WORKING PAPERS
Revisiting the Last Mile: The Development Effects of a Mass Electrification Program in Kenya (with D. Kassem, and E. Uzor), Accepted (December 2025), Energy Economics
3IE RIDIE Registration ID 1025
Access to electricity remains a central priority in development policy, particularly in Sub-Saharan Africa (SSA), which accounts for over half of the global population without reliable power. This study examines the socio-economic impacts of Kenya’s Last Mile Connectivity Project (LMCP), a large-scale electrification initiative aimed at expanding grid access by connecting households and businesses within 600 meters of distribution transformers. Using a spatial regression discontinuity design, we evaluate the program’s effectiveness in increasing electricity access and its broader development outcomes. Our findings show that the LMCP significantly improved household connectivity, leading to a substantial shift from off-grid energy sources to grid electricity. While this transition resulted in modest yet positive economic effects—most notably, increased household consumption—there was no significant impact on employment, income, or women's empowerment. These limited economic gains appear to stem from persistent challenges such as unreliable power supply and financial constraints that hinder productive electricity use. While our results highlight the role of electrification in improving living conditions, they also suggest that access alone may not be sufficient to drive broader economic transformation.
Climate Risk Perception and Adoption of Sustainable Agricultural Practices: Insights from Kenya's Central Highlands (with P. Rajkhowa, M. Buisson, M. Kinyua). December 2025 (Submitted)
Smallholder farmers are increasingly exposed to climate variability that threatens agricultural productivity and household livelihoods. Sustainable agricultural practices (SAPs) are widely recognized as a key pathway for sustaining yields, conserving natural resources, and enhancing climate resilience. Identifying the factors that shape adoption decisions is therefore critical for designing effective adaptation strategies. While existing literature has largely emphasized socio-economic determinants of technology adoption, the role of behavioral factors, particularly climate risk perceptions, remains comparatively underexplored. Using cross-sectional survey data from the Central Highlands of Kenya and a Poisson regression framework, this study examines how farmers’ climate risk perceptions are associated with the adoption of SAPs and identifies which types of practices are most responsive to climate concerns. The results reveal a non-linear relationship between climate risk perception and adoption, with significant increases in adoption occurring only at high levels of perceived climate concern. Importantly, exposure to climate shocks alone does not appear sufficient to induce behavioral change; rather, adoption responds to the perception of these shocks as serious threats. We further highlight heterogeneity in these relationships across gender of decision maker and plot size: adoption is more responsive to climate concerns on male-managed plots and among smaller farms. In addition, practices such as organic input use and soil and water conservation are more sensitive to climate perceptions than more resource-intensive practices, including irrigation, intercropping, and no-tillage. While not causal, the findings highlight the role of behavioral and perceptual factors in adoption and offer practical guidance for context-sensitive extension and outreach.
Beyond Irrigation: Valuing Multiple Water Uses and Services of Small Reservoirs in Northern Ghana (with M Buisson, S Appiah, G Asare). December 2025. (Submitted)
Small reservoirs play an important role in supporting rural livelihoods in northern Ghana by providing water for a range of domestic and productive activities. This paper examines how reservoir water is used, who benefits from access, and how engagement in reservoir water use affects household economic and welfare outcomes. Using household survey data collected in September–October 2023 from approximately 1,100 households around 72 small reservoirs in the Upper East Region, we document patterns of water use and analyze their relationship with income, food security, nutrition, and health. We find substantial heterogeneity in access and use. Female-headed households are less likely to engage in multiple uses, while households with leadership roles exhibit broader engagement. Proximity to reservoirs strongly shapes access, and wealthier households rely less on reservoir water for domestic purposes. Using an instrumental variables strategy based on distance to the reservoir, we find that engagement in reservoir water use generates positive income gains, driven primarily by productive activities such as irrigation. In contrast, effects on food security and nutrition are weaker and not robust. Domestic uses involving direct contact with reservoir water are associated with higher illness incidence, highlighting potential health externalities. Overall, the results emphasize the need for valuation and policy approaches that account for both economic benefits and risks associated with small reservoirs.
Discrimination Against Workers from Slums: A Case Study in Argentina (with W. Zanoni, P. Acevedo, and H. Hernandez), January 2025. (Submitted)
This research examines discrimination in the labor market faced by urban slum dwellers in Latin America, focusing on a case study in Buenos Aires, Argentina. Using an artifactual field experiment that simulates being an actual job, we hired human resource recruiters to evaluate job applicants online and found that slum dwellers were selected at a significantly lower rate than non-slum dwellers, and received lower wages. Specifically, slum dwellers were selected only 4.2 times out of every 10 applicants, received wages that were 1.3% lower than non-slum dwellers, and were consistently rated poorer in assessments of job fitness. Our study also tested an intervention to increase awareness of discrimination rates, which resulted in recruiters changing their behavior in favor of the slum dwellers. These findings shed light on the challenges faced by slum dwellers in the labor market and provide insights into potential interventions to address labor market discrimination in urban areas of Latin America and the Caribbean.
Group-based Reminders and Loan Repayment: Evidence from Maize Farmers in Kenya (with T. Harigaya, M. Kremer, and M. Thomas), June 2024. (draft available upon request)
There are several studies on the effect of individually targeted nudges on loan repayment, but none have analyzed the effect of group-based nudges, which have a theoretically ambiguous effect on repayment in a group liability setting. Group reminders can generate positive peer pressure or discouragement depending on their content and how message recipients use the information. Using a randomized controlled trial with over 360,000 smallholder farmers, we study the impact of individual and group text message reminders on on-time agricultural loan repayment. We find that individual reminders are cost-effective and increase the probability of on-time repayment by 2%. In contrast, group reminders have a small but significant adverse effect. Consistent with a discouragement effect, heterogeneity analysis shows that this effect was driven by farmers with low outstanding balances in underperforming groups at the time that group reminders were sent.
Sharing a Workforce: The Effect of Agricultural Productivity Shocks on Industrial Performance, May 2018
PUBLICATIONS
In peer-reviewed journals
Digital Information Provision and Behavior Change: Lessons from Six Experiments in East Africa, (with R. Fabregas, M. Kremer, M. Lowes, and R. On), American Economic Journal: Applied Economics, 2025.
Socio-Economic Valuation of Water Reallocation: Framework Development and Application in South Africa (with M. Buisson, B. van Koppen, P. Sithole, E. Mapedza, and J. Dini), Water Policy, 2025
The contribution of livestock to household livelihoods in Tanzania and Uganda. Measuring tradable and non-tradable livestock outputs, (with U. Pica-Ciamarra), Tropical Animal Health and Production, 2021.
Other
Poverty and Vulnerability in Urban Angola. (with L. Sousa ), Equitable Growth, Finance & Institutions Note. World Bank. 2024.
Enumerator bias in yield measurement: A comparison of harvest versus allometric measurement of coffee yields (with V. Hoffmann, M. Murphy, E. Rwakazooba, C. Angebault, and G. Kagezi), IFPRI Discussion Paper, February 2022.
Impact Evaluation of the Chimborazo Rural Development Program – Rural Roads Component (with L. Corral), Technical note, Inter-American Development Bank (IDB), January 2021.
Impact Evaluation of SU-L1009: Support to Improve the Sustainability of the Electrical Service (with L. Corral), Technical note, Inter-American Development Bank (IDB), November 2020. (Blog)
Chimborazo Rural Development Program: Irrigation Component Impact Evaluation (with L. Corral), Technical note, Inter-American Development Bank (IDB), June 2020.
Measuring coffee yields: a comparison of one-time harvest and allometric methods (with V. Hoffman, G. Kagezi, and E. Rwakazooba), Project note, International Food Policy Research Institute (IFPRI), December 2019.
Firm Capabilities and Economic Growth (with N. Bloom, G. Fischer, I. Rasul, A. Rodriguez-Clare, T. Suri, C. Udry, E. Verhoogen, and C. Woodruff), Evidence paper, International Growth Centre (IGC), December 2013.