Text

theses

My dissertation investigates economic incentives relevant to consumer financial services. I focus on two types of unsecured consumer loans primarily for credit-constrained consumers: student loans and consumer overdrafts. Currently in the U.S., the federal government supplies most student loans and applies strong disclosure rules for consumer overdrafts to address market failures due to asymmetric information. However, high student default rates and heavy overdraft fees demand improvements in these government interventions. Nevertheless, beneficial policy changes are hampered by insufficient understanding of student loan repayment behavior and market forces for overdraft fees. Chapter 1 helps fill this gap by demonstrating that stronger repayment incentives reduce defaults on federal student loans, and Chapters 2 and 3 reveal that competition is a key determinant of overdraft fees.

In Chapter 1, I examine how Tennessee's occupational license suspension policy affects student loan repayment. Using program changes in 2009 and 2013 as a quasi-experiment, my difference-in-differences and synthetic control estimation employing the College Scorecard Data shows that the license suspension program led to a surprisingly large reduction in federal student loan defaults: default rates fell by about 30% in Tennessee after the program. I also find suggestive evidence that these effects were largest among women and degree non-completers. While most studies have attributed high student loan default rates to low earnings or inability to repay, my study demonstrates that the incentive to repay is also important in explaining student loan defaults. This finding suggests that policymakers take into account the importance of repayment incentives to improve the sustainability of federal student loan programs.

In Chapter 2, I explore how competition affects overdraft fee revenue of U.S. banks. Linking newly available overdraft fee revenue data in Call Reports to a competition index derived from branch deposits data, I demonstrate that banks facing higher competition generate more overdraft fee revenue per account than their peers facing lower competition. I use a historical competition index as an instrument for the current competition index to affirm the causality from competition to per-account overdraft fee revenue. These findings are consistent with a theory of bank risk-taking, developed by Marcus (1984) and Keeley (1990): an increase in competition reduces bank charter value, encouraging the bank to take more risk. In my study's context, the increased risk-taking involves supplying more overdraft credits or charging overdraft fees more aggressively at the expense of default or legal risk. In contrast to a general idea that competition benefits consumers, my study shows that competition among banks drives them to generate more overdraft fee revenue, likely hurting low-income consumers who are disproportionately exposed to overdraft fees.

In Chapter 3, I investigate a theoretical explanation for why competition among banks may reduce general deposit fees but raise overdraft fees. I theorize the effects of competition on the two types of deposit fees using a version of the Salop model where consumers are fully rational. In my model, banks provide a deposit account service and an overdraft service to high- and low-income consumers. I assume that low-income consumers are more likely to overdraw and are more sensitive to a difference in fees between competing banks than high-income consumers. My model predicts that the equilibrium overdraft fee rises as the number of banks increases. The economic logic is straightforward: since low-income consumers are more responsive to changes in fees and are more likely to pay the overdraft fee than high-income consumers, banks set the overdraft fee below the overdraft service costs in an equilibrium. As the number of banks in the market grows, each bank's revenue falls, and the banks respond by raising their overdraft fees in a new equilibrium.

ETD_10784

Ph.D.

Includes bibliographical references

doi:10.7282/t3-3t8t-wf22

ETD doctoral

The author owns the copyright to this work.