"Overpaid"? "Underpaid"? How Teaching Pay Compares to Teachers' Next-Best Options
This paper explores the positive question underlying the debate over whether teachers are "overpaid" or "underpaid": How much are teachers paid relative to their next-best options, and why is there a difference, if any? Identifying teachers' next-best options is inherently difficult due to selection in and out of teaching and there being limited data on who is interested in teaching and where they work aside from teaching. We construct a linked administrative dataset that tracks the employment and teaching-related milestones for the near-universe of workers in Kentucky, in which we can identify teachers and individuals who express an interest in teaching. Using novel quasi-experimental variation from an entry exam test score cutoff and event studies around teacher exits from the profession, we find that teachers have substantially lower-paying next-best options in our setting. The combination of the teaching premium, the hiring difficulties reported by Kentucky schools, and the backdrop of the state's labor market conditions suggests that the teaching job holds low-amenity value even relative to teachers' lower-paying next-best options.