Research

This paper examines how technological change in production affects social cohesion in ethnically diverse societies. I study the early expansion of the electric grid in the United States between 1900 and 1940, when electrification transformed manufacturing and large-scale immigration reshaped the labor force. Using newly digitized maps of the U.S. high-voltage transmission network linked to full-count census data, I exploit the staggered rollout of electrification across counties to estimate its causal effects on the integration of immigrant and native workers. I find that electrification increased the share of immigrant workers in manufacturing and reduced ethnic segregation across occupations within industries. Electrified industries became more diverse and more integrated, especially among groups culturally distant from the native-born population. These effects extend beyond the workplace. Electrification is associated with lower residential segregation among manufacturing workers and a partial attenuation of the negative relationship between immigrant presence and local public service provision. Together, these results suggest that electrification fostered greater social cohesion in communities where manufacturing played a central role. Finally, I find weak evidence of effects on cultural assimilation. Intermarriage between immigrants and natives rose modestly in electrified counties, but naming patterns among immigrant families did not change systematically. Overall, I find that, in this context, technological change reshaped the social fabric by promoting integration both at work and within local communities.

Presentations:

ASREC Europe Conference 2025 (University of Copenhagen); ASSA Annual Meeting 2026 (Philadelphia); CAE Annual Meetings 2025 (Université du Québec à Montréal); CEPR-EBRD-KCL 7th Workshop on the Economics and Politics of Migration (Sapienza Università di Roma);  CESifo Venice Summer Institute 2026 – Workforce Adjustments to Technological Change; CLEF Conference 2026 (Université du Québec à Montréal); CNEH Conference 2025 (Université de Montréal);  Collegio Carlo Alberto (Turin); CUNEF (Madrid); EEA-ESEM Annual Meeting 2026 (University College Dublin); EHA Annual Meeting 2025 (Philadelphia, poster); EHS Annual Conference (London School of Economics); Hebrew University of Jerusalem; INCAE (Costa Rica); Max Planck Institute for Innovation and Competition (Munich); NBER Summer Institute 2025 – Development of the American Economy (Cambridge, poster); SIOE Annual Conference 2026 (INSEAD); UEA North American Meeting 2025 (Université du Québec à Montréal); Universidad Carlos III (Madrid); Univeristy of British Columbia internal seminars (Vancouver);  University of California, Berkeley (Economics); University of Maryland, Baltimore County; University of Southern Denmark (Odense); University of Vienna; Università Cattolica (Milan); Virtual Economic History Workshop 2025 (online).

This paper studies how frictions within ethnic groups shape immigrant assimilation. While most research focuses on interactions between immigrants and the native majority, we examine how successive waves of co-ethnic immigrants affect assimilation outcomes. We study this question using historical Jewish migration to the United States, where earlier German Jewish immigrants interacted with later arrivals from Eastern Europe who differed in socioeconomic and cultural characteristics. We develop a novel method to identify Jewish individuals in the full-count U.S. censuses between 1880 and 1930 using distinctively Jewish surnames. We then exploit two Jewish-led relocation programs that redistributed Eastern European Jewish immigrants from New York to smaller cities across the United States. Using variation in the timing and intensity of relocations, we implement an event-study and difference-in-differences design to estimate causal effects. We find that the arrival of newcomers permanently displaced incumbent Non-Eastern Jews. This increased the share of Eastern European Jews and reduced cultural assimilation, as reflected in more distinctively Jewish names and lower intermarriage between Eastern and Non-Eastern Jews. At the same time, locations receiving newcomers experienced substantial growth in religious, charitable, and educational institutions. These findings point to a multifaceted response to co-ethnic inflows and show that assimilation is shaped not only by majority–minority interactions, but also by dynamics across successive immigrant waves.

Presentations:

ASREC Conference 2026 (Chapman University); ASREC Europe Conference 2026 (University of Warwick); NBER Conference 2025 – Linking Historical Data Sources for Small Populations; University of British Columbia Dev/PE Lunch Seminar (Vancouver).

This paper studies how immigrant workers adjusted to the onset of the Great Depression using linked individual-level data from the 1920 and 1930 U.S. Censuses. I compare the response of first-generation immigrants to that of comparable U.S.-born individuals with foreign-born parents across multiple margins: geographic mobility, unemployment, and changes in industry, occupation, and occupational rank. I find that immigrants were more responsive to the shock. They were more likely to move across states, become unemployed, change jobs, and move to lower-ranked occupations. These patterns indicate that adjustment combined labor reallocation with economic downgrading. Responses vary strongly with time spent in the United States. Recent immigrants were substantially more mobile than long-term immigrants across most margins, while long-term immigrants were more attached to urban areas and less likely to relocate, increasing their exposure to unemployment during the downturn.

This project studies the long-run impact of historical conflict on social preferences in Europe. We geolocate nearly 2,800 battles between 1 and 1900 CE and manually classify them by actors involved, prevailing party, type of violence, motives, and intensity. Combining this information with individual-level measures from the Global Preference Survey, we investigate how variation in historical conflict exposure across regions is related to patience, risk-taking, prosociality, and trust today. We find that regions historically more exposed to conflict exhibit greater patience but lower prosociality and risk-taking, with the effects being strongest for intense conflicts, such as sieges, retaliatory battles, and territorial conquests. These results suggest that repeated exposure to conflict becomes embedded in local collective memory and continues to shape cultural traits and individual behavior over the long run, contributing to broader discussions on the cultural and psychological legacy of conflict.

Presentations:

ASREC Conference 2025 (George Mason University); CESifo Venice Summer Institute 2025 – Poverty, Persistence and Policy: Comparative Development and the Deep Roots of Prosperity; University of British Columbia Dev/PE Lunch Seminar (Vancouver).

Conflict remains a persistent feature of Africa’s recent history. While often linked to ethnic divisions and resource disputes, anecdotal evidence suggests that religion also plays an important role. Yet, formally identifying the role of religion is difficult, as religious divisions frequently overlap with ethnic cleavages, historical rivalries, and socioeconomic inequalities. To better understand the religious component of violence, this project uses the ACLED database, which records about 350,000 geolocated violent events across 58 African countries between 1997 and 2023. We identify religiously related events in two ways: (i) by manually classifying the more than 18,000 actors listed in ACLED, defining as “religious” those whose names or secondary sources indicate a religious affiliation; and (ii) by analyzing event descriptions, which often provide additional clues about religious dimensions beyond actor labels. To capture this, we manually label a random sample of 5,300 event descriptions and construct a list of religious keywords. We then apply this keyword list to the full database, classifying an event as religious if its description contains at least one of these terms. Combining both approaches, we find that 24.1% of violent events reported in Africa between 1997 and 2023 have a religious dimension. Of these, 69.35\% involve at least one actor related to Jihad, while only 3.11% involve a political actor with a religious dimension. Only a small share, 3.70%, involves two religious actors; within this, in just over half of these cases (52.11%), both actors belong to the same religion (intra-faith violence).

Presentations:

Max Planck Summer School 2024 – Political Economy of Conflict and Redistribution (Berlin).

This project studies how assimilation policies targeting American Indians in the late 19th and early 20th centuries affected individualism, ethnic identity, and attitudes towards the U.S. government, as measured through the clauses and language used in over 350 newly digitised tribal constitutions and charters. We first demonstrate that there is meaningful variation in the types of clauses (e.g., blood quantum or residence restrictions on membership) and language (e.g., emphasising the “tribe” or the “individual”) across constitutions. We then examine how tribal characteristics and historical exposure to government policies – such as Indian schools and land allotment – affected constitutional features. Finally, we employ natural language processing techniques to measure the extent to which these characteristics affected trust and cooperation with the U.S. government during the Indian New Deal, as measured by semantic similarity between constitutions written in the 1930s and a ‘model’ constitution distributed by the Indian Office at the time. This project contributes to our understanding of how assimilation policies influenced Indigenous institutional development and political culture, highlighting the long-term effects of government intervention on tribal sovereignty, collective identity, and intergovernmental relations in American Indian Nations.

We study how the urban density of the areas where firms are located impacts their stock returns. Using high-resolution satellite images and image segmentation algorithms, we develop measures of building density for the most populated metropolitan statistical areas (MSAs) in the US. We find that firms in urban areas with a higher potential for density increase are perceived as less risky in the stock market because they can feasibly sustain their growth, leading to lower stock returns. This effect is stronger for smaller and younger firms.

Presentations:

University of British Columbia TRIOS Lunch Seminar (Vancouver).

We study the effects of urban characteristics on access to commercial credit. For that purpose, we construct a dataset about the development potential of land in different cities, which takes into account building density, as well as natural geographic constraints. We use these data to assess how the development potential of urban centers affects the impact of bank concentration on financing. We apply our method to Latin America, a region in which banks play a crucial role, and many loans require real estate collateral. In line with the information-based hypothesis, we find a positive relation between bank concentration and firms’ credit accessibility; and we show that the development potential of the cities in which firms are located moderates this positive relationship and helps to explain firm-level differences in access to capital.

This paper explores the relationship between culture and negotiation, identifying clusters of negotiation patterns and assessing whether the negotiator’s culture affect the probability of adopting one model over the others. We use latent class analysis on a sample of 2099 observations of negotiation behaviors in 69 countries, to identify negotiation clusters and we obtain three negotiation prototypes. One prototype is oriented to personal relationships, the expression of emotions, and a flexible agenda for a polychronic procedure. Another is formal, facts-focused, and monochronic, maximizing economic value but disregarding personal relationships. The third prototype is harder to interpret and might indicate contexts where different negotiation behaviors coexist. Culture, defined taking into account multiple sources, is a significant predictor of negotiation prototypes (beyond other socio-economic factors). Our results highlight the importance of behaviors along the relational-transactional dimension for international negotiations, as well as the advantages of using mixed qualitative-quantitative methods in negotiation research.

Presentations:

IACM Conference 2019 (Dublin) 

Integrative and distributive negotiation strategies are a key paradigm of practice, teaching, and research. Are these US-formulated negotiation prototypes valid in the rest of the world? Adopting a cross-cultural view, we analyze a sample of 214 foreigners who detailed the negotiation behavior they faced in Italy (134) and in the United States (80). Implementing latent class analysis, we identify three clusters of negotiation prototypes. Our findings show how the Country is a predictor for cluster membership, and peculiar cultural traits of the two groups contribute to explain the differences in negotiation strategies. Three prototypes emerged: a typically distributive, an emotional integrative (mostly Italian), and an impersonal integrative (mostly American). Results show how the handling of emotions is a crucial part of the interaction for Italian negotiators, regardless of their orientation toward negotiation strategies, implying a cultural influence toward handling emotions in negotiations.

This study examines how extreme weather events affect the sustainability conduct of small businesses based in the Global South. The premise is that firms affected by extreme weather suffer damage. What remains unclear is whether such damages prompt firms to cut costs on sustainability practices or, conversely, the experience with extreme weather encourages practices that contribute positively to environmental sustainability. We argue that coffee suppliers affected by severe weather shocks are more likely to adopt risk-internalizing strategies, specifically by increasing their reliance on family labor and implementing recycling initiatives. However, the extent to which these strategies are utilized depends on the suppliers' proximity to the multinational enterprises (MNE) that buy their products. Suppliers more closely integrated into MNEs’ global value chains demonstrate a weaker association between weather shocks and reliance on family labor, but a stronger tendency to adopt recycling initiatives. Using the La Niña weather phenomenon from 2008 to 2012 as a quasi-experimental setting, we analyze 11,153 farm-year observations, complemented by 15 follow-up interviews within the Colombian coffee industry. Our findings offer new insights into the adaptive capacities of small firms within global value chains, uncovering the novel role of MNE relationships in shaping resilience and sustainability outcomes.