8/6/2019 Kohaneinterdependence

1/8

8/6/2019 Kohaneinterdependence

2/8

2 International institutions: caninterdependence work?Robert 0. Keohane(1998)

To analyze world politics in the I990s is to discuss international institutions:the rules that govern elements of world politics and the organizations thathelp implement those rules. Should NATO expand? How can the Un it edNations Security Council assure UN inspectors access to sites where Iraqmight be conduct ing banned weapons activity? Under what conditionsshould China be admit ted to the World Trade Organ izat ion (WTO)? Howmany billions of dol la rs does the Interna tional Monetary Fund (IMF)need at its disposal to remain an effective "lender of last resort" for countriessuch as Indonesia, Korea, and Thailand that were threatened in 1997 withfinancial collapse? Will the tentative Kyoto Protocol on Climate Change berenegotiated, ratified, and implemented effectively? Can future Uni tedNations peacekeeping practices -- in contrast t o the UN fiascoes in Bosniaand Somalia be made more effective?These questions help illustrate the growing importance of internationalinstitutions for maintaining world order. In 1985 (Foreign Policy 60: 148-(7)

Joseph Nye and I gave "two cheers for multilateralism," pointing Ollt thateven the administration of President Ronald Reagan, which took otflce illdisposed toward international institutions, had grudgingly come to acceptthe ir va lue in achiev ing American purposes. Superpowers need generalru les because they seek to inf luence events around t he world. Even anuncha llenged superpower such as the United Sta tes would be unab le toachieve its goals through the bilateral exercise of influence: the costs of suchmassive "arm-twisting" would be too great.International institutions are increasingly important, but they are not

always successful. Ineffective ins ti tu tion s such as the Uni te d Nat ionsIndustrial Development Organization or the Organization of African Unityexist alongside effectual ones such as the Montreal Protocol on Substancestha t Dep le te the Ozone Layer and the European Union . In recent years,we have gained insight into wha t makes some insti tu tions more capab lethan ot her s - how such i ns ti tuti ons best promote cooperation amongstates and what mechanics of bargaining they use. But our knowledge isincomplete, and as the world moves toward new forms of global regulation and governance, the increasing impact of international institutions

8/6/2019 Kohaneinterdependence

3/8

28 Interdependence and institutionshas raised new que st ion s about how these insti tu tions themselves a regoverned.

Theory and reality, 1919-89Academic "scribblers" did not always have to pay much attention to international institutions. The 1919 Versailles Treaty constituted an attempt toconstruct an institution for multila teral diplomacy - the League of Nations.But the rejection of the League Covenant by the US Senate ensured that untilWorld War II the most important negotiations in world po li tics - f rom thesecret German-Russian deals of the 1920s to the 1938 Munich conference took place on an ad hoc basis. Only after the United Nations was foundedin 1945, with strong support f rom the United Sta tes and a multiplicity ofspecialized agencies performing different tasks, did international institutionsbegin to command substantial international attention.Until the late I960s, American students of international relations equated

international institutions with formal international organizations, especiallythe United Nations. Internatiunal Organizatiun, the leading academic journalon the subject, carried long summaries of UN meetings until 1971. However,most observers recognized long before 1972 that the United Nations did notplaya central role in world politics. Except for occasional peacekeeping missions .- of which the First UN Emergency Force in the Middle East between1956 and 1967 was the most successful _. its ability to resolve hostilities wasparalyzed by conflicts of interest that resulted in frequent superpower vetoesin the Security Counc il . Moreover , the influx of new' postcolonial stateshelped turn the General Assembly into an arena for North-South conflictafter 1960 and ensured that the major Western powers, especially the UnitedStates, would view many General Assembly resolutions as hostile to theirinterests and values - for example, the New International Economic Orderand the Zionism is Racism resolutions of the 1970s. Analysts and policymakers in Europe, North America, and much of Asia concluded that international institutions were marginal to a game of world politics still driven bythe traditional exercise of state power. The UN - called "a dangerous place"by former US representative to the UN Daniel Patrick Moynihan - seemedmore a forum for scoring points in the Cold War or North-South conflictsthan an instrument for problem-solving cooperation.In reality, however, even the most powerful states were relying increasingly

on international institutions- not so much on the UN as other organizationsand regimes that set rules and standards to govern specific sets of activities.From the late 1960s onward, the Treaty on the Non-Proliferation of NuclearWeapons was the chief vehicle for efforts to prevent the dangerous spreadof nuclear weapons. NATO was not only the mos t successful multilateralalliance in history but also themost highly institutionalized, with a secretarygeneral, a permanent staff, and elaborate rules governing relations amongmembers. From i ts founding in 1947 through the Uruguay Round that

Intenwtio/w! institutions 29concluded in 1993, the General Agreement on Taritl's and Trade (GATT)presided over a ser ies of trade rounds that have reduced import tariffsamong industrialized countries by up to 90 percent, boosting internationalt rade. A ft er a shaky start in the I940s, the IMF had _. by the 1960s become the centerpiece of efforts by the major capitalist democracies toregulate their monetary affairs. When that function atrophied with the onsetof flexible exchange rates in the 1970s, it became the ir leading agent forfinancing and promoting economic development in Africa, Asia, and LatinAmerica. The sheer number of intergovernmental organizations also rosedramatica lly - from about 30 in 1910 to 70 in 1940 to more than 1,000by 1981.The exchange rate and oil crises of the early 1970s helped bring perceptions

in line with reality. Suddenly, both top policymakers and academic observersin the United States realized that global issues required systematic policycoordination and that such coordination required institutions. In 1974, thensecretary of s ta te Henry Kissinger , who had paid li tt le a tten tion to international institutions, helped establish the International Energy Agency toenable Western countries to deal cooperatively with the threat of future oilembargoes like the 1973 OPEC embargo of the Netherlands and UnitedStates. And the Ford administration sought to construct a new internationalmonetary regime based on flexible rather than pegged exchange rates. Confronted with complex interdependence and the efforts of slates to manageit, political scientists began to redefine the study of international institutions,broadening it to encompass wha t they cal led "international regimes" structures of rules and norms that could be more or less informal. Theinternational trade regime, for example, did not have strong formal rules orintegrated, centralized management; rather, it provided a set of interlockinginstitutions, including regular meetings of the GATT contracting parties,formal dispute settlement arrangements, and delegation or technical tasks toa secretariat, which gradually developed a body of case law and practice.Some interna tional lawyers grumbled tha t the pol it ical scient is ts weremerely using other terms to discuss international law. Nevertheless, politicalscientists were once again discussing how international rules and norms affectstate behavior, even if they avoided the" L-word."In the I980s, r esearch on int ernat iona l regimes moved from at tempt s

to describe the phenomena of interdependence and international regimes tocloser analysis of the conditions under which countries cooperate. How doescooperation occur among sovereign states and how do international institut ions affect it? From the standpoint of political realism, both the relianceplaced by states on certain international institutions and the explosion intheir numbers were puzzling. Why should international institutions exist atall in a world dominated by sovereign s tates? Thi s ques tion seemedunanswerable if institutions were seen as opposed to, or above, the state butnot if they were viewed as devices to help states accomplish their objectives.The new research on interna tional insti tu tions broke decisively with

8/6/2019 Kohaneinterdependence

4/8

30 Inll'rdependence and institutionslegalism -- the view that law can be effective regardless of political conditions-as well as with the idea lism assoc ia ted with the f ield's or igins. Ins tead ,scholars adopted the assumptions of realism, accepting that relative statepower and competing interests were key factors in world politics, but at thesame time drawing new conclusions about the influence of institutions on theprocess. Institutions create the capability for states to cooperate in mutuallybeneficial ways by reducing the costs of making and enforcing agreements what e conomist s refer to as "transaction costs. " They rarely engage incentralized enforcement of agreements, but they do reinforce practices ofreciprocity, which provide incentives for governments to keep the ir owncommitments to ensure that others do so as well. Even powerful s tates havean interest, most of the time, in following the rules of well-established international institutions, since general conformity to rules makes the behavior ofother states more predictable.

This scholarship drew heavily on the twin concepts of uncertainty andcredibility. Theorists increasingly recognized that the preferences of statesamount to "private information" ... that absent full transparency, s tates areuncertain about what their part ners and rivals value at any given time. Theynaturally respond to uncertainty by being less willing to enter into agreements , s ince they are unsure how their partners will later interpret the termsof such agreements. International institutions can reduce this uncertainty bypromoting negotiations in which transparency is encouraged; by dealing witha series of issues over many years and under similar rules, thus encouraginghonesty in order to preserve future reputation; an d by systematically monitoring the compliance of governments with their commitments.

Even if a government genuinely desires an international agreement, it maybe unable to persuade its partners that it will, in the future, be willing and ableto implement it. Successful international negotiations may therefore requirechanges in domestic institutions. For instance, without "fast-track" authorityon trade, the United States' negotiating partners have no assurance t hat Congress will refrain from adding new provisions to trade agreements as a condition for their ratification. Hence, other states are reluctant to enter into tradenegotiations with the United States since they may be confronted, at the endof tortuous negotiations, with a redesigned agreement less favorable to themthan the draft they initialed. By the same token, without fast-track authority,no promise by the US government to ab ide by negot ia ted terms has muchcredibility, due to the president's lack of control over Congress.

In short , this new school of thought argued that, rather than imposingthemselves on states, international institutions should respond to the demandby s ta tes for coopera tive ways to fulfill the ir own purposes. By reducinguncertainty and the costs of making and enforcing agreements, internationalinstitutions help states achieve collective gains.

International institutions 31Yesterday's controversies: 1989-95This new ins ti tu tional ism was not without its cri tics, who focused theirattacks on three perceived shortcomings: first, they c la imed tha t international institutions are fundamentally insignificant since states wield theonly real power in world politics. They emphasized the weakness of efforts bythe UN or League of Nations to achieve collective security against aggressionby great powers, and they pointed to the dominant role of major contributorsin international economic organizations. Hence, any effects of these international institutions were attributed more to the efforts of their great powerbackers than to the institutions themselves.

This argument was overstated. Of course, great powers such as the UnitedStates exercise enormous influence within international institutions. But thepolicies that emerge from these ins ti tu tions are different from those tha tthe United States would have adopted unilaterally. Whether toward Iraq orrecipients of IM F loans, policies for specific situations cannot be entirely adhoc but mus t conform to general ly appl icab le rules and principles to beendorsed by multilateral institutions. Where agreement by many states isnecessary for policy to be effective, even the United States finds it usefulto compromise on substance to obtain the ins ti tu tional seal of approval.Therefore, the decision-making procedures and general rules of internationalinstitutions matter. They affect both thesubstance of policy and the degree towhich other states accept it.

The second counterargument focused on "anarchy": the absence of aworld government or effective international legal system to which victimsof injustice can appeal. As a result of anarchy, critics argued, states preferrelative gains (i.e., doing better than other states) to absolute gains. They seekto pro tect the ir power and status and will resist even mutually beneficialcoopera tion if the ir par tners a re likely to benefit more than they are. Forinstance, throughout the American-Soviet arms race, both sides focused onthe ir relat ive pos it ions - who was ahead or threatening to gain a decisiveadvantage - rather than on their own levels of armaments. Similar dynamicsa pp ea r o n certa in economic issues, such as the fierce Euro-Americancompetition (i.e., Airbus Industrie versus Boeing) in the production of largepassenger jets.

Scholarly disputes about the "relative gains question" were intense butshort-lived. It turned o ut t ha t the question needed to be reframed: not, "d ostates seek relative or absolute gains?" but "under what conditions do theyforego even mutually beneficial cooperation to preserve their relative powerand status?" When there are only two major players, a n d o n e side's gains maydecisively change power relationships, relative gains loom large: in armsraces, for example, or monopolistic competition (as between Airbus andBoeing). Most issues of potential cooperation, however, from trade liberalization to climate change, involve multilateral negotiations that make relativegains hard to calculate and entail little risk of decisive power shifts for one

8/6/2019 Kohaneinterdependence

5/8

8/6/2019 Kohaneinterdependence

6/8

34 Interdependence and institutionsThe procedures and rules of international insti tutions create infor

mational structures. They determine what principles are acceptable as thebasis for reducing conflicts and whether governmental actions are legitimateor i llegitimate. Consequently, they help shape actors' expectations. Forinstance, trade conflicts are increasingly ritualized in a process of protestingin the WTO - p romis ing tough ac tion on behal f of one's own industries,engaging in quasi-judicial dispute resolution procedures, claiming victory ifpossible, or complaining about defeat when necessary. There is much soundand fury, but regularly institutionalized processes usually relegate conflict tothe realm of dramatic expression. Institutions thereby create differentiatedinformation. "lnsiders" can interpret the language directed toward "outsiders" and use their own understandings to interpret, or manipulate, others'beliefs.Finally, students of international insti tutions conlinue to try to under

stand why some insti tutions are so much more effective than others. Variation in the coherence of institutional policy or members' conformity withinstitutional rules is partial ly accounted for by the degree of commoninterests and the distribution of power among members. Institutions whosemembers share social values and have s imilar pol it ica l systems - such asNATO or the European Union - are likely to be s tronger than those suchas the Organization for Security an d Cooperation in Europe or the Association of South East Asian Nations, whose more diverse membership doesnot necessari ly bave the same kind of deep common interests. Additionally,the character of domestic politics has a substantial impact on internationalinstitutions. The distribution of power is also important. Insti tutionsdominated by a small number of members _. for example, the IMF, with itsweighted vo ting system - can typically take more decisive acti on thanthose where influence is more widely diffused, such as the UN GeneralAssembly.

Overcoming the democratic deficitEven as scholars pursue these areas of inquiry, they are in danger of overlooking a major normative issue: the "democrat ic def ic it" that exists inmany of the world's most important international institutions. As illustratedmost recently by the far-reaching interventions of the IM F in East Asia, theglobalization of the world economy and the expanding role of internationalinstitutions are creating a powerful form of global regulation. Major international institutions are increasingly laying down rules and guidelines thatgovernments, if they wish to attract foreign investment and generate growth,must follow. But these international institutions are managed by technocratsand supervised by high governmental officials. That is, they are run by elites.Only in the most attenuated sense is democratic control exercised over majorinternational organizations. Key negotiations in the WTO are made in closedsessions. The 1MF negotiates in secret with potential borrowers, and it only

Internlltional institutions 35began to become more t ranspa rent about the condit ions it imposes onrecipients after the world financial crisis of 1997-98.The EU providesanothercase in point. Its most important decision-makingbody is its Council of Ministers, which is composed of government representatives who perform more important legislative functions than the membersof the European Parliament. The council meets behind closed doors and untilrecently did not publish its votes. It also appoints members to the EuropeanCommission, which acts as the EU executive, whose ties to the public are thusvery indirect indeed. The European Parliament has narrowly defined powersand little status; most nat iona l par liaments do not closely scrutinizeEuropean-level actions. How much genuine influence do German or Italianvoters therefore have over the council's decisions? Very little.

The issue here is not one of state sovereignty. Economic interdependenceand its regulation have altered notions of sovereignty: few states can stilldemand to be complete ly independent of external authori ty over legalpractices within their terri tories. The best most states can hope for is to beable to use their sovereign authority as a bargaining tool to assure that othersalso have to abide by common rules and practices. Given these changes, theissue here is who has influence over the sorts of bargains tnat are struck?Democratic theory gives pride of place to the public role in deciding on thedistributional and value tradeoffs inherent in legislation and regulation. Butthe practices of international institutions place that privilege in the hands ofthe elites of national governments and of international organizations.Admittedly, democracy does not always work well. American politicians

regularly engage in diatribes against international institutions, playing on thedismay of a vocal segment of their electorates at the excessive number offoreigners in the United Nations. More seriously, an argument can be madetha t the IMF, like central banks, can only be eRective if it is insulatedfrom direct democratic control. Ever since 1787, however, practitioners andtheorists have explored how authoritative decision making can be combinedwith accountabil ity to publics and indirect democratic control . The USConstitution is based on such a theory- the idea that popular sovereignty,though essential, is best exercised indirectly, through rather elaborateinstitutions. An issue that scholars should now explore is how to devise international insti tutions that are not only competent and effective but alsoaccountable, at least ultimately, to democratic publics.One possible response is to say that all is well, since international institut ions are responsible to governments - which, in turn, are accountablein democracies to their own people. International regulation simply addsanother link to the chain of delegation. But long chains of delegation, inwhich the public affects act ion only at several removes, reduce actualpublic authority. If the terms of multilateral cooperation are to reflect theinterests of broader democratic publics rat l)er than just those of narrowelites, traditional patterns of delegation will have to be supplemented byother means of ensuring greater accountability to public opinion.

8/6/2019 Kohaneinterdependence

7/8

36 Interdependence and inSlillilionsOne promising approach would be to seek to inv igorate t ransna tional

society in the form of networks among individuals and nongovernmentalorganizations. The growth of such networks - of scientists, professionals invarious fields, and human rights and environmental activists - has been aidedgreatly by the fax machine and the Internet and by institutional arrangementsthat incorporate these networks into decision making. For example, naturaland social scientists developed the scientific consensus underlying the KyotoProtocol through the Intergovernmental Panel on Climate Change (lPCC)whose scientific work was organized by scientists who did not have to answerto any governments. The Kyoto Protocol was negotiated, but governmentsopposed to effective action on climate change could not hope to renegotiatethe scientific guidelines set by the IPCC.

The dramatic fall in the cost of long-distance communication willfacilitate the development of many more such transnational networks. As aresult, wealthy hierarchical organizations - multinational corporations aswell as s tat es - are likely to have more difficulty dominating transnationalcommunications. Thirty years ago, engaging in prolonged intercontinentalcommunication required considerable resources. Now individuals do so onthe Internet, virtually free.Therefore, the future accountability of international institutions to theirpubl ics may rest only part ly on delegation through formal democraticinstitutions. Its other pillar may be voluntary pluralism under conditions ofmaximum transparency. International policies may increasingly be monitored by loose group ings of scientists or other professionals, or by issueadvocacy networks such as Amnesty lnternational and Greenpeace, whosemembers, scattered around the world, will be l inked even more closely bymodern information technology. Accountability will be enhanced not onlyby chains of official responsibility, but by the requirement of transparency.Official actions, negotiated among state representatives in internationalorganizations, will be subjected to scrutiny by transnational networks.Such transparency, however, represents nongovernmental. organizationsand networks more than ord inary people, who may be as exc luded from

elite networks as they are from government circles. That is, transnationalcivil society may be a necessary but insufficient condition for democraticaccountability. Democracies should insist that, wherever feasible, international organizations maintain sufficient transparency for transnationalnetworks of advocacy groups, domestic legislators, and democratic publics toevaluate their actions. But proponents of democratic accountability shouldalso seek counterparts to the mechanisms of control embedded in nationaldemocratic institutions. Governors of the Federal Reserve Board are, afterall, nominated by the president and confirmed by the Senate, even if theyexercise great authority during their terms of office. If Madison, Hamilton,and Jay could invent indirect mechanisms of popular control in the FederalistPapers two centuries ago, it should not be beyond our competence to devisecomparable mechanisms at the global level in the twenty-first century.

InlC/,/Wlilll1i1! in,l'liiUtions 37As we continue to think about the normative implication::. of globalization,

we should focus simultaneously on the maintenance of robust democraticinstitutions at home, the establishment of formal structures of internationaldelegation, and the role of transnational networks. To be effective in thetwenty-first century, modern democracy requires international institutions.And to be consisten t with democra tic values, these insti tu tions must beaccountable to domestic civil society. Combining global governance witheffective democratic accountability will be a major challenge for scholars andpolicymakers alike in the years ahead.

Want to know more?The best single source for academic writings on internalional institutions isthe quarterly journal Inlernaliona! Orguni::alion, published by MIT Press.A special issue, published in Autumn 1998, reviewed the lasl 30 years ofscholarship in the field.

The sophisticated realism of the I970s, which largely ignored internationalinstitutions, is best represented by Kenneth Waltz's T l u ~ o r y of World Politics(Reading, MA: Addison Wesley, 1979). Fordata on numbers of internationalorganizations, see Cheryl Shanks, Harold Jacobson, and Jeffrey Kaplan's"Inertia and Change in the ConsteUation of International GovernmentalOrganizations, 1981-1992" (fnternmiona! Organi:::alion, Autumn 1996). Forstatements of institutionalist theory, see Robert Keohane's After Hegemony:Cooperation and Discord in the World Political Economy (Princeton, NJ:Princeton University Press, 1984) and Kenneth Oye, ed., Cooperation underAnarchy (Princeton, NJ: Princeton University Press, 1986). For a reflectionon this literature by an international lawyer, see Anne-Marie Slaughter[Burley], "International Law and International Relations Theory: A DualAgenda" (American Journal o( Interrullional LLlIr, April 1993).

On theUnited Nations and multilateralism, see Daniel Patrick Moynihan'sA Dangerous Place (Boston , MA: Lit tle, Brown and Company, 1978);Keohane and Joseph Nye, J1'. 's "Two Cheers for Multilateralism" (ForeignPolicy, Fall 1985); and John Ruggie's Winning the Peace (New York, NY:Columbia University Press for the Twentieth Century Fund, 1996).

The "relative gains debate" is thoroughly reported in David Baldwin, ed.,Neorealism and Neoliberalism: the Contemporary Debllte (New York, NY:Columbia University Press, 1993).

On bargaining and distributional issues, see Stephen Krasner's "GlobalCommunications and National Power: Life on the Pareto Frontier" (WorldPOlilics, April 1991); James Morrow's "Modeling the Forms of InternationalCooperation: Distribution versus Information" (lntel'l1aliollal Orguni::ation,Summer 1994); and James Fearon's "Bargaining, Enforcement and International Cooperation" (lnternationa! Organi::ation, forthcoming).Work on the legalization of international institutions is just beginning;my commen ts in this ar ti cl e reflect an ongo ing project on this subject

8/6/2019 Kohaneinterdependence

8/8

World polit ics both creates opportunities for modern go"imposes constraints on the range of act ions tha t it is feasibpursue. One way to think about these opportunities and ccanalyze the operation of the contemporary international posystem, or the world polit ical economy, and to consider howaffect state action. Much of the modern study of internatjOJdevoted to this task. Yet another perspective on the impact ofon s ta tes can be gained by asking how percept ive observers (reflected on these issues in the past. This approach, which lookof political thought for insights into contemporary internatioJbe pursued here. Although the form and intensity of the c