Association for Public Policy Analysis & Management

There is growing interest in strengthening teacher incentives by tying pay to performance measures based on student achievement. Yet, while the theory concerning the design of performance incentives is well developed, there is little empirical evidence on how teachers may respond to specific design features of performance pay schemes. In particular, theoretically appealing but complex schemes may not outperform less appealing but simple schemes in practice. In this paper, we present the results of a randomized trial designed to test alternative approaches of mapping student achievement into rewards for teachers. Math teachers across 216 schools in western China were randomly assigned to participate in relative performance pay schemes in which teacher rankings were determined by one of three different methods of defining teacher output as a function of student achievement on standardized exams. We find that teachers offered pay-for-percentile incentives (specifically designed to elicit effort that contributes to achievement gains for all students based on theory in Barlevy and Neal, AER, 2012) outperform teachers offered two more simple schemes based on year-end class average achievement levels or average gains over the course of a school year. Moreover, we find that teachers behave in line with theory and respond to incentives by allocating effort across students according to returns determined by contract structure. Our findings have implications beyond education and suggest generally that individuals can respond to relatively complex features of reward schemes.