Public Schools Can Improve Student Outcomes: Evidence from a Natural Experiment in India

I exploit a natural experiment in education policy in India to examine the effects of creating high-quality public schools. The "model" schools program established schools that admit students through an entrance exam. I estimate the effect of model schools on educational outcomes using a fuzzy Regression Discontinuity Design based upon the entrance exam cutoffs. With a data set of over 63,000 students, I consider three dimensions: (i) academic achievement; (ii) educational attainment; and (iii) career choice. For academic achievement outcomes, attending a model school increases test scores in math by 0.38 standard deviations, in science by 0.26 sd, and in social science by 0.26 sd on average. Attending a model school also increases the probability of obtaining an A in tenth-grade by 20 percentage points. For educational attainment indicators, model schools increase the probability of joining pre-university by 11.5 percentage points. However, attending a model school has no effect on the choice of major in pre-university college. Furthermore, I estimate multiple local average treatment effects and find that model schools have a similar positive effect for students across the ability distribution. Lastly, the per-pupil expenditure in model schools is comparable to that of traditional public schools. Overall, this paper provides suggestive evidence that the quality of public schools can be raised but other barriers persist.