberal institutionalism (Grieco 1988), regime theory has become the theoretical point of departure for the bulk of all academic writing and research on the institutional developments of the 1980s and 1990s.⁵

The purpose of the next four chapters is to subject this literature to critical theoretical scrutiny. While others have engaged in similar efforts, much of their criticism has focused on what I regard as secondary issues. Thus, proponents of the new institutionalist paradigm are often taken to task for failing to adequately integrate the role of ideas, identity formation, and culture into their larger explanatory framework.⁶ While a source of concern, however, this is not, it seems to me, a crippling defect. Nor is the problem that international institutions do not really matter, in the sense of having a major (independent) impact on the flow of world events.⁷ Insofar as the institutionalist paradigm is deficient, the fault lies primarily, I will argue, in its theoretical one-sidedness: Why assume that multilateral institutions always and everywhere facilitate mutual gains? Though it is true that states throughout the world are scurrying to join such institutions—or, when shut out, are rushing to create new, smaller-scale institutions of their own—one cannot assume that the resulting progression from anarchy to organization is therefore welfare enhancing. Undergirded by rational choice assumptions about state behavior, the model of international cooperation I will introduce in this part of the book raises the possibility that the past several years’ frenzy of supranational activity, rather than fostering peace and prosperity, will end up working to the great disadvantage of some (perhaps most) participating nations.

BACK TO BASICS: WHY DO STATES (SOMETIMES) ENGAGE IN COLLECTIVE ACTION?

The publication of After Hegemony (1984) could not have been better timed. Emboldened by their success in revolutionizing the study of economic organization, rational choice theorists were about to launch a series of forays into other disciplines. In the field of international relations, Keohane’s justly celebrated book paved the way.

The starting point for the book’s argument is a simple observation: Policies and events originating in one nation often have large—albeit indirect—effects on the welfare of citizens and governments in other nations.⁸ Sometimes these spillover effects are mutually beneficial: If German economic policy makers succeed in boosting Germany’s domestic living standards, German consumers have more money to spend on goods and services, and French workers who specialize in producing exports for the German market enjoy a windfall. Other times, however, these spillovers are unwanted, as when a currency devaluation in the United States reduces the competitiveness of U.S. imports from Europe, prompting European countries to devalue their own currencies, thereby precipitating a further devaluation by the United States, and so on.

In situations where independent policy making would give rise to what economists call negative externalities, Keohane drew the not unreasonable inference that nations would be better off actively coordinating their policies. Thus, the political leaders of country A could agree not to pursue policies harmful to country B so long as their political counterparts in country B refrained from undertaking policies harmful to country A, and vice versa. At bottom, then, cooperation is about reciprocity. As Keohane (1988, 380) writes: "When cooperation takes place, each party changes his or her behavior contingent on changes in the other’s behavior."⁹ The result of this exchange—of what Helen Milner (1997, 7–8) characterizes as the adjustment of one state’s policies in return for, or anticipation of, the adjustment of other states’ policies—is that all parties end up better off.

But if all parties benefit, why haven’t state leaders always been as willing to cooperate as they are today? Drawing on Keohane’s work, neoliberal theorists give two answers. The first is that nations today are more interdependent than they used to be, and thus the potential for negative externalities is greater—as, it follows, are the prospective benefits of multilateral cooperation—than was the case in previous years.¹⁰ That said, these benefits may not be sufficient, at least not in themselves, to induce the kinds of mutually beneficial policy adjustments that would leave all states better off. While there are many reasons for this, Keohane and his followers emphasize the potential for ex post defections: A state that suspects its cooperative overtures will not be reciprocated will rationally avoid making these overtures in the first place.¹¹

This line of analysis suggests a second reason for the recent proliferation of new multilateral arrangements. Part of the explanation may be that the knowledge of how institutions work—and what makes them work well (Keohane 1988, 393) is becoming more prevalent. By using this knowledge to design better institutions, national leaders have ensured that the one-shot payoff available to opportunistic defectors amounts to something less than the ongoing stream of collective benefits available to long-term cooperators, the result being utility gains for everyone involved.

FIGURE 2-1. The Positive-Sum Logic of Neoliberalism

Neoliberal theorists can be credited with putting forward a cogent explanation for recent developments, one that is grounded in clear, deductive reasoning. Before we accept that explanation’s universal validity, however, it is worth taking a moment to consider the second link in its implicit causal chain (Figure 2-1). Is it not conceivable that institutionally-mediated cooperation, rather than resulting in the realization or capture of collective gains, works exclusively to the benefit of the actors who set the process in motion, increasing their power and prosperity at the expense of other actors who participate only because they have no better option?

Not only is this conceivable—it is often, I would argue, exactly what happens. Thus far, however, neoliberal theory’s supposedly tough-minded realist critics have taken a different line of attack. Instead of questioning whether multilateral cooperation does in fact help states manage globalization, realists have devoted their energies to showing that—the presence of international institutions notwithstanding—cooperative interstate relations are next to impossible to sustain, at least on a voluntary basis, for any significant length of time. Realists, in other words, have taken issue with the first arrow shown in Figure 2-1, not the second.¹²

THE REALIST ASSAULT ON NEOLIBERAL THEORY

In the years following the publication of After Hegemony, Keohane’s claims about the role of institutions in facilitating cooperation among nations emerged as the subject of intense controversy, inspiring a lively, sometimes rancorous debate between neoliberal scholars, who were broadly sympathetic to Keohane’s thesis, and realist scholars, who took the view that international institutions were essentially irrelevant. At first, as mentioned earlier, those in the latter camp felt little need to justify their position. Again, realists have always been more interested in understanding what states can do to preserve the status quo than in exploring how, by devising appropriate multilateral institutions, they might be able to make things better (and, as an empirical matter, none of the multilateral arrangements created in the immediate aftermath of the Second World War seemed to be having an especially large impact anyway).¹³ While numerous countries participated in institutional arrangements like the GATT, the European Community, and the UN, realists doubted that any of these arrangements would ever be permitted to encroach upon their member states’ freedom of maneuver—at least not in any serious or meaningful way.¹⁴ By the mid-1980s, however, these and other multilateral institutions had begun playing a significant role in the flow of regional and global developments, and realists seemed increasingly out of step with the world around them.¹⁵

At that point, realist critics of neoliberal theory could have retreated from their hard-line, anti-institutionalist position in favor of a more nuanced perspective. After all, the acknowledged father of neorealism, Kenneth Waltz, had never seen any reason why the logic of anarchy should prevent countries from behaving in neoliberal fashion, even if the resulting rules, institutions, and patterns of cooperation [were all likely to be] limited in extent and different from what they might otherwise be.¹⁶ But engaging in the messy business of distinguishing institutions that effectively promoted cooperation from those that were merely epiphenomenal was not what most realists had in mind. Instead, they continued to insist that international institutions had little (independent) impact on world politics, and that liberals who argued otherwise were relying upon faulty logic.¹⁷

Where, exactly, did liberals go wrong? According to one line of realist criticism, it was in neglecting the importance of relative gains.¹⁸ So much attention has been lavished on the relative-gains critique that the liberal-realist debate of the past several years is sometimes equated with the controversy over relative gains.¹⁹ Casting the debate in this way is a mistake, however, for this was but one flank in a larger realist offensive.

Are International Agreements Enforceable?

For many realists, an even bigger problem with neoliberal theory was its failure to explain how cooperating nations overcome the practical difficulties of treaty enforcement and compliance.²⁰ According to Keohane, multilateral cooperation actually required little more than that each partner be able to distinguish fellow cooperators from free riders. Once these distinctions were clear, the assumption was that each participant would take it upon itself to reward the former and penalize the latter, all in a strictly decentralized fashion.²¹ In fact, this assumption was crucial, for if prospective free riders did not expect to be punished, the whole cooperative system would break down, preventing everyone—free riders and cooperators alike—from enjoying the ensuing benefits of policy coordination.

To be effective as a deterrent, however, sanctions (against free riders) often need to be leveled by a large number of (cooperating) countries. Naturally, this raises the possibility that each member of the coalition of potential sanctioners will attempt to pass the buck, thereby impeding the decentralized enforcement mechanism so fundamental to the success of long-term cooperation. And the danger of buck-passing aside, what incentive would good-faith cooperators have to impose retaliatory sanctions anyway? By the