Economic Brief

Bank examinations, like the recent "stress test," yield information of interest to the market. Releasing those results may increase transparency. For routine annual bank exams, however, doing so could impede a supervisor's ability to collect information.

Every year, bank supervisors in the United States collect information on the safety and soundness of banks. They examine each bank's assets, balance sheet, operations, and management. The information is summarized in a report that is then used to help determine if changes are needed at a bank or, in worst cases, whether the bank should be shut down. Not only is this process expensive to the government, it is even more expensive to the banks. They spend a lot of money gathering and reporting information to regulators. Anyone who does business with a bank or buys its securities is interested in having the information contained in the report.

Bank supervisors just finished a different form of bank audit. The 19 largest banks, all over $100 billion in assets, underwent a "stress test" to see if they had enough capital, loan loss reserves, and future earnings to survive a continued severe economic shock.¹ Banks that did not have enough of a "buffer" have been required to raise more capital, preferably from the public, but from the government if need be. Right now, investors are closely reviewing the results.

The stress test was a new sort of bank examination and it had a different goal than the regular annual exam. The test was part of a strategy of giving confidence to investors and creditors in the strength of the banks in the hope that this will encourage banks to lend more and contribute to an economic recovery. A routine examination, on the other hand, is geared toward protecting taxpayers from the liabilities that can result from deposit insurance.

Both exams contain information investors and creditors are interested in and they are expensive to conduct. The stress test results have been shared with the public, so why not share the routine exam results with the public too? The public won't have to duplicate the supervisor's work. After all, the information is a pure public good in the economist's sense – use of it by one person does not reduce the use by another.

This logic makes sense for the stress test. The purpose of the test and any ensuing capital injections is to provide confidence to investors, creditors, and counterparties. The only way to do this is to credibly disclose the information that the government learns.

But this logic does not necessarily make sense for routine exams. Edward Simpson Prescott lays out this case in a winter 2008 article published by the Federal Reserve Bank of Richmond in Economic Quarterly, "Should Bank Supervisors Disclose Information about Their Banks?"² In it, Prescott concludes that there are sometimes good reasons to keep the final report of routine annual bank examinations from the public. In fact, a case can be made that, in some circumstances, not only should the bank supervisor be forbidden from releasing this information but so should the bank itself.

The Methods of the Bank Examiner

To help put the above question into context, it's necessary to understand how routine banks exams are conducted. According to federal law, all banks must have a formal on-site exam conducted at least once every year, though under certain conditions those with less than $250 million in assets can be examined once every 18 months. On-site examinations are not the only form of direct supervision. Bank supervisors also monitor banks by analyzing a variety of data, often called "off-site surveillance." Furthermore, supervisors maintain offices in the headquarters of large banks throughout the year to allow them to generate a constant flow of information. In all cases, however, some of the information that supervisors use in their examination process is provided by the bank itself. The final exam report integrates the information obtained by all of these methods.

An exam is broad in its scope and is meant to rate the bank on the following measures:

• Capital Adequacy
• Asset Quality
• Management and Administrative Ability
• Earnings Level and Quality
• Liquidity Level
• Sensitivity to Market Risk

Combining the first letter of each component creates the acronym (CAMELS) that provides the name for the final rating awarded by the supervisor. Each component is assigned a rating of one to five, with one being the best and five being the worst. The components are then combined to create an overall CAMELS rating for the bank. The overall rating uses the same scale as the components.

The final report also contains detailed comments from the supervisor about the condition of the bank. As current law stands, the exam report is confidential and cannot be disclosed by the bank without the permission of the supervisor.

A Bank's Incentive to Disclose Information to Its Supervisor

Should supervisors disclose their information to the public? To answer this question, Prescott argues that first you need to figure out what incentives a bank has to disclose information to a supervisor. A supervisor cannot see everything that goes on in a bank, so is necessarily dependent on what information the bank provides.

He argues that if a supervisor discloses the results, the bank faces two costs to providing information to the supervisor: dealing with the supervisor and the market reaction. When the bank is doing well, this is not a problem, but when it is doing poorly it can be. The cost of the market reaction is an additional cost to the bank and if it is high enough, the bank will be reluctant to share its information with the supervisor. However, if the supervisor does not disclose the result, there is no additional cost, so the bank will be more willing to provide the information to the supervisor. This will be a problem precisely when the supervisor most needs the information. This is similar to the situation in many police television shows where the police officer does not want to reveal his source because doing so would cause that flow of information to dry up.

Voluntary Disclosure Is the Same as Mandatory Disclosure

What about voluntary disclosure? Why not let banks who want to disclose do so? Other banks could keep their information quiet. Prescott argues that voluntary disclosure is the same as disclosure by the regulator.

Why is this the case? Imagine yourself as the CEO of a bank that is doing well. If you do not disclose the information, the market will lump you in with the other banks, which includes some not so good ones. But if you disclose the result of your report – and we are assuming that you cannot issue a fake report – the market will know that you are one of the "good guys." Wouldn't you wave your good CAMELS rating around? Of course you would and so would every other good bank.

This leaves the weaker banks as the only ones not disclosing, but even then the strongest of the weak are going to want to avoid being lumped in with the really weak, so they disclose too. In the end, everyone basically discloses, except maybe the weakest.

What does this mean for supervisory disclosure? It means that if the supervisor does not want to disclose its report then it better stop all the banks from disclosing. In fact, they prevent banks from disclosing CAMELS ratings by force of law.

It is worth noting that the same can be said for the results of the stress test. If the government allowed the banks to release the results on their own, the same unraveling would occur. Of course, as we argued earlier, the point of the stress test was to get the information out to the public, so whether it came out voluntarily or mandatorily should not really matter. It seems simplest to just have the government release it all at once, which is what happened.

Conclusion

The analysis above describes the tradeoffs to disclosure of bank examination reports. The main argument is that if supervisors need the cooperation of a bank to receive information, disclosure will increase the cost of cooperation to the bank. This increased cost either reduces the quality of information the supervisor receives or it requires the supervisor to spend more of his resources collecting the information.

Whether these reports would be released by the supervisor or the banks doesn't matter. Either disclosure scenario makes it costlier for a bank to be completely honest with the supervisor. Thus, the only scenario in which the supervisor receives reliable information during the audit is if neither he nor the bank is legally able to release the exam rating at the end of the process.³

In other words, it's important to remember that routine disclosure of information can negatively affect the ability to collect it. In the world of financial regulation, disclosure is often viewed a bit like baseball, motherhood, and apple pie – you can never have too much of it. But as this article argues, when it comes to disclosure, more is not always better.

Edward Simpson Prescott is a senior economist and vice president in the Research department of the Federal Reserve Bank of Richmond. Stephen Slivinski is senior editor of the Bank's quarterly magazine, Region Focus.