Changing the education workforce? The relationships among teacher quality, motivation, and performance pay

Abstract

Background/Context: With a growing body of evidence to support the assertion that teacher quality is vital to producing better student outcomes, policymakers continue to seek solutions to attract and retain the best educators. Performance-based pay is a reform that has become popular in K-12 education over the last decade. This strategy potentially produces positive impacts on student achievement in two ways: better alignment of financial incentives with desired outcomes and improved the composition of the teacher workforce. While evaluations have primarily focused on the former result, there is little research on whether the longer-term implementation of these polices can attract more effective teachers. Purpose: In this study we aim to provide evidence for potential long-term impacts that performance- based pay can have on the composition of the teacher workforce by addressing two questions: Does performance-based pay attract fundamentally different individuals, as measured by their risk preferences, to the teaching profession? Are stated preferences for a particular pay format correlated to measures of teacher quality? Research Design: We apply methods from experimental economics and conduct surveys with 120 teachers from two school districts who have experienced performance pay. We compare the risk preferences of teachers hired under the two pay formats to test the hypothesis that performance-based pay attracts individuals with different characteristics to the profession. We also analyze teachers' survey responses on their preferences for performance-based pay to determine their relationships to two measures of teacher quality: student test-score gains and principal evaluations. Conclusions/Recommendations: We find mixed results regarding the ability of performancebased pay to alter the composition of the teacher workforce. Teachers hired with performancebased pay in place are no different from their colleagues. However, teachers claiming to seek employment in districts with performance-based pay in place appear significantly less risk averse. Surprisingly, additional analyses indicate that teachers' value-added scores and performance evaluations do not predict a positive disposition towards merit pay. Thus, while these results indicate the possibility for performance-based pay to attract different individuals to teaching, they do not provide evidence that such change would necessarily improve the composition of the workforce. Policymakers should take this potential tradeoff into consideration when considering the expansion of performance pay policies.