Hugh Macartney from Duke University will present "Incentive Design in Education: A Distributional Analysis” (joint with Rob McMillan and Uros Petronijevic).
This paper provides the first empirical analysis to examine how different incentive schemes in education affect the full distribution of student outcomes. Our starting point is an asymmetric hump-shaped relationship between teacher effort and incentive strength, estimated semi-parametrically using exogenous incentive variation and rich administrative data. We recover the primitives underlying this effort function, showing they can be identified by estimating a flexible effort-choice model. The model and resulting estimates are key inputs to a counterfactual framework we propose for tracing the unstudied effects of alternative accountability systems on the entire test score distribution, allowing effort to adjust endogenously. We find that widespread fixed-target systems give rise to a steep performance-inequality tradeoff; further, existing schemes can be modified to reduce test score inequality while improving average student performance at no extra cost. Overall, the combined estimation-simulation approach opens up new possibilities for incentive design in practice.