Income Inequality and Household Debt: Assessing the Relationship Using Household Panel Data

Abstract

Is income inequality positively associated with household debt? Previous scholarship suggests that widening income inequality could stimulate household debt, particularly in advanced economies. Furthermore, this potential link could be associated with negative consequences including reduced GDP and secular stagnation, financial fragility and crisis, and decreased capabilities, mobility, and well-being. Yet, until recently, this “income inequality/household debt” link has been understudied. The first chapter discusses two theoretical bases for the link and synthesizes the burgeoning empirical research to identify gaps. Theoretically, income inequality could bolster credit supply because marginal propensity to save increases with income. But income inequality could also stimulate credit demand through the relative income hypothesis. Empirically, income inequality is positively associated with rising debt-to-income ratios (DTI) in developed countries. However, this empirical research generally lacks controls for an important omitted variable (financial deregulation) and an alternative explanation (wealth effects). Moreover, few studies explore changes in the intensive margin of household DTI, which has been the largest contributor to rising overall DTI and is most likely linked to negative consequences. The second chapter examines the relative impacts of various measures of state-level income inequality, financial deregulation, and wealth effects on both the intensive margin and overall DTI of United States households using household panel data from the Panel Study of Income Dynamics (PSID). Results indicate that income inequality is significantly, positively associated with household DTI, but only for the Gini coefficient and the top 10 percent income share. Yet for these measures, income inequality has the strongest and most significant impact on the intensive margin of household DTI as compared to alternative explanations. The third chapter explores a potential microfoundation for the income inequality/DTI link: whether subjective well-being is associated with household DTI. A long literature notes how income inequality and relative income preferences affect subjective well-being. If subjective well-being also affects debt, it could function as a measure of credit demand. Again using data from the PSID and a fixed effects method, results show that subjective well-being has a positive, significant, contemporaneous association with household DTIs. Such a positive association could indicate a “tunnel effect” -- that households with a positive, aspirational outlook take on increasing amounts of debt. However, reverse causality is a potential issue, and additional models designed to address reverse causality with a two-year lag do not find significant results, perhaps due to data limitations for lag specifications. The key finding of this dissertation is that income inequality is positively associated with household DTI, and, in fact, contributes more to driving indebted households deeper into debt than alternative explanations. Yet when viewed from the lens of subjective well-being, happier households are more likely to take on additional debt. These findings suggest that, in the United States, income inequality may lead some households to go deeper into debt as they ostensibly chase the “American dream” and aspire to the same living standards as higher income earners.