Social Interactions and Schooling Decisions
利用墨西哥PROGRESA随机干预项目，研究儿童的教育选择是否受同伴影响。结果显示，有资格获得补贴的儿童上学更频繁，而无资格儿童也因同伴效应增加了教育投入，社会互动效应几乎与直接补贴效应同等重要。
The aim of this paper is to study whether a child's schooling choices are affected by the schooling choices of other children. Identification is based on a randomized targeted intervention that grants a cash subsidy conditional on school attendance to a subgroup of eligible children within small rural villages in Mexico (PROGRESA). This policy change spills over to ineligible children if social interactions are relevant. Results indicate that the eligible children tend to attend school more frequently, and the ineligible children acquire more schooling when the subsidy is introduced in their local village. Moreover, the overall effect of PROGRESA on eligible children is the sum of a direct effect due to cash transfers and an indirect effect due to changes in peer group schooling. Interestingly, the social interactions effect is almost as important as the direct effect. Copyright by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.