Principal Quality and the Role of Personnel Management in Public Education
Principals shape almost every aspect of schools, assigning teachers to classrooms and matching teachers and students with resources. This paper quantifies the impact of principal quality on student achievement and documents the correlates of principal effectiveness. Using a variance decomposition that exploits principal transitions across schools in the North Carolina public school system, I find that differences in principal quality explain approximately 5% of the observed variation in test scores. To identify effective principals, I construct principal value-added (PVA) estimates and provide the first evidence that they are forecast unbiased. I use these individual-level estimates to examine the correlates of PVA and the mechanisms through which principal effects operate. My results suggest that previous effective teaching strongly predicts subsequent PVA. To elucidate potential mechanisms, I employ an event-study design around principal transitions, finding that more effective principals excel at attracting better teachers and retaining their best staff. Furthermore, they are more likely to assign their school's best teachers to larger classrooms, which increases overall student learning. School survey data allow me to unpack why effective principals attract and retain high-quality teachers. I document a robust relationship between PVA and various measures of leadership and teacher empowerment, suggesting that test score--boosting principals also possess certain soft skills that make them more appealing supervisors relative to others.
The Impact of Tenure Removal on Teachers' Labor Supply Responses
Non-pecuniary benefits, such as job stability, are central to the workforce. Yet, evidence on whether they affect worker effort and labor supply decisions is limited, as output is typically hard to observe and strict collective bargaining agreements complicate overhauling worker contracts. This paper overcomes these challenges by studying a statewide reform in North Carolina that abruptly eliminated tenure paths for recently arrived and newly hired teachers. Using administrative data, I link teachers to classrooms and examine how decreased job security affects teacher impacts on student test scores. I find that productivity does not decline after receiving tenure, as teacher value-added parallels the preceding years. Regarding labor supply, I find that teachers entering after the reform were less effective than their older peers, even after controlling for experience, with differences in quality widening over time. Teachers entering after the policy were 4% and 2% of a standard-deviation less effective on math and reading value-added, respectively. I argue that these declines reflect teacher selection, as observable characteristics a restable across cohorts.