Has realism become cost-benefit analysis?

sysop

Has realism become cost-benefit analysis?  

Has realism become cost-benefit analysis? A review essay

International Security v. 26 no2 (Fall 2001) p. 132-54

Lloyd Gruber, Ruling the World: Power Politics and the Rise of Supranational Institutions.
Princeton, N.J.: Princeton University Press, 2000.

Contemporary realism is at a crossroads. If realist claims are defined specifically--predicating particular responses by nations to defined patterns of international power--they cannot be fully sustained. Nations do not respond predictably when confronted by superior power; nor do they always act as if a zero-sum game existed between countries. The amount of cooperation in the international system is not a constant. If, on the other hand--as has been happening recently--realist claims are defined more generally--and preferences, beliefs, and institutions are admitted to have a causal role--the distinctive contribution of realism is lost in a welter of other variables. Defined generally, realism is in danger of being reduced to cost-benefit criteria in which empirical outcomes are unclear.

In Ruling the World, Lloyd Gruber furthers these generalist conceptions by interpreting modern international and supranational institutions in a new way in which national self-interest, rather than power compulsions, determines the outcome.(FN1) His synthesis seeks to explain why institutions should be growing in power and membership and yet--from his point of view--large numbers of participants should be dissatisfied, preferring a now-unobtainable status quo ante.

The points I seek to make are as follows: (1) Realism is moving away from relatively specific, clear, and testable claims about the nature of the international system to a more generalist formulation in which outcomes are increasingly indistinct or opaque; (2) the original formulation--"specific realism," as I shall call it--laid great stress on the balance of power and balancing operations to sustain the system; (3) in the new generalist realism--of which Gruber’s work on institutions is the latest and most developed example--consolidation, agglomeration, and bandwagoning rather than power oppositions are characteristic; (4) realist explanations, however, are still there--based not on bullying and the threat of force but on manipulations of state interest; and (5) Gruber is right about the decline of power oppositions and the rise of institutions but wrong about its causes and likely results.

SPECIFIC REALISM

The early realist thinkers, such as Thomas Hobbes and Niccolò Machiavelli, understood that conflict, deception, bullying, coercion, and war were the crux of international politics. The fifteenth-century military leader Cesare Borgia--an adept practitioner of realist principles--might dupe or kill an opponent. Small duchies might well be annexed or wiped out by larger states in quest of national grandeur. Great powers would survive because while men in the state of nature lived in fear of violent death, states did not necessarily do so.(FN2) They could outlast the demise of a ruler or the loss of a province. Twentieth-century realists such as Bertrand Russell, E.H. Carr, and Hans Morgenthau also believed that nations had to balance against the power of an aggressor.(FN3) If they did not, one state might eventually achieve hegemony and abolish the state system altogether. All of these "specific realists" believed that international relations were a kind of state of nature in which the rule of the stronger prevailed. In that formulation "clubs are trumps," and those who did not acknowledge this dictum fell by the wayside. According to specific realists, there was no law or division of labor among nations, so independence and territorial integrity depended on self-help.(FN4) But the essential ingredient in specific realism was the existence of a balance of power or a tendency in that direction. Kenneth Waltz contends that a balance of power is the natural outcome of interstate rivalry. States might not actively seek a balance. Rather, "balances of power tend to form whether some or all states consciously aim to establish and maintain a balance."(FN5) Thus the balance of power emerges from the pattern of opposition among states.(FN6) In Glenn Snyder’s more recent formulation, nations have to balance one another’s power, because "with each successful conquest, the aggressor presents a greater threat to the security of the remaining states because of its increasing capability and increasing confidence that it will be unopposed. Hence the defenders’ collective good in blocking further encroachments increases dramatically with each round.... At some point, the cost of resistance will be assessed as lower than the cost of allowing the aggression to succeed."(FN7).

Some specific realists, however, believe that nations cannot rely on the competitive nature of the international system to foster balance by itself.(FN8) Defenders not only have to resist if attacked, but have to take action to create a balance. States need to favor the weaker side in any competition.(FN9) As Morton Kaplan phrased it in rule 4 of the balance-of-power system, states have to "act to oppose any coalition or single actor which tends to assume a position of predominance with respect to the rest of the system."(FN10) Michael Nicholson went even further, stating that balance-of-power principles "require the actors to deviate from normal self-interest to having goals for the system as such."(FN11) These specific realists thus posit particular behavior on the part of states. If states deviate from establishing a balance, they and the system will suffer.(FN12) In fact, it is the presence of a balance and balancing behavior to sustain it that resides at the heart of specific realism.(FN13).

GENERALIST REALISM

Recently, however, a new variety of realist thought, which I term "generalist realism," has emerged that challenges the older specific realist formulation.(FN14) In this view, nations may join institutions to forward their national positions, and the balance of power is no longer characteristic of international relations. Institutions, however, are subject to "collective goods" problems--that is, once created, those goods are available to all. This may cause some countries to "free ride" rather than cooperate in the institutional setting. Nonetheless, nations join institutions for reasons of self-interest, and these institutions can become important determinants in world politics. Balancing may or may not be in the interest of a country, and in institutional and other contexts bandwagoning may become the preferred strategy.(FN15) Alternatively, a distinct type of bandwagoining may be necessary to create a knife-edge balance of power.(FN16) The development of this new breed of realist theories poses important questions both for the antecedent tradition and for the opposition between liberalism and realism.

Unlike specific realism, generalist realism makes few assumptions about the responses of states in international relations. States will behave rationally, but not necessarily seek to perform balancing tasks. Influenced by this tradition, Snyder is uncertain about how far nations will go in upholding the system against attack.(FN17) In earlier years Paul Kecskemeti’s study Strategic Surrender outlined conditions in which countries, seeking to preserve their underlying social and economic orders, preferred surrender to fighting.(FN18) Such conclusions should not be surprising. The nature of the game presumed by specific realists is typically different from that predicated by generalist realists. To sketch the differences between specific and generalist realists, it is necessary to make a brief detour into their respective game-theoretic assumptions.

GAME THEORY, BALANCING, AND COOPERATION

In this section I seek to show that as the opposition of interests between states declines (as one moves from zero- or constant-sum games to those of variable sum), balancing becomes both less likely and less frequent. The prisoner’s dilemma game still retains balancing elements, because each side is willing to defect in face of the other’s cooperation. But in more cooperative games (stag hunt and chicken, for example), cooperation is a more likely outcome, and balancing (a negative response to a player’s actions) is less so. The net result is that bandwagoning, buck-passing, and cooperation become more characteristic as variable-sum outcomes emerge. In this sense, specific realism gives way increasingly to generalist realism as games evolve.

Specific realists are clearly partisans of the zero- or constant-sum game. The winners win what the losers lose, and common interests are absent. Constant-sum games foster balancing behavior, depending on the number of participants. The formulation.

A + B = 0.

clearly elicits balancing because if A goes up, B must go down. So does.

A + B = k (a constant).

On the other hand, if.

A + B + C + D = k (a constant),.

it is not clear who suffers from A’s increase. It may be B, C, or D. This is the essential insight that makes balancing less certain under multipolar, as opposed to bipolar, conditions. Nevertheless, with the sum of national positions equal to a constant, international cooperation will be limited and balancing will frequently occur.

If the sum of interests is a variable (providing for the possibility of an increasing-sum game), balancing becomes more uncertain, as in.

A + B + C + D = x (a variable).

In games of variable sum, the opposition of interests may be absent altogether, because an increase in A’s position may be taken up by an increase in x, rendering B’s action unnecessary. However, there may still remain a problem of cooperation. The prisoner’s dilemma game is variable sum, but it does not obviate conflict. Under conditions of the prisoner’s dilemma, players have an incentive to sanction one another even though cooperation offers the highest mutual payoff.(FN19) As incentives to sanction one another decline, as they do in chicken and stag hunt, however, cooperation becomes more likely, and the prospect of balancing declines. There is no dominant strategy, however, in either of these games.(FN20) The assurance II game leads to a dominant strategy of cooperation because the payoff to defection is lower. If, therefore, more cooperative variable-sum games have recently become more characteristic of international relations, bandwagoning has begun to displace balancing, thus leading theoretical orientations to move from specific realism to generalist realism. This does not suggest that the system has overall become more cooperative or that chicken, stag hunt, and assurance II now characterize the only relevant relationships of interest in the system. Prisoner’s dilemma remains an important feature of present-day international relations. But national action is much more unspecified and uncertain, depending on particular relationships in particular circumstances and varying conditions of cost and benefit.

These more general conclusions have found representations in the realist literature. Recent work by Emerson Niou, Peter Ordeshook, and Gregory Rose in addition to Robert Powell would put them in the category of generalist realists.(FN21) Niou, Ordeshook, and Rose have provided a formal analysis in which bandwagoning, rather than balancing, provides the last critical step to the formation of an effective balance of power against aggression. Until one potential aggressor has amassed 50 percent of the resources in the system, a balance against it is not certain, and system outcomes remain indeterminate. Powell has offered an intriguing analysis of alignment decisions in which bandwagoning or waiting are typically superior to balancing. Powell writes: "A state is less likely to be part of a winning coalition if it aligns with the weaker side and more likely to be part of a winning coalition if it aligns with the stronger side. This consideration makes bandwagoning more appealing."(FN22) Basing their analyses on non-zero-sum games, generalist realists frequently find bandwagoning and not balancing to be in the interest of interstate system members.

LLOYD GRUBER AND THE INSTITUTIONALIST ARGUMENT

Gruber’s account of international institutions contrasts with the existing literature. His conclusions combine generalist realism (featuring the accumulation rather than the diffusion of power) with a belief in the growing strength of international and supranational organizations. In this sense, his empirical findings are like those of neoliberal institutionalists, while his explanations are entirely of a realist variety. Existing realist views of international institutions are well known. At least two different realist views vie for influence. The first sees institutions as ineffective; the second concedes them some independent influence. In the first category, John Mearsheimer asserts that international institutions offer a "false promise" if they expect to control the power drives of nations.(FN23) It follows that member countries of the North Atlantic Treaty Organization (NATO), the European Union (EU), and the International Monetary Fund (IMF) do what is in their interest. They would pursue that interest whether the institutions existed or not. Recently, Kenneth Waltz argued that NATO survives primarily, if not wholly, as an agent of U.S. power. Other member states have little influence, and NATO’s rationale as a security alliance has been undermined.(FN24) In the 1930s the realist argument would claim that the League of Nations and its affiliated organizations did nothing to alter the basic conflicts among rivaling states. In regard to more recent institutional experience, Joseph Grieco depicts the negotiations between the United States and Europe concerning nontariff barriers in the aftermath of the Tokyo Round as a relative gains competition.(FN25).

Yet realists and others must be aware that despite such conflicts, the multilateral trade negotiations did not falter, and a successful end to the Uruguay Round was reached at the beginning of the 1990s with the creation of the World Trade Organization (WTO). At the EU meetings in Nice, both France and Germany sought to improve their relative positions in budget assessments and voting strength, but a treaty still emerged governing enlargement.(FN26) New institutions such as the North American Free Trade Agreement (NAFTA) and Asia-Pacific Economic Cooperation (APEC) have come into being, and a free trade agreement of the Americas and even a free trade arrangement (FTA) linking Europe and the United States (Trans-Atlantic Free Trade Area) have been proposed. If the International Monetary Fund did not handle the 1998 monetary crisis in the Far East with exemplary skill, no one can complain about its lack of influence or power. In addition, the continuing enlargement of both NATO and the EU testify to these institutions’ political and diplomatic vitality. The major problem in both cases has been keeping prospective adherents out, not in failing to attract new members. These outcomes are in diametric opposition to the role of the League of Nations and other institutions in the 1930s--when they either failed to restrain conflict or collapsed altogether.

Realists can of course claim that international and supranational organizations act only in their members’ interests, contributing nothing to international outcomes. Thus one can seek to see German domination of the EU or U.S. domination of NATO and the IMF as an explanation of the continued strength of such institutions. But this conclusion overstates national influence and downplays the constitutionalizing effect of institutions themselves. If this conclusion were true, there would be no difference between unilateral imperial domination and the outcomes of multilateralism. In fact, countries are prepared to join with Germany or the United States so long as these leading nations’ policies are constrained by and embedded in institutions. They would not be prepared, however, to follow the naked dictates of German or U.S. power. Indeed, if NATO is simply the expression of increasing U.S. strength, it should lead, on realist grounds alone, to a balance-of-power reaction against it. In fact, in regard to both NATO and the EU, centripetal influences have heavily outweighed centrifugal ones.(FN27).

This recognition has led to a new and distinct realist response to institutions. Joseph Grieco, Charles Glaser, and Randall Schweller--realists all--have come to see institutions as an independent influence in international relations, one that cannot merely be reduced to the policy of their leading member. Grieco has recently come to see a new "binding hypothesis" as a means of enabling weak states to improve their positions in international organizations even at the expense of established members. Weaker states "bind" themselves to the institution because they achieve greater voice within it. Glaser believes that multilateral institutions in arms control can help overcome the security dilemma. In explaining institutional persistence and power, Schweller and David Priess have testified to other influences--such as intentions, democratic governments, and interdependence--contributing to their position.(FN28).

Gruber also believes that international and supranational institutions exert an independent influence over state behavior, but he does not explain it on the bases of other variables. His generalist realism offers a completely realist synthesis even though there is no coercion or bullying by great powers. Moreover, he argues that bandwagoning rather than balancing is the typical result of the development of new institutions. Countries flock to join new and powerful organizations in Asia, Europe, and North and South America even in the absence of a bipolar opponent. But despite their evident popularity, institutions do not confer commensurate and symmetrical benefits. Gruber writes:.

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 ... 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. (p. 16).

In this way, Gruber’s book represents an attack on the neoliberal theory of institutions, particularly on the work of Robert Keohane, Lisa Martin, and Helen Milner, with Keohane as the primary target.(FN29) Liberal institutionalists have seen institutions arising from the "negative externalities" stemming from uncoordinated national behavior. In institutions reciprocity is fostered, and countries change their behavior on the basis of adjustments made by other nations.(FN30) This leads to the neoliberal pattern in which institutions produce voluntary cooperation, and this in turn results in utility gains for each cooperating member state. International agreements were supposedly enforceable because members could distinguish compliance from noncooperation and would sanction the latter. Even realists accepted a portion of this argument, asserting that mutual benefit could stem from institutional behavior.(FN31) In neoliberal theory, however, there remained the problem of enforcement. Sanctions might have to be levied against free-riding consumers of a collective good, and potential sanctioners might instead pass the buck, deflecting the task to others or neglecting it altogether. Deviant behavior would then go unpunished. It was therefore difficult for both realists and liberals to believe that international institutions would always work effectively in an anarchic environment. To overcome prisoner’s dilemma incentives, Keohane moved away from strictly rational arguments to support international organization.(FN32).

Yet Gruber, more than many neoliberals, believes that international institutions are effective and growing in strength. He derives this conclusion both from theory and from a study of institutions--particularly the European Monetary System (EMS) from 1979 to 1994 and the North American Free Trade Agreement in the late 1980s and 1990s. These institutions do not appear to be prone to collective action problems. Members do not cheat or free ride. They abide by the rules or are forced to withdraw. How is this possible?

Essentially it is because the formation of the organization by the founding members has changed the status quo for subsequent members. Many or most of the latter would have preferred the status quo ante--with tariff protection or capital controls, for example--but once major powers have formed an organization to limit these restrictions, outsiders cannot afford not to join. They need the markets and the capital from which they would be excluded. Thus the issue for Gruber is not the distribution of welfare gains within a range of Pareto optimality, as in Figure 1.

Assuming Y is the dominant founding member and X the latecomer, X is not allowed to choose among favorable positions on the contract curve, mn. In Figure 1, X does not have the choice of proceeding along the line segments OC, OB, or OA, as we shall see below. Even point A is denied to X (though it is the least favorable of the three points on the curve). The outcomes that X faces, according to Gruber, are actually much more unfavorable than those in Figure 1. Figure 2 depicts Gruber’s analysis of the situation.

A lesser potential member state, X, is at SQ1 at the beginning of play. The formation by Y and others of a new organization reduces X’s utility to SQ2 with all the disadvantages of being left out, including lesser access to markets, capital, and so on. Thus while SQ1 would still be X’s preferred option, X is forced to join the organization and move to point K. Point K, of course, still ensures losses as compared to the original status quo. In contrast, Y’s utility (that of the founding member) increases throughout the two-step process.

Does this depiction capture important aspects of international institutional reality? Gruber believes it does. Left-out states know that the international arrangements they dislike are capable of functioning without them. They understand that other, perhaps larger, countries benefit, and that major countries would continue to support them even if no others joined. Gruber claims: "The fact that the winners would still realize welfare gains relative to the initial status quo affords them a kind of ’go-it-alone power, the importance of which--indeed, the very existence of which--has been overlooked by previous scholars" (p. xiv).

Gruber is careful to observe that his account of the processes of formation of new international organizations is not the only one possible.(FN33) First, there might be no initial disagreements among the actor’s underlying preferences, in which case all members might obtain sizable and mutual benefits. Second, after the organization has been formed, the spillover effects of not joining might not be both negative and large. If they are, however, those left out will have a huge incentive to join. Third, smaller states might enjoy some holdup potential, preventing founding members from forming an organization on their own. Fourth, for Gruber’s argument to apply, the founders cannot be worse off if additional actors consume the good they produce.(FN34).

These potential limitations, however, turn out in Gruber’s analysis to be less restrictive than they might initially appear. That left-out countries might fear negative spillovers is attested by the alacrity with which they seek to join. Holdup potential is not widespread. Key countries in both Europe and America have been able to go it alone if others did not join. Usually the economic clubs in Europe and North America could be expanded without diluting the gains of the major players. Finally, even where major goals (such as the lowering of tariff barriers) are common among potential institutional participants, there will be and are differences over which goods should be privileged. Thus the possible qualifications do not appear to present a major threat to Gruber’s analysis. If so, Gruber’s argument might seem to provide a satisfactory explanation for the development and persistence of international organizations.

In Gruber’s analysis, power factors (though not coercion) provide the answer. They operate subtly and indirectly by depriving states of previous options. Military power plays no role, but previously recalcitrant countries are still effectively forced to join. Thus the European Monetary System included a hesitant Italy.(FN35) The U.S.-Canadian Free Trade Agreement brought in reluctant Mexico to form NAFTA, and the EU has a legion of weak applicants for membership. Aspiring member countries might have preferred the status quo ante, but when key nations formed the new organization, they could not afford to stay outside.

Realists had previously argued that cheating would undermine institutions founded on public goods principles. Countries might desert or free ride. Thomas Schelling, Duncan Snidal, and Howard Raiffa had assumed that organizations faced a multiparty prisoner’s dilemma--that is, after the public goods had been provided by founding members, latecomers could benefit and free ride.(FN36) In contrast, Gruber believes that late joiners will stay committed over time without free riding: "What keeps ... the losers from opting out? The answer, put simply, is their fear of being left behind. What tips the balance for these participants--what motivates them to integrate their policies with those of the winners--is not the prospect of mutual gain so much as the absence of any better alternative" (p. 47). In addition, the founding members (the winners) do not establish a fund of transferable benefits that latecomers can preempt. They keep the benefits to themselves. In a golf club, the founding members distribute their costs among new members without appreciably (or at all) reducing their benefits on the fairways or greens. Without fully realizing it, Gruber finds "club goods" to be the means of avoiding free riding. If goods are fully collective, free riding in their use is available even to nonmembers. But the goods in his cases are not collective goods; instead they are club goods. His cases are organizational clubs--which gives an entirely new meaning to the chestnut "clubs are trumps." In Gruber’s clubs, the good is available only to those who are members.(FN37).

The existence of a club and club goods, then, is the primary reason why nations cannot expect to benefit without joining. In addition, even the losers find that domestic politics makes it difficult for them to exit the organization. This is partly because of the influence that the new organization exerts on the internal politics of new members. "The prime movers ... design their regime specifically to ensure that its domestic beneficiaries included a wider-than-usual cross-section of each signatory’s population. A widely employed method for broadening a regime’s clientele involves the interweaving of numerous sector-or industry-specific agreements into a single comprehensive package. In addition to economizing on transaction costs, the linking of disparate agreements in this fashion has the effect of enlarging the potential numbers and geographic diversity of the new stakeholders" (p. 87).

Equally important, a new institution creates its own constituencies. The longer an institution is in existence, "the more likely it is to have spawned powerful new actors who either did not exist prior to the arrangement’s entry into force (e.g., new government agencies created specifically to oversee a member country’s involvement in the arrangement) or who did exist but lacked political clout (e.g., industries and investors who benefit directly from the arrangement)" (p. 86).

Thus the purpose of supranational governance is "to deter the enacting coalition’s successors from mounting a serious challenge to the new regime if they should ever get the opportunity" (p. 90). As long as the stream of benefits relative to the available alternatives is large, Gruber says, "participants will find it in their own interest to comply with institutional rules and norms" (ibid.). Elites of latecoming members do not withdraw from international organizations. But their successors may. This is ultimately what scuttled the European Monetary System. Both France and Britain were unable to sustain the exchange-rate regime when the D-mark increased in value--the British opting out and the French changing the bands of fluctuation vis-à-vis Germany. Prime Minister John Major dismantled British membership, and President Francois Mitterand broke temporarily with EMS parties. Italy also withdrew.

According to Gruber, the U.S.-Canadian Free Trade Agreement and NAFTA evinced the same tendencies, though no founding member defected later. The signing by Canada and the United States (then led by Prime Minister Brian Mulroney and President Ronald Reagan) of a Free Trade Agreement, created a problem for Mexico. Mexico City had long favored higher tariffs. Yet FTA seemed to give Canada an exclusive advantage in the U.S. market that Mexico could not tolerate.(FN38) When President George Bush and President Carlos Salinas de Gortari began negotiating a similar arrangement for Mexico, Canada also had to participate in the talks, leading ultimately to the three-party North American Free Trade Agreement.(FN39).

In both cases Gruber seeks to show that latecomers would have preferred an alternative arrangement (an informal exchange relationship or higher tariffs), but were forced to join the new organization when it created special benefits for its founders. In sum, Gruber contends that new international organizations keep their lesser member states in a condition of permanent inferiority.

GRUBER’S CONTRIBUTIONS

Gruber has offered a powerful if not fully convincing new view of international institutions and has contributed to the development of realist thought. As a generalist realist, he has seen that institutions now bestride the world, and far more than the territorial state, are expanding in both jurisdiction and power. No previous realist has understood this development, and Gruber, while attempting to accommodate its evolution to realist explanations, gives it full recognition. Even if he does not completely acknowledge that latecomers to international institutions also benefit, his own argument lends credence to that overarching point.

Another outcome of Gruber’s analysis is that the process of creating powerful regional institutions may effectively reverse the balance of power, giving greater scope to generalist realism while diverging from specific realism. As Gruber gives eloquent testimony, the basic process of founding regional institutions is one of accumulating and centralizing power. While "bandwagoning" is extensively treated in his text, the index does not even contain a reference to the "balancing of power" (see p. 307). The countries initially left out of new organizations do not balance against the more integrated centralization of economic power; rather they seek to join it.(FN40) The creation of the new institution, as Gruber correctly says, has changed the status quo ante in fundamental ways. Of course, one might claim that the formation of new institutions on different continents might be regarded as a form of regional or bloc balancing against each other. This would be true if membership were distinct and not overlapping. But unlike in the 1930s, such institutions are not exclusive or sectarian in either outlook or membership. In fact the regional clubs, as Arthur Stein and I seek to show, are connected by "overlapping clubs."(FN41) NAFTA and the EU are linked together by the Group of Eight, the Organization for Economic Cooperation and Development, NATO, and other influential organizations. The Association of Southeast Asian Nations (ASEAN) and NAFTA are linked by APEC, the WTO, and the IMF. In contrast, the clubs of the 1930s contained little or no overlap.(FN42) Gruber has given eloquent testimony to the strength and power of new international organizations as clubs.

INSTITUTIONALISM AND LLOYD GRUBER--A RECONSIDERATION

As reflected in Gruber’s work, realism is in process of evolution. The original formulation of specific realists that involved intermittent or systemic balancing by nations in response to challenge or superior force has been modified.(FN43) As the nature of the game has shifted with variable-sum outcomes replacing constant-sum results in many instances, the amount of cooperation in the system has correspondingly increased, and the amount of balancing has declined. Realists have taken account of these shifts, and a new category of generalist realism de-emphasizing balancing as the typical mode of adjustment has emerged. These changes were in process prior to Lloyd Gruber’s study of institutions. Gruber has added to their force and content by stressing that balancing does not take place in regard to organizations that huge numbers of outsiders still want to join. The peripheral and excluded powers do not seek to form opposing coalitions; they flock to get into the central coalition. In this way, at least in the United States and Europe--Gruber’s twin arenas--centripetal or centralizing forces outweigh centrifugal or balance-of-power influences.

But this leaves us with a quandary if Gruber is right: The latecomers do not benefit from joining the organizations. They would rather have stayed out, but the old status quo is no longer available. They are forced to join--not by bullying but by the changed nature of the incentives they face. This presents a dilemma: If the situation is really so bad for latecomers, why do they not seek to prevent the formation of the organization in the first place? Or, if collective goods are involved in the creation of the organization, why can they not stay out and still benefit from their provision? As I seek to show below, the second choice is removed because there are no collective goods in Gruber’s formulation. What is involved are club goods. Second, nations join organizations and stay in not because they have no alternatives to a decline in their enjoyment of benefits, but because they do expect to benefit over time. Gruber’s argument--based as it is on counterfactuals--cannot exclude this possibility, and in important historical cases the benefits are in fact realized. As Albert Hirschman shows, members can profit from achieving "voice" in organizations even when they do not earn immediate benefits.(FN44) Member countries may express their views and vote in organizations they do not fully control.

There are a number of deficiencies that limit the significance of Gruber’s findings. First, a key point is that Gruber’s analysis cannot apply generally to international politics and to the problem of collective goods. The huge Olsonian debate over free riding is not resolved by Gruber.(FN45) Keohane’s and Robert North’s means of preventing free riding are still relevant to most international organizations.(FN46) Where public goods exist, there will be cheating and free riding unless such behavior can be sanctioned or curbed through attitudinal change. Gruber’s argument does not solve nor and does it try to solve the public goods problem in organizations because according to Gruber, there are no collective goods. In the organizations that Gruber treats, a state gets the benefits only if it is a member. If this were not true, there would be no overarching pressure to join, and Gruber’s argument would fall of its own weight. Equally the latecoming members would benefit from previously created public goods. Once again, his analysis would not hold.

If truly public goods were involved, an outsider could wait, expecting to benefit and free ride from the public goods created by the organization. As I have shown above, Gruber is dealing with club goods. His analysis would not generally apply to nongeographic organizations such as the IMF and WTO because to greater or lesser degrees even nonmembers can benefit from their operation through most-favored-nation clauses and international funding. A country does not have to be a member of the IMF to benefit (or suffer) from IMF rescue operations; the IMF merely needs to have an important international stake in the state’s financial sector.(FN47) This limits the applicability of Gruber’s findings to exclusive, largely regional clubs.

The debate over collective goods in international institutions will continue. But this debate cannot be resolved by focusing solely on regional institutions and club goods. General international institutions can be indefinitely expanded. Regional organizations cannot. Although the EU and NAFTA--regional providers of club goods--are capable of considerable growth in membership, they cannot admit every state. As James Buchanan, the leading analyst of club goods, argues, there are limits to membership in clubs that do not exist in general purpose organizations In clubs, there is a cost curve that must decline toward zero as more members join and contribute. But there is also a benefits curve that rises initially with new members, but ultimately declines as they overwhelm the organization’s facilities. Golf clubs do better with 300 members than with 10, but they do worse with 3,000 members crowding their courses. So clubs cannot indefinitely be expanded. The club good cost curve may be declining, but (at some point) the benefit curve begins to decline as well, as additional members join.(FN48).

Second, members of organizational clubs probably do considerably better economically and politically than Gruber suggests. Founding members do not fare as well, as he believes, and latecomers do not suffer as much. Gruber can reach his conclusions because he does not fully take account of how cost and benefit emerge in the uninterrupted flow of day-to-day interactions among states. Although it is possible to cut into this interaction sequence at a particular point in time, what Gruber regards as cause may simply be the effect of prior interactions or causes. His argument depends on starting and concluding at particular points along the continuum of action. In the formation of organizations, he begins with the action of founders (or "winners") and carries the analysis only to the immediate suffering of the "losers" who join later. To have a more complete theoretical view--one that changes the respective benefits of founder and latecomer--both ends of the continuum need to be extended. This reviewer takes exception to Gruber’s conclusion that founders benefit and latecomers suffer.

Why did the winners form the organization? Gruber telescopes the analytic process by not looking at the causes that led (or even coerced) his founders to propose the new international organizations in the first place. In the case of the European Monetary System, U.S. currency problems were dragging Europe down, and key European states wanted a stronger and separate currency base. Even so-called winners might have preferred a status quo ante that was not available to them--a stable dollar-based system. In other words, the founders suffered and then formed the organization to improve the situation. Equally in regard to NAFTA, Canada and the United States might have preferred a Mexico that was stable economically and politically secure. Then new foreign trade initiatives might not have been necessary. But that was not an available alternative in the late 1980s and the early 1990s. Bush administration officials were worried about Mexico’s long-term future and were committed to taking action. Mexico had an undemocratic government, an inefficient rural sector, and a protected industry. It was a country likely once again to be beset by problems similar to the Mexican debt crisis of 1982. Then Mexico borrowed too much and did not have the capacity to repay at the higher interest rates of the time. If the U.S. market were open, on the other hand, Mexico could repay with exports. If government subsidies and tariff inefficiencies were swept away, foreign direct investment coming into Mexico would compensate for Mexican indebtedness. Thus the potential for Mexican economic and political collapse was an important reason for President Bush’s sudden willingness to construct NAFTA. Even American so-called winners in this process might have preferred an outcome that provided for Mexican stability without giving concessions in the U.S. market. American officials knew that low-priced Mexican imports would force some U.S. factories to close. But this alternative was not available. Gruber overlooks the causes that led his "winners" to propose the new international organizations.

This point is redoubled in force when one looks at the longer-term outcomes for so-called losers or latecomers that are supposed to suffer over time. Gruber takes the case of New York and the ratification of the U.S. Constitution. The enacting "winners" were Massachusetts, Pennsylvania, New Jersey, Connecticut, and Georgia. New York was the coerced "loser." As Gruber writes: "New Yorkers did not like the new constitution to which they agreed to adhere. Quite the contrary, the state’s government would have been perfectly happy to continue charging other states a fee for the ’privilege of using New York ports. Once it was clear that New York was in the position of our actor D, however, and that Massachusetts (actor C) was fully prepared to join Pennsylvania, New Jersey, Connecticut, and Georgia (the A’s and B’s) in forging a new nation, it would have been foolish for New Yorkers not to jump on board the Federalist bandwagon" (p. 269, emphasis in the original). Indeed it would have been. Whatever the short-term advantage to high-tariff Massachusetts, the long-term trading and financial advantages deriving from New York’s position under the U.S. Constitution and the federal system can scarcely be denied.

Similarly, Mexico might have been worried about the consequences of joining NAFTA, but both Mexico and Canada have benefited at least as much as the United States and perhaps more. It is perhaps too early to assess the record in regard to the European Monetary System and the European Monetary Union, but it does not appear that latecomers suffered egregiously at the hands of the founders. France and Germany were Gruber’s A and B countries, yet their growth has not outstripped that of the United Kingdom, and Italy’s prospects are scarcely grim. By foreshortening his vision, Gruber sees only short-term outcomes and takes no account of longer-term evolutions. In fact, using his terms, a strong case can be made that A’s and B’s initially benefit, and then C’s and D’s gain at the expense of later E’s and F’s. What is clearly not true is that "losers would still be happier if the original, noncooperative status quo had never been disrupted" (p. 8). New York does not today seek independence from the federal union, and even the American South (another partial latecomer) is reconciled to it. Here Gruber might usefully have taken into account Mancur Olson’s argument that economic growth may depend on the stimulus resulting from sudden immersion into a free-trade environment. Because they are likely to have higher tariff walls to begin with, latecomers will find the greater stimulus and potentially the greater growth from joining.(FN49) Alexander Gerschenkron’s "advantages of backwardness" thesis would be a useful corrective to Gruber’s pessimism.(FN50).

Third, Gruber acknowledges but does not fully take account of another powerful reason for joining organizations: Latecomers may benefit because constitutional arrangements frequently leave important matters to be decided by the membership as the institution evolves. "Voice" as Albert Hirschman contended, is a primary means of obtaining "loyalty."(FN51) Organizations may achieve a higher degree of integration benefiting newcomers. As Gruber himself understands, founding members typically formulate institutional designs that embody "’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" (p. 9). This process has already taken place in the EU, NAFTA, and ASEAN. Latecomers have sometimes been able to translate voice into greater influence and higher benefits. Consider the role of Spain in the European Union as an example.

Finally, Gruber does not fully recognize that in the never-ending chain of stimulus and response (leading to another stimulus), particular actors do not determine the long-term result. Their actions are themselves the result of prior causes. Reading Gruber’s book, the A and B so-called founders of international organizations would rush to explain how they were forced to act, perhaps against their will, to initiate the new institution. Yet to give a comprehensive account, the chain of investigation must begin at a prior stage. It must also extend to a later one. In addition, as Gruber acknowledges, the whole force of his analysis depends on the certainty of the counterfactual--that is, the coerced member would have been better off staying out of the new institution if it could have done so (p. 10). New York would offer a powerful counterexample here, to say nothing of latecomers to the EU.

CONCLUSION

Gruber’s conclusions in Ruling the World suggest that realism is increasingly becoming a form of cost-benefit analysis. Countries act not because they must balance against aggression to survive but because they receive concrete incentives. They do what seems to be in their best interests. Some countries fought in World War II. Some stayed neutral or declined to fight. It is not transparent that the latter fared worse than the former.(FN52) As Robert Powell points out, nations have strategies of waiting, balancing, or bandwagoning in cases of potential danger.(FN53) Depending on the nature of the game, the first and third strategies may be as appropriate as the second. Variable- or increasing-sum games induce the more cooperative responses. Among major powers at least, these are more generally characteristic than they were during the Cold War. It is therefore not surprising that the tenets of specific realism--with balancing as its central preoccupation--have increasingly yielded to a more generalist form of realism.

Institutional research follows a similar pattern. There cannot be any doubt that in contrast to the 1930s, institutions are growing in both number and power. But it is still uncertain who benefits and who loses from institutional development. Collective goods approaches might suggest that the nonmember (and noncontributor) benefits most. Gruber sees latecomers improving their position, though not as compared to the pre-organizational status quo. A new intellectual industry is needed that traces the incentives of institutional precursors and then examines institutional results down the line. Have Gruber’s winners actually lost, and do his losers actually win? Gruber has usefully identified a form of institutional statics. But to show whether he is right or wrong and to point the direction that institutions are actually taking, we need a new institutional dynamics.

Added material.

Richard Rosecrance is Professor of Political Science at the University of California, Los Angeles, and Project Director of UCLA’s Carnegie Study of Globalization and National Self-Determination.

The author would like to thank the members of the International Colloquium at the University of California, Los Angeles, for helpful comments.

Figure 1. Pareto-Optimal Outcomes.

Figure 2. Non-Pareto-Optimality Outcomes.

FOOTNOTES

1. Lloyd Gruber, Ruling the World: Power Politics and the Rise of Supranational Institutions (Princeton, N.J.: Princeton University Press, 2000), chap. 1. Further references to this work appear parenthetically in the text.

2. See Michael W. Doyle, Ways of War and Peace: Realism, Liberalism, and Socialism (New York: W.W. Norton, 1997), pp. 116-117.

3. Bertrand Russell, Power: A New Social Analysis (New York: W.W. Norton, 1938); E.H. Carr, The Twenty Years’ Crisis, 1919-1939 (London: Macmillan, 1939); and Hans. J. Morgenthau, Politics among Nations: The Struggle for Power and Peace (New York: Alfred Knopf, 1948). In addition, Morgenthau believed that the struggle to balance power would ultimately lead to an attempt to obtain the maximum available under the circumstances.

4. See Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979), pp. 111-118.

5. Ibid., p. 119.

6. Waltz went so far as to claim that "if there is any distinctively political theory of international politics, balance-of-power theory is it." Ibid., p. 117.

7. Glenn H. Snyder, Alliance Politics (Ithaca, N.Y.: Cornell University Press, 1997), p. 51.

8. See Doyle, Ways of War and Peace, pp. 170-172.

9. Waltz, Theory of International Politics, p. 126, writes: "Because power is a means and not an end, states prefer to join the weaker of two coalitions.".

10. Morton A. Kaplan, System and Process in International Politics (New York: John Wiley, 1957), p. 23. The other rules appear on p. 23 as well.

11. Michael Nicholson, Formal Theories of International Relations (Cambridge: Cambridge University Press, 1989), p. 26. Waltz and some others would dissent from this formulation, arguing only that failure to balance would render the system unstable and jeopardize the security of individual states.

12. See Waltz, Theory of International Politics, pp. 126-127.

13. With two main participants, even zero-sum games foster balancing (as shown below). If A’s and B’s positions sum to 0 or a constant, each must prevent any increase in the other’s position (otherwise its own will be diminished).

14. There are a series of oppositions in realist thought. Defensive realism versus offensive realism and neorealism versus postclassical realism are two among many. The key distinction here between specific realism and generalist realism, however, is between a concept of realism based on conflict and material power capabilities conjoined with balancing on the one hand, and a much broader and inclusive notion of realism that involves no necessary balance of power on the other.

15. See Robert Powell, In the Shadow of Power: States and Strategies in International Politics (Princeton, N.J.: Princeton University Press, 1999); and Randall L. Schweller, Deadly Imbalances: Tripolarity and Hitler’s Strategy of World Conquest (Cambridge: Cambridge University Press, 1998).

16. See Emerson M.S. Niou, Peter C. Ordeshook, and Gregory F. Rose, The Balance of Power: Stability in International Systems (Cambridge: Cambridge University Press, 1989).

17. Snyder, Alliance Politics, p. 51, argues that "cooperative resistance is likely, although not certain, since some states might still be tempted to free ride if they thought their own capabilities were not essential to defeating the aggressor. Fully collective resistance becomes certain only when the attacker would gain control of more than half the power resources in the system with one more successful attack." In practice and historical experience, however, this point is seldom if ever reached. Individual aggressors have always been far short of attaining a majority of power in the system as a whole.

18. Paul Kecskemeti, Strategic Surrender: The Politics of Victory and Defeat (Stanford, Calif.: Stanford University Press, 1958).

19. In the prisoner’s dilemma supergame, cooperation remains a possibility, depending on the amount of iteration and the discount parameter attached to future payoffs. See Robert M. Axelrod, The Evolution of Cooperation (New York: Basic Books, 1984); and Robert M. Axelrod, The Complexity of Cooperation: Agent-Based Models of Competition and Collaboration (Princeton, N.J.: Princeton University Press, 1997).

20. See Richard Rosecrance and Chih-cheng Lo, "Balancing, Stability, and War: The Mysterious Case of the Napoleonic International System," International Studies Quarterly, Vol. 40, No. 4 (December 1996), pp. 479-500.

21. See Niou, Ordeshook, and Rose, The Balance of Power; and Powell, In the Shadow of Power.

22. He concludes that states generally bandwagon unless there are large returns to scale. See Powell, In the Shadow of Power, p. 178.

23. John J. Mearsheimer, "The False Promise of International Institutions," International Security, Vol. 19, No. 3 (Winter 1994/95), pp. 5-49.

24. Kenneth N. Waltz, "Structural Realism after the Cold War," International Security, Vol. 25, No. 1 (Summer 2000), pp. 5-41.

25. Joseph M. Grieco, Cooperation among Nations: Europe, America, and Non-Tariff Barriers to Trade (Ithaca, N.Y.: Cornell University Press, 1990).

26. Initially rejected in the Irish referendum in spring 2001, the treaty will be resubmitted for later approval.

27. See Richard Rosecrance, "The European Union: A New Type of International Actor," in Jan Zielonka, ed., Paradoxes of European Foreign Policy (The Hague: Kluwer, 1998), chap. 1.

28. Joseph M. Grieco, "States Interests and Institutional Rule Trajectories: A Neorealist Interpretation of the Maastricht Treaty and European Economic and Monetary Union," in Benjamin Frankel, ed., Realism: Restatements and Renewal (London: Frank Cass, 1996); Charles L. Glaser, "The Security Dilemma Revisited," World Politics, Vol. 50, No. 1 (October 1997), pp. 171-201; and Randall L. Schweller and David Priess, "A Tale of Two Realisms: Expanding the Institutions Debate," International Studies Quarterly, Vol. 41, No. 1 (May 1997), pp. 1-32.

29. See Robert O. Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N.J.: Princeton University Press, 1984); Lisa L. Martin, Coercive Cooperation: Explaining Multilateral Economic Sanctions (Princeton, N.J.: Princeton University Press, 1992); and Helen V. Milner, Interests, Institutions, and Information: Domestic Politics and International Relations (Princeton, N.J.: Princeton University Press, 1997).

30. Milner, Interests, Institutions, and Information, pp. 7-8; and Keohane, After Hegemony, p. 12.

31. See Grieco, Cooperation among Nations, pp. 233-234.

32. See Keohane, After Hegemony, chap. 7.

33. See ibid., pp. 45-47.

34. Alternatively, one can think of a "club good" in which the founding members would seek to prevent a certain number of outsiders from enjoying the fruits of membership.

35. Guido Carli commented, "If Italy is too weak to participate in the EMS, it is also true that she is too weak not to participate," Quoted in Gruber, Ruling the World, p. 184.

36. The benefits resulted from the initial formation of a small k-group. See Thomas C. Schelling, Micromotives and Macrobehavior (New York: W.W. Norton, 1978); Duncan Snidal, "The Limits of Hegemonic Stability Theory," International Organization, Vol. 39, No. 5 (Autumn 1985), pp. 579-614; and Howard Raiffa, The Art and Science of Negotiation: How to Resolve Conflicts and Get the Best Out of Bargaining (Cambridge, Mass.: Harvard University Press, 1982). See also Richard Rosecrance and Jennifer Taw, "Japan and the Theory of International Leadership," World Politics, Vol. 42, No. 2 (January 1990), pp. 184-209.

37. This of course contradicts his assumption that there are no barriers to membership and that founding members always benefit from the addition of new members. In a golf or tennis club, for example, this is true only up to a point.

38. Jaime Serra Puche, Mexico’s NAFTA negotiator, noted, "Our exports to Canada and the United States will be shut out as a result of the preferential treatment they give each other." Quoted in Gruber, Ruling the World, pp. 131-132.

39. It is intriguing that the Liberals in Canada had strongly opposed FTA, yet ratified and supported NAFTA. Ibid., p. 158, writes: "In the end, however, the Liberals concluded that they were better off ratifying NAFTA anyway. And, indeed, they were not wrong about this: if Prime Minister Chrétien had gone ahead and rejected the accord, the United States would have been only too happy to negotiate a separate Mexico-U.S. agreement. That, after all, had been Salinas’s initial request.".

40. See Rosecrance, "The European Union.".

41. See Richard Rosecrance and Arthur A. Stein, "The Theory of Overlapping Clubs," in Rosecrance, ed., The New Great Power Coalition: Toward a World Concert of Nations (Boulder, Colo.: Rowman and Littlefield, 2001).

42. See Kerry Chase, "Sectors, Firms, and Regional Trade Blocs," Ph.D. dissertation, University of California, Los Angeles, 1998.

43. This formulation includes balancing in response to power (à la Waltz) and balancing in response to threat (à la Walt). See Stephen M. Walt, The Origin of Alliances (Ithaca, N.Y.: Cornell University Press, 1987).

44. See Albert O. Hirschman, Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States (Berkeley: University of California Press, 1970).

45. See particularly Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, Mass.: Harvard University Press, 1965). Olson contended that, in large organizations, the problem of free riding would be difficult if not impossible to solve.

46. Keohane, After Hegemony, chap. 7, claimed that nations accept "bounded rationality" and cooperate even when fully rational decisions might involve conflict with one another. Douglass C. North, Structure and Change in Economic History (New York: W.W. Norton, 1981), contended that property rights in Western economic history have depended on ideological consensus.

47. One would not want to claim, however, that membership in such functional clubs conveys no benefits. In the IMF case, outsiders probably do not have the same access to capital. In the WTO case, tariffs may well be lower for members. But because these organizations also create public goods, many excluded nations benefit without joining them.

48. See the diagrams in James Buchanan, "An Economic Theory of Clubs," Economica, Vol. 32 (February 1965), pp. 1-14.

49. See Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities (New Haven, Conn.: Yale University Press, 1984).

50. See Alexander Gerschenkron, Economic Backwardness in Historical Perspective: A Book of Essays (Cambridge, Mass.: Harvard University Press, 1962).

51. Hirschman, Exit, Voice, and Loyalty.

52. Dan Reiter, Crucible of Beliefs: Learning, Alliances, and World Wars (Ithaca, N.Y.: Cornell University Press, 1996).

53. Powell, In the Shadow of Power.