Essays on High-Cost Consumer Credit

This dissertation studies high-cost consumer credit markets. High-cost consumer lending, e.g., payday loans in the U.S., typically charge an APR of 322\% for small, short-term loans. Often, policymakers and academia discuss whether they cause more harm than good. In Chapter 1, I make two contributions. Firstly, I document in a nationally representative survey that households that take out payday loans have low wealth and low-liquidity levels; relatively low income, although there are payday borrowers across the income distribution; high demand and rejection rates for traditional credit sources; and are more likely to experience expenditure shocks or unemployment spells. Secondly, I develop a model of banking and payday lending that delivers, in equilibrium, an interest rate and loan size spread between these lenders which we observe in the data.

In chapter 2, I study the welfare consequences of regulations on high-cost consumer credit in the U.S., such as borrowing limits and interest rate caps. For some borrowers, it is desirable to borrow at high interest rates when they experience adverse shocks (e.g., to their health). However, others have preferences with self-control issues that may induce them to overborrow. I estimate a heterogeneous-agents model with risk-based pricing of loans that features standard exponential discounters and households with self-control and temptation. To identify different household types, I use transaction-level payday lending data and the literature's valuations of a no-borrowing incentive. I find that one-third of high-cost borrowers suffer from temptation. Although individually targeted regulation could improve the welfare of these households, I find that noncontingent regulatory borrowing limits and interest-rate caps—like those contained in typical regulations of payday loans—reduce the welfare of all types of households. The reason is that lenders offer borrowers tight individually-targeted loan price schedules that limit households' borrowing capacity so that noncontingent regulatory limits cannot improve welfare over them.