Why do some autocrats survive for decades, and others fall soon after taking power? The authors argue that when authoritarian rulers need to solicit the cooperation of outsiders or deter the threat of rebellion, they rely on political institutions. Partisan legislatures incorporate potential opposition forces, giving them a stake in the ruler's survival. By broadening the basis of support for autocrats, these institutions lengthen their tenures. An analysis of all authoritarian rulers in power during the 1946-1996 period provides evidence of the effect of nominally democratic institutions on their political survival.

What makes political regimes rise, endure, and fall? The main question is whether the observed close relation between levels of economic development and the incidence of democratic regimes is due to democracies being more likely to emerge or only more likely to survive in the more developed countries. We answer this question using data concerning 135 countries that existed at any time between 1950 and 1990. We find that the level of economic development does not affect the probability of transitions to democracy but that affluence does make democratic regimes more stable. The relation between affluence and democratic stability is monotonic, and the breakdown of democracies at middle levels of development is a phenomenon peculiar to the Southern Cone of Latin America. These patterns also appear to have been true of the earlier period, but dictatorships are more likely to survive in wealthy countries that became independent only after 1950. We conclude that modernization need not generate democracy but democracies survive in countries that are modern.

Dictatorships are not all the same: some are purely autocratic but many exhibit a full panoply of seemingly democratic institutions. To explain these differences, we develop a model in which dictators may need cooperation to generate rents and may face a threat of rebellion. Dictators have two instruments: they can make policy concessions or share rents. We conclude that when they need more cooperation dictators make more extensive policy concessions and share fewer rents. In turn, when the threat of rebellion is greater, they make larger concessions but also distribute more spoils. Assuming that policy concessions require an institutional setting of legislatures and parties, we test this prediction statistically for all dictatorships that existed between 1946 and 1996. Copyright 2006 Blackwell Publishing Ltd..

No abstract

No abstract