Does private school competition improve country-level student achievement? A cross-country analysis using PISA 2012.

Abstract

This paper investigates the relationship between private competition and student achievement and whether countries with higher shares of privately-managed schools have better standardized test scores. It uses the cross-country OLS and instrumental variable estimation approaches and a covariate-rich dataset, PISA 2012, which measures math and reading literacy of 15-year-olds in more than 30 OECD and non-OECD countries. The privately-managed school share coefficient is not statistically significant at any conventional level after controlling for large subsets of student and family characteristics, school inputs, and country-level controls. These null results suggest the earlier positive cross-country evidence should be interpreted with caution and further research recommendations are briefly discussed.