Michael Brosnan is one of Washington’s leading authorities on the systemic risk issues facing the national banking system. He serves as deputy comptroller for risk evaluation at the Office of the Comptroller of the Currency and chairs the agency’s national risk committee. He also oversees the OCC’s treasury and market risk division, which is responsible for developing regulatory policy and helping monitor trading activities, liquidity, securitization and interest rate risk at national banks. He spoke with editor Joe Kolman in September.

Derivatives Strategy: What are the main risks you see banks take on with derivatives?

Michael Brosnan: A lot of people have a misconception about how banks use derivatives. They think that most banks are making their money betting on positions or through proprietary trades. But the truth of the matter is that proprietary trading is a relatively small part of it, and, to the extent that meaningful positioning occurs, it is generally with cash-market instruments because of greater liquidity.

The biggest financial risk associated with bank derivatives dealing is credit risk. At the end of the second quarter, there was around $360 billion of credit risk in U.S. banks after netting. There are two sets of large numbers that represent the different elements generating this exposure figure. The first is the positive mark-to-market value for derivatives contracts—which is how much customers owe a bank at a particular point in time. That total number is approximately $150 billion. The second set of numbers estimates how much bigger the exposure might become over time. That’s called potential future exposure, and it’s approximately $210 billion. These figures are based on the formulas dictated under the risk-based capital framework. I think it is important to note that the amount of credit risk from over-the-counter derivatives transactions in the U.S. banking system would be much larger if not for the positive benefits of netting agreements.

In the aggregate, the credit risk numbers we come up with for potential future exposure using the admittedly imprecise risk-based capital framework aren’t far off from the calculations generated by internal bank models. That’s what gives me confidence that the Office of the Comptroller of the Currency’s quarterly derivatives report, which we make available to the general public, contains information that is reasonably reflective of the credit risk being taken on by U.S. commercial banks.

When you have tens of thousands of contracts, the errors tend to smooth out. I strongly believe that bank practices of measuring the add-on exposure are much more sophisticated and accurate than what we are using for regulatory capital purposes. But at the end of the day, it often works out to be quite similar.

DS: So you seem to be confident that the top dozen or so banks have their credit risk measurement under control.

MB: Yes. Our examiners have determined, and regularly affirm, that the credit risk measurement process is reasonably good at the large dealer banks.

DS: Is the methodology pretty standard or are there still significant differences?

MB: I’m not a quant, but if you gave each bank a 15-year swap, they would probably come up with slightly different numbers, depending on what data history they’re using, the confidence interval and so on. But the numbers are generally pretty close.

On average, the aggregate amount of credit risk from OTC derivatives is about three times the size of bank capital at these large dealer banks. The question is, since we know it’s an extremely big number, how well do these banks manage this risk? Along these lines, our examiners focus on the process from beginning to end—starting with an assessment of the bank’s customer selection process, how risk is measured, and trade and legal documentation, and then moving on to how the credit relationships are administered over time.

DS: How do the derivatives holdings compare with other bank holdings?

MB: Overall, the composition of OTC derivatives exposures are of high quality generally, and certainly in comparison with typical commercial loan holdings. When I first started working at the OCC, dealer-bank derivatives portfolios were probably an AA or AA- quality. Today, given the broad acceptance of risk management tools by many corporates as well as other financial institutions, banks have significantly expanded their target markets. But this has led, not unexpectedly, to a decline in overall portfolio quality so that it now averages A or A-. This is still a pretty good credit risk.

DS: Do you expect this migration in overall credit quality to continue?

MB: Yes. The customer base for derivatives has expanded well beyond the small pool of highly rated clients that dominated this business when I was first cutting my teeth on capital markets examinations. In the future, as more and more entities use derivatives or risk management purposes, it would not be unreasonable to expect increasing numbers of transactions to take place with the more typical commercial and industrial clientele—which is generally not investment grade.

DS: That always seems to be how people get in trouble—going down the credit scale.

MB: We have learned, time and again, that complacency, coupled with short-term compensation schemes, is a recipe for slippage and, eventually, losses. However, increasing risk per se is not bad in and of itself—the key is responsible risk management. So if banks continue to take on more business that is sub-investment grade, it will be essential that there be a commensurate pickup in controls, reserves, capital coverage, disclosures and so on. Absent a sound credit risk management framework, unchecked downward migration could be a problem.

I’m paid to scrutinize things that trend on 45-degree angles. Having said that, we don’t now see, or expect to see, a rapid decline in the quality of credit risk. But we are seeing, and expect to continue to see, further migration toward the typical quality of commercial loan portfolios. I don’t expect the average quality of a derivatives counterparty portfolio ever to reach that of a commercial loan portfolio. Derivatives portfolios should remain better.

DS: What questions do you usually ask banks about counterparty credit quality?

“Any time you have less-liquid positions being marked or otherwise controlled by traders, you have exposure because of moral hazards.”

MB: The customer selection issue becomes important from two major angles: First, how has the bank determined the client’s repayment capacity? Bankers need to do some good old-fashioned credit homework to answer this. Second, what steps has the bank taken to satisfy itself that the transaction is appropriate? This is linked with the assessment of repayment capacity. What is the purpose of the transaction? Does it fit the client’s risk management objectives as they are understood by the bank? These questions are particularly important because, as the leverage and complexity of a transaction increase, so does reputation risk.

There are some well-publicizedinstances in which dealers allegedly took part in transactions that were inappropriate for their clients, and those transactions subsequently became the subject of legal proceedings. In other cases, we have seen banks do their credit homework effectively but, after an adverse market move, some customers have leveraged off bank fears of litigation and publicity, and sought forgiveness of some or all of the amounts owed. This second angle is clearly complex and goes to the heart of the customer selection process, which includes developing some comfort that a “relationship” exists in which each party will perform as agreed. In addition, does the internal control process ensure that bank personnel conduct their business in a principled and ethical manner?

DS: I guess you’d call that traditional credit risk management—knowing who you’re loaning money to. But these days, there are all sorts of new techniques to manage credit using credit derivatives and securitizations of one kind or another. Will those techniques become more mainstream in the next few years?

MB: I think securitization is essential, particularly for the larger banks. The best performers will have securitization in their tool kit and will be able to use it effectively to manage liquidity, interest rate risk, capital positions and, to some extent, credit risk. But not all who have used this technique have used it well.

DS: Are there other areas in derivatives that concern you?

MB: I worry quite a lot about valuations. Every dealer out there has illiquid positions, whether they’re longer-term contracts, infrequently traded cash positions, or out-of-the-money options. Invariably, the question you ask is, “Where does the price, rate or volatility come from?” In most cases, the trader is the source. So we have to make sure that we always know what those transactions are—and that bank managers know what they are, so they are held out in the sunshine as a place of potential vulnerability.

That leads me to another subject. While I don’t have any major concerns about compensation schemes, we have to recognize that this is a mark-to-market business where traders get paid based on performance. One of the most obvious trends in the derivatives market is the rapid growth of longer-tenor transactions. Any time you have less-liquid positions being marked or otherwise controlled by traders, you have exposure because of moral hazards. If you pay bonuses in December based on up-front profits for a transaction that will last five years, you have to account for the risk of something going wrong. Our examiners pay close attention to the volume of positions for which the bank simply can’t get independent pricing factors, and how the bank validates the factors used. And we make assessments about whether the compensation program adequately aligns the trader’s interest with the bank’s.

DS: Can you give me an example of something you’ve noticed at a bank that made you uncomfortable and inspired some aggressive action?

MB: Probably the area where we’ve had the most concerns involves the quality of systems and management information. Many of the large dealer banks have long had geographically dispersed front- and back-office operations because of the global nature of their customer bases. Since these trading rooms were set up at different times and for varying reasons, few had common systems.

This resulted in major difficulties in measuring and communicating firm-wide risk levels for credit, market and liquidity risk, among others. Obviously, the line managers have shown they can effectively manage their respective activities, since they have come through a number of volatile market environments. But cost-conscious bank executives are not all that anxious to invest in technology and upgraded systems in advance of real-life examples of embarrassing losses that demonstrate why it is necessary. So fragmented systems make it quite difficult for non-traders, especially senior management, at many banks to truly understand the bank-wide risk profile arising from derivative activities.

When we issued our policy on modern risk management practices via Banking Circular 277 in 1993, we set a high standard for senior bank management and their boards. With the policy and resulting standards in place, we have pushed hard for improvements, particularly in the management information arena. Consequently, over the years since 1993, we have had quite a bit of success in getting banks to spend the necessary funds and commit the right personnel to these areas. While our job is not yet complete in some institutions, it is clear to me that without the pressure by our examiners, who are on-site in the larger banks, as well as the continued support from the comptroller, many banks would not have the value-at-risk, stress testing or credit risk management processes they now have. And I am certain that the banks in question would be much more vulnerable to adverse surprises than they are today.

DS: What about credit losses? What kind of counterparties have caused losses, and how were they rated?

MB: Many of the losses involve nonrated clients. The bank will rate them internally, of course, but the clients won’t have an independent rating. Foreign entities, in particular, often don’t have audits—no matter what GAAP they have. So, because of competitive conditions, the bank makes credit decisions, and accepts credit risk, based on lower-quality information.

“Without the pressure by our examiners, many banks would not have the value-at-risk, stress testing or credit risk management processes they now have.”

Then events occur—let’s say a currency devaluation—and you have some sort of economic turmoil that affects the industries and the customers within those regions. They can’t perform, so you do a postmortem. It’s not difficult to look back and say, “What the heck were you thinking of when you picked these customers?”

We had a big spike in charge-offs in the third quarter of 1998. We also had some in the fourth quarter in 1997. Net-net, losses from derivatives aren’t that great, but it also wasn’t ever expected that they would get that high. In most cases, the charge-off decisions were made before the customer even had a chance to go past due. It was simply known that they couldn’t perform, so the bank just charged it off, which is good accounting. But why do banks get into that box? Because they get complacent and overly aggressive in selecting customers during bull markets. This is really the same phenomenon we see in some aspects of the commercial lending business right now.

DS: Do you think banks that have had their knuckles rapped have learned from this experience?

MB: Yes. There’s a lot of explaining to do with charge-offs—first, because it’s real money, and second, because there’s publicity, and it’s not the type of publicity you want.

“Banks that have taken poor credit risks on the derivatives side suffer from us, but more important, they suffer from their own senior management and shareholders.”

When it comes to hedge funds and other highly leveraged investors, banks are getting better at gathering information and assessing it, but at the end of the day it’s important to have a really good understanding of who these people are and what their strategy is. It’s also important to have the right amount of collateral, because the trading books of these guys have the chance for rapid gains and rapid losses.

DS: Specifically, what do you think about credit derivatives as a tool to manage credit exposure?

MB: Credit derivatives exposure is relatively small when you look at notional amounts. You get a $220 billion round number for U.S. commercial banks. A few years ago, though, that number was zero. All we are counting is default and total-return swaps. We don’t count credit-linked notes because, in the Call Reports where this information gets reported, these are cash securities.

We are seeing the activity pick up both in notional amounts and in rhetoric, based on the number of calls that we get. Our examiners are asking questions about risk management expectations. And banks are asking, “How are you guys going to treat this from a risk-based capital basis?”

Let’s face it. Almost every firm has some form of credit risk it needs to manage. So, I think there is the prospect for significant growth in this market. But, having said that, the market is still small and it is confined to a handful of players.

DS: Do you think the use of credit derivatives should be encouraged?

MB: That’s a difficult question for a regulator to answer. Credit derivatives give credit managers more tools to use. We don’t want to appear to endorse derivatives, but we don’t discourage them either. I worry that eventually somebody will use credit derivatives improperly, and that we’ll get a backlash such as occurs periodically with other financial derivatives.

In 1996, we issued a banking bulletin outlining our risk management expectations on credit derivatives. We are now in the midst of working on a more detailed handbook. We take a neutral stance on these products, recognizing they are going to come into the market whether we like them or not. If banks use them properly, credit derivatives can address one of the oldest and most difficult problems in banking—credit concentrations. So I think credit derivatives have great potential to make credit portfolios more efficient. Let’s put it this way: I endorse people who use derivatives—including credit derivatives—properly. But if you don’t have the personnel, systems and technology in place to adequately understand, measure and control the risks, then my advice is don’t use them at all.

Was this information valuable? Subscribe to Derivatives Strategy by clicking here!