Quarterly Journal of Political Science > Vol 11 > Issue 4

The Altruistic Rich? Inequality and Other-Regarding Preferences for Redistribution

Matthew Dimick, SUNY Buffalo, USA, mdimick@buffalo.edu , David Rueda, University of Oxford, UK, david.rueda@politics.ox.ac.uk , Daniel Stegmueller, Duke University, USA, daniel.stegmueller@duke.edu
Suggested Citation
Matthew Dimick, David Rueda and Daniel Stegmueller (2017), "The Altruistic Rich? Inequality and Other-Regarding Preferences for Redistribution", Quarterly Journal of Political Science: Vol. 11: No. 4, pp 385-439. http://dx.doi.org/10.1561/100.00015099

Publication Date: 20 Feb 2017
© 2017 M. Dimick, D. Rueda, and D. Stegmueller
Public economics,  Elections,  Comparative political economy,  Electoral behavior,  Formal modelling,  Political economy,  Voting behavior,  Voting
RedistributionPreferences for redistributionAltruismSelf-interestInequalityTaxation


Download article
In this article:
A Model of Altruistic Preferences for Redistribution 
Model Results 
Testing the Model 
Robustness Checks 
A Proofs 
B Alternative Models of Preferences 
C Deriving the Full Estimating Equation 
D Descriptive Statistics 
E Bootstrap Standard Errors 
F Envy 
G CPS Income Data 
H National Inequality 
I Importance of National Redistribution for Individual States 


What determines support among individuals for redistributive policies? Do individuals care about others when they assess the consequences of redistribution? This article proposes a model of other-regarding preferences for redistribution, which we term income-dependent altruism. Our model predicts that an individual’s preferred level of redistribution is decreasing in income, increasing in inequality, and, more importantly, that the inequality effect is increasing in income. Thus, even though the rich prefer less redistribution than the poor, the rich are more responsive, in a positive way, to changes in inequality than are the poor. We contrast these results with several other prominent alternatives of other-regarding behavior. Using data for the United States from 1978 to 2010, we find significant support for our claims.