A reexamination of Card and Krueger's state-level study of the minimum wage

Card and Krueger's (1995) difference-in-difference study of the 1990-1991 federal minimum wage hikes compared states by the proportion of workers directly affected by the minimum wage. They found "no evidence that the increase in the minimum wage significantly lowered teenage employment rates more in highly affected states" and they concluded the minimum wage did not reduce employment. Their paper was highly influential and convinced many that the minimum wage did not reduce employment. However, when I apply their model to the 1996-1997 federal minimum wage hike, I find that increases in the minimum wage significantly lowered teenage employment rates more in highly affected states. Using Card and Krueger's interpretation, this implies the minimum wage did reduce teenage employment.