Conclusion
The recent crisis has highlighted the limitations of fixating on ROE as banks’ central performance measurement metric. In particular, critics argue that banks have tended to increase ROE by increasing financial leverage and undertaking risky lending and other non-traditional banking activities. Accordingly, we test whether investors can better screen bank stocks by employing fundamental analysis in addition to using traditional summary measures of profitability. We ex-ante identify fourteen bank fundamental signals related to overall profitability, components of profitability, prudence in banking practices, and growth, to create an index of bank fundamental strength (BSCORE). We first document that BSCORE is positively associated with one-year-ahead change in profitability measures (ΔROE and ΔROA), over and above the current profitability changes. Further, the stock market only partially incorporates the information in BSCORE in current returns, leading to a positive association with future returns. A long-short strategy based on deciles of BSCORE yields positive industry-adjusted hedge returns of 9.9% in the 1994–2014 period. The results are consistent across a variety of partitions related to information environment and implementability. Inconsistent with a risk-based explanation, positive hedge returns are obtained for all but two years during the sample period, and the hedge returns are especially strong during the financial crisis period. The hedge returns persist after controlling for a variety of potential risk factors in asset pricing tests. Lending credence to a mispricing-based explanation, we observe a positive relation between BSCORE and both future analyst forecast surprises and excess returns around subsequent earnings announcements, and a negative relation between BSCORE and future performance-related delistings.
