Income inequality has risen sharply in the United States over the past forty years, yet there remains substantial uncertainty about the consequences of income inequality on social life. This dissertation advances research on these consequences by focusing on mechanisms through which inequality may matter and on the methods by which the effects of income inequality are determined.
I primarily draw on data from the 1973–2014 General Social Surveys linked to administrative data of the demographic and economic characteristics of each respondent’s state. This includes state-level income inequality, for which I utilize a new annual series based on household income tax returns. I also conduct an online survey experiment that manipulates perceptions of state-level income inequality.
First, numerous scholarly accounts posit that as income inequality rises, individuals will be less satisfied with their own finances as they feel increasingly deprived relative to others—driving individuals to try to spend more as they engage in positional competition and increasing their anxieties as position in the income distribution becomes ever more crucial. I find that higher state-level income inequality decreases financial satisfaction overall, and that this effect is especially pronounced for those in the middle of the income distribution. Counterfactual simulations suggest rising inequality explains a substantial portion of the over-time decline in financial satisfaction.
Second, concerns about rising income inequality are frequently linked to discussions about opportunity and mobility, yet little research explores if and how this inequality affects people’s economic optimism, something with far reaching implications for life satisfaction, public opinion, and real economic mobility. Both the survey analysis and the experiment show that higher income inequality decreases economic optimism. The survey shows that the rate of change in inequality moderates the effect of the level of inequality, and that household income further moderates the effects of the level and change in income inequality on economic optimism. There was no evidence of this moderation in the experiment. Key differences between the two methodological approaches are discussed.
Third, although both popular and scholarly accounts have argued that income inequality reduces trust, some recent research has been more skeptical, noting these claims are more robust cross-sectionally than longitudinally. Furthermore, although multiple mechanisms have been proposed for why inequality could affect trust, these have rarely been tested explicitly. I find little evidence that states that have been more unequal over time have less trusting people. There is some evidence that the growth in income inequality is linked with a decrease in trust, but these effects are sensitive to how time is accounted for. While much of the inequality and trust research has focused on status anxiety and feelings of relative deprivation, this mechanism receives the weakest support, and mechanisms based on societal fractionalization and exploitation receive stronger support.
Finally, social comparisons of income have far-reaching consequences for individual decision-making and public policy, yet there persists a significant gap between “true” relative income and what Americans perceive. Although one compelling explanation is that reference groups affect what people perceive as “average,” there is little consensus about who people compare themselves with. Previous research has proposed reference groups based on both geographic proximity and on sociodemographic similarity, but few studies have considered multiple reference groups systematically or simultaneously. I find that the effect of reference group income depends on both egoist and fraternal comparisons: higher median incomes of large reference groups and those with weak status hierarchies increases perceived relative income, while higher median incomes of small reference groups and those with strong status hierarchies decreases perceived relative income. These results have important implications for how reference groups are used in research on neighborhood effects, residential segregation, and income inequality.
