Dan Slater offers thoughtful and incisive comments. We respond here to three of his points. The first is that by limiting our study to the post–Cold War period, we convert it into a “period piece,” akin to studies of fascist and communist regimes. Although this may be true, a historically bounded analysis is essential because of the changing character of the international environment. World historical time powerfully shapes regime outcomes. The prospects for democracy and authoritarianism during the Cold War, which was marked by global superpower rivalry, differed dramatically from those during periods of Western liberal hegemony. During the Cold War, for example, nearly all military coups ushered in authoritarian rule; after 1989, nearly 70 percent of coups led to multiparty electionsFootnote ¹ In 1989, single-party rule predominated in Eastern Europe and sub-Saharan Africa; five years later, it had disappeared.

The importance of world historical time is clearly seen in Slater's own work. The Cold War international environment encouraged and reinforced protection pacts in Malaysia, Indonesia, and Singapore, and the end of the Cold War contributed to the unravelling of protection pacts in Indonesia, South Africa, and Taiwan. In another world historical moment, then, Slater's stories might have played out differently. A changing international environment poses problems for universal theories of democratization. Rather than build universal theories that have difficulty withstanding empirical scrutiny, we prefer a midrange theory that gets it right.

A second criticism is that our theory offers no domestic route to democracy. We make clear in the book that democratization is not always externally driven. Indeed, it is usually domestically driven. The absence of a domestic route to democracy in our study is a product of the particular nature of our cases. With few exceptions (e.g., Mexico, Taiwan), countries that were competitive authoritarian in the early 1990s—more than a decade into the Third Wave—lacked favorable domestic conditions for democratization. Because of underdevelopment and/or decades of state socialism, nearly all our cases were characterized by small private sectors, underdeveloped civil societies, and weak, fragmented opposition movements—conditions that rarely generate a robust domestic push for democracy. Unlike successful Third Wave cases like Brazil, Spain, or South Korea, then, democratization in countries like Albania, Cambodia, Guyana, Malawi, Mozambique, Nicaragua, and Romania required a strong external push.

A third, and related, criticism is that we underpredict democratization in cases of low organizational power. We sought to build a parsimonious theory capable of explaining regime trajectories in thirty-five countries across five world regions. Unlike most small and medium-n analyses, ours includes cases that do not fit our theory. Although we correctly predict (and demonstrate the causal process behind) outcomes in 80 percent of the cases, our theory has outliers. Several of these are cases of low organizational power. According to Slater, our theory “needlessly” predicts turnover without democratization in these cases and thus “unnecessarily” generates false negatives (e.g., democratization in Benin, Mali, Peru, and Ukraine). In his view, the “democratic transitions literature … provides ample theoretical reason to expect democratization” in such cases. Here Slater is simply wrong. These cases are characterized not only by weak civil societies and domestic oppositions but also by state and party weakness. With the exception of the early transitions literature, which is widely regarded as overly voluntarist, no major theory of democratization provides any reason to expect democratization in cases such as Benin, Mali, Georgia, Madagascar, Malawi, or Moldova. On the contrary, there is ample theoretical reason to expect that democracy will not emerge from transitions in such cases, and that if it does emerge, it will not survive. Indeed, of our “false negative” cases, Mali is borderline competitive authoritarian and Ukraine had already reverted to competitive authoritarianism by 2010. In 2011, Benin's presidential election was marred by fraud charges and Peru slid toward a possible democratic crisis. Thus, although our theory does in fact underpredict democracy, these predictions are theoretically sound—and they may yet be borne out!