My research focuses on two themes: (1) how relations between ethnic groups and the roles of ethnic minorities shape political attitudes and long-run economic development; and (2) how religion (broadly defined), through the beliefs they transmit and the social networks they create, shape economic outcomes and cultural norms.
Job Market Paper
Irreplaceable Middlemen: Evidence from the Expulsion of Indians from Uganda
Abstract: Middleman minorities often dominate intermediation between producers and markets, but can their economic function be replaced when the group is removed? I study the 1972 expulsion of Uganda's Indian minority, which comprised less than one percent of the population but controlled the trade linking farmers to markets. Using digitized census and road-network data, I measure exposure to the loss of Indian middlemen and exploit quasi-random variation from delays in a road-upgrading program. Decades later, more-exposed households obtain less crop value from their farmland. Satellite data show a lasting reduction in agricultural production that was visible already from 1981. Effects are largest for the most marketed crops, which depend most on intermediation services, and these households market less output and sell less to traders. Altogether, the evidence points to persistent disintermediation and the non-replacement of middleman-minority trading networks.
Presented at In Time and Space Workshop (York), ERINN Workshop (Namur), Building for the Future (Louvain-la-Neuve), Minority in Economics Mentoring Workshop (Bristol), Ifo Dresden Workshop on Regional Economics (Dresden), Australasian Public Choice Conference (Sydney)*, Annual Conference of the Australian Society for Quantitative Political Science (Melbourne)*, AASLE (Beijing)*
*scheduled
Working Papers
The Political Economy of Intergroup Contact: Evidence from Malaysia (with Gedeon Lim, Danial Shariat, Abu Siddique, and Shunsuke Tsuda)
Abstract: Are there particular social structures that allow ethnic diversity to coexist with political stability and economic development? This paper examines the long-run effects of interethnic proximity using quasi-random variation from a colonial-era program that forcibly relocated over 500,000 ethnic minority Chinese into mono-ethnic villages across Malaysia. Ethnic majority Malays residing closer to these villages exhibit lower electoral support for the ethnonationalist coalition, potentially reflecting a moderation of political identity. We observe moderately positive impacts on local economic development. Political effects are stronger in regions with initial, historical interethnic complementarities--even without persistent economic prosperity. Malays report greater contact with Chinese, higher interethnic trust, and weaker zero-sum beliefs. Effects are stronger (reversed) in areas with interethnic complementarities (competition). Throughout, effects on social integration remain muted. These findings suggest that the nature of the underlying economic relationship can have a persistently important role in shaping the long-run effects of contact.
Media: VoxDev
Presented at BÆM Development Workshop (Bristol)*, Development Economics Workshop (Manchester)*, Diversity and Inclusion Workshop (Milan-Bicocca)*, Essex*, SWET*, Migration and Development Conference (PSE)*, NEUDC*, Leicester*, Early Career Workshop QPE (KCL), Development Economics Workshop (Manchester)* (all 2025), UBC*, Tufts*, Brown*, PacDev (Stanford)*, UCSD*, UC Irvine* (all 2024), ASREC Australasia* (Monash), Korean Economic History Association* (Seoul), LKYSPP*, Warwick CAGE Summer School, Australasian Public Choice Conference (all 2023), Monash, NUS*, SMU*, Royal Holloway*, DWI Berlin* (all 2022), Hong Kong Empirical Microeconomics Workshop* (HKU-HKUST), Virtual Historical Political Economy Workshop* (2021), Boston University DRG* (2020)
*Presented by co-author
The Visibility Premium: Article Order and Citations (with ⓡ Francesco Lippi and ⓡ Sascha O. Becker) [R&R at Economic Inquiry]
Abstract: We exploit a natural experiment to identify the causal effect of the order in which articles appear within a journal issue on citations. Using data on 1,456 articles published in the Economic Journal between 2010 and 2024, we find that articles appearing earlier in an issue receive substantially more citations throughout the sample period. Articles ranked 7-9 receive 20% fewer two-year citations than those ranked 1-3, with effects growing to 28% over five years. Between December 2015 and February 2025, the journal switched from chronological to alphabetical ordering, allowing us to disentangle rank (visibility) effects from alphabetical (name) effects. Analysis of views and downloads confirms a visibility mechanism, with important implications for editorial practices.
Publication
Gender and Religion: A Survey (with Sascha O. Becker and Jeanet Sinding Bentzen) [Journal of Demographic Economics, forthcoming]
Abstract: This paper surveys the literature on gender differences in religiosity and on how religion shapes gender-related economic and social outcomes. Part I examines why women tend to be more religious than men, reviewing leading explanations from sociology, economics, and psychology. Part II analyzes how religion affects gender norms and attitudes, education, labor market participation, fertility, health, legal institutions, and discrimination. Across domains, we distinguish between effects driven by individual religiosity—such as beliefs and religious practice—and those driven by religious denomination. We emphasize studies that employ credible causal identification strategies, including natural experiments, instrumental variables, and policy reforms, while also reviewing correlational evidence for context. Overall, the literature suggests that religious teachings and participation often reinforce traditional gender roles, influencing women’s education, labor supply, and fertility decisions, though important heterogeneity and exceptions exist. We also highlight instances in which secular reforms or religious movements have altered these outcomes. The survey concludes by identifying gaps in the literature and outlining priorities for future empirical research.
Work-in-progress
Pray together, Prosper Together? Experimental Evidence on Church Networks in Ghana (with Anthony Amoah, Prachi Jain, and Amma Panin)
Abstract: Faith-based organizations are among the largest providers of social networks in many low- and middle-income countries, yet little is known about how much of their economic value comes from the social connections they create among members. In partnership with a large Pentecostal church in Ghana, we will randomly assign congregants to new social connections with fellow members. First, we will examine the impact on a range of economic outcomes, including business activity and revenue, informal transfers and risk-sharing, access to credit, employment, and savings. Second, we use features of the design to study what makes church ties economically valuable and for whom they matter most. The results will help explain why faith-based networks persist and grow amid economic development and what they offer that secular networks may not.
Status: Rolling out pilot survey soon
Confucian Elites and Peasant Rebellions (with Cheongyeon Won)
Abstract: Rulers rely on religious institutions to secure obedience, but because these institutions command their own authority and following, they can also organize dissent. We study whether religious institutions pacify and when they mobilize. In Joseon Korea, the state relied on the local gentry, the class of Confucian elites who governed rural society, to secure order in the countryside. The gentry was placed between the state and the peasantry. They could petition the court, negotiate leniency, and take responsibility on behalf of the community. We have compiled data on the locations of Confucian temples and the presence of the local gentry, and we are in the process of building a novel dataset on peasant unrest spanning several centuries. For identification, we exploit weather shocks that affected harvest failures to study whether the presence of the local gentry amplified or attenuated the effect of economic hardship on peasant rebellion. We will further explore the conditions under which the gentry encouraged or discouraged rebellion. Our findings will show that the effect of Confucianism on conflict depends not only on its doctrine, but also on its institutions, which can either secure obedience or give voice to dissent.
Status: Collecting data
Folklore and Gender Norms (with Joseph Gomes, Karine Moukaddem, and Amma Panin)
Inter-ethnic Proximity and Competition in Southeast Asia (with Gedeon Lim, Abu Siddique, and Shunsuke Tsuda)
Language, Nation-building, and Economic Development: Evidence from Indonesia (with Arya Gaduh and Gedeon Lim)