B. Enke, M. Polborn, and Alex Wu
NBER Working Paper No. 30001
April 2022
JEL No. D03,D72
Abstract
This paper develops a theory of political behavior in which moral values are a luxury good: the relative weight that voters place on moral rather than material considerations increases in income. This idea both generates new testable implications and ties together a broad set of empirical regularities about political polarization in the U.S. The model predicts (i) the emergence of economically left-wing elites; (ii) that more rich than poor people vote against their material interests; (iii) that within-party heterogeneity is larger among Democrats than Republicans; and (iv) widely-discussed realignment patterns: rich moral liberals who swing Democrat, and poor moral conservatives who swing Republican. Assuming that parties set policies by aggregating their supporters’ preferences, the model also predicts increasing social party polarization over time, such that poor moral conservatives swing Republican even though their relative incomes decreased. We relate these predictions to known stylized facts, and test our new predictions empirically.
Conclusion
This paper has shown that the simple idea of income-dependent utility weights – which is bolstered by a large body of evidence on “modernization” or “postmaterialism” – generates a host of new testable predictions and sheds light on various widely-emphasized stylized facts about the nature of political polarization and realignment patterns in the U.S. In particular, our approach offers a new lens through which the increased salience of moral and cultural dimensions of political conflict can be understood.
One aspect of polarization that we only briefly and informally touched upon is affective polarization: the stylized fact that people’s dislike of supporters of the other party has strongly increased over time (Iyengar, Sood and Lelkes, 2012; Iyengar et al., 2019). In our interpretation, this reflects that the distribution of moral values of Republican and Democrat voters have diverged over time as a result of sorting processes that are triggered by our account of morals as luxury goods. However, while much psychological research suggests that people get angry if others don’t share their basic moral convictions (Haidt, 2012), more research is needed to establish a direct link between increased voter sorting based on moral values and affective polarization.
