Financial regulations are often born out of financial crises, a response intended to reduce the impact of future upheavals. A pertinent question for our times is: how much of a difference do they make?

Ryan LaFond and Haifeng You ponder this question in looking at a recent study of the Federal Deposit Insurance Corporation Improvement Act (FDICIA), which was created after the savings and loan crisis in the U.S. in the late 1980s. About 1,150 banks failed then, so the FDICIA introduced a number of reforms to improve financial management of banks. One of these, Section 36, aimed to reduce information asymmetry between banks and their stakeholders by improving the quality and oversight of financial reporting.

While the previous study, by J. Altamuro and A. Beatty, found Section 36 had positive results, such as increases in the validity of loan-loss provision and earnings persistence, the reporting requirement did not cover all banks. Only those with total assets greater than $500 million were required to comply. And there, argue LaFond and Yau, is the rub.

The exemption may have ostensibly been introduced over concerns about the cost to smaller banks, but it also reflected regulators' views that smaller banks are less important to the system. This implied that the findings of FDICIA's positive results could be attributable to a differing application because regulators believed certain banks were too big to fail.

The authors also point to the fact that disclosure of internal control deficiencies (ICDs) is voluntary under FDICIA, and that smaller firms are more likely to report ICDs than larger ones (this disclosure became mandatory under another piece of legislation born out of financial crisis, the Sarbanes-Oxley Act of 2002). Without knowing the full extent of internal control problems, it is not possible to link them directly to financial reporting quality.

"Whether internal control-related regulations cause the changes in financial reporting quality, or if instead the regulation and enforcement of FDICIA as a whole causes the changes, is open to interpretation," the authors say.

The authors bear all these factors in mind as they put Section 36 to their own tests. They are most interested to see if the impacts of the Altamuro-Beatty study could be due to changes within sub-groups, rather than the sample group as a whole.

They look at thousands of bank-year observations from 1986-1992 (pre-FDICIA) and 1995-2001 (post-FDICIA), focusing on two categories: loan-loss provision and earnings persistence. The results are mixed.

Improvements in loan-loss provision quality appear to be driven mainly by improvements in the financial reporting quality of complying banks (assets greater than $500 million), and no change or improvement for non-complying banks. On the other hand, changes in earnings persistence are driven by a decrease for non-complying banks and no change for complying banks.

The authors say the inconclusive results highlight the need to look at regulations in conjunction rather than isolation. For example, a more careful study of ICDs could focus on whether there were benefits from increased regulation under SOX or shortcomings in FDICIA's regulations and enforcement. "This could shed light on whether writing new regulations or enforcing old regulations is more effective," they say.

The findings also have relevance to the current financial crisis because of the likelihood of calls for further regulation. "One potential conclusion is that after 30-plus years of internal control regulations, internal controls are still lacking. However, a different conclusion is that there are limitations on what internal controls can achieve. They cannot prevent bad decisions or fraud, but they can make them more difficult to carry out," they say.

The authors also call for more research into the "elephant in the room" - the costs of regulations so as to get a fuller understanding of how to interpret and assess regulatory reforms.