Year

This thesis provides empirical evidence on heterogeneity in returns on wealth among U.S. households. Chapter 1 provides a summary of how return heterogeneity has been theoretically and quantitatively shown to be necessary in explaining wealth inequality. It discusses the limited empirical evidence driven primarily by a lack of adequate panel data on household returns. Chapter 2 presents a new method to construct panel data on household-level rates of return in the United States using the newly revised Panel Study of Income and Dynamics (PSID) data. Returns are calculated for total household assets and wealth as well as within five asset classes. This panel data provides the first return measure for the total household portfolio, and it is the first to include net investment in the measure of capital gains. Empirical evidence is documented on the degree of return and risk-adjusted return heterogeneity and returns' correlation with wages. Chapter 3 uses fixed effects with empirical Bayes shrinkage to document permanent heterogeneity in returns across households, even within asset classes. The majority of the permanent heterogeneity in the returns to wealth are driven by heterogeneity in the degree and cost of borrowing. Permanent heterogeneity in leverage across households is also documented. Evidence is provided on how returns differ across the wealth distribution. Chapter 4 documents the degree of idiosyncratic asset return risk, its serial correlation, and its correlation with wage risk. Sizeable transitory idiosyncratic return risk is documented for total household assets and for returns within each asset class. Idiosyncratic shocks to wages are found to be correlated with idiosyncratic shocks to returns in specific asset classes. Return and wage risk correlation and return serial correlation are found to depend on a household's age and homeownership. By providing a new source for panel data on rates of return, this thesis presents the first empirical evidence on the nature and degree of return heterogeneity in the United States. Chapter 5 summarizes the empirical evidence. It highlights the implications for the literature on portfolio choice and for the structure and calibration of return heterogeneity in quantitative models to study wealth and income inequality..”