Do State College Promise Programs Increase Enrollment? Evidence from Staggered Adoption

Abstract

State "Promise" programs that provide free community college tuition have spread rapidly across the United States, reaching 20 states by 2021. Despite the policy's popularity, aggregate enrollment effects remain unclear. Using a staggered difference-in-differences design with Callaway-Sant'Anna estimators and state-level enrollment data from the American Community Survey (2010–2023), I find that Promise programs have no statistically significant effect on aggregate college enrollment (ATT = $-0.0136$ log points, SE = 0.0102, $p > 0.05$). Event study estimates show no differential pre-trends and no significant post-treatment effects through seven years after adoption. However, power analysis with 20 treated states and 31 control states reveals a minimum detectable effect of 29%, suggesting the null finding may reflect insufficient statistical power rather than true absence of effects. These results highlight the limitations of state-aggregate data for evaluating Promise programs and suggest that previous findings of large enrollment increases may derive from compositional shifts (community college vs. four-year) rather than net enrollment gains.