This study jointly evaluates the effects of the U.S. Treasury's Troubled Asset Relief Program (TARP), the Federal Reserve's Discount Window (DW), and Term Auction Facility (TAF) on bank syndicated lending during the 2007–2009 financial crisis, using a unique data set that tracks the exposure of each lender in each syndicated credit facility in each year. By comparing lending changes within a group of banks that lend to the same facility of the same firm in the same year, it eliminates the impacts of demand-side factors that often bias the results of empirical studies on bank credit supply. Overall, I find that TARP, DW, and TAF played only a marginal role in increasing bank syndicated lending. By examining lending changes at the facility-lender and firm-lender levels, this study is less prone to the reverse causality problem that is inherent in studies using bank-level data. Therefore, this study complements studies using bank-level data and provides policymakers with a balanced view on the effects of these programs.

We present an empirical study of stress testing for portfolios of auto loans. We find that loans aged five years or more have significantly higher default probabilities. This finding raises concerns about the increasing maturity of auto loans in recent years. A challenge in stress testing is the instability of the estimated coefficient of macroeconomic variables, which raises questions on the reliability of stress test results. For this reason, it is important for model developers to perform sensitivity analyses and make conservative adjustment to minimize model risk.

We present a comprehensive analysis to calculate the Basel III liquidity coverage ratio (LCR) and the net stable funding ratio (NSFR) of U.S. commercial banks using Call Report data over the period 2001–2011, and provide indirect empirical evidence on net cash outflow rates of certain liability categories. In addition, we examine potential links between Basel III liquidity risk measures and bank failures using a model that differentiates between idiosyncratic and systemic liquidity risks. We find that while both the NSFR and the LCR have limited effects on bank failures, the systemic liquidity risk is a major contributor to bank failures in 2009 and 2010. This finding suggests that an effective framework of liquidity risk management needs to target liquidity risk at both the individual level and the system level.