Discrimination, trust and social capital: Three essays in applied public economics

In the first essay. "Competition to Default? Racial Discrimination in the Market for Online Peer-to-Peer Lending", I investigate the incidence and causes of racial discrimination in an online lending market. Competition ameliorates discrimination, by reducing interest rates for black borrowers at twice the rate compared to whites. However, competition does not appear to drive interest rates towards the economic costs of lending to blacks, thus, racial prejudice does not appear to drive disparate treatment. I conclude the market appears to possess an inefficient degree of statistical discrimination. The second essay, "A Matter of Trust: Understanding Worldwide Public Pension Conversions" and written with Kent Smetters, provides a novel explanation for the recent wave of public pension reforms. Why have so many countries reformed when traditional systems are likely superior in design? Why have these reforms typically have been larger in developing countries facing less severe demographic problems? We show that an overlapping-generations median voter model can help answer both questions. Larger reforms are motivated by a fundamental breakdown of intergenerational trust. Smaller reforms are motivated by a lack of trust in the ability of the government to save resources for smoothing demographic shocks. In the third essay, "Am I My Brother's Firefighter? Social Capital and the Voluntary Provision of Local Public Goods", I study the social factors affecting the quality, and provision of volunteer firefighting services. Residents protected by one of the 18.575 all-volunteer fire departments in the United States depend on the willingness of volunteers to sacrifice their time, and their lives, in the service of their neighbors. Using data from the National Fire Incident Reporting System (NFIRS), I find that residents of more homogenous communities have higher propensities to volunteer, and such communities are more likely to use volunteers departments. I find evidence of crowd-out of pure volunteers when volunteers are paid for their service, but no crowd-out effects from professional firefighters. More generally, this study contributes by providing the first direct evidence on whether social capital affects the outcomes of volunteer efforts rather than just volunteer participation.