The bad end of the bargain?: Revisiting the relationship between collective bargaining agreements and student achievement

Abstract

This paper revisits the relationship between teacher collective bargaining agreements (CBAs) and student achievement. Using a district-level dataset of California teacher CBAs that includes measures of overall and subarea contract strength linked to district-level panel data, we build on prior work by controlling for unobserved fixed and time-varying confounders. This study demonstrates that naïve pooled OLS estimates of student achievement on overall CBA strength are larger and more negative than lagged achievement and within-district estimates, signifying a negative bias in the naïve levels models. When controlling for time invariant and time-varying unobservables, the relationship between CBA strength and student achievement is persistently negative and small, or null, but never significantly positive. This relationship extends to specific CBA subareas and to subgroups of students. These findings have important implications for new reforms designed to weaken teacher collective bargaining rights.