Explaining State Unemployment in the U.S.: Cross-national Versus Political Predictors
William A. McCracken, J. Norman Baldwin

Abstract
This article endeavors to determine (1) whether political variables or traditional predictors of cross-national economic growth are better predictors of state unemployment and (2) whether national or state political and economic factors are better predictors of state unemployment in the United States (U.S.). Twenty-three years of panel data were gathered on the 48 states in the continental United States. All models were tested through generalized least squares (GLS) regressions with panel corrected standard errors. The findings indicate that state unemployment and unemployment growth decrease as four-year colleges and highway expenditures increase and increase with Democratic Congresses and as savings and national unemployment rates increase. The state equivalents of cross-national variables prove to be the most consistent predictors of unemployment. However, a limited number of national economic and political variables must also be considered in pursuing remedies to unemployment because they explain a substantial amount of variance in state unemployment and unemployment growth.

Full Text: PDF