The Pursuit of Happiness: Quasi-Experimental and Experimental Evidence for the Causal Effect of Election Outcomes on Well-Being

Abstract

This study measures the causal effect of U.S. presidential, congressional, and gubernatorial election outcomes on self-reported happiness using both quasi-experimental and experimental research designs in four parts. In the first part, I use data from 1973 to 2014 to show that Republicans are happier when there is a Republican president as opposed to a Democratic president. This effect is driven entirely by increased happiness of higher-income Republicans; lower-income Republicans do not see increased happiness when there is a Republican president. I find no effect on happiness for Democrats when there is a Democratic president. Similar effects are present when controlling for the party of the president of the year prior, even to the extent that lower-income Republicans are less happy when the president the year prior was Republican. In the second part, I extend this analysis to the House and Senate to show that higher income Republicans and Democrats alike are happier when there is a Republican House. I find no such effect for Senate election outcomes. In the third part, I use a regression discontinuity (RD) design and data on 141 close gubernatorial elections to help establish the causal effect of election outcomes on self-reported happiness. Using the RD design, I show that Democrats are happier when there is a Democratic governor. I find no impacts on happiness for Republicans at the state level. The fourth part of the paper sheds light on why lower-income Republicans vote in the way that they do using a novel randomized survey experiment. In particular, I show that priming people with the Republican Party leads to a monetary allocation away from a hypothetical individual with an Islamic name, and priming people with Trump leads to greater support of restrictions on entry to the United States for people coming from countries with high Muslim populations, potentially identifying a fear-based voting rationale for lower-income Republicans. The results from the survey experiment provide insight into why lower-income Republicans vote for the Republican Party, even though the first three parts of the paper establish that a Republican election outcome has no causal effect on their happiness.