Unraveling the relationship between presidential approval and the economy: A multi-dimensional semi-parametric approach

Empirical studies analyzing the determinants of U.S. presidential popularity have delivered quite inconclusive results concerning the role of economic variables by assuming linear relationships. We employ penalized spline smoothing in the context of semi-parametric additive mixed models and allow for flexible functional forms and thus possible non-linear effects for the economic determinants. By controlling for the well-known politically motivated covariables, we find strong evidence for non-linear and negative effects of unemployment, inflation, and government consumption on presidential approval. Additionally, we present new results in favor of non-parametric trivariate interaction effects between the macroeconomic covariables.