Introducing a performance-based school grant in Jakarta : what do we know about its impact after two years ?

This paper evaluates the early impact of introducing a performance component into Jakarta's school grant program on learning outcomes. Using administrative data, it applies difference-in-differences and regression discontinuity approaches to identify the impact of the grant by exploiting differences in program coverage over time, as well as by comparing changes in test scores between schools that received the additional performance award with schools that did not. The paper finds that the introduction of the performance component had different impacts on government primary and junior secondary schools. The program improved learning outcomes for primary schools at the bottom of the performance distribution and narrowed performance gaps across schools. However, improvements in equity were also driven by negative impacts of the program on better performing primary schools. Overall, the program reduced primary examination scores albeit by a small amount. In contrast to the results at the primary level, the performance component improved examination scores in government junior secondary schools. However, the impact seemed to be greatest among better performing schools and has therefore widened performance gaps. The findings also suggest that program impact was largely through competition between schools to receive the performance component. There is little evidence that the additional resources schools received from the award had any additional impact. The evaluation utilized preexisting administrative data and the paper offers some suggestions on how education information systems can be strengthened to create more robust feedback loops between research and policy.