Merit‐based incentives are a topic of growing interest in labor economics due to their potential to increase performance for private and public employees. Following this argument, such pay schemes have been applied in numerous countries to provide incentives to teachers and schools based on their students’ achievement scores and other performance metrics. However, because of the multi-task, multi-principal and multi-period nature of education, they present several caveats. Observational and experimental research provides ambiguous conclusions about their impact. This paper contributes to this literature by evaluating the effects of the Mexican PECD, a program that since 2010 has provided salary bonuses to teachers in primary and secondary public schools based on their national standardized tests scores. Nearly 30,000 schools and 300,000 teachers have benefited from this program in its first implementation. Given its characteristics, the PECD provides an ideal ground for a Regression Discontinuity Design (RDD) and for a Multiple Rating‐Score Discontinuity (MRSD). Combining these quasi‐experimental techniques with a Hierarchical Linear Model (HLM), I show that the effect of this program on student performance is null. If any, this program seems to harm student performance; this negative effect is more evident for indigenous schools.