Research

Peer Reviewed Publications

Monitoring war destruction from space using machine learning
Published in PNAS June 8, 2021 118 (23)
With Hannes Mueller, Andre Groeger, Jonathan Hersh, and Joan Serrat
​
Satellite imagery is becoming ubiquitous. Research has demonstrated that artificial intelligence applied to satellite imagery holds promise for automated detection of war-related building destruction. While these results are promising, monitoring in real-world applications requires high precision, especially when destruction is sparse and detecting destroyed buildings is equivalent to looking for a needle in a haystack. We demonstrate that exploiting the persistent nature of building destruction can substantially improve the training of automated destruction monitoring. We also propose an additional machine-learning stage that leverages images of surrounding areas and multiple successive images of the same area, which further improves detection significantly. This will allow real-world applications, and we illustrate this in the context of the Syrian civil war.

Book Chapters

Historical Data: Where to Find Them, How to Use Them – Joint with Paola Giuliano in Handbook of Historical Economics, edited by Alberto Bisin and Giovanni
Federico

The Use of Archaeological Data In Economics – Joint with Luigi Pascali in Handbook of Historical Economics, edited by Alberto Bisin and Giovanni Federico

Working Papers

The Ant and the Grasshopper: Seasonality and the Invention of Agriculture
Review and Resubmit at Quarterly Journal of Economics

During the Neolithic Revolution, seven populations independently invented agriculture. In this paper, I argue that this innovation was a response to a large increase in climatic seasonality. Hunter-gatherers in the most affected regions became sedentary in order to store food and smooth their consumption. I present a model capturing the key incentives for adopting agriculture, and I test the resulting predictions against a global panel dataset of climate conditions and Neolithic adoption dates. I find that invention and adoption were both systematically more likely in places with higher seasonality. The findings of this paper imply that seasonality patterns 10,000 years ago were amongst the major determinants of the present day global distribution of crop productivities, ethnic groups, cultural traditions, and political institutions.

All Along the Watchtower: Military Landholders and Serfdom Consolidation in Early Modern Russia – joint with Timur Nathkov

We study the emergence of extractive institutions induced by external military threats.Using the case of early modern Russia, we explore the consolidation of serfdom under the pressure of landholding military elites who gained political influence due to the prolonged struggle with steppe nomads. To contain nomadic raids, the Russian state erected defense lines on the southern frontier, and granted lands in the area to soldiers in charge of its defense. The soldiers could not farm while on defensive duties, nor could they compete in the market for peasant labor because the lands had been selected for their defensive rather than agricultural value. The system was therefore only sustainable by gradually binding peasants to the land. Using newly digitized 17th century population data, we showa higher prevalence of serfs and military landholders in districts on the defense line. We also find a higher prevalence of small estates – up to 25 serf households – sufficient to support a warrior and his family. Placebo tests reveal that these patterns do not hold for non-serf peasants and in other, defensively non-optimal, locations. To ensure causality,we develop a novel algorithm that reconstructs the optimal invasion routes for nomads and pinpoints the optimal location of the defense line using topographic data. Our results suggest that military considerations – rather than the high land-labor ratio – were among the key factors of serfdom formation. This sheds new light on the possible mechanisms of institutional divergence between Eastern and Western Europe in the early modern period.