Although a large body of literature has proposed various models to identify an impending financial crisis by studying systemic risk and contagion, scarce previous research has considered the possibility that banks can protect themselves during a financial crisis and therefore affect the propagation of losses through financial linkages, such as the interbank market. Drawing upon a subset of U.S. bank accounting ratios from 1992Q4 to 2011Q4, the thesis investigates banks’ preemptive actions by analysing significant structural shifts in response to crises at the aggregated bank level. We’ve found Bank size does matter in context of applicability of banking accounting ratios serve as early warning signals. The results show that certain indicators such as ‘leverage’ and ‘coverage’ ratios are appropriate indicators for the detection of banking system vulnerabilities all banks. And nonperforming loans ratio (NPLs) additionally serves as an indicator for the timing of a crisis. The thesis also finds that whereas capital levels were closely monitored, heavy reliance of banks on wholesale funding is often overlooked. Banks accumulate liquidity to protect themselves from liquidity shocks and therefore contribute to (or mitigate) the onset of a crisis. Therefore, the impact of bank size and interbank lending on bank risk-taking are carefully examined; and a nonlinear (U-shaped) relationship is found. It adds empirical weights to the ‘too big to fail’ phenomenon. In addition to this, preemptive actions of large banks are found in the interbank market during the financial crisis. In other words, interbank lending is associated with substantially lower risk taking by borrowing banks in financial crisis, which are consistent with monitoring by lending banks. Finally, the thesis considers banks’ liquidity creation during the interbank lending crunch. The author finds those same factors leading to precautionary liquidity hoarding also contributed to a decline in interbank lending: banks with net interbank borrowing positions rationed lending due to self-insurance motives and they offered higher rates to attract external funding; net lenders hoarded liquidity due to heightened counterparty risk. The author also proposes two on-balance proxies for liquidity risk: (i) the unrealized security loss ratio and (ii) the loan loss allowance ratio. Banks choose to build up liquidity in anticipation of future expected losses from holding assets. On the policy frontier, besides credit and securities lending programs targeted at the interbank market, the author proposes interbank lending subsidization.
|Date of Award||22 Mar 2018|
|Supervisor||Ania Zalewska (Supervisor) & Andreas Krause (Supervisor)|