Three essays on traditional and nontraditional banking and portfolio decisions

This dissertation proposal contains three coherent essays in the field of banking. The first paper aims to understand whether and how banks' shift from traditional to nontraditional income sources contributed to the failure of hundreds of U.S. depository institutions between 2008 and 2010. A key ingredient of our methodology is the recognition that different fee-generating or noninterest activities have different production and risk-return characteristics, and hence are likely to have different impacts on the probability of insolvency. We find that even after controlling for the traditional drivers of insolvency risk identified in past studies nontraditional banking activities had economically meaningful effects on the probability of bank failure during the crisis. Nontraditional fee-for-service income (insurance sales, securitization, loan servicing) helps commercial banks preventing failure. Yet, nontraditional stakeholder income (investment banking, venture capital) is positively associated with the likelihood of failure, particularly at distressed banks. Focusing solely on traditional activities, the second research paper derives a theoretical model and investigates the loan portfolio decisions at small commercial banks that are subject to capital market imperfections. The findings indicate that net business lending at smaller banks increases with the expected profitability and risk tolerance, generally decreases with the illiquidity and the size of overhanging business loan stock, and is responsive to cross-sector return covariances. The effects get stronger during the financial crisis, which is consistent with the binding market imperfections during the financial turmoil. Taking into consideration the findings and the implications of these two essays jointly, the last essay investigates what impact illiquid stake-based asset positions have on loan portfolio and business lending decisions at bank holding companies that actively integrate traditional and nontraditional banking business models. The results show that illiquid overhanging stakeholder assets lead banks to reduce business lending and contribute to the pro-cyclicality of business loan supply. Consistent with the theory, the effects get stronger as the market imperfections get severe.