Date of Graduation


Document Type


Degree Name

Doctor of Philosophy in Education Policy (PhD)

Degree Level



Education Reform


Jay P. Greene

Committee Member

Patrick J. Wolf

Second Committee Member

Scott Eidelman


Education, Compensation, Education policy, Education reform, Merit pay, Teacher quality


Teacher quality has a significant impact on both student learning gains and later life outcomes. With this is mind, policymakers implement reforms to attract and retain more effective educators. A major obstacle for designing these policies is that the ingredients for training, as well as initially identifying, effective teachers remain largely a mystery. However, there are strong theoretical arguments for certain education policy reforms producing improvements in the quality of the teacher workforce. One increasingly popular example is performance-based pay. Performance pay has the potential to better align teachers' incentives to produce increases in student achievement. Paying teachers based on students' learning gains, rather than years of experience, increases lifetime earnings for effective teachers which could attract and retain higher-caliber educators. However, since performance-based pay programs rarely last longer than a few years and tend to be small in scale, researchers have not been able to evaluate how changing the compensation structure affects the composition of the teacher workforce.

I provide preliminary evidence for how implementing a performance-based pay program could impact the teacher workforce. Based on analyses from three studies, I conclude that individuals who enter the teaching profession are significantly more risk averse than individuals entering other professions. This finding supports a common stereotype about teachers and possibly provides an explanation for their resistances to education reforms such as merit pay, even when such policies are fairly popular with the rest of the general public. In a follow-up study, the preferences of teachers with a preference for performance-based pay are compared to those of other teachers. The hypothesis of this study is that changing a central component of teacher compensation, the step and lane pay scale, is likely to attract less risk-averse employees. The evidence from this study is inconclusive but suggests that performance pay might alter the composition of the teacher workforce by either attracting more risk-loving teachers to the classroom or deterring relatively risk-averse individuals from entering the profession. Finally, I link teachers' risk and performance pay preferences to measures of teacher quality. I find that teachers who are the least supportive of performance pay are actually more effective in the classroom. This result contradicts the argument that a compositional shift from performance-based pay will necessarily improve teacher quality. This finding is somewhat counterintuitive but is possibly explained by prior research on the negative relationship between intrinsic or mission-driven motivations and the introduction of performance-based financial incentives.