Working paper from World Bank- A kleptocrat's survival guide : autocratic longevity in the face of civil conflict;
Summary: Autocratic regimes are quite often short-lived kleptocracies formed and maintained through force and used to appropriate wealth from subjects. Some of these autocracies collapse after only a year or two of plundering while others manage to survive for 15 or 20 years. This paper asks why some autocratic regimes survive while others fail. A database of political regimes from 1960 to 2003 is introduced and accompanies the paper in an appendix. A model of political survival suggests that autocrats exchange constraints on their executive power for their continued survival. The relationship between payouts from successful rebellion and ease of rebellion determines how willing kleptocrats are to extend the political franchise and protect their power. Results show that extremely oppressive regimes and great expenditures on security are likely to accompany the most difficult environments for defense of the state. The model is used to identify the costs of pervasive political conflict and to decompose the "civil peace dividend" enjoyed by inclusive democracies that do not suffer from the malady of kleptocratic rule. Finally, the model suggests that slow democratization pushed by the autocratic elites to guarantee their survival, accompanied by stable development, may be the best path toward a democratic future for many fragile states.
On the graph; "Figure 1 depicts both the general trend that more developed countries are more democratic and the nuanced reality that economic development alone cannot account for either the successes or the failures of recent democratic transitions. Of course, these results do not speak to causality at all, simply the relationship between development and democracy. What can be concluded from Figure 1 and the statistics so far is that countries that are able to build lasting democracies and enjoy some prosperity seem to pool at a stable equilibrium among other successful performers including the OECD countries and a few eastern European transition economies as suggested by Przeworski, Alvarez, Cheibub and Limongi (2000)"