ABSTRACT
Previous literature has demonstrated that individuals in the US report greater happiness when the president is of the same party that the individual identifies with. It has likewise been demonstrated that individuals with more extreme political views, be they liberal or conservative, report higher happiness than those with moderate political views. This article demonstrates that the relationship between these two dimensions and happiness is not separable. In fact, congruence with the president, in terms of party and ideology, matters the most for those with extreme political views. This is demonstrated by estimation of the interaction effect between political extremism and measures of presidential congruence.
Acknowledgments
The author would like to thank session participants at the 2018 meetings of the Public Choice Society as well as comments received at a research seminar at North Dakota State University. A portion of this research was completed while the author was serving as a visiting fellow of the Eudaimonia Institute at Wake Forest University. This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
2 The data and methodology used to estimate presidential ideology are described in McCarty and Poole (Citation1995).
3 The variables in each of these matrices of control variables are consistent with those common in the happiness literature and are the same as those used in Jackson (Citation2019). Individual controls include indicator for political party, income, sex, age, age2, indicator for four year college degree, indicator for student, occupation, number of children. State level time variant controls include gini coefficient on income inequality, percentage of adults (over 25) with at least a bachelor’s degree and personal income.
4 This procedure has been previously used by Stevenson and Wolfers (Citation2008), Geishecker (Citation2012) and Jackson, Crawford and Pritchard (Citation2018).
5 Only coefficients for variables of interest are shown. Full tables of results are available upon request from the author.