Exit from International Organizations is a timely and tremendously important book. It offers a comprehensive theoretical account of withdrawals and suspensions from International Organizations (IOs) and features an extraordinarily rich and sophisticated empirical analysis based on both quantitative and qualitative data. The authors rigorously derive hypotheses from their theoretical framework and subject them to careful empirical testing. This book will surely remain the definitive reference on IO exit for many years to come.
The book establishes three central insights. First, withdrawals and suspensions from IOs are shown to be part of ordinary political bargaining, broadly consistent with rational institutionalist expectations. Second, the authors convincingly demonstrate that exit from IOs is costly. Third, the book challenges alternative explanations that portray IO exits primarily as manifestations of a broader rejection of multilateralism and intrusive global governance.
It would be easy to use the limited space of this review exclusively for praise. Instead, I will engage more closely with these three claims. All of them are valid and persuasively supported; two of them, however, warrant caution against overgeneralization. In particular, I maintain that the analysis cannot rule out the possibility that a qualitatively new development is currently unfolding—of which some IO exits are a part—and that cannot be fully understood as politics as usual. Such a fundamental change may generate exits that are not reducible to routine bargaining but instead reflect a broader rejection of the post-1990s international order. In this sense, the “exit as business as usual” argument may be somewhat overstated.
Arguably, the book’s most significant contribution is somewhat hidden behind the overarching claim that exit is business as usual: the demonstration that exit from IOs is costly. Withdrawals are associated with worsened political risk scores, declining investor confidence, a lower probability of being elected to the UN Security Council, and a reduced likelihood that remaining members conclude treaties with the withdrawing state in the near term. Empirically establishing these effects is a major achievement, and it confirms assumptions long held by theorists of international institutions.
Von Borzyskowski and Vabulas also demonstrate that material effects, measured in terms of trade with IO members and foreign direct investment (FDI) as a share of GDP, are considerably weaker. This qualification is important, and it underscores the sophistication of their empirical strategy. Still—this is a minor point—the juxtaposition of reputational and material effects may be somewhat misleading. A decline in investor confidence has material consequences, even if these do not always immediately appear in aggregate FDI figures. In the case of Brexit, for instance, both GDP and FDI declined, suggesting that reputational and material effects are more closely intertwined than the distinction might imply.
For von Borzyskowski and Vabulas, IO exit constitutes a routine instrument for negotiating institutional and policy change among rational actors. The authors test this claim by deriving hypotheses from a bargaining model in which threats of exit are employed when actors are dissatisfied with negotiation outcomes, and actual exit occurs if renegotiation fails. These additional costs associated with exit help explain why withdrawals remain comparatively rare.
Empirically, the authors demonstrate that threats to withdraw are observed about 10 times more often than actual withdrawals, which they interpret as evidence that exit threats function as routine bargaining tools. By contrast, actual withdrawal reflects a process of “self-selection.” From the perspective of rational design and negotiation theory, however, one might ask why self-selection should not also operate at the level of threats, given that unfulfilled threats can be reputationally costly as well.
More fundamentally, to claim that exits reflect “regular struggles over differing preferences” requires a clearer specification of what kind of conflicts are at stake, in order to go beyond the obvious claim that conflicts about IOs underlie IO exit. It implicitly relies on distinctions well-established in the literature: between interest conflicts and value conflicts as described by Vilhelm Aubert in the 1960s, between games and fights (including debates), between Stephen Krasner’s distributional conflicts on the Pareto frontier and the effort to get to the Pareto-optimum, or between John Ruggie’s distinction between interests and social purpose.
Von Borzyskowski and Vabulas operationalize the bargaining logic by measuring voting distance in the General Assembly (GA) between the withdrawing state and the other member states of a given IO. It is precisely this operationalization that raises concerns of conceptual stretching. Differences in policies and institutional design are indeed reflected in GA voting—but so too is a more fundamental rejection of an IO’s social purpose. As a result, the measure conflates cases in which states seek movement along the Pareto frontier with cases in which they aim to redefine—or abandon altogether—the purpose of the institution. The withdrawals of Germany and Italy from the League of Nations, or the wave of exits by Eastern European states in the 1950s, had little to do with distributional conflicts within existing frameworks.
The authors convincingly show that most IO exits are embedded in regular distributional politics. However, their data do not allow the inference that withdrawals and suspensions are never expressions of more fundamental conflicts over the purpose of IOs. The authors’ tendency toward overgeneralization is particularly visible in their interpretation of Brexit as a case of regular distributive conflict. Undeniably, Cameron’s attempt to renegotiate the UK’s terms of EU membership fits the bargaining model, and the failure of this renegotiation contributed to the subsequent withdrawal. Yet this perspective captures only the surface dynamics. Cameron’s strategy was itself reactive: it was driven by sustained domestic pressure from UKIP, Nigel Farage, and increasingly Eurosceptic factions within the Conservative Party. This pressure reflected not a dispute over the distributional consequences of EU membership, but a more fundamental rejection of the EU’s political and social purpose. Brexit, therefore, cannot be adequately understood as a failed bargaining episode; rather, it exemplifies how foundational conflicts over institutional purpose can generate exit, with bargaining serving as a contingent and ultimately unsuccessful attempt to contain them.
Due to the overgeneralization implied by measuring preference divergence for policies and institutional features via GA voting distance, the claimed rebuttal of alternative explanations is less conclusive than suggested. These alternative accounts do not argue that nationalist, populist, or sovereigntist motives explain most instances of IO exit. Rather, they claim that certain exits—and particularly certain peaks in exit activity—can be linked to broader political transformations, notably the formation of the bipolar system in the late 1940s and 1950s and the growing contestation of the liberal international order following the global financial crisis.
Demonstrating that most exits occur for reasons other than these claims does not invalidate them. By analogy, showing that most killings in a country occur for private reasons does not refute the argument that a surge in killings during a particular period is driven by political upheaval. That said, Exit from International Organizations convincingly demonstrates that these peaks are smaller than alternative theories would predict, especially when the growing number of IOs is taken into account. Most importantly, the overall number of exits declined again after 2007. Yet the recent announcement by the Trump administration of its intention to withdraw from 66 IOs raises a crucial question about the relationship between informal neglect of IOs and formal withdrawal. What is the sequence linking contestation, neglect, threats, and exit?
Understanding these sequences—both for exits as expressions of routine interest politics and for exits as manifestations of foundational struggles over the purpose of the international order—constitutes the most important task for future research building on the achievements of von Borzyskowski and Vabulas. Their book provides the indispensable foundation for this next generation of work.
