Working Papers

  

When Does Inequality Lead to Political Demands? How Attribution Matters (under review)

Meir Alkon

Abstract: When does inequality lead to political demands? I argue that attribution determines how inequality affects public preferences. When inequality is attributed to the government, individuals are more likely to demand policy reform. I test this theory on three experimental populations in China, where inequality is highly salient. Combining a randomized factorial design with conjoint analysis, I show that priming inequality increases demands for reforming central government corruption, but these demands are attenuated when inequality is attributed to the global economy. A second experiment replicates this effect, shows that it is inequality-specific, and distinguishes attribution from displacement. I then illustrate the mechanisms of attribution using structural topic models of open-ended responses from a third population, showing respondents implicitly attribute inequality to corruption and central government policies. These findings contribute to understanding the importance of attribution in determining inequality-induced political demands generally, and to understanding the politics of inequality in contemporary China.

PDF

Authoritarian Environmental Governance Cycles: Centralization-Decentralization Dynamics in China's Power Sector (under review)

Meir Alkon and Audrye Wong

Abstract: We develop a theory to explain the persistence of tensions between decentralized delegation and centralized control of environmental governance in authoritarian regimes. Economic benefits from decentralization -- information, competition, and efficiency -- conflict with environmental goals of centralized policy harmonization and management of inter-jurisdictional externalities. Decentralization to local government actors can facilitate economic growth but also empowers them in ways that undermine environmental governance. Persistent tensions between decentralized and centralized imperatives generate cycles in environmental and energy systems governance. We test our theory of authoritarian environmental governance cycles using the case of China's power sector, drawing on evidence from primary source documents, field interviews, and multiple data sources on the development and distribution of energy generating capacity. We focus on two policy areas -- coal-fired power and wind energy -- that are integral to central government efforts to improve the quality of environmental governance. This research explains the puzzling alternations in the locus of governance, and contributes to understanding inter-governmental relations and environmental politics in authoritarian regimes.

PDF 

  

Local Sociotropism: How Community Variation in Trade Exposure Affects Voter Demands

Meir Alkon

Abstract: Recent research demonstrates the electoral importance of trade-induced employment loss for determining voting in presidential elections. Voters respond to localized job losses by punishing the incumbent. However, the mechanism underlying these effects remain unclear. What explains these large, localized electoral effects? I develop a theory of local sociotropism, in which individual's identification with their local community catalyzes a feeling of community threat from trade and lowers sociotropic assessments of trade's overall effects and of economic growth overall. I test this theory using data from an original survey of 1,780 respondents. Using data on the share of manufacturing jobs lost to rising imports (the so-called ``China Shock''),  my analyses show that respondents living in areas with more job losses due to trade are less likely to believe that trade has been good for the country, less likely to favor trade, to have lower subjective perceptions of economic trajectory, and also to be more likely to vote for Donald Trump in the 2016 election. I also show that priming respondents with information about different government compensation programs has no effect on their attitudes towards trade or their assessments of government performance. Finally, I detail the next steps in this research project, using data on job losses, 2016 county-level vote returns, and the 2016 American National Election Survey (ANES). Using this nationally representative survey data, with questions that are identical to those used in the survey above, I test for local sociotropism and assess the relationship holds between local sociotropism and 2016 presidential voting.