Does increasing student access to charter schools lead to more equitable outcomes for all? More specifically, does increasing the number of students who attend charters in a district lead to a narrowing achievement gap for all or do charters simply help students who are already ahead? To address these questions, I take advantage of a plausibly exogenous raise in the Michigan charter cap. Combining school- level test score data from the Michigan Educational Assessment Program and district-level data from the National Alliance for Public Charter Schools, I regress measures of the achievement gap and student achievement on the fraction of district students enrolled in charters. I find a large and significant negative relationship between charter enrollment and the achievement gap with estimates indicating a one standard deviation (SD) increase in charter enrollment decreases the racial achievement gap between 0.10 and 0.17 SD, a 9.3 to 15.6 percent decline in the average racial achievement gap in the sample. However, this relationship is no longer significant when using models that isolate quasi-random variation indicating there is likely no causal effect of charter school enrollment on the achievement gap. This paper emphasizes the difficulty of measuring the causal effect of charter schools on all students due to charters ability to selectively locate.

Read the full paper here.