Poster Paper: Do Peer Effects Influence Household Decision Making? Evidence from Child Food Intake in India

Friday, November 7, 2014
Ballroom B (Convention Center)

*Names in bold indicate Presenter

Eeshani Kandpal, World Bank and Kathy Baylis, University of Illinois, Urbana-Champaign
Peer effects in social norm-driven behaviors may go beyond informational spillovers to directly influence outcomes. We use primary data from Uttarakhand, India to quantify the impacts of peer effects on household decision making. Using participation in a female education program, Mahila Samakhya, to identify changes in empowerment, we examine how friends’ program participation affects (1) a woman’s physical mobility, (2) access to outside employment, (3) likelihood of working outside the household, and (4) her children’s food intake. We then use an extension of a spatial weighting technique that relies on friends of friends to identify peer effects on bargaining power and child food intake, while also instrumenting for endogenous group formation and program participation. Results show that peer effects have a significant effect on all proxies of female bargaining power, including children’s food intake. We find that Mahila Samakhya participants whose friends also participate are more likely to (1) leave the house without permission, (2) have access to the National Rural Employment Guarantee Scheme (NREGS), and (3) work for pay outside the household. At the same time, we find that non-participants in Mahila Samakhya are less likely to have access to NREGS although they are more likely to work outside the household, suggesting an occupational switching mechanism. Examining children’s food intake, we find an overall positive effect of the mother’s friends’ empowerment levels on her children’s food intake. In addition, we find that girls whose mothers have more empowered friends eat a better quality diet relative to the boys in the same household. This last finding corresponds to the finding in the literature that empowered women invest more in their daughters than in sons. Finally, combining the Nash bargaining framework with the demographic diffusion literature and identity economics, we define and provide suggestive empirical evidence on three ways in which networks function: (1) information, (2) influence, and (3) identity. While this analysis refrains from making welfare conclusions, our results highlight the presence of significant and complex peer effects in household decision making.