Multilateralism, Bilateralism, and Exclusion in the Nuclear Proliferation Regime

I use the nuclear proliferation regime to show that dyadic diplomacy is not necessarily incompatible with the building of a multilateral regime; bilateralism is not the opposite of multilateralism, but an efficient component thereof+ Although this point will not be new to most students of institutions, no general rationale has so far been offered on the complementarity of bilateral and multilateral diplomacy+ Starting from a characterization of proliferation as the result of a large number of prisoner’s dilemmas played out between states engaged in local dyadic rivalries, I demonstrate that it is possible for the superpowers to design an optimal mix of threats and bribes in which states with low compliance costs join the regime on the terms of the multilateral treaty alone; states with intermediate compliance costs need additional customized incentives, delivered through bilateral agreements; and states with high compliance costs are not only left out of the regime but also punished for nonparticipation+ I draw a few comparative statics that I systematically test on Nuclear Proliferation Treaty ~NPT! membership data+ I discuss the applicability of the model to the currency, trade, and aid regimes+ The nuclear proliferation regime has design features that are interesting to the institutionalist research agenda+ A key feature is the complementarity of a multilateral instrument—the Nuclear Proliferation Treaty ~NPT!—with superpower bilateral diplomacy+ The United States and the Soviet Union did more than negotiate a text that was agreeable to other countries; they also bribed and threatened some of their respective clients into signing the NPT+ This study suggests that the NPT regime, along with several other important regimes, is neither bilateral nor multilateral, but a combination of both+ The bilateral-multilateral dichotomy is a staple of the International Relations literature+ It is prominently featured in the grand debate pitting constructivists My thanks to Alex Downes, Jing Han, Massimo Morelli, Alex Thompson, Ching-Jen Sun, the editors of International Organization: Lisa Martin, Emanuel Adler, and Louis Pauly, three anonymous reviewers for comments, and T+ Marion Anderson and Byungwon Woo for research assistance+ Research was financed by a grant from the Mershon Center for International Security Studies+ 1+ See Koremenos, Lipson, and Snidal 2001+ International Organization 62, Summer 2008, pp+ 439–76 © 2008 by The IO Foundation+ doi:10+10170S0020818308080156 against realists+ Ruggie characterizes the last seventy years as a progressive shift away from bilateral and hegemonic regimes, in which relations among states are compartmentalized into dyads and obligations are specific to each dyad, toward multilateral regimes, characterized by equal treatment and universal participation+ Realists such as Mearsheimer counter that the spread of international institutions changes nothing in the way states have been interacting for centuries, for these institutions merely reflect the more powerful states’ calculations+ Using a more fine-grained approach to institutional reality, students of the trade and aid regimes have found that multilateral and bilateral obligations actually coexist, but in neither case has this finding led to a formalization that would make it possible to predict the actual proportion of multilateral and bilateral instruments+ The institutionalist literature in general holds that multilateral institutions perform better in issue areas involving bargaining and coordination, while bilateral institutions do better in areas involving enforcement through retaliation, but this literature stops short of arguing that the two instruments are necessary complements+ In this article, I show that both bilateral and multilateral instruments are optimally used in the nonproliferation regime, and for reasons other than the bargainingenforcement dichotomy+ I offer a general rationale for the complementarity of multilateral and bilateral diplomacy+ Regimes are like contracts in which a group of founders purchase a good from signatory states in exchange for a price and0or the nonimposition of a sanction+ Such contracts can be multilateral, offering uniform terms across eligible participants, or they can be bilateral, customizing the offer to reflect each state’s peculiar circumstances+ The multilateral strategy has the advantage of saving on transaction costs—there is only one deal and it is the same for everyone—but has the drawback of being expensive: participants are offered an incentive that is calculated to be sufficient to elicit the participation of the individual state that is burdened with the highest cost of compliance with the proposed regime+ In contrast, the bilateral strategy allows the founders to save on resources by giving each state the incentive it needs to participate and no more+ However, the bilateral strategy offers drawbacks as well, multiplying transaction costs, since a brand new contract has to be written for each new participant+ Consequently, I argue, for states with a low cost of compliance, a multilateral instrument should suffice, 2+ Ruggie 1992+ 3+ Mearsheimer 1994–95+ 4+ On the importance of relative power, see also Krasner 1991+ For an argument that the U+S+ hegemonic decline has caused a shift away from multilateralism to miniand bilateralism, see Yarbrough and Yarbrough 1992+ 5+ See Rodrick 1996; and Milner 2006 on aid; and Pahre 2001 on trade+ 6+ See Oye 1986; and Koremenos, Lipson, and Snidal 2001, 783+ See also Keohane 1984, 90; and Martin 1992+ The bargaining-enforcement dichotomy is not the only rationale thought to favor multilateralism+ Snidal ~1991! argues that increasing the number of states mitigates the bargain’s zero-sum properties+ Fearon ~1998, 298! cites multilateral bargains as a solution to the hold-out problem+ Taking exception with the bargaining-enforcement dichotomization are Pahre 1994; and Lohmann 1997+ 440 International Organization