Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Abstract We investigate the effect of the Russian-Ukrainian war on the research productivity of scholars affiliated with around 15,000 Ukrainian research institutions. Using the 2014 Russian invasion as a quasi-natural experiment, we apply a difference-in-differences estimator on a sample of half a million journal articles collected from Scopus. Researchers affiliated with institutions located in the occupied parts of Donetsk and Luhansk and the whole of Crimea form our treatment group, and those affiliated with institutions in the unoccupied regions of Ukraine, our control group. We document a significant decline in quantity and quality of research, measured by the average number of papers/citations, produced by authors based in the Donetsk/Luhansk occupied regions with active hostilities. By contrast, we observe a rise in the quantity of papers published by authors based in annexed Crimea. Yet, this pattern, which can be explained by increased funding by Russian authorities towards institutions located in Crimea, is driven by articles published in Russian journals and by scientists with relatively low productivity. Our results are robust when using different control groups and estimation methods, including causal machine learning tools, and when controlling for publication lags.