I’m giving a talk at Texas A&M next week on military coalitions, specifically when they provoke opposition from outsiders and when they can keep their targets isolated. The United States is prolific in its use of coalitions for coercing other states, and as military budgets tighten in the future, it doesn’t take much to imagine that acting with “friends and allies” against foreign opponents is going to remain the order of the day. In fact, multilateralism in general and coalitions in particular are often considered the sine qua non of signaling restraint and preventing counterbalancing.
This paper, though, takes a different approach. Coalitions have plenty of benefits, from aggregating power to burden-sharing to salving public opinion, but the fact remains that they can be aggregations of power more threatening as a whole than their individual parts. In fact, 17% of the time, they provoke counter-coalitions, which not only widen wars but can also undermine whatever other benefits to multilateral action we might expect.
To boil the model and its implications down to the essentials, powerful coalitions that are likely to disband—and therefore diminish potential threats to outsiders—are less likely to provoke the formation of counter-coalitions than similarly powerful coalitions that are likely to act together in the future. Observers who fear future coalitional action—think Iran in 1991 or 2003—can look at the diversity of interests in the coalition to judge its durability, and in fact more diverse coalitions are far more likely to keep their targets isolated than more homogeneous coalitions. Here’s one of the better graphs from the paper (I may post slides later), plotting the predicted probability that an American-led coalition provokes opposition (i.e. a counter-coalition) as a function of the diversity of the coalition’s interests. (Controls include the number of states, power, regime type, and UN support/opposition, among others.)
As you can see, that’s a fairly substantial drop in the probability of provoking opposition as the coalition grows more diverse. (As well as a source of relief for me after hoping that the implications of the theory would come out favorably.)
Granted, not all coalitions are matters of immediate choice based on this factor. For example, when alliance commitments explain the formation of a coalition, concerns over provoking opposition may be moot. So there may not be too much useful for partner choice here (after all, in 1991, working without the Saudis was an obviously dominated strategy, whatever other factors might’ve been in play), but in terms of what to expect, in terms of identifying the coalitions most likely to provoke wider, longer, or harder-to-win wars, it could be useful.
Let’s just hope my audience is inclined to agree.