Income Inequality and Political Polarization: Time Series Evidence Over Nine Decades

Abstract

Rising income inequality and political polarization have led some to hypothesize that the two are causally linked. Properly interpreting such correlations is complicated by the multiple factors that drive each of these phenomena, potential feedbacks between inequality and polarization, measurement issues, and statistical challenges for modeling non-stationary variables. We find that a more precise measure of inequality (the inverted Pareto-Lorenz coefficient) is statistically related to polarization while a less precise one (top 1% income share) is not, and that there are bi-directional feedbacks between polarization and inequality. Findings support a nuanced view of the links between polarization and inequality.JEL Codes: D72, D63, D31

Income Inequality, Media Fragmentation, and Increased Political Polarization

“…The better fits of the House models may stem from its composition being more closely tied to very recent media or inequality conditions in contrast to the Senate whose more slowly shifting composition reflects conditions over the three preceding election cycles. Third, the TeaParty variable is more statistically significant in the Senate models, where it raises the political polarization index by about 8 points—similar in magnitude to the results found in Duca and Saving () who use a longer sample and focus on examining long‐run inequality effects on polarization . The difference between the House and Senate findings may reflect the fact that that more Senate seats were represented by moderates in the pre‐Tea‐Party era (perhaps because Senate “districts” cannot be gerrymandered).…”

“…We follow Duca and Saving () by including a control variable for the “Tea Party” phenomenon ( TeaParty ), equal to 1 for the 2011 and 2013 Congresses and 0 otherwise. Skocpol and Williamson () find that the Tea Party is associated with greater turnout among conservative portions of the electorate and has driven moderates of both parties out of Congress.…”

“…This implies cleaner evidence of causality running from Cable to polarization than from Gini to polarization. Nevertheless, there are reasons why bidirectional long‐run causality might arise between inequality and polarization (see Duca and Saving ; McCarty ) and bidirectional causality is a central theme in several polarization studies (e.g., McCarty, Poole, and Rosenthal , ).…”

Falling Apart or Flocking Together? Financial Crises, Inequality and Left-Right Polarization in the OECD

“…The author also argues that as inequality goes up, poor voters participate less in elections (not examined explicitly) 8 , which leads to the diminishing vote shares of the left parties and forces them to move to the center. Duca and Saving (2016) show for the U.S. that elite polarization and income inequality coevolve. McCarty et al (2003, p. 4) observe in a detailed empirical study an increasing partyincome divide in the U.S. electoral landscape: "...we can see that the stratification of partisanship by income has steadily increased over the past 40 years, leading to an increasing cleavage between the parties" (see also McCarty et al, 2015).…”

“…More recently, McCarty et al (2003, 2015 have noticed that economic inequality and political polarization have increased dramatically across the United States since the 1980s. Duca and Saving (2016) have shown in an econometric analysis that changes in inequality precede the changes in economic polarization in the U.S. although there is also some evidence for a reciprocal effect. Kuhn et al (2020) and others have shown additionally that wealth inequality has grown after the collapse of the Lehmann Brothers, a result that is also observable for currency crises in the developed world (Rübsam and Schneider 2020).…”