The Minimum Wage and the Great Recession: A Response to Zipperer and Recapitulation of the Evidence

Clemens and Wither (2014) find that minimum wage increases contributed to employment declines among low-skilled individuals during the Great Recession. Zipperer (2016) argues that Clemens and Wither's estimates are biased. This paper assesses what underlies the difference between Zipperer's estimates and Clemens and Wither's estimates. I first show that Zipperer's control sets significantly attenuate the relationship between Clemens and Wither's ``treatment indicator'' variables and states' minimum wage rates. Scaling for this dilution of the underlying treatment accounts for nearly half of the difference between Zipperer's estimates and Clemens and Wither's estimates. Second, I show that the within-region variation on which Zipperer focuses attention biases his estimates towards positive values. Employment and income aggregates, as well as housing and construction indicators, reveal that within-region comparisons are prone to considerable upward bias. Florida, for example, experienced a far more severe housing decline than the regional neighbors for which several of Zipperer's specifications use it as the primary control. I show that Zipperer's estimates are quite sensitive to removing states with extreme housing crises from the sample, while the original Clemens and Wither estimates are not. I further show that Zipperer's specifications have implausible implications for the minimum wage's ``effects'' on employment within high skilled population groups. I conclude by recapitulating the basic facts underlying Clemens and Wither's assessment of the evidence.