Full-time schools, policy-induced school switching, and academic performance

While many developing countries are extending their school day from half- to full-time, little is known about the effects of such programs on student outcomes. Mexico recently extended its school-day duration from 4.5 h to 8 h through a full-time schools program. Using administrative student-level panel data, I exploit variation in the rollout of this program to identify its effect on math and language test scores. Controlling for endogenous switching between schools, I find that full-time schooling improves students’ math and language test scores, respectively, by 2.4 and 1.5 percent of a standard deviation in the first year of implementation. These results persist, growing to 13.7 and 10.8 percent of a standard deviation after five years. Moreover, the main results indicate that failure to control for endogenous school switching inflates the effects of five years of exposure to full-time schooling by a factor of 1.2 on math test scores and by a factor of 1.6 on language test scores.