Peer-reviewed journal articles
[1] “Spillover Effects and Diffusion of Savings Groups.” 2024. World Development. Volume 173, 106377
Joint with Rossa O’Keeffe-O’Donovan
We estimate the spatial spillover effects of a program to promote savings groups in Uganda and Malawi. We use geolocation data from a cluster-randomized experiment to compare the outcomes of households in areas where a high proportion of nearby villages were randomly assigned to treatment to those in areas where a low proportion of nearby villages were assigned to treatment. We have four key results. First, take-up of the program is highest in villages where promotion by NGO-employed field officers was concentrated — promotion by peers was about half as effective, and ‘organic’ replication was rare. Second, we find evidence of large positive between-village spillover effects on total income, food security, and business outcomes in control villages. Third, we estimate positive direct effects of the program on business outcomes and financial inclusion — these estimates are qualitatively similar to a previous evaluation of the program (Karlan et al., 2017) which did not account for spillovers, though our estimated impacts on business outcomes are larger. Finally, we estimate that 23%–28% of the overall benefits of the program are not captured by an analysis which does not account for spatial spillover effects.
Working papers (* = Job Market Paper)
*[1] “Labour Market Effects of India’s Termination from the United States’ GSP.” 2025. IDS Working Paper 615, Brighton: Institute of Development Studies, DOI: 10.19088/IDS.2025.008. Featured in the Financial Times and Sky News
Joint with Amrita Saha
This paper investigates the sudden 2019 termination by the United States Generalized System of Preferences (GSP) of trade with India as a quasi-natural experiment. Given the re-election of Donald Trump and global concerns about the future of global trade, understanding the distributional consequences from past trade shocks can inform strategies to balance trade preferences with social equity. The impact of the GSP termination is examined with a triple difference-in-differences specification that controls for both country- and product-level export changes to estimate the impact of the termination on India’s exports to the US deriving a shock index by sector. We then use the shock index as a treatment variable in a second model, utilising a difference-in-differences approach to measure the effect of the shock on monthly individual-level worker incomes and employment. Our findings suggest that the GSP termination had a considerable impact on Indian exports to the US: on average, exports of affected products dropped by 2 per cent in 2019, with exports of transportation manufacturing declining more than 20 per cent. These reduced trade flows from India to the US had knock-on economic implications for workers in India that we argue were likely driven by a strong countercyclical response from large firms that resorted to occupational restructuring, which saw companies prioritising professionals at the expense of day labourers. These findings have important implications for economic inequality as earners in the top income quintiles arguably capture the largest income gains and demonstrate the greatest employment resilience in the immediate aftermath of the termination.
[2] “TV, TEDx, and Tweets: Measuring the Impacts of a Multi-pronged Edutainment Program in Kyrgyz Republic and Tajikistan.” 2025. World Bank Policy Research Working Paper No. 11088.
Joint with Ben Fowler, Marrium Khan, Renuka Pai, and Sakshi Varma
The evidence so far is mixed as to whether educational entertainment (or “edutainment”) can create sustainable changes in financial attitudes and behaviours, and there are few studies that test such hypotheses in Central Asia. This paper utilizes a statistical matching methodology to measure the impact of three edutainment interventions in Kyrgyz Republic and Tajikistan: A television series, a TEDx-style talk, and an interactive social media video. This was coupled with an encouragement design where we randomly select 2,187 respondents from 14 cities across the two countries to participate in the RCT. Respondents were sent text messages encouraging them to view the three edutainment interventions, disseminated nationally in the respective countries. Using a midline survey conducted a few weeks after concluding the campaigns in both countries, and an endline survey three months later, we are able to measure immediate effects of the campaigns as well as those that last into the medium term. We find that campaign consumption spurred changes in financial attitudes and behaviours. Treated beneficiaries were more likely to save money in formal accounts, more likely to open a new financial account, and more likely to transact on these accounts. These impacts remained present through endline. We do not find effects on personal beliefs consistent with existing research and with the theory of change developed during the pre-analysis phase. While the program had significant effects on older women’s awareness of societal-driven, finance-related social norms, these impacts were not conferred to youth in our sample and were not present for any demographic at endline. Our study suggests that edutainment campaigns can be used to create lasting effects on financial behaviors, but more research is needed to establish the chain of psychosocial beliefs that drive this change in behavior.
Peer-reviewed book chapters
[1] “From De-Industrialization to Job Creation: New perspectives on African Growth.” 2024. In: “New Pathways to Job Creation and Transformation in Africa: The Promise of Industries Without Smokestacks.” Rowman & Littlefield. [read, order]
Joint with Richard Newfarmer and John Page
This chapter utilizes an analytical framework emerging from the literature on structural transformation and that synthesizes data from all eight case studies that appear in the book. The opening section reviews recent literature on the roles of manufacturing, agriculture and services in contributing to unconditional convergence in labor productivity growth rates between low- and high-income countries. It highlights activities that have the potential to drive income convergence and analyzes ways Africa’s pattern of transformation differs from other regions. The second section drills down on the many activities within sectors in Africa beyond manufacturing that hold the promise of higher productivity, and then sketches the role of these IWSS activities in a comparative analytic framework for the eight countries in this study. It reviews their growth, employment, and export performance. A third section elaborates on the role of IWSS activities for productivity growth. The analysis underscores that productivity gains within IWSS sectors are an essential part of Africa’s transformation.
[2] “Policies to Create Productive Jobs in Africa: Unleashing Industries Without Smokestacks.” 2024. In: “New Pathways to Job Creation and Transformation in Africa: The Promise of Industries Without Smokestacks.” Rowman & Littlefield. [read, order]
Joint with Richard Newfarmer and John Page
Engaging the full power of Africa’s young labor force in the process of development is a major challenge across the continent. As this book has shown, employing Africa’s young labor force requires integrating them into both traditional growth-driving sectors like manufacturing as well as agro-processing, horticulture and export crops, ICT, business and financial services and other industries without smokestacks (IWSS). The case studies describe in detail how IWSS sectors are at the center of efforts across the continent to accelerate growth, increase trade, and create new jobs. Policy has an important part to play in shepherding Africa’s economic transformation. This chapter lays out policies that emerge from the case studies to create jobs in the transformation process and in IWSS sectors in particular. To set the stage, the chapter begins with a brief summary of key trends emerging from the case studies and then a look ahead at prospective job creation over the next 10-15 years. It then considers policies on both the demand side of the labor market (those spurring overall growth) and on the supply side of the labor market (those developing skills). It also advances some specific policies to promote IWSS. It then concludes by pointing to ways the international community can support Africa’s transformation.
Joint with Gracelin Baskaran, Mma Amara Ekeruche, Aloysius Ordu, and Lemma W. Senbet
Amidst rapidly escalating climate crises, there is an urgent need to enable African countries to remodel existing financial infrastructure with the aim of strengthening climate resilience and developing green infrastructure. As the world scrambles towards decarbonisation, the G20 nations, which represent the largest source of wealth, with 85 percent of global GDP, are well equipped to support vulnerable countries. This Policy Brief proposes three mechanisms whereby the G20 can support African countries: providing grant funding and technical assistance to the Programme for Infrastructure Development for Africa (PIDA) to increase the number of high-quality bankable projects and mobilise financiers; strengthening the coordination of climate financing from the G20 countries to the continent; and unlocking financial technology and entrepreneurship to mobilise financing for bankable projects. This Brief further recommends that the G20 should use its technical capacity, financial muscle, and convening power to put African countries on the path towards climate resilience.
Work in progress
This paper applies an empirical model derived from neoclassical growth theory to a newly created 12-sector database to estimate unconditional convergence in these sectors over the last five decades. This paper offers four key findings. First, using a new dataset, it affirms the literature’s recent harmony of findings that economy-wide unconditional convergence is present and has been for several decades. Second, many sectors display unconditional convergence, especially highly-tradeable services sectors like business (including ICT), finance and real estate. Interestingly, the last few decades have witnessed unconditional convergence in the aggregate agricultural sector but not in aggregate manufacturing. Third, middle-income economies from East Asia and Pacific and South Asia have served as the pivotal accelerators of unconditional convergence in recent decades, while middle-income economies from Africa converge more slowly than the global trend. Though resource-rich middle-income economies showed weaker unconditional convergence in the 80s and 90s than other middle-income countries, they appear to have narrowed the gap somewhat since 2000. Despite heterogeneity in convergence behavior, all middle-income country groupings studied show clear evidence of unconditional convergence since the 1990s. Fourth, while unconditional convergence is alive, it is not well: this paper estimates that it will take nearly 85 years for poor countries to halve the productivity gap with rich countries. The unconditional convergence half-life is smallest in sectors like business, finance, and real estate, where the gap is on track to be halved in just a few decades.
Joint with Lorraine MacKenzie
In 1986, an explosion occurred at the Chernobyl Nuclear Power Plant (NPP), sending an unprecedented wave of radiation across Europe---and a wave of trepidation toward other nuclear power plants across the world. We exploit this plausibly exogenous event, using gridded population data and nighttime lights data to understand how the Chernobyl disaster impacted labor productivity growth in areas close to the Chernobyl NPP (the direct effect) and areas close to non-Chernobyl NPPs (the perception effect). We have three key results. First, we find that areas closer to a non-Chernobyl NPP saw less growth in labour productivity; in other words, Chernobyl had a perception shock on labor productivity growth inversely related to distance from the nearest NPP. We demonstrate that this result is driven both by a negative effect on economic growth as well as a positive effect on population growth in cells close to non-Chernobyl NPPs. Second, in areas within 25-75km of Chernobyl, we find a positive direct effect, even after accounting for the perception effect, on labour productivity growth, driven primarily by large increase in economic activity, but also by out-migration. Finally, both the direct and perception effect diminish in size as the distance to the NPP approaches 150km. Research has long recognized, but rarely quantified, the impacts of the Chernobyl disaster, and the results of this paper provide insights into broader development challenges faced by former planned economies as well as developing and developed countries across the world, where the spectre of Chernobyl continues to shape economic trajectories today.