Working Paper: NBER ID: w16021
Authors: Robert T. Jensen
Abstract: Gender differences in health and education are a concern for a number of developing countries. While standard theory predicts human capital should respond to market returns, social norms (e.g., disapproval of women working outside the home) may weaken or even sever this link for girls. Though many studies have examined the link between women's wages or labor force participation and investment in girls, two significant problems are the possibility of omitted variables bias and reverse causality, and difficulty in identifying which of several mechanisms (returns, bargaining power, income, etc.) link the two. To overcome these problems, we provided three years of recruiting services to help young women in randomly selected Indian villages get jobs in the business process outsourcing industry. Girls in treatment villages were more likely to be in school and had greater measured BMI. We argue that the design of the experiment (providing opportunities almost exclusively for young, unmarried women rather than current mothers) allows us to rule out that mechanisms other than increases in the returns explain our results.
Keywords: gender differences; human capital; employment opportunities; India; experimental evidence
JEL Codes: I00; J16; O12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Increased employment opportunities for women (J21) | Human capital investments in girls (J24) |
| BPO recruiting services (L84) | School enrollment for girls (I24) |
| BPO recruiting services (L84) | Health outcomes (BMI) for girls (I12) |
| Increased awareness of employment opportunities (J68) | Human capital investments in girls (J24) |
| BPO recruiting services (L84) | Perceived returns of investing in girls' education (I26) |