January 24, 2017
News /
Posted by NCID

On January 13th, Maria Lombardi, a PhD candidate from Universitat Pompeu Fabra, presented a paper titled: “Teacher Performance Pay and Student Learning: Evidence from a Nationwide Program in Peru,” which studies Bono Escuela, a nationwide pay-for-performance program in Peru. It is well-known that teaching quality is imperative for good educational attainment and that several countries’ payment schemes involve flat-salaries that are closely tied to seniority. It is in this context that the Peruvian program offers a school-level monetary incentive in an effort to improve student achievements. However, assessments of similar programs, in both developed and developing economies, are not conclusive on their effectiveness.

The program was introduced in public secondary schools in 2015. It consists of a tournament where schools are ranked depending on the average performance of their eighth graders on nationwide math and language standardized tests. The top 20% schools in each tournament are rewarded economically. It is a collective prize, such that all teachers are paid. An identification strategy is obtained by exploiting the fact that only eighth graders take the test and that it is the first year of implementation.

The authors study the impact of internal grades as opposed to the standardized measurements to avoid biases. That is, teachers could focus on test content and test-taking strategies to the detriment of broader learning. This effect can be measured, as it is not student grades, but standardized test results that are factored in the ranking.

The authors find no discernible impact on students’ math and language grades. An online survey was carried out to study this further. It seems that the size or nature of the incentive, misinformation or changes in teaching practices were not responsible for this outcome. Nonetheless, there is evidence that aspects of the standardized test may prevent teachers from exerting effort: students have no stakes in the test, and teachers may not know what pedagogical practices would improve performance. Likewise, teachers could be discouraged from investing more efforts if they estimated their school’s ranking to be particularly high or low. Finally, teachers might need more time to adapt and react to the program, as the policy itself was implemented and concluded within the confines of one academic year.