Abstract
Studies of the effects of economic conditions on elections reach a mixed verdict. Some find significant effects of unemployment, real income, and/or inflation, while others do not. Stigler contends that macroeconomic factors are not important but that distributional issues are. In this article the effects of real income growth and inflation on U.S. presidential elections are studied using a pooling of state-level and time-series data. The data are analyzed using ordinary least squares, variance components, and other estimation techniques, since any single technique is not wholly satisfactory for the analysis of this data set. The effect of real income growth is found to be insignificant over time but significant across states. This suggests that Stigler's contention is correct. The effect of inflation is not as clear since high t-values for the full-sample inflation term may be due to conincidental correlation with time-dependent factors outside the model.
Get full access to this article
View all access options for this article.
