A note on the robustness of Card and Krueger (1994) and Neumark and Wascher (2000)

This note adds to the discussion originating from Card and Krueger (Am Econ Rev 84:772–793, 1994 ) and Neumark and Wascher (Am Econ Rev 90:1362–1396, 2000 ) by re-evaluating their results using the semiparametric difference-in-differences estimator introduced in Abadie (Rev Econ Stud 72:1–19, 2005 ). The note addresses the question of whether the differences in the original datasets can explain the differences in their conclusions. Although we observe heterogenous employment effects together with the differences in the distributions, the results show that the original findings on the overall employment effect of a minimum wage are fairly robust and just slightly changed when the differences in the distributions of the observed covariates are taken into account.