On his Foreign Policy blog today, political scientist Stephen Walt weighs in on a problem that keeps recurring in countries hit with popular uprisings in the past year: how do you convince the power-holders in an authoritarian regime to exit gracefully when the elected governments that follow might punish them for their crimes or strip them of their wealth?
A central issue is the familiar problem of credible commitment. In order to convince unpopular rulers to leave power (or at least to give up a lot of their current privileges), you have to convince them that they are not signing their own death warrants or ensuring their own financial ruin. But it is difficult for a successor regime to guarantee that they won’t go after the old elites at some point in the future; they cannot “credibly commit” to leave the old rulers alone once they have the power to prosecute them.
Walt’s proposed solution to this difficult problem is a bit fuzzy: maybe constitutions, maybe amnesties, mostly reassurances. “My own view,” he writes in conclusion, “is that entrenched elites need to [be] reassured that their immediate privileges won’t be dramatically curtailed, even if they give up a lot of political power. In essence, current rulers need to believe that they will be able to live out their lives in reasonable comfort, and that their immediate families won’t be ruined.”
As it happens, scholars of comparative democratization have spent a lot of time on this very topic. In their “little green book,” published in 1987, Guillermo O’Donnell and Philippe Schmitter contemplate the same problem and reach a conclusion similar to Walt’s. Where Walt talks in game-theory-speak of “credible commitments,” O’Donnell and Schmitter talk about pacts, which they define as “an explicit, but not always publicly explicated or justified, agreement among a select set of actors which seeks to define (or better, to redefine) rules governing the exercise of power on the basis of mutual guarantees for the ‘vital interests’ of those entering into it.”
In O’Donnell and Schmitter’s view, pacts aren’t a prerequisite to democratization, but they help. “We are convinced,” they write on p. 39, “that where they are a feature of the transition, [pacts] are desirable–that is, they enhance the probability that the process will lead to a viable political democracy.” They also observe that pacts are more likely to occur when “conflicting or competing groups are interdependent, in that they can neither do without each other nor unilaterally impose their preferred solution on each other.” In short, stalemates encourage pacting, and pacting is good for democratization.
When it was published, O’Donnell and Schmitter’s work was so influential, in part, because it showed how the outcomes of transitions weren’t predetermined by the structural conditions that had dominated earlier thinking about how and why democracy develops. In their view, the mode of transition mattered at least as much as the preconditions, and pacting was identified as the mode most likely to produce durable democracy. O’Donnell and Schmitter’s argument drew on a much earlier essay by Dankwart Rustow, and it was amplified in ensuing work by Terry Lynn Karl and Giuseppe di Palma, among others.
The claim that pacting helps democratization is certainly plausible, but it has not gone unchallenged. Without a shared sense of what kinds of negotiations and agreements constitute pacting, the concept has sometimes been stretched to cover almost any kind of elite interaction, which occurs to some extent in every instance of national political transformation. Democratic regimes have fallen in some iconic cases of pacting, such as Venezuela, while they have survived in many others where explicitly negotiated agreements were absent. These various outcomes mean that pacts are neither necessary nor sufficient to explain why some authoritarian openings produce democratic governments while others don’t.
Another way to try to get some purchase on this problem is to ask what differentiates the pacts that stick from the ones that fall apart. Like Walt’s recommendations, the evidence on this question is unclear. The theorizing on the subject implies that pacts work because they promise powerful groups protection from future harm, but it says little about what gives those promises the credibility they need to induce cooperation.
Democratization scholar Alfred Stepan tries to answer the question by framing it as a two-level bargaining problem. He posits that successful pacts “have two requirements: first, leaders with the organizational and ideological capacity to negotiate a grand coalition among themselves; second, the allegiance of their political followers to the terms of the pact.” These requirements are ultimately more descriptive than they are explanatory, however, in that they leave unanswered questions about how leaders acquire that capacity to coalesce and why political followers would give their allegiance.
Barry Weingast makes more headway on the matter by using game theory to reframe the problem as a matter of incentives. He shows that pacts succeed when they are self-enforcing, and to be self-enforcing, pacts must do two things. First and most obvious, parties to the bargain must believe they are better off under it than not. Second, “elites and their followers must be willing to punish those who seek unilateral defections from the pact.” These two conditions are sometimes difficult to distinguish in practice—after all, why would a party to a pact be willing to suffer the cost of punishing a defector unless he thought he would gain something from doing so?—but they usefully focus our attention on the need for gains from cooperation and for credible threats of punishment for actors who violate the terms of the pact.
Of course, in many real-world situations, compliance with a commitment will not obviously be aligned with an actor’s self-interest—in other words, the commitment will not be self-enforcing. In these situations, promises can still be made credible by vesting monitoring and enforcement capacity in some third party not susceptible to the same temptations. For this arrangement to work, that third party must itself be regarded as credible and capable. If that third party is a domestic organization or agency, it must be reliably rewarded for behaviors that uphold the bargain, such as running a clean election or ruling impartially on legal matters, and it must be thoroughly insulated from partisan manipulation. If it is an outside actor, such as a foreign government or international organization, the relevant domestic groups must believe that it is in that outside actor’s self-interest to enforce the terms of the deal. As Weingast notes, though, these conditions are not easy to meet. Consequently, self-enforcing pacts will often be difficult, if not impossible, to devise.
Considering all of this theory and evidence together, I think it’s fair to say that pacting looks more powerful than it really is. The kinds of assurances Walt proposes offering to members of the ancien regime could grease the hinges on the door to their exit from power, but it’s very hard to make those assurances credible enough to assuage the fears that are often keeping elites in their seats. And that’s without even considering the effects those backroom bargains might have on other groups who want the very things a pact promises they can’t get, like justice for crimes of state repression or a share of the wealth produced by their past labors. In the end, pacting turns out to be one of those solutions that looks quite promising on paper but is actually pretty flimsy in practice.