We analyze the influence of IMF and World Bank programs on political regime transitions. We develop an extended version of Acemoglu and Robinson's [American Economic Review 91, 2001] model of political transitions to show how the anticipation of new loans from international financial institutions can trigger political transitions which would not otherwise have taken place. We test this unexplored implication of the theory empirically. We find in a world sample from 1970 to 2002 that the anticipation of receiving new programs immediately after a political regime transition increases the probability of a transition from autocracy to democracy and reduces the probability of democratic survival.