Measurement error, legalized abortion, and the decline in crime: A response to Foote and Goetz

We are grateful to Foote and Goetz for noting that the final table of Donohue and Levitt (Quarterly Journal of Economics, 116 (2001), 379-420) inadvertently omitted state-year interactions. Correcting our mistake does not alter the sign or statistical significance of our estimates, although it does reduce their magnitude. Using a more carefully constructed measure of abortion that better links birth cohorts to abortion exposure (by using abortion data by state of residence rather than of occurrence, by adjusting for cross-state mobility, and by more precisely estimating birth years from age of arrest data), we present new evidence that abortion legalization reduces crime through both a cohort-size and a selection effect.