Dissertations, Theses, and Capstone Projects

Date of Degree

5-2019

Document Type

Dissertation

Degree Name

Ph.D.

Program

Business

Advisor

Linda Allen

Committee Members

Lin Peng

Yi Tang

Dexin Zhou

Subject Categories

Business Administration, Management, and Operations | Corporate Finance | Finance and Financial Management

Keywords

Systemic Risk, Managerial Styles, FinTech, Mortgage, Peer-to-Peer Lending, Social Interaction

Abstract

This dissertation consists of three chapters that span managerial styles, financial technologies, and social interactions.

Chapter 1 Banks increase credit risk-taking in syndicated bank loans when their systemic risk increases; however, the interrelationship across risks depends on bank managerial styles. Using a connectedness sampling method to differentiate patterns of business policy styles and systemic risk-taking among managers, I find that credit risk-taking is more sensitive to the bank's systemic risk if the manager exhibits a preference for systemic risk. Asset-innovating managers (exhibiting a preference for non-traditional forms of income and assets) take higher credit risk in their loan portfolios, but marginally reduce their credit risk during systemic crises. In contrast, liability-innovating managers (relying on non-traditional funding sources) generally take less credit risk, but increase their credit risk during systemic crises. Bank-level differences cannot explain the observed heterogeneity across banks.

Chapter 2 This chapter studies whether FinTech mortgage lenders fill the credit gap left by non-FinTech lenders (i.e., traditional banks and non-FinTech shadow banks). Using natural disasters as shocks to local mortgage demand, I find different reactions between FinTech and non-FinTech lenders. First, FinTech lenders and traditional banks expand lending after demand shocks, while non-FinTech shadow banks do not. Second, non-FinTech lenders tighten lending standards after demand shocks, whereas no evidence shows that FinTech lenders change lending standards or risk-taking. Third, non-FinTech lenders tend to ``cherry pick'' good borrowers after demand shocks, and no similar behavior is observed on FinTech lenders. However, there is little support that FinTech loans originated after demand shocks perform worse. These results suggest that the adoption of financial technologies allows FinTech lenders to meet local credit demand more efficiently.

Chapter 3 I examine the effects of social connectivity on the demand for and supply of consumer and small business loans on peer-to-peer (P2P) FinTech sites such as LendingClub. P2P loan demand increases when geographically distant, but socially connected areas have large amounts of past P2P borrowing activity. Both approval rates and quality (as measured by loan grade and interest rates) are higher the greater an area's aggregate online social connections. Performance (i.e., reductions in defaults or delayed payments) is enhanced by social connectivity indicating that information diffusion through online social networks improves lending outcomes for both high and low risk borrowers.

Share

COinS