Though I am a guest scholar at Brookings, I do not of course speak for Brookings, or for anyone other than myself.
I should like to answer your questions first, and then propose a somewhat different set of priorities for the reform of banking legislation, looking forward to the realities of financial services in the next century rather than backward to the debates that have been with us for sixty years.
I am opposed to affiliations between banks and nonfinancial entities, and quite belligerently opposed as long as the banks continue to receive the support of a federal safety net administered by banking regulators. The Federal Reserve demands “entity regulation” because bank regulators must supervise everything that touches on the payments system. But the failure of a nonfinancial enterprise owned by a bank or a bank holding company would also endanger the “confidence” depositors and lenders must feel when they provide funds to a bank. Thus the bank would turn itself inside out to support a failing subsidiary—and would do so with the support of the banking regulators, for the bankruptcy of a bank-owned company would surely spook some part of the community of funders. What we would find, indeed, is that the banking regulators would eventually demand to supervise the nonfinancial as well as the financial subsidiaries of the bank or bank holding company. The breaches in the firewalls that always occur in a crisis—remember Continental and First Options—would quickly make the bank regulators interfere in the operations of non-bank affiliates or subsidiaries.
The tale of what the examiners were doing in Citicorp during the gory years of 1990-1992 touches on much more than the operations of Citibank, which did not by any means want to shed all the affiliates it sold. The Banque de France got involved with the non-financial businesses of Credit Lyonnais up to and including MGM Studios, and the Banco d’Espagne was plagued with problems relating to the nonbank holdings of Banesto. One of the reasons the Japanese banks are in such desperate straits is the tight linkage formed between them and their keiretsu partners following MacArthur’s breakup of the zaibatsu.
Closer to home, it would be worth your while to call up the data on the Bank of the United States in San Diego and its affiliation with the Westgate projects in that city. Or the sad history of American Savings & Loan of Stockton, CA., and the Federal Home Loan Bank Board’s close and ineffective supervision of all its activities. Or the sad story of Charley Keating. One notes as the mark of Cain on the current proposals by the Comptroller the fact that one of the first applications by a bank seeking to go into a non-banking business was for authorization to enter the field of real estate development, where the business cycle closely tracks the business cycle of banking, implying an intensification rather than a diversification of risk.
Deposit insurance, of course, is a small part of the safety net. In the big banks, insured deposits are a minor fraction of total liabilities. But the safety net is under the entire bank, which will mean more than just the bank if you have entity regulation. Despite FDICIA, the banking regulators still believe in their hearts in the doctrine of too big to fail, first enunciated in hearings I believe before this committee by Mr. Ludwig’s predecessor Todd Conover, and enthusiastically endorsed by L. William Seidman when he was chairman of the FDIC. When First Republic of Texas was failing, Mr. Seidman sent its chairman a letter to be published, announcing that the FDIC would pay back everybody who put money into First Republic whether the form in which that money was received was a deposit or any form of loan. When National Bank of Washington collapsed, the FDIC bailed out all the creditors of its Bahamas branch—which had never paid deposit insurance premiums. Several billion dollars of FDIC money was spent to pay off non-insured creditors of Continental-Illinois.
Over and over again, losses to the deposit insurance fund (and potentially to the taxpayer) have been heightened by the habit of the Federal Reserve to take all the good assets out of a failing bank as collateral for the discount window loans that keep the place afloat after it should have been permitted to sink. The prompt closure commands of FDICIA are by no means accepted at the Fed to this day, and it should of course be kept in mind that the bank examiners can determine whether or not a bank has sufficient capital to continue in business by the valuation they assign to the assets. I print in my book the change in the procedures for valuing commercial mortgages that Robert Clarke instituted when he became Comptroller.
It is in part because I think the Fed should be the regulator of the holding company that I believe the mix of banking and commerce should be prevented. In addition, the securities company in the holding company must be separated out for regulatory purposes and ruled by the SEC. If the Fed believes that the failure of the securities affiliate could cause systemic trouble in the banking system, then a tightening of the Glass-Steagall controls should be the recommendation, not their elimination. I don’t believe that, and I don’t think the Fed does, either: I think the Fed is grabbing turf.
Securities companies do fail, and must be allowed to fail. A week ago I testified with former SEC Chairman Richard Breeden before your colleague Mr. Oxley, and he told the story of Drexel Burnham. He consulted the Fed before exercising his authority to close down the brokerage house, and the Fed urged him to forbear. Drexel was a debtor to the Government of Portugal—how would the markets react to the failure of a debtor to the Government of Portugal? Mr. Breeden said he told the Fed he thought the markets would giggle.
Now, the truth of this matter is a little more complicated, because hard work had to be done to make sure that the Drexel failure made ripples rather than waves. But because Mr. Breeden’s agency understood markets, he knew he could close Drexel without significant systemic risk. Bankers, college professors and bank regulators do not understand risk very well. The Fed is simply risk averse when it deals with banks—and always will be. Paul Volcker strongly resisted closing Penn Square. But markets require risk-taking. Markets need regulation, rules that all the players must follow, and not supervision, which by definition is at least partly individual and arbitrary. The Fed supervises. Indeed, the Fed has never been much good at supervising the markets in which it is now closely involved—look at the Salomon Brothers scandal, Drysdale Securities, Lombard-Wall, Gibson Greeting Cards, Procter & Gamble, Daiwa Bank. There is reason to believe that the Joe Jett-Kidder Peabody matter got as messy as it got because the Fed knowing that Kidder had misreported its Treasury strip and mortgage tranche positions nevertheless permitted an enormous and continuing buildup of the miscategorized holdings.
On the other hand, when bank regulators look at the market activities of banks they tend to be too tolerant, because they assume that the banks know what they’re doing. Indeed, banks that must submit all their loans however complicated to examiner scrutiny are permitted to value their own derivative positions without hindrance on the grounds that they know better than the examiners. With the approval of the Fed, the Bank of England handled the Barings matter, which really did present serious systemic risk, as though it were simply the failure of a medium-sized bank. The banking regulators both in England and here did not understand that the failure of Barings’ derivatives subsidiary might—and almost did—provoke a failure at the Singapore commodities exchange clearing house which is joined at the hip with the Chicago Merc. Thank God it was Mary Schapiro at the Commodities Futures Trading Commision and not the Fed that had the regulation of the commodities division of Merrill Lynch. When I look at the Fed’s regulation of the government bonds market and contemplate extension of its activities to the securities and commodities markets I feel as William Schwenk Gilbert felt about the House of Lords—that noble fingers should not itch/ to interfere in matters which/ they do not understand.
More seriously, there is a real culture clash here, which the Congress ignores at great peril to the financial sector and the economy at large. Banks are run to a large extent in secret, and banking regulators enforce secrecy. Examiners reports are so secret that a bank acquiring another bank, and performing a due diligence that requires a look into every nook and cranny of the institution to be acquired, is theoretically forbidden to see its examination report. I am authoritatively informed by counsel to banks acquiring other banks that this law is honored in the breach, like some other bank regulatory rules (supervision, as noted, is individual and arbitrary), but it’s the law. Secrecy is required, the banking regulators say, to maintain confidence in the payments system.
Markets, by contrast, are public information systems, and the thrust of our securities laws is disclosure and transparency. Not the least of my fears in the extension of banking regulators’ authority over securities houses is that they would sabotage in the securities markets as they have sabotaged at the banks the work of the Financial Accounting Standards Board, which draws its authority from the Securities and Exchange Commission and sets the rules of Generally Accepted Accounting Principles that are the bedrock of all securities valuation in this country. (I should note that I am currently engaged in writing an article about FASB for Institutional Investor magazine.) The major source of the authority of banking regulators over banks today is their power to value the assets. Alan Greenspan has been the most consistent warrior against the mark-to-market ethos of the securities markets. In the derivatives context, the Fed has actually approved procedures by which banks can price their derivatives position without the intention or even the willingness to do business at the prices they assert for accounting or regulatory purposes. The Chairman’s recent suggestion of a regulator-approved form for expressing a bank’s gross derivatives exposure is clearly prompted by the need to counter the SEC’s already announced and much more revealing disclosure standards for such instruments.
The securities markets activities of financial institutions must continue to be regulated by the SEC, though obviously the Commission will always wish to be in consultation with the Fed on the many matters in which they have common interests. As Chairman Greenspan’s recent speech to the Chicago Bank Structure conference indicates, the Fed is interested only in the profitability and stability of the banks and the banking system. It is the need of the regulator to see all the aspects of the holding company, not the need for the public and the markets to receive and evaluate information, that drives the Chairman’s analysis. “Regulation,” he argued, “must fit the architecture of what is being regulated.” But in a world where the capital markets dominate the credit markets, to steal Dennis Weatherstone’s admirable phrase, that architecture cannot be dictated by bank regulators. Indeed, if there was to be a single dominant regulator of the financial services holding company, we would be better off with the SEC as the final arbiter for the holding company, and we must leave open a path to get there, though for the time being the bank regulators must retain ultimate control of the banks.
I say for the time being, because within the next few years Real Time Gross Settlement processes will take care of the banking end of systemic risk. The European Central Bank is coming, and with it comes the Target system to provide RTGS for Europe. The Japanese are on the same track and the Swiss are already there. Real Time Gross Settlement means that nobody will be able actually to enter a payment in the system until he has the money, and it is the credit to the payee, not the debit to the payor, that moves through the system. Thus credit risk can arise only when the payor must borrow to make his payment, a process clearly to be managed in the private sector, with the central bank present and alert as a potential lender of last resort. The present New York Clearing House CHIPS system, with its limits on exposure and its collateralization rules, can be adapted by the smart people who run it to operate as a continuous flow. EDIBANX at the Chicago Clearing House settles every morning at nine, but by its nature it is a continuous flow operation.
The most important implication of RTGS is that it will complete the process of severing the payments system from the lending system, which is the most important development in modern banking—with which, I must note, HR 10 makes no contact at all. These improvements to the payments system—and, equally important, the expansion of the Electronic Data Interchange component of payments processes—will require privatization. The Fed would of course continue to regulate and even perhaps to supervise the operations of the privatized payments system, but it should no longer be a competitor in providing those services. The enormous expenditures on East Rutherford, for a system not designed to accommodate EDI, is a symbol of what’s going wrong. Experts tell me that there are only 35 banks in this country today that can handle Electronic Data Interchange, and the obvious reason for this disaster is that the Fed is fat and happy with its own systems and has not wished to encourage its member banks to enter any activity not invented at the Fed.
The most likely scenario for the future is that the content providers—the First Datas and FiServs and EDS’s—will take over the business and use the banks as their marketing facades. This process has gone much further than most of you realize. First Data, with 40,000 employees has issued something like a hundred million credit cards, nearly all of which have the name of a bank on them somewhere, and its activities in the payments area generated last year a revenue stream approaching six billion dollars. Last I looked, EDS owned something like thirteen thousand ATM machines, nearly all of them with bank logos on them, did the complete nightly accounting for a couple of thousand banks, and had an revenue stream of about two billion dollars a year from banking services. The banks and their regulators have left something like 15 million Americans without bank accounts—and the Treasury and the Defense Department are now mandated by law to make all their payments by electronic funds transfer begining January 1, 1999. That legislation—and it is imperative not only for budgetary reasons that it be retained intact in this Congress—will work immense and beneficial changes in our banking system. HR 10 doesn’t know it exists.
To the extent that technology reduces systemic risk through RTGS, the safety net can be diminished. Deposit insurance for ordinary people—one account per head, please, with the maximum no greater than $100,000—must of course be retained, but beyond that, nothing. The safety net can once again become, as intended, a support for people, not for banks and their regulators. If systemic risk worries are reduced, bank secrecy can be reduced. A dozen years have passed since Bill Isaac on his way out as FDIC chairman suggested to this committee that examiners’ reports should be published. As ever-increasing fractions of bank portfolios are in the form of securitized rather than directly administered loans, the privacy aspect of bank secrecy rapidly declines in importance. Certainly, accounting standards at financial services institutions should be standardized. To say that a securitized loan has one value at a bank and another at a brokerage house, which is where we stand today, is not an intelligent way to regulate financial services.
The accounting problem is far and away the most urgent. A recent study for the International Accounting Standards Committee has proposed that accounting for financial instruments should be at fair value whether those instruments are part of a bank portfolio or part of a securities firm’s portfolio. “Understandability and comparability must suffer,” the study argues, “if measures of reported financial position and income are the result of adding together numbers that are determined on different bases or that have little relationship, or an unclear relationship, to accepted economic realities of value and risk.” HR 10 permits American bank regulators to continue their idiosyncratic accounting rules, making it far more difficult for a market to judge whether a given bank is a suitable investment or depository. Indeed, the banking regulators hope to keep for themselves as much as possible judgments on whether banks are acceptably managed. At a time when the capital markets are so much stronger than the credit markets, this King Canute approach to the future of financial services is improper and dangerous.
Having reduced systemic risk and the safety net and having improved accounting procedures, we can invite true market discipline into the financial services industry. At that point, we can approach more rationally the juncture of different financial services activities into a single entity and indeed the mixture of financial and non-financial activities. In any event, financial services modernization should be seen in a context of changing technology and institutional response, not in a context of whose ox is gored today, who will oppose if he doesn’t get this goody, who can be tempted to support if he does get that goody, which regulator is to occupy which turf. The recent telecom legislation and its results—higher prices to consumers, no service improvements, declining stock prices despite the bull market for the telephone and cable companies that lobbied so hard for the bennies they thought they got—should stand as a warning to those trying to write banking legislation that can be blessed by the lobbyists.