Democracy is widely accepted today, perhaps as never before, as the most suitable form of government. But what is democracy, and does it always produce good government? There is a tradition in popular thinking and in political science that associates democracy with the existence of competitive elections. This follows the tradition of Schumpeter (1950) and Downs (1957) in public choice. Competitive elections are undoubtedly a necessary condition for democracy. But they are not sufficient for true democracy or for democracy to function reasonably well. For example, the fathers of the American Constitution expressed their fear of factions and of mob sentiments. One problem with competitive elections is that they provide no protection for minorities. A second is that they may not express the long-term interests of the electorate itself, or their attitudes in sober second thought. Yet a third is the well-known Arrow problem of cyclical majorities – perhaps one of the central discoveries of modern public choice. A fourth is the recurring tension between the two principles of government by majority on the one hand and the rule of law on the other.

Contemplating the failures of democracy in theory, one early tradition in public choice theorizing was based on the idea that, whatever its failings in theory, democracy works well in practice. This puzzle arose early in public choice theory, and the question, “Why so much stability?” that is, specifically, “Why do democracies appear so much more stable in practice than in theory?