Research

More than one in four women worldwide report having experienced physical or sexual intimate partner violence (IPV), yet few policy interventions have proven effective and those that have all operated at the community level rather than targeting households at risk. However, it is unclear whether community-wide treatment is a necessary component of success, and, if so, what is the mechanism through which engaging the larger community promotes behavior change among those at risk. Models of social norms enforcement suggest a potential mechanism, but rigorous evidence on their role in reducing IPV is scarce. To test whether social norms enforcement can reduce the incidence of IPV, we evaluate Peru’s flagship prevention program using a field experiment randomized across 250 communities. High-risk households are identified at baseline and offered 8–12 interactive sessions led by community health workers, combining prevention curricula with telenovela-style edutainment. In a randomly chosen half of communities, 30% of couples at low risk of violence are also invited to participate. Consistent with previous findings, the program substantially reduced IPV reported by high-risk women, with effects concentrated among the 50% of active participants. In addition, effects on at-risk women are significantly larger in communities that also deliver the program to low-risk households. The expanded program achieves greater reductions in IPV by generating program effects among at-risk households that are most difficult to reach: in villages that targeted households at risk, IPV falls exclusively among program participants, while in villages that also engaged non-violent households, IPV also falls among high-risk non-participants even absent direct exposure to the program. Moreover, the benefits of the expanded program are observed even when delivered to couples at home rather than in a group setting in which participants can interact. Overall, these patterns indicate that engaging non-violent households activates community-level norms enforcement, providing novel evidence that collective norms activation is central to extending the reach and efficacy of IPV prevention efforts.

In many settings, gender norms constrain women’s agency and discourage them from pursuing fruitful economic opportunities. To study this issue, we run a randomized control trial in 33 solar-powered and 33 unelectrified villages in rural Myanmar. We test a psychological intervention that combines an edutainment component featuring role models with a visualization, goal-setting, and planning workshop. The treatment aims to increase women's empowerment, encourage higher aspirations, and provide participants with a blueprint to pursue them effectively. To test whether the effects of the intervention vary by the level of energy access, we stratified the village-level randomization by region and electrification status. The intervention increased women’s business and financial empowerment, including by raising the share of women who had their own savings, banking, and mobile money account, with larger effects for less empowered women at baseline. Effects on aspirations are concentrated in unelectrified villages, where we find positive impacts on a novel measure of aspirations for productive appliances and on aspirations for children's education. Consistent with the shift in aspirations, we also see a parallel increase in education expenditures, the time women spend on childcare, and the probability that households took out a business loan in these villages. Our findings show that interventions targeting aspirations interact with the local economic environment and are most effective when they provide relevant role models for pathways such as education, which are particularly valuable where other investment opportunities, including productive uses of electricity, are constrained.

Presented at: 

2026: IV Workshop on Development Economics in Spain 

2025: AERE @ the Western Economic Association International (WEAI) Annual Conference, Environment for Development (EfD) Annual Meeting, International Population, Health and Development @ Duke University, International Conference on Development and Environmental Economics: SANDEE@25, NOVAFRICA Seminar @ Nova School of Business and Economics

2024: Sustainable Energy Transitions Initiative (SETI) Annual Workshop, Southeastern Workshop on Energy & Environmental Economics and Policy (SWEEEP), Southern Economic Association (SEA) Conference

Coverage:

Economics That Really Matters, The Economic Misfit

The formation of aspirations depends on both social factors and personal circumstances and experiences. Socially determined aspirations create behavioral incentives to improve one's standing but can also have a frustrating effect, motivating the hypothesis of a non-monotonic relationship between aspirations and forward-looking investments. Moreover, marginalized groups facing societal constraints on the pursuit of their life ambitions, such as women, may manage their aspirations by adjusting them to a level they can satisfy. In this paper, we draw on primary surveys with women in rural Myanmar to study these issues. First, we document positive associations between aspirations, wealth, and several measures of forward-looking expenditures, including health, education, and energy. Underlying these positive associations, we find evidence of a non-monotonic relationship between aspirations and investment that is consistent with aspirations failure at both the lower and higher ends of the distribution. Second, we provide novel evidence on the complex relationship between women's empowerment and aspirations. Specifically, we show that aspirations are an S-shape function of empowerment: aspirations are largely unresponsive to empowerment in the lower half of the distribution but, as empowerment grows, the relationship turns steeply positive. Overall, we provide a new perspective on how empowerment may constrain women's long-run outcomes.

Climate change threatens rural livelihoods, yet evidence on the effectiveness of resilience investments and households’ valuation of improved recovery remains limited. We examine these questions among farming and fishing households in Kenya using household panel data and a contingent valuation experiment. First, we apply the Ability to Recover (ATR) framework and estimate a difference-in-differences model with household fixed effects to assess changes in resilience associated with access to two market-based interventions that leverage solar energy to provide irrigation and cold-chain services, respectively. Estimates are consistently positive across alternative resilience measures but generally imprecise, providing suggestive rather than definitive evidence of improved recovery. Second, we elicit willingness to pay for a hypothetical resilience-enhancing intervention offering either a 30% or an 80% improvement in recovery from climate-related disruptions at randomly assigned monthly prices. Parametric estimates imply average monthly willingness to pay of KES 1,324 in farming communities and KES 1,785 in fishing communities, equivalent to 4.4% and 16.6% of median monthly household income, respectively. Willingness to pay increases significantly with the magnitude of recovery benefits and is consistently higher in fishing communities across parametric, Turnbull, and Kriström estimators. Among farmers, valuations are higher in counties that experienced wetter conditions than in those that experienced warmer and drier conditions. These findings demonstrate that households value improvements in recovery and distinguish between benefit levels, while highlighting the need for stronger evidence on intervention effectiveness and financing arrangements that account for households’ capacity to pay.

This study uses a difference-in-differences design on a matched sample to evaluate the impacts of solar-powered irrigation provided by SunCulture (a Kenyan social enterprise providing small-scale farmers with solar-powered irrigation solutions through an affordable "pay-as-you-go" credit model). We study how access to solar-powered irrigation influences various aspects of smallholder farmers' daily livelihoods, including financial and economic outcomes, food security, farm productivity, social capital, household wealth, and overall resilience. An important component of our evaluation focuses on assessing the resilience of participating households, particularly their capacity to recover from economic and environmental shocks. This includes measuring how households perceive and respond to risk, and how these perceptions relate to their broader adaptive strategies.

This study aims to assess the impacts of a cold chain intervention by Keep IT Cool (KIC), a local company offering cold chain services to fisherfolk in Kenya, that connect suppliers at beach landing sites to urban consumers. It uses data collected through a panel survey of fisherfolks operating in five Beach Management Units (BMUs) - which are a type of fishing cooperative - around Lake Turkana. Baseline data were collected in 2023, as KIC was first establishing operations in the area, and endline (2024), at which point KIC had established relationships with a subset of the surveyed BMUs. It combines both panel and cross-sectional observations and employs a difference-in-differences strategy for the identification of the intervention's causal impacts on fishing outcomes, income, resilience, and food security. We find that working with KIC reduced fish spoilage and losses for fisherfolk in partner BMUs, which contributed to enhancing food and income security. The intervention also contributed to economic empowerment, increasing the ability of fisherfolk to purchase their own fishing equipment. This financial autonomy was, in turn, associated with increased commitment to fishing. Finally, though catch did not significantly change, beneficiary fishermen appeared more capable of investing in building alternative sources of income, thus diversifying their livelihoods. These positive results notwithstanding, it is evident that there is a need to strengthen the adaptive capacity of local communities in the face of the adverse impacts of climate change, including their ability to cope with, absorb, and respond to shocks. Future interventions could focus on supporting similar local businesses and small and medium enterprises, particularly by providing training on the adoption and use of innovative technological solutions.

The upcoming decades will be defined by how societies are able to navigate the structural transformations associated with the deployment of automation-related technologies and the energy transition required to effectively fight climate change. Potential increases in unemployment will be a recurrent feature of such economic transformations. Retraining programs are a policy tool that can be used to mitigate negative effects on affected workers. Despite evidence that retraining programs can lead to higher earnings and better employment outcomes for the unemployed, particularly in the medium and long term, the history of retraining programs has been characterized by low uptake rates. This paper reviews the literature on displacement and job training programs and explores two bottlenecks to increasing workers’ uptake of training opportunities: (i) information and job expectations and (ii) locus of control and self-efficacy.

Can reproductive health interventions reduce adolescent childbearing, and can conventional surveys detect their effects? We study a cluster-randomized trial in 120 secondary schools in rural Ayacucho, Peru, comparing a mobile clinic providing private reproductive health counseling, the clinic combined with teacher-led sex education, and a control group. Linking students to administrative birth and school enrollment records, we find that the mobile clinic reduces births occurring 9–23 months after intervention exposure by 9.4 percentage points among students aged 18–19 and reduces school dropout by 19.5 percentage points in this group. Teacher-led sex education provides no detectable additional benefit. Yet self-reported outcomes fail to detect corresponding reductions in childbearing. We document two overlapping measurement problems: women not reached at endline have administrative birth rates more than twice those of respondents, and approximately one-third of respondents with confirmed births report never having been pregnant, while nearly half deny ever having had sex. Building on these findings, we are also piloting a randomized survey experiment to test whether financial support conditional on verified motherhood increases truthful disclosure. Administrative records provide an independent benchmark for distinguishing changes in reporting from changes in underlying behavior. Together, the project evaluates a reproductive health intervention and investigates how selective survey participation and concealment of stigmatized outcomes can obscure program impacts, while testing an approach to improve their measurement.

In this study we will (i) explore deviations from the standard timeframes employed by the DHS to measure IPV by asking survey items for different timeframes and experimentally varying the order in which we ask each set and (ii) test two novel methods for the measurement of IPV that address the privacy and broader psychological limitations of DHS questionnaires with the goal of generating actionable insights for improving the measurement of IPV. 

In this study we will evaluate an enhanced marketing intervention to promote the uptake of solar irrigation technologies in rural Kenya. Specifically the study will target farming household that do not currently have access to the technology through a traditional door-to-door marketing approach (control condition) and a marketing "plus" approach (treatment condition). The treatment will will include the following additional elements: (i) use of an edutainment style component combining farmer testimonials with a digestible summary of the results of a prior evaluation, and (ii) the invitation to attend "rainmaker" parties with local model farmers.

This paper uses data from a baseline survey of a randomized controlled trial, several experimental games played between spouses, and a Vickrey auction to explore women's bargaining power in rural Myanmar and its relation to survey measures to woman empowerment, energy-related investments and consumption, and willingness to pay for a selected appliance. To do so, we construct two alternative indices of woman empowerment - using Anderson (2008)'s approach on a selected set of agency sub-indices and Principal Component Analysis on an extension of this set. Finally, we also conduct an exercise where we re-create the PCA approach on solar mini-grid villages and unelectrified villages to explore whether the structure of empowerment itself differs according to the level of energy access.

Pastoralism is an important economic activity in Mongolia. Steppe rangelands have been decreasing in productivity, in part due to increases in livestock heads, and due to an increase in the frequency and severity of both drier summers and harsher winters (dzuds). Consequently, fodder availability, especially during the winters, is increasingly becoming a challenge. Livestock heads that survive the winters are particularly emaciated, thus increasing grazing pressure on grasslands during the summer months and consequently reducing grassland productivity and its regeneration capacity. Mongolia's capacity to improve the productivity of its grasslands and increase forest area relies on meeting fodder requirements through alternative methods. This study will use the Household Socioeconomic Survey 2023, which is a nationally representative survey of households in Mongolia, to understand the scope for meeting fodder requirements by supporting agropastoralism. 

Electricity access is key for growth and poverty alleviation. Despite being a longstanding priority on the development agenda, about 1 in 10 people worldwide, most of which in the rural Global South, still lack access. Off-grid mini-grids are a potentially scalable strategy to expand electricity access in remote areas. However, financial sustainability for developers is a serious challenge to scaling mini-grid solutions and realizing their development potential. Ex ante demand estimates are often inaccurate, and over-optimistic, and lead to mini-grid sizing that is inconsistent with local needs, weakening their economic performance. The same low demand that raises mini-grids’ appeal when compared to the economics of grid expansion is also an obstacle to their sustainability. Sites are characterized by low incomes and electricity demand, leading to low revenues that make it difficult to cover operational costs. The study aims to evaluate COMET - a customer engagement workshop and simulation platform for optimizing mini-grid design via better demand estimates, increasing demand, and improving financial sustainability at the project and sectoral level. The study would be located in Uganda and is the result of a partnership that includes COMET, the solar developer Equatorial Power, and Environment for Development - Uganda (EfD-Uganda).