How Does Internal Control Regulation Affect Financial Reporting?

In our paper How does internal control regulation affect financial reporting? which was recently accepted for publication in the Journal of Accounting and Economics, we examine the financial reporting effects of the Federal Depository Insurance Corporation Improvement Act (FDICIA) internal control provisions. The internal control provisions of the FDICIA provide exemptions that allow us to separately identify the effects of these provisions. In particular, the FDICIA exempts institutions with assets less than $500 million from its internal control monitoring and reporting requirements. More specifically, these institutions are exempted from FDICIA’s requirements that management issue a report on the effectiveness of internal controls over financial reporting, and that their independent public accountant attest to management’s report.

We compare annual and quarterly financial reporting of banks affected by FDICIA’s internal control provisions to that of unaffected banks. Specifically, we examine changes in the validity of the loan-loss provision, earnings quality, benchmark-beating and accounting conservatism. We analyze two samples: (1) a sample of US public and private banks included in the Fed Form Y9-C Regulatory Filing database and (2) a sample of publicly-traded banks included in the COMPUSTAT database. Our difference-in-differences research design isolates the effects of the FDICIA internal controls provision by controlling for changes in financial reporting unrelated to those provisions. We validate our control samples by testing for differences between the affected and unaffected firms in the pre-regulation period.

We compare the change in financial reporting for our affected and control firms in the 7-year periods before and after the passage of FDICIA. First we examine the properties of the annual financial reports. We find that the FDICIA-mandated internal control requirements lead to improvements in the validity of the loan-loss provision. Specifically, the association between the loan-loss provision and actual loans written off for affected banks strengthened in the period after the enactment of FDICIA. This improvement addresses the GAO’s (1994) concern “that banks’ loan-loss allowances included large supplemental reserves that were not linked to analysis of loss exposure or supported by evidence.” We find a corresponding increase in both earnings’ persistence and ability to predict cash flows, and a reduction in the use of earnings management to report positive earnings growth, suggesting that reducing supplemental reserves generally improves reporting quality. However, we also find that earnings conservatism declines for affected versus unaffected banks in both samples. This reduction in conservatism is also consistent with a reduction in supplemental reserves.

Next we examine the properties of quarterly reports to determine whether the effects are larger in the interim quarters relative to the fourth quarter, when an increased auditor presence might substitute for improved internal controls. Consistent with this hypothesis, we find that the improvements in the validity of the loan-loss provision, and the increase in earnings persistence and predictability of future cash flows, are all larger in the first three quarters than in the fourth quarter.

Taken together, these results suggest that the FDICIA-mandated internal control provisions resulted in the average bank exercising less reporting discretion. This reduced discretion creates a greater association between current reported accrual numbers and future cash flow numbers. However, as a result of this improved association, the reported accrual numbers also became less conservative. Thus, the conclusion about how this regulation affected the quality of financial reports depends on one’s definition of quality.

The full paper is available for download here.