Essays on Accounting for Financial Stability in the Banking Industry

The dissertation at hand focuses on the role of accounting in the aftermath of the 2007-2009 financial crisis. Particularly, concerns were raised about the impact of accounting rules, accounting discretions, and dividend payouts on bank behaviour. Consequently, the International Accounting Standards Board (IASB) and the Financial Accounting Standards Board (FASB) set forth their revisions to accounting standards. Furthermore, there is an increased intervention by both the supervisors and regulators. The first working paper titled "Dividends, Loan Loss Provisions, Lending: Early Evidence from European Banks" studies an important accounting choice by banks, that has triggered the change in accounting rules following the financial crisis. In it, I empirically examine whether the relationship between loan loss provisions (LLPs) and lending is influenced by banks’ dividend payouts. The joint effects of binding of capital requirements and financing frictions induce banks to reduce lending. However, curbing cash dividends is another way that banks could resort to enhancing their capital ratios. Nevertheless, banks have chosen to distribute considerable dividends and thus lower bank capital. Using the results of the 2014 Asset Quality Review in Europe to identify delayed expected loan loss recognition (DELR), I find that the greater payout banks experienced significantly larger lending reduction associated with greater DELR. Payout policy also plays a significant role in maintaining lending when banks’ equity experiences an adverse effect. The second working paper, "Do Financial Statements Inform of Bank’s Resilience: Evidence from EBA Stress Test?", proposes an accounting-based measure of bank’s resilience to complement banking supervision (co-authored with Joshua Ronen). The measure is based on fair value of net assets, defined as the difference between fair value of assets and book value of liabilities. Using the ECB stress test 2014, we find that within-country our measure is significantly and positively associated with the stress test results, measured as the scaled shortfall or surplus of common equity tier 1 (CET1) capital (under the adverse scenario). Going further, we compare the predictive ability of the loan component of our measure and stress test result in predicting one-year-ahead credit losses (proxied by net charge-offs). Results show that our measure performs equivalent to the stress test results in the full sample. Further analysis reveals that our measure performs better in the passed subsample while stress test results perform better in the failed group. We recommend using our measure as a complement to stress testing in identifying the bank’s resilience to reduce operational constraints on stress testing framework.