Summary: Attracting and retaining high-quality teachers has a large social benefit, but it is challenging for schools to identify good teachers ex-ante. This paper uses teachers’ contract choices and a randomized controlled trial of performance pay with 7,000 teachers in 243 private schools in Pakistan to study whether performance pay affects the composition of teachers. Consistent with adverse selection models, we find that performance pay induces positive sorting: both among teachers with higher latent ability and among those with a more elastic effort response to incentives. Teachers also have better information about these dimensions of type than their principals. Using two additional treatments, we show effects are more pronounced among teachers with better information about their quality and teachers with lower switching costs. Accounting for these sorting effects, the total effect of performance pay on test scores is twice as large as the direct effect on the existing stock of teachers, suggesting that analyses that ignore sorting effects may substantially understate the effects of performance pay.
Inducing Positive Sorting through Performance Pay: Experimental Evidence from Pakistan Schools
Citation: Brown, Christina and Tahir Andrabi. 2022. “Inducing Positive Sorting through Performance Pay: Experimental Evidence from Pakistan Schools”, Working Paper.
Christina Brown
Tahir Andrabi
Teachers are the most important input in the education production function, but schools imperfectly observe teacher quality, making it hard to effectively screen teachers. The characteristics available to schools, such as experience, college grades, credentials, and interview scores, are poor predictors of future performance, explaining less than 5% of the variation in teacher value-added. This challenge is not unique to schools. The majority of firms cite challenges in hiring and retaining high-quality employees.
Schools face a challenge in identifying who will be a good teacher. The characteristics available to schools at the time of hiring, including interview scores, explain less than 5% of the variation in teacher quality. However, while employers might have a hard time knowing who is a good teacher, teachers (or potential teachers) themselves may know a lot about their own ability. Using teachers’ “private information” about their own skills could help schools identify who are the great teachers. One way to elicit this information and allow better teachers to self-select into teaching could be to offer performance incentives, which, in theory, should be more attractive to high performers.
We test whether performance pay contracts allow schools to attract and retain better teachers. We conducted a two-year experiment with 7,000 teachers in 243 schools across Pakistan. The experiment proceeded in two phases. First, we offered teachers the opportunity to choose their contract for the coming year, selecting between a flat raise versus a performance-based raise. We let teachers know their choice would be implemented with some probability. Next, we randomly assigned schools to one of three conditions: performance-based raises, flat raises, or the teacher’s choice.
Study Design and Findings
Performance pay attracts higher-performing teachers
We can test whether teachers who chose performance raises are systemically different. If teachers have information about how they will fair under incentives versus flat pay, then we would expect higher-quality teachers to be more likely to select performance pay. To do this, we compared teachers’ baseline performance for teachers who chose performance versus flat raises. While there is a lot of overall variation in teacher quality, high value-added teachers are more likely to select performance pay, and low value-added teachers are more likely to select flat pay. We also find that schools that are assigned performance pay attract and retain better teachers in the long run. These effects are mostly driven by high value-added teachers moving from control to treatment schools and low value-added teachers moving from treatment to control schools.
Performance pay attracts teachers who respond better to incentives
We may also expect that teachers who chose performance pay contracts “respond” better to them as well. In other words, teachers may vary in how much harder they work under incentive pay and this effort response might depend on what type of contract the teacher wanted. To test this, we compared the effect of being assigned a performance pay contract versus a flat pay contract for teachers who wanted performance pay and for those who did not want it. Teachers who chose performance pay contracts during the baseline choice exercise have nearly nine times the effect of performance pay on test scores as compared to those who chose flat pay. This suggests, perhaps unsurprisingly, that incentives work much better for those that want them.
Combined, this evidence suggests that schools would benefit from introducing performance pay or allowing teachers to choose what type of contract they receive. Since performance pay works particularly well for those who want it, allowing teachers to choose could be both politically palatable and substantially beneficial in terms of improving student outcomes. This evidence also suggests that previous estimates of the effect of performance pay, which focus on existing teachers’ effort but ignore this sorting channel (that performance pay may attract and retain better teachers), are missing a big piece of the impact of these policies. Since the use of performance incentives for teachers has doubled globally in the last ten years, understanding the full effects of these policies has large-scale implications.
Study Resources
The following resources are for public use in presentations, papers, lectures, and more under the Creative Commons license BY-ND. Click the images below to view or download individual images, or use the button to download all.
As a condition of use, please cite as: Brown, Christina and Tahir Andrabi. 2022. “Inducing Positive Sorting through Performance Pay: Experimental Evidence from Pakistan Schools”, Working Paper.
Distribution of teacher value-added for teachers who chose performance versus flat pay
Effect of performance pay treatment on test scores by teachers' contract preference
Effect of performance pay treatment on teaching strategies by teachers' contract preference
Performance Pay and Sorting Presentation