The challenge of designing effective teacher performance pay programs: experimental evidence from Tanzania

Teacher performance pay systems can improve student learning outcomes, particularly in settings where existing mechanisms for teacher accountability are weak. We use a field experiment in Tanzanian public primary schools to directly compare the effectiveness on early grade learning of two different teacher performance pay systems: a pay for percentile system and a system that rewards teachers based on student proficiency levels. Pay for percentile is more complex but can (under certain conditions) induce optimal effort among teachers. However, a threshold system is easier to implement and it may be easier for teachers to figure out how to react optimally to the incentive scheme. Both systems improve student test scores. However, despite the potential to induce optimal effort of the pay for percentile system, the proficiency system is at least as effective in boosting student learning. Moreover, we find suggestive evidence that the pay for percentile system favors students from the top of the performance distribution, highlighting the challenge of designing incentives that can deliver optimal and equitable learning gains for all students.