Vote Mobilization at Work: How Employers Subvert Elections

Abstract : Scholars have long documented how party activists mobilize voters during elections, but politicians use a range of other actors to help get out the vote as well. Despite increasing interest among scholars in non-party based vote mobilization, there is much that we do not know. We explore how politicians use employers and party activists to get voters to the polls and present three findings. First, looking at a range of middle-income countries, we show that workplace mobilization of voters by employers is common, even in countries where parties are relatively strong. Second, using data from Russia, we show that workplace mobilization presents a trade-off. It helps to mobilize workers to vote, but also reduces support among the general public who generally disapproves of this tactic. Third, we demonstrate that media freedom can help reduce the incidence of workplace mobilization by increasing the likelihood of detection by the general public. These results suggest the need for greater attention to employers as political agents during elections and to the normative implications of different voter mobilization strategies.

---

Production Networks and War: Evidence from Ukrainian Railway Shipments

Abstract : How do severe shocks, such as an armed conflict, alter a country's economy? We study how the production network of a low-income country is affected by a devastating but localized conflict. We use the Ukrainian railway shipment data for 2013--2016, complemented by the administrative data on firms, to document the propagation of a conflict shock throughout the supply-chain network. First, we document substantial spillover effects of conflict on interfirm trade outside of the conflict areas. Our estimates suggest that the magnitude of the second-order effects of conflict is about a third of the magnitude of the first-order effects. Second, we study the consequences of an exogenous change in the network structure. Firms that, for exogenous reasons, became more central in the production network after the start of the conflict, received a boost to their revenues and profits. However, losers from such changes in the production network structure outweigh the winners. According to our back-of-the-envelope calculations, conflict decreased sales by at least $5.8\%$ outside of violent areas due to the two effects. Finally, we document that if the firms were not allowed to adjust their trading networks, counterfactual losses from the exogenous network shock would have been three times larger than the observed losses.

---

The Soviet Great Famine, 1931--33

Abstract : This paper documents several new facts about the causes of the Soviet Great Famine, 1931–33. First, bad weather was not an important contributor to the famine. Second, excess mortality was much higher for ethnic Ukrainians, the largest minority group, than ethnic Russians, even outside of the Republic of Ukraine. Third, this cannot be explained by differences in weather conditions, historical productivity, or political variables. Fourth, Soviet policies were implemented more zealously for Ukrainians, which resulted in more collectivization of agriculture, lower grain productivity, and higher procurement during the famine for Ukrainians. Both ethnic Ukrainian and non-ethnic Ukrainian Party leaders increased Ukrainian mortality. The results are consistent with the presence of ethnic bias in famine-era Soviet policies.

---