Asset Pricing with Revealed Utility.

Abstract

This dissertation consists of two essays that interpret crime as revealed marginal utility of heterogeneous consumers, and investigates its implications for asset pricing.

The first chapter proposes crime as a revealed response of individuals that derive utility from relative wealth. Using daily reported crime incidents from over 2,500 law enforcement agencies across 27 states from 1991-2012, a contemporaneous relationship between daily stock returns and various types of crimes are found. Market changes also impact investors’ and non-investors’ utility differently and this is interpreted as evidence in support of envy models such as Abel (1990) that individuals care about their own wealth relative to others. For example, daily stock market increases are associated with decreases in violent crimes in high income locations, while market increases are associated with increases in violent crime in low income locations.

The second chapter builds upon using crime as revealed marginal utility. Having established a relationship between violent crime and the stock market in the first chapter, the second chapter proposes violent crime growth as a measure of revealed marginal utility growth of heterogeneous consumers in incomplete markets to price the cross-section of stock returns. Consumer heterogeneity is measured using the cross-sectional average and cross-sectional variance of crime growth exploiting a monthly panel of reported crime incidents from over 10,000 law enforcement agencies across the United States from 1975-2012. Consistent with heterogeneous consumer models such as Mankiw (1986), the cross-sectional average and variance of violent crime growth are found to explain the cross-section of stock returns. Specifically, investors pay a premium for assets that have higher betas to the violent crime growth moments.