Labor surplus and punishment : testing for a direct effect of unemployment on imprisonment

The relationships between labor surplus, crime and punishment of criminal offenders have generated considerable controversy and research. Conflict theorists contend that punishment varies directly with economic conditions, particularly the size of the labor surplus. The effect of labor surplus on punishment is, however, often attributed to variation in crime rates. The primary objective of this study is to determine the effect of change in labor surplus, as measured by unemployment, on change in imprisonment while controlling for crime and other theoretically relevant variables. Residual-change regression is applied to data from a sample of 269 U.S. counties. The analysis supports the independent contribution of labor surplus to imprisonment. Supplementary analyses using cross-sectional equations for each year 1980 to 1990 and a regression of static imprisonment on lagged imprisonment and static independent variables, find no support for a direct relationship of labor surplus and imprisonment. The strength of the relationship between labor surplus and imprisonment is contingent on the technique used to examine it.