National Conference for Teachers of Advanced Placement Economics
Federal Reserve Bank of Richmond
It's a pleasure to be with you tonight. Bryce has welcomed you on behalf of the Powell Endowment. Let me add my welcome on behalf of the Federal Reserve Bank of Richmond. We're delighted and honored to have you here in our city, and we are very happy to play a role in this important conference. And when I say 'important' I really mean it. I've been an economic policy adviser and, more recently, a policymaker, in the Fed for almost 30 years, and I can tell you that from my perspective economic literacy - or more precisely, the lack of it - is a major problem in our country. Teachers like you truly are on the front lines in the battle to correct this problem. We at the Fed salute you; we're proud to know you; and we hope this conference will help you in the worthy work you are doing.
We are especially pleased to co-sponsor this conference with the Powell Endowment. The late Angus Powell, who created the Endowment, was the chairman of this Bank's board of directors a number of years ago, and I can tell you that no board member in my years at the Bank appreciated the crucial role economic literacy plays in fostering healthy economic policy and a healthy economy more than Angus. So this conference is a particularly appropriate way to honor him.
I believe the conference program indicates that I will be speaking tonight on the current economic situation from a policymaker's perspective.' With your indulgence, I would like to make a switch. After the program was printed and mailed out, I learned that my colleague, Governor Ned Gramlich of the Federal Reserve Board, would be addressing you tomorrow on fiscal and monetary policy, which obviously is a big part of any discussion of current economic conditions from a policymaker's perspective. So rather than subjecting you to two speeches on very closely related topics, I would like to focus my remarks tonight on a different topic: namely, the extraordinary recent developments in the banking industry and some major banking policy issues arising from these developments. As you know, the Fed is a bank supervisor and regulator as well as a monetary policymaker and, therefore, is vitally interested in banking conditions. And I have no doubt that you are interested as well, given the crucial role banks play in the economy and the fact that banks arguably are the most intriguing of all commercial enterprises for many Americans. After all, in the famous movie 'It's a Wonderful Life,' Jimmy Stewart played a banker; he wasn't an automobile company executive or the manager of a department store.
The trend toward increased concentration in the banking industry has been apparent for a quarter of a century now, and its impact has been profound. In 1980 there were 14,400 banks in the U.S.; today, there are only about 8,700 - a 40 percent decline. In 1980 the largest bank in the country, the Bank of America, had assets totaling $110 billion; today, the largest is the new Bank of America, with about $570 billion. Moreover, both the House and the Senate have passed bills that would repeal the old Glass-Steagall Act and permit the mixing of commercial banking, investment banking and insurance activities in the same company, which would broaden the concentration beyond banking to include much of the rest of the financial sector of the economy. Anticipating passage of this legislation, Citicorp, one of the largest banking companies, and Travelers Group, a huge insurance and securities company, have already merged into Citigroup.
To many Americans, of course, the most obvious sign of change in banking are all the new bank signs in our neighborhoods and towns. I saw a cartoon recently that showed a fellow replacing a bank sign as a bunch of other sign-changers waited in line for their turns. Proud cities like Dallas, Philadelphia and Richmond no longer have any major, large private banks based in their metro areas, while other cities - most notably Charlotte in this region - have become major banking and financial centers. Many individual bank customers now find themselves dealing with larger, and what they may regard as more impersonal, banks. They may face higher fees for some services. To be sure, there are other important current developments in banking - such as ongoing advances in the automation of banking transactions and services. But the continuing consolidation in the industry poses the greatest challenges for public policy toward banks, at least for now.
I've touched on the most visible aspects of bank consolidation. But the consolidation trend raises deeper and more fundamental questions that I think all Americans, not just bankers, bank regulators and a few economists, should be aware of. Specifically, what happens if one of the mega-banking organizations gets into trouble? And, in particular, how can we prevent such a situation from imposing large losses on taxpayers while still protecting depositors with government guaranteed FDIC insurance? In other words, how can we avoid a repetition of the S&L disaster of the late 1980s, which ultimately cost American taxpayers - that's you and me - about $130 billion, or $520 for every man, woman and child in the country?
Let me say something about FDIC insurance. It was established in 1934 in the wake of the Great Depression and the failure of 15,000 banks - almost half of all U.S. banks - between 1920 and 1933. Initially, the FDIC insured individual deposits up to $2,500; today up to $100,000 is covered. Deposit insurance has been hugely successful with respect to what most Americans regard as the bottom line: no insured depositor has ever lost a penny in a depository institution in the 65 years deposit insurance has been around.
All well and good. But as you economics teachers know perfectly well, economics is all about incentives. Economic incentives affect economic behavior and hence the way the economy works, for better or worse. And deposit insurance creates a problem with incentives. It reduces the incentive individual bank depositors otherwise would have to monitor the financial strength of their banks. In the days before deposit insurance, depositors routinely monitored their banks carefully. Because if their banks failed, they would lose all or part of their money. Further, because the banks knew their customers were monitoring them, they had a strong incentive to lend prudently and to keep themselves healthy. They knew that if they were not prudent, their depositors might withdraw their deposits, and in extreme cases possibly 'run' the bank as in 'It's a Wonderful Life.'
Deposit insurance, for all the good things it does, weakens these benevolent incentives. The depositor has less reason to monitor the condition of his or her bank, and banks have less incentive to be prudent, which puts the insurance funds, and hence taxpayers, at risk. Indeed, when an insured bank gets into real difficulty, it has a strong incentive to 'bet the house' and make loans that earn high interest but are risky. If the risky loans pay off, the bank wins and restores its health; if they don't, the bank fails and the insurance fund - and perhaps also the taxpayer - loses. Your deposit in the bank may be safe, but the bank's risk-taking may hit you in your wallet as a taxpayer. If you think this possibility is mainly academic, let me assure you that this is exactly what happened with many of the S&Ls in trouble in the 1980s.
Now at this point, some of you may be thinking: 'But wait. The $100,000 deposit insurance limit may be pretty high, but some rich folks and even some fairly small businesses may have more than that in the bank, which would not be insured. Won't they have an incentive to monitor the bank?'
Good question. In general the answer is 'Yes,' but there is still a problem. Specifically, the public has come to expect that the government will bail out the largest banks if they get into difficulty. This expectation even has a name: 'TBTF,' which stands for 'Too Big To Fail.' Historically, at least since the Depression, the government has stepped in rather than letting big banks fail outright, because there's been concern about the economy-wide, so-called 'systemic' fallout of a big bank failure. In particular, uninsured depositors and at least some other creditors have generally been protected from loss in these cases.
An especially prominent example of Too Big To Fail occurred in 1984 when fallout from weakness in petroleum and other energy markets strained banks exposed to this sector. One such bank was Continental Illinois, a big Chicago bank, which had a heavy concentration of loans to energy companies. Many other banks had correspondent deposits in Continental, and the regulators feared that the failure of Continental might trigger a chain reaction of bank failures. Ultimately the authorities assisted Continental rather than letting it fail. The Comptroller of the Currency at the time, Todd Conover, summed it up well:
'We debated at some length how to handle the situation. In our collective judgment, had Continental failed and been treated in a way in which depositors and creditors were not made whole, we could very well have seen a national, if not international, financial crisis the dimensions of which were difficult to imagine. None of us wanted to find out.'
The Continental bailout was a watershed in the recent history of bank regulation. Prior to the bailout, in the early 1980s, the FDIC had begun allowing uninsured depositors to suffer losses when banks failed. But the Continental bailout created a strong expectation, which persists to this day, that the government will not let large banks fail, and, most significantly, that it will make uninsured depositors as well as insured depositors of these banks whole when the banks get into difficulty. In these circumstances, uninsured depositors - at least of the large banks - like insured depositors, have much less incentive to monitor their banks and penalize them for excessive risk-taking.
What I have been talking about is a common-sense phenomenon which has a fancy name: 'moral hazard.' The basic idea is that if someone is insured against something, he may be less careful to avoid the kind of behavior that might trigger the event he is insured against. A standard example from bygone days is fire insurance. If you had fire insurance on your house, you were more likely to smoke in bed. (For the benefit of the younger members of the audience, some people actually did smoke in bed in bygone days.)
Let me give you another example outside the realm of formal insurance that I think all of you can appreciate - especially those of you who are parents of college students. Your daughter is off to college in September, and you have given her a certain amount of spending money for the entire semester. On October 15 she calls up; she's out of money. Do you bail her out? If you don't, you gain valuable credibility for next semester. But then again, she may have to get a part-time job and her grades may go down, which would diminish the value of your investment in her tuition. So you bail her out. The problem is your daughter figured all this out the day she left home. She saw you as her insurer. That's moral hazard.
The point here is that explicit or implicit insurance changes behavior. Moral hazard in banking is especially dangerous to taxpayers and others. Beyond our own S&L crisis in the U.S., the recent Asian crisis is an excellent example. Many economists believe that moral hazard in the banking sectors of key Asian countries contributed to the severity of the crisis, a crisis that produced sharp downturns in several Asian economies and even affected our own economy and financial markets.
We can agree, then, that moral hazard is a problem for the FDIC and ultimately for taxpayers. How can we deal with it in practice? Well, for starters, it is a good idea for the Fed to maintain price stability in conducting monetary policy. Inflation and disinflation can destabilize the banking system. The S&L crisis is often associated with deregulation in the 1980s, but its origins were in the inflationary monetary policy of the late 1970s. Once inflation was unleashed, the Fed was forced to confront it by tightening policy sharply, which drove short-term interest rates to unprecedented heights - 20 percent and above in some instances. S&Ls had to pay these high rates for the deposits they needed to fund the long-term mortgages they had booked earlier at much lower rates. The result was a negative cash flow for an extended period for many S&Ls that depleted their capital. It is largely because many S&Ls had little or none of their own money at stake by the mid-1980s that moral hazard became such a serious problem in the industry.
The upshot is that we might have avoided the S&L crisis altogether if we had contained inflation with monetary policy. But the low inflation we are enjoying currently will not insulate the banking system fully; moral hazard is a problem even when banks are well capitalized. For now, we confront it primarily in the standard way professional insurers try to manage their risks and protect themselves: by regulating the risk-taking behavior of the insured, and, in this context, 'insured' means the banks. The FDIC shares this task with the Comptroller of the Currency, an arm of the U.S. Treasury that regulates national banks, and with the Fed and state bank regulators. All of us send teams of examiners into banks, traditionally to inspect the quality of loans, but increasingly to inspect the effectiveness of the banks' own risk management practices and procedures. Further, we assess the adequacy of the bank's equity capital, which is the first line of defense should a bank have problems, so we can restrict the bank's activities - or close it - if capital falls to a point where the bank has little to lose by taking big risks: in other words, when moral hazard becomes very severe.
Supervision and regulation are certainly useful in countering moral hazard in banking. But they are far from a guarantee against excessive risk-taking, as we saw in the S&L crisis. This is not to denigrate either our examination staff or our examination procedures. On the contrary, our Federal Reserve Bank, the other 11 Reserve Banks around the country, and other bank regulatory agencies employ highly competent people who work very hard to monitor bank risk as accurately as possible. And we are taking innovative steps - with the help of recent advances in information technology - to make our processes more efficient and bring them more in sync with evolving bank practices. High quality, detailed supervision and regulation will always be a crucial part of the mechanism we use to contain bank risk, and I believe that currently, we are effectively monitoring and controlling banking risks using this traditional approach.
In the future, however, I believe it would be a mistake to rely solely on supervision and regulation to cope with the risks posed by banking organizations that are likely to become even larger and ever more complex, especially if the financial modernization legislation I mentioned earlier passes.
Here's why. First, to rely on supervision and regulation alone would be costly. We already have 150 employees in our examination department at the Richmond Fed. In the absence of any other changes we would have to enlarge our staff, probably substantially, to keep up, and other regulators would have to do the same. Moreover, we would have to hire more than a few specialists to monitor the increasingly sophisticated operations of large banks, in derivative markets for example. These people command higher salaries than we have paid historically. These high costs ultimately would be paid by you and me, either as bank customers or taxpayers, since bank examinations are funded either by fees charged banks, which they pass on to their customers, or from public funds. Some of these higher costs could be offset by increased efficiency in our supervisory operations and by sharing resources with other Federal Reserve Banks, but probably not all.
Second, the increased supervision and regulation required for us to meet our responsibilities would inevitably add to the already substantial regulatory burden imposed on banks. That, too, would be costly, not only in the direct sense I just noted, but also in terms of the increased costs of innovation as banks hire more accountants, lawyers and maybe even a few economists to ensure that new products will not raise regulatory issues. Bottom line: as I have already suggested, we are rapidly moving toward a situation in this country where relying on examinations and regulations alone to contain moral hazard in banking is just plain too expensive and too cumbersome. We need to supplement it.
Supplement it with what? Well, several proposals have been made, and I want briefly to describe a couple of the most prominent. As you will see, the common thread in these proposals is that they would change the structure of incentives in the relationships among banks, their depositors and the FDIC, by creating truly uninsured depositors and investors in large complex banks that would be exposed to appreciable losses if a bank became insolvent. These depositors and investors would have a strong incentive to monitor their banks closely and withdraw their funds early on if the bank began to take excessive risks, which in turn would motivate banks to avoid excessive risk-taking in the first place. This is where economic literacy comes in. To be accepted, the reform proposals I'm about to summarize must be understood and supported by the public. An economically literate public, thanks to you, can understand the crucial role incentives play in the banking system and the highly undesirable consequences of not structuring them properly.
The first proposal would require all large banks to fund a certain minimum proportion of their lending by issuing subordinated debt. I recognize that subordinated debt is not something you want to hear a lot about after a fairly heavy meal. But, briefly, subordinated debt is simply a type of bond that, in this context and in the event of difficulties, would be paid off last after all depositors, insured and uninsured, and other creditors were paid off. Because potential buyers of subordinated bonds would be least likely to recover their investment if a bank failed, they would be especially sensitive to the way in which their bank managed its risk. The proposal would require banks to issue subordinated debt several times a year. If a bank were not managing its risk well, investors either would not purchase the bonds or would do so only if they were paid elevated interest rates to cover their relatively high risk. Hence the bank would bear the cost of the increased risk. Further, problems in issuing subordinated debt would be a clear signal to regulators and others of possible problems at a bank. I am sure you can see how such a market signal could be a useful supplement to direct bank examination.
The second proposal has been made by the Federal Reserve Bank of Minneapolis, and it is aimed squarely at the Too Big To Fail problem. Under this proposal, all uninsured depositors would be required to suffer some loss when a bank failed, no matter how large the bank, but up to a maximum 20 percent of an individual depositor's deposit. The idea here is twofold. First, since uninsured depositors would lose a 20 percent haircut, they would have a much stronger incentive to monitor their banks than they do now, when they've been led to expect to be fully covered. Second, since uninsured depositor losses would be capped at 20 percent, bank regulators would be more inclined to let even a large bank fail, since they would be less worried about systemic, economy-wide fallout than if uninsured depositors lost all or a more substantial portion of their funds. In addition, the more credible threat that even a large bank might be allowed to fail would reinforce bank incentives to operate prudently.
In my view, both of these proposals merit serious consideration. And note that they are not mutually exclusive. We could do both. Both aim at using market discipline based on appropriate incentives to reduce moral hazard and the accompanying risks to taxpayers and the economy. But I need to say that neither of these proposals, nor others I have seen, are panaceas. Neither guarantees that the government will not decide for one reason or another to intervene in a particular situation and make all uninsured depositors whole. Obviously, if this happened regularly the reforms would lose their credibility, and incentives, behavior and moral hazard would all return to their original levels.
Moreover - and this is a somewhat less discussed but critical point - as you may know, the Fed can lend funds to banks on a short-term basis through our so-called discount window facility. This facility is a line of temporary credit that allows us to assist healthy banks when they face temporary liquidity problems; it is not designed to be used when a bank's solvency is in question. In practice, however, it is sometimes hard to draw a firm distinction between temporary liquidity problems and true insolvency in particular cases. In these circumstances, the Fed occasionally has lent to banks for fairly extended periods before they eventually failed. In some cases, this has given uninsured depositors and other creditors time to pull their funds out of a troubled bank leaving the FDIC to foot the whole bill when the bank ultimately went under. Here again, if we continued to do this, the proposed reforms would be undermined, and perverse incentives and moral hazard would remain serious problems.
Bottom line: if reforms of the sort I've described are instituted, regulators, including the Fed, will have to validate them by credibly refusing to bail out uninsured depositors and other creditors of large banks routinely. That won't be easy. In a crisis situation, bailing out a large bank may well look like the most prudent thing to do - and even a basic duty for regulators. Continuing this practice, however, by nourishing moral hazard and fostering excessive risk taking, over time could undermine the strength of our banking system and ultimately the economy. Because of this we must strengthen our resolve to allow at least some losses at large banks. To accomplish this, we need to build a broader public appreciation of the potential costs of increasing moral hazard and a wider recognition that some failures and losses are a natural feature of healthy dynamic industries, including banking.
Creating broad public support for a tougher but more balanced approach to dealing with problems at large banks will be challenging, to put it mildly. But I believe it is doable with an economically more literate public that understands the importance and power of viable economic incentives. That's where you come in. Hang in there. You're performing a great service, and again we salute you and welcome you to this conference.