Can We Predict Bank Failure in 2009 Using Financial Data?

• In Event: Financial Accounting and Reporting 11

Abstract

This study is to answer the question whether we can use a bank’s past financial data to predict bank fail in 2009. Using a bank failure list from the Federal Deposit Insurance Corporation (FDIC) database, I match the failed banks in 2009 with a control sample based on geography, size, the ratio of total loan to total assets, and age of banks. The model suggested by this paper can predict the bank fail 85.7% correctly and not fail bank 96.9% correctly and overall 93.4% correctly (p=0.5). Specifically, the predictors for bank failure suggested by a stepwise logistic regression are proxies for capital adequacy, assets/loan risk, efficiency, earnings, and liquidity risk.

Author

• Shirley Liu, Florida Atlantic University