What is Good for the Goose is Good for the Gander: The Effects of Child Care Provision in Mexico

In 2007, seeking to increase female labor force participation and more generally ease burdens on working women, the Mexican government introduced one of the most ambitious child care programs in an emerging economy: Estancias Infantiles para Apoyar a Madres Trabajadoras (EI). EI covers 90% of the cost of enrolling a child under the age of four at a formal child care center. The program is intended to benefit women who are looking for work, in school, or working -with the exception of those who already have access to child care because their job is covered by Mexico’s social security system (IMSS). The roll-out of EI was so aggressive that by 2010 it enrolled 340,000 participants, more than double that of the 25-year-old IMSS child care program. However, EI was also rolled-out unevenly, owing to idiosyncratic variation in how quickly local offices processed applications from child care centers. I exploit the variation in the program’s availability across time, across municipalities, and between eligible families and very similar ineligible ones. The essence of my approach is difference-in-difference-in-differences, but I extend this method in several ways –most importantly by adapting the Synthetic Control Method to my repeated cross-section data so that the ineligible comparison group is highly credible. Most of my results are intention-to-treat effects but I also show pseudo treatment-onthe-treated effects in which I assume that all the effects flow through families in which EI-eligible women significantly reduced how many hours they spent caring for their own children. I find that EI increased women’s probability of working and reduced the time they devoted to child rearing. EI caused women to obtain more stable jobs, and it increased their labor incomes. EI also had effects -probably unintendedon men. Affected husbands spent less time on child rearing and housework, and they were more likely to switch to a better-paid job. However, husbands of EI-eligible women who were initially unemployed were less likely to re-enter the labor force. ∗June 2012. I am really grateful to Caroline Hoxby, Giacomo De Giorgi, and Luigi Pistaferri, and participants in seminars for their helpful comments and suggestions. †Address: Encina Hall, 616 Serra St, Stanford, CA. 94305-6055. Email: [email protected].