In an effort to enhance the quality of early childhood education (ECE) at scale, nearly all U.S. states have recently adopted Quality Rating and Improvement Systems (QRIS). These accountability systems give providers and parents information on program quality and create both reputational and financial incentives for program improvement. However, we know little about whether these accountability reforms operate as theorized. This study provides the first empirical evidence on this question using data from North Carolina, a state with a mature QRIS. Using a regression discontinuity design, we examine how quasi-random assignment to a lower quality rating influenced subsequent outcomes of ECE programs. We find that programs responded to a lower quality rating with comparative performance gains, including improvement on a multi-faceted measure of classroom quality. Programs quasi-randomly assigned to a lower star rating also experienced enrollment declines, which is consistent with the hypothesis that parents responded to information about program quality by selectively enrolling away from programs with lower ratings. These effects were concentrated among programs that faced higher levels of competition from nearby providers.