Retesting Selectorate Theory: Separating the Effects ofWfrom Other Elements of Democracy

Abstract

Kevin Clarke and Randall Stone (2008) offer a methodological critique of some of our tests of the selectorate theory inThe Logic of Political Survival(Bueno de Mesquita et al.2003). We accept their critique of residualization for control variables in those tests, but reject the contention that the size of the winning coalition does not predict the provision of public goods and private benefits. We present new tests that control for elements of democracy other than W and that do not use residualization. These new tests show that selectorate theory is strongly and robustly supported. Our measure of the size of the winning coalition is in the theoretically predicted direction and is statistically significant for 28 out of 31 different public goods and private benefits. Aspects of democracy not contained in the selectorate theory explain less of the variance than does the theory's core factor, namely, winning coalition size, for 25 of the 31 public goods and private benefits.