We present evidence from the first three years of a randomized controlled trial of a government-administered pilot teacher performance pay program in Punjab, Pakistan. The program offers yearly cash bonuses to teachers in a sample of 600 public primary schools with the lowest mean student exam scores in the province. The bonus is linked to the change in the school’s average student exam scores, the change in the school’s enrollment, and the level of student exam participation in the school. Bonus receipt and size are randomly assigned across schools according to whether or not the teacher is the school’s head. The program increases student exam participation rates in the second and third year and increases enrollment in grade 1 in the third year. We do not find that the program increases student exam scores in any year. Mean impacts are similar across program variants. The absence of positive impacts on test scores may be due to weaknesses in the program’s incentive structure and/or limitations in the program’s administrative data.