A Critique of "The Impact of Legalized Abortion on Crime"

After reading the critique of "The Impact of Legalized Abortion on Crime" I would have to side with the authors of the critique. It is however possible that Donohue and Levitt came out with a response to the critique arguing their side. But several factors impact my assessment of the original paper. As the critique points out, there was a coding mistake in DL's concluding regressions, which identified abortion's effect on crime by comparing the experiences of different age cohorts within the same state. Another convincing theoretical argument is that statewide crime rates are influenced by other factors besides abortion, and that crime in one state is not necessarily caused by the same crime committed in another state. The final nail in coffin in my mind comes from the fact that DL's final regression uses total number of arrests instead of per capita number of arrests. This would certainly distort the regression results and would incorrectly measure abortion's effect on crime.