How to do Development Economics Research that is Valuable to Policymakers
Paper Session
Sunday, Jan. 5, 2025 10:15 AM - 12:15 PM (PST)
- Chair: Elie Hassenfeld, Givewell
How Does Research Affect Philanthropy?
Abstract
GiveWell directs hundreds of millions of dollars per year to global health interventions, much of which is based on research by development and health economists. It is also increasingly funding new research to help address key uncertainties in its cost-effectiveness models. In this panel, we explain how GiveWell thinks about the value of information from new research, outline how recent research grants have influenced its subsequent funding decisions, and highlight areas where we would like to see more academic research.Learning How to Deliver More Impact While Using Fewer Resources
Abstract
The cost side is currently often a missing piece in development economics policy decisions. We illustrate how to make progress in this space with examples from recent work at CEGA. In particular, we integrate cost research with two randomized evaluations of large-scale $50-75 M, multi-year development interventions in Malawi and Zimbabwe. Transaction-level expenditure data are used to separately cost-out 10-12 sub-activities offered as a full package of services to eligible beneficiaries. We explore the cost structure of the intervention, asking: what is the cost to deliver one additional sub-activity – through the consortium or a sub-implementing partner – and pinpoint the amount of expenditure that directly reaches beneficiaries, relative to the cost of activity delivery. Where feasible, we combine cost per beneficiary with estimates of treatment impact from the RCT to measure the incremental cost-effectiveness of treatment on gains in household food security, consumption, income, assets, and well-being.Targeting Women with Work: Experimental Evidence from Four Countries
Abstract
Understanding how to promote women’s autonomy is a major priority for public policymakers in low and middle income countries but there is limited evidence on how to do so. Can women’s participation in public works affect their agency and wellbeing in the household? We randomize participation in and participant gender for World Food Programme cash-for-work programs that target poor households in four countries (El Salvador, Haiti, Kenya, and Rwanda). We use this variation to estimate how program impacts for women differ when they are targeted for work compared to men. When men are assigned to work, we see a retrenchment in women's agency over consumption, but a reduction in reported incidences of IPV and an improvement in women's psycho-social wellbeing. This retrenchment in agency persists, as do women's reported improvements in wellbeing, a few months after the program concludes. Assigning the women to work induces a substantial shift in women's labor force participation. This reverses the retrenchment in women's agency that we observe when men are targeted. However, targeting women increases instances of backlash through an uptick in regressive attitudes and IPV, translating into slight reductions in women's psycho-social wellbeing. Women's agency remains stronger a few months after the program ends, and men become more accepting as instances of backlash through regressive attitudes and IPV dissipate. Heterogeneity analysis suggests that the positive impacts of targeting women for cash for work programs are concentrated among women with relatively low agency at baseline, who would otherwise have been less likely to participate in the labor market. These results highlight the nuances of targeting different members of the household and suggest that programs need to be sensitive to intra-household dynamics when they do.Discussant(s)
Constantine Manda
,
University of California-Irvine
Diana Moreira
,
University of California-Davis
Julie Faller
,
Givewell
Edward Miguel
,
University of California-Berkeley
JEL Classifications
- O0 - General
- H0 - General