Essays on financial contracts and financial intermediation

Abstract

In the third chapter, I adopt the stochastic frontier analysis (SFA) to evaluate banks' relative lending performances and examine why some banks make greater profits in the corporate loan market. The uniqueness of SFA is that it allows us to construct a "frontier" indicating the maximum loan price that a bank can charge for any given combination of borrower and loan characteristics. The difference between any observed loan price and the "frontier" price would be the result of banks' inefficiency in lending. I relate banks' differences in lending performance to market power, focus versus diversification lending strategies, syndicate partner selection, and lending style, controlling for banks' governance and rating level and applying bank fixed effect and year fixed effect in our estimation models. Using a sample of 22,868 loan facilities lend by 118 bank holding companies between 1988 and 2009, I find that banks with higher market power and focused lending strategy in the bank loan market perform better. In addition, banks that choose a syndicate partner who has more similar lending specialization perform better. Finally, the way banks allocate their lending portfolios has a significant impact on their performance. On average, banks' lending performances are better if they allocate more weight on first-time loans and risky rating group loans.

In the second chapter, I examine the effect of information intermediaries, specifically financial analysts, on bank loan contracting. I find that higher analyst coverage is, on average, associated with lower bank loan costs. The cost reductions are more pronounced for loans that require more intense due diligence and monitoring (i.e., first-time loans, loans from one lender, loans to small borrowers, or loans to borrowers without credit ratings). Controlling for the size of analyst coverage, more experienced analysts have stronger effects on the cost of bank loans. I also show that higher analyst coverage is associated with shorter loan maturity, less restrictive covenants, a lower likelihood of collateral pledges, and smaller chances of having performance-pricing provisions. For low quality loan, high analyst coverage reduces the concentration of syndicate structure.

Over the past decade, the use of bank loans as a source of financing by banks has become pervasive and is growing in importance. In the first chapter, I examine the factors that determine the cost of this particular funding source of banks. Specifically, using a sample of 1,854 syndicated loans issued to 530 commercial banks from 1995 to 2009 in 42 countries, this paper reveals that the cost of loans is significantly influenced by bank regulations, market structure, institutional qualities, and their relative differences in borrower and lead lender countries. The results further demonstrate that prudential bank regulations are effective in reducing banks' funding costs, especially when the borrowers are relatively risky and in need of more intensive monitoring. Additionally, prudential regulations reduce banks' borrowing costs more when the banking industry is highly concentrated or country has weak institutional system.

This dissertation consists of three distinct but related essays in financial contracts and financial intermediations. The first essay examines how bank regulations, market structure, institutional qualities, and their relative differences between borrowing and lending countries impact the cost of bank-to-bank loans. The second essay assesses the important role of information intermediaries, specifically financial analysts, in the bank loan market. The third essay develops a new performance evaluation matrix in the syndicated loan market and examines the determinants of banks' lending performance.