Research
Research
Working Paper
Who Needs to Know? Intra-household Differences in Responses to Health Information (with T.V. Ninan)
Supported by the Weiss Fund
Presentations: Pacific Conference for Development Economics (PacDev) 2026, University of California, Davis; Northwest Development Workshop 2025, Reed College; Pacific Northwest Labor Day Workshop 2025, University of Washington.
Exposure to household air pollution from cooking with solid fuels such as charcoal and firewood is a major health risk in many low- and middle-income settings, particularly for women. This paper examines whether providing information about these risks changes beliefs and behaviors, and whether the identity of the information recipient within the household matters. We test this using a cluster randomized controlled trial with 2,000 households in rural India, where we inform either the (male) household head or the (female) primary cook about the actual health risks of cooking with solid fuels. Sixteen weeks after the intervention, we find that information provision increases primary cooks’ perception of risk by 0.21 standard deviations but not household heads’. Despite the increase in awareness, we find no average effect on fuel use. Heterogeneity analysis reveals two channels through which behavior does respond. First, primary cooks who have a say in LPG refill decisions reduce solid fuel use when informed, consistent with decision-making authority over the relevant expenditure being important for translating beliefs into action. Second, more educated primary cooks, who are not necessarily more empowered in our setting, both updated their beliefs more and reduced solid fuel use at the intensive margin. Together, these findings suggest that both low salience of information and limited say over fuel purchase decisions constrain the effect of the information intervention on household fuel choice, while education enables primary cooks to act on new information along margins that may not require explicit bargaining with the household head.
Work in Progress
Pilot complete (supported by PEDL)
This project documents differences in social norms, intra-household bargaining power, and business outcomes between women microentrepreneurs from two neighboring tribes — Bamoun and Bamileke — in the West Region of Cameroon, despite sharing the same markets and geographic proximity. Using primary survey data collected from 270 married women microentrepreneurs in rural Cameroon, we find that Bamoun women face significantly more restrictive gender norms than their Bamileke counterparts. Bamoun women are more likely to require spousal permission across a wide range of economic activities, from using a bank account and taking loans to working with others in the main market. These differences in autonomy are accompanied by differences in time spent on business and annual revenue. That differences in autonomy occur alongside differences in enterprise outcomes within a shared economic environment is consistent with a framework in which social norms shape intra-household bargaining and, in turn, business outcomes. Additionally, we document asymmetric network segregation along tribal lines: 57.7% of Bamoun women report zero interaction with Bamileke women in the previous week, and 23.1% report no interaction even with other Bamoun women. In contrast, only 15.7% of Bamileke women report no contact with Bamoun
women, and just 0.7% report no interaction with other Bamileke women. Despite this segregation, 85% of respondents express willingness to interact with women from the other tribe, suggesting that segregation is not driven by fixed preferences alone. Given that Bamileke women face fewer norm-based restrictions and maintain denser networks, these findings motivate a randomized intervention to test whether facilitated inter-tribe interactions can shift norms and improve women’s business outcomes.
Unintended Effects of a Multifaceted Intervention to Empower Young Women in India.
This paper evaluates India’s Rajiv Gandhi Scheme for the Empowerment of Adolescent Girls (RGSEAG or ‘SABLA’), a large-scale program that provided nutritional support, health interventions, life-skills training, and vocational education to girls aged 11–18 across 200 pilot districts in India between 2010 and 2015. Using a difference-in-differences strategy, with district and age fixed effects, I find that girls exposed to the program had 0.118 fewer years of completed schooling, 1.5% lower than those not exposed to the program. They also had 0.023 more children, 2.1% more than the control mean of 1.10. Both findings are consistent with an unintended perverse incentive: conditioning the nutritional supplement on out-of-school status for girls aged 11-14 may have induced families to withdraw in-school girls. Critically, SABLA had no detectable effect on any of its intended empowerment outcomes — women’s say in household decisions, mobility, labor force participation, and attitudes towards intimate partner violence.