Does Accumulated Other Comprehensive Income “Accumulate” Risk?

Abstract

Our paper links the dynamics of other comprehensive income (OCI) with crash risk. Our study is motivated by the fact that crash risk research in the banking sector is sparse and the findings about the relevance of other comprehensive income (OCI) are inconclusive. We focus on AOCI because, first, it is the accumulation of different market risks that banks are exposed to, and second, it contains information about managers’ discretion and banks’ opacity. Using a sample of COMPUSTAT banks, we test whether the market risks and managers’ discretion contained in AOCI have predictive power on future crash risk. By comparing banks and non-banking firms, we find that AOCI is predictive of stock price crash risk in one-year-ahead for banking sector, whereas such association does not exist for non-banking firms. Then we show that the strength of this association varies. We find that the association between AOCI and crash risk in the banking sector is more pronounced when banks have extreme OCI or medium AOCI and when banks suffer more accumulated unrealized losses. We provide evidence that the significant association between AOCI and future crash risk only exists for commercial banks, not for savings banks. We find that the association between AOCI and future crash risk is more pronounced when commercial banks operate in extreme interest rate environment, have higher leverage ratio, and have more opacity. We further find that the association tends to be weaker when external monitoring is highly effective. Our paper contributes to a better understanding of the risk relevance of OCI and the unique effects of financial reporting on crash risk in the banking sector.