Thank you for that kind introduction, and thank you, ladies and gentlemen, for joining me this afternoon.
I noticed in your conference program that my talk today rates an “SK”—for specialized knowledge. I confess that I am flattered—because it strikes me that you’re the ones with the SK that really matters. Professor Alfred Rappaport, one of the leading proponents of the shareholder value theory of management, has observed that while cash flow is a fact, profit is an opinion. If he’s right—and I think he is—then ultimately it is your opinion that’s behind much of the strength we’ve seen in the consumer and business sectors. Perhaps we ought to switch places.
Another place where it helps to have some specialized knowledge is up on Capitol Hill. As you know, the 106th Congress has been considering financial reform legislation for some months now, and I am pleased that the financial reform bill will be signed into law.
Still, financial modernization legislation occupies just one place along a continuum of reform. To be sure, it is the most important one, as the public policy choices made by lawmakers can impede or enhance the operation of financial markets. But today, I’d like to look beyond any particular reform measures and to consider the continuum itself. I’d like to think about the way financial markets have changed in the last few years and what those changes suggest about the principles that ought to guide lawmakers and regulators at the Fed, the OCC, the OTS and the FDIC and in the states.
So I’ll begin with the observation that the financial services marketplace moves much more quickly than lawmakers and regulators. In the last year or two, the most obvious example has been Citibank’s merger with Traveller’s Insurance. That deal proceeded on the assumption that the underlying legal and regulatory structure would be amended so that the insurance operation could be retained. The market acted—and policymakers reacted.
But the truth is that this is not a particularly new development—because the market began undermining the American financial regulatory structure nearly a quarter century ago.
Beginning with the Great Depression, the central objective of American financial policy was preventing competition between financial sectors. The Glass-Steagall Act erected legal firewalls between the sectors, and regulators maintained the firewalls by reducing competition within sectors. If banks didn’t have to compete too hard with other banks, for example—if they were more or less guaranteed a geographic market—then they wouldn’t have to compete in other nonbanking financial markets.
But the firewalls between financial sectors began to crumble quickly in the 1970s, when data processing and information technologies made it possible for consumers to do something about the opportunity costs presented by high inflation and interest rates. The unveiling of Merrill Lynch’s cash management account in 1978 was a signal event. Previously, if you wanted to earn interest on your deposits, you had to be wealthy enough to open an offshore eurodollar account. With the cash management account, smaller depositors were able to earn market rates of interest through money market mutual funds. The upshot, of course, was that consumers left banks in droves, and policymakers were eventually forced to abandon Regulation Q and to permit banks to offer interest on most types of retail transaction accounts.
Twenty years later, we know that the Reg Q capitulation was the first of many to follow, and today its significance seems clear: market developments undermined public policy. It’s a pattern we’ve seen over and over again with the demise of interstate banking prohibitions, the end of interstate branching prohibitions, the expansion of Section 20 powers and, of course, the impending enactment of the financial reform legislation.
Still, it’s not enough to understand that the financial services industry has changed for good. If supervisors are going to continue to pursue safety and soundness—as I’ll argue shortly that they should—they also have to understand why things have changed. Specifically, they must understand the role technology played in the transformation of financial services.
At its most fundamental level, technology has allowed financial institutions to split financial claims into their constituent parts. Credit risk, interest rate risk, currency risk—all can now be assessed, divided, priced and sold to investors willing to incur them. Now, obviously, the technology that enabled credit default swaps has evolved significantly over the technology that allowed Merrill to divide Treasury bills among smaller investors. Still, the result is the same: Institutions that were legally distinct—and therefore restricted to offering products that were also legally distinct—now are able to offer products that are functionally equivalent.
To summarize, then: New technology transformed financial products; new products transformed financial institutions; institutions transformed financial markets. And markets—they simply overwhelmed many long-standing policies.
In this technology-powered, market-driven environment, it would be tempting to conclude that supervisors are irrelevant—tempting, but wrong. For one thing, the market’s rules and incentives are not necessarily aligned with the interests of taxpayers and consumers. For another, the market’s corruption-prevention record is less than stellar (although, in fairness, it should be mentioned that supervisors’ record is not much better).
But by far the most compelling reason for regulation is the one offered by Senator Glass and Congressman Steagall 66 years ago: the safety and soundness of the U.S. financial system. We need not go back very far to remind ourselves that shocks in the financial system can adversely affect the real economy: In 1987 and again last year, the FOMC was compelled to adjust monetary policy to shield the real economy from domestic and international financial shocks. Ongoing supervision and regulation is essential both to reduce the probability of future shocks and to help us manage those shocks that do occur.
So the first challenge for the Fed and other supervisors is to redefine the term “safety and soundness.” The goal, I suspect, will always be the mitigation of systemic risk. But it may no longer be possible or desirable to mitigate systemic risk by preventing the failure of individual financial institutions. Containment, I believe, will be the preferred approach, but even that’s easier said than done.
The second challenge is how we work, or how we do the containing. Traditional supervisory approaches—as we’ve already established—cannot keep up with innovations in the financial services marketplace. Even the latest regulatory innovation, the 1988 Basel Accord’s risk-based capital model—which took so many years to gain broad acceptance—is completely inadequate as a gauge for determining capital requirements for many very large institutions.
The question, then, is how examiners—equipped with paper tools in an electronic age—keep up with the market. The answer is that we don’t. The answer, I think, is that we enlist the market to help regulate financial firms by shifting most of the risk of failure back to the private sector. But if markets are to reliably safeguard behavior, then banks need to become more transparent. Because if private financial markets and institutions are going to bear a greater risk of failure, they need a much clearer picture of the potential risks they face from other market participants.
If you read the British magazine the Economist, you know that an economic practitioner can play any one of five oral trump cards to win an economic argument. And “transparency,” the Economist says, is one of two jokers in the deck. Now, as an engineer who’s risen through the Federal Reserve System, I confess that I admire many of my colleagues’ facility with economic trump cards. But I also have to admit that there’s something to transparency. So for purposes of keeping the relatively simple relatively simple, let me suggest another way of thinking about the dilemma we bank examiners face and what I mean by transparency.
Suppose the Federal Aviation Administration decided that the best way to bring order to the skies was to move its air traffic controllers from the towers to the planes. Instead of hunkering down before banks of radar screens that monitor the real-time location of every craft in the sky, controllers would ride one to a plane, keeping a sharp eye on the instruments and the sky—and hoping for the best. Well, I’m not an aviation historian, and as far as I know, the FAA has never advocated this approach. Still, in the very early days of aviation—when air traffic meant migratory birds—it might have worked.
Ladies and gentlemen, the American financial system is just as dynamic and innovative and fast moving and technologically advanced as the American civil aviation system. And yet, in 1999, we are trying to bring safety and soundness to the financial system the way we were half a century ago: by riding shotgun with the banks. It simply won’t work any more. It’s not just that there are so many more airplanes in the sky or that they can go anywhere in the world. It’s that the market gets bigger and faster and more complicated every day.
What supervisors ought to do is get out of the cockpits. We cannot and should not stop an institution from failing if it isn’t fundamentally sound (though I should add that with deposit insurance, we can still save a few of the passengers). As for the safety and soundness of the financial system as a whole, the best thing we can do is help the market police itself. We do that by ensuring that all institutions disclose where they are and where they’re going. That, of course, is the job of supervisors, and while it will still require on-site examinations, the model of transparency I’ve just described would represent a philosophical shift—out of the cockpit.
Still, I do recognize the limits of my analogy. I realize that while establishing aviation transparency may be as simple as erecting radar and installing locator beacons, assessing the condition of financial institutions in markets that change by the second is a far more complicated matter altogether. How, indeed, do we go about establishing transparency in financial markets?
I think it makes sense to start with the easiest question: who directs traffic? Obviously, that’s the job of the supervisor. But here I think it’s worth emphasizing one thing: directing traffic is not the same thing as determining who goes where or when and how parties depart or arrive. All of those are market decisions. The job of supervisor is to know the location and condition of every market participant and to help them avoid each other—to avoid a systemic collapse—during the voyage.
Well, if it’s our job to know where market participants are, the next issue is whose assessment—or whose coordinates—to use. Or to put it another way, who knows better about an institution’s condition: the supervisor or the institution? Increasingly, I think the answer is the institution. Here again, the dilemma for supervisors has to do with the speed of markets and the rapid evolution of technology. At large, complex banking organizations—our clever acronym in the S&R world is LCBOs—risk changes by the second. At the Fed we are just beginning to move away from linking capital requirements for LCBOs to so-called “risk buckets.” We will instead increasingly rely on an institution’s own internal risk assessment models.
Which is not to say that this approach will work for every bank. In the first place, most banks do not require and have not developed internal risk assessment models: only the biggest banks—those that hold the largest share of the nation’s assets—do. In the second place, internal risk assessment models will not be sufficient in the absence of external review. Until banks establish a more formal system of external review for their proprietary risk assessment models, supervisors ought to continue to require that banks provide additional information about their activities to the market.
To summarize so far: Because markets move so much faster than supervisors, safety and soundness ought to become more a function of markets, and less a function of supervisors. The job of supervisors will be to require transparency—to ensure that markets get the information they need from individual institutions.
The third question, in my opinion, is the most difficult one: What information constitutes transparency? Put another way, what information facilitates market discipline? There are two issues here. The first is finding the right balance between releasing information that helps the market evaluate risk and not requiring disclosure of information that might help competitors. I concede that this could be difficult. In the case of LCBOs, for example, it’s possible that public disclosure of an institution’s own internal risk assessment could reveal something about the models used to produce that assessment, which could in turn reveal something about the market position of that institution.
Still, the interest-balancing issue of the information disclosure issue pales in comparison to the second issue, which has to do with the quality of the information being disclosed.
The biggest banks in the United States are a conglomeration of dozens of separate and distinct financial businesses. They operate 24 hours a day, seven days a week in hundreds of countries around the world. Their financial condition changes by the second. The critical issue for supervisors and accountants alike is assessing the condition of these institutions on an ongoing basis. To borrow the air traffic control analogy one last time, the way things stand today—with static, paper balance sheets that present the condition of an institution at a single moment in time—banks may as well be transmitting their bearing and location every two or three hours. This is, to put it mildly, not particularly useful information.
If we’re serious about the marketplace regulating the financial services industry, then the market’s got to have better information about the institutions it’s policing. Institutions that fly close to the envelope, with minimal capital for the level of risks, ought to disclose their condition more often. Institutions that are well within the envelope, with lots of capital relative to risk, may be monitored with much less frequency and with greater reliance on after-the-fact reviews. The question is how we do this.
Ultimately, it seems to me that this is a question best suited for the accounting profession. But until a more permanent solution begins to materialize, I think that some items are basic enough to begin to fill the bill. I’m thinking in particular of information on the fair value of bank portfolios, internal credit ratings and risk management practices governed by fair value. Markets might also be better served through disclosure of those categories in which an institution faces credit risk, the general nature of credit composition and the potential exposure an institution faces in its securitization practices.
All of this is probably all too familiar to you, as yours is fundamentally the business of assessing information and passing it along to markets. You’ve been at this a very long time, whereas we supervisors are really just beginning. As we increasingly rely on markets to bring safety and soundness to the financial system, we stand to learn a lot from the accounting profession.
Thank you for your attention.