It is a pleasure to be with you this morning. The theme of this session is “How Banks Compete.” I want to develop a variation on this theme and consider how the intensity of competition in banking has increased over the years, and some of the challenges this change presents.
In the 33 years I have worked at the Richmond Fed banking has changed immensely. A salient feature of this change — perhaps the single most important feature — has been expanded competition. Today Chicago banks, for example, can own branches in any state, pay market-determined interest rates on deposits and charge market-determined rates on loans, and offer a continuum of financial products to their customers. Twenty years ago all of these powers were restricted. As regulatory reforms have eased the constraints that previously held banking competition in check, the industry has become more efficient and consumers now enjoy far wider choices.
While such regulatory changes are certainly welcome, they have made the job of the bank supervisor more difficult. Today’s marketplace demands that banks take advantage of every profit opportunity. This is appropriate, and drives much banking innovation and efforts to improve operational efficiency. Nonetheless, given the presence of the financial safety net, competitive pressures also may induce banks to act in ways that distort markets, degrade the safety and soundness of the banking system, and put taxpayers at risk. Actions taken by savings and loan companies during the 1980s provide the most notorious example of such behavior. The obvious implication is that bank supervisors — aware of the greater incentive to take advantage of the safety net — must be far more diligent now than at any time in the past.
Before the Great Depression, restrictions on banks were few and government guarantees were minimal. Consequently, banks were free to compete aggressively but enjoyed little in the way of a safety net to protect depositors in case of bank failure. Following the widespread failures of the 1920s and early 1930s, as you know, legislators established a safety net in the form of deposit insurance. At the same time, they imposed tight restrictions on banks’ ability to compete, which limited the opportunity to take inappropriate advantage of safety net guarantees.
When I came to work for the Richmond Fed in 1970 the banking system was operating in a regulatory structure largely unchanged since the Depression. Beginning in the late 1970s, however, shifts in economic conditions and in the financial marketplace forced the removal of many of the most severe restrictions. As a result, banks can now compete more aggressively, and the banking industry is far more efficient today than it was. But while banks now compete vigorously, and have every reason to take advantage of profit opportunities the safety net might afford, the net remains as wide as ever. An examination of events leading to the rejection of the earlier restrictions, and the resulting growth in pressure to take advantage of the safety net, is key to identifying changes that can improve the regulatory environment going forward.
A prominent example of restrictions established during the Depression were Regulation Q ceilings on deposit interest rates. These rules, put in place in 1933, gave the Federal Reserve the authority to set limits on the deposit rates paid by all banks, thereby preventing banks from competing for deposits on the basis of price, at least above the ceilings. The fact is, of course, that banks continued to compete, though in a highly inefficient manner — by offering “free” banking services, and in some cases merchandise premiums such as toasters, in lieu of interest payments.
Rising inflation during the late 1970s and early 1980s, drove up market rates of interest and made restrictions unsustainable. Nonbank financial institutions not subject to Regulation Q ceilings emerged and began paying market rates, rates much higher than those paid by Regulation Q-inhibited banks. Money market mutual funds were one of the important nonbank competitors. Legislators faced a dilemma: either impose Regulation Q ceilings on nonbank financial institutions, or remove them from banks. Otherwise, banks, unable to provide sufficient free services to match the high interest payments made by nonbanks, would quickly lose depositors and shrink. Legislators chose to remove the ceilings, and phased them out over the first half of the 1980s.
With the removal of Regulation Q ceilings, most interest rates are now set in the market, eliminating inefficient efforts to end-run interest rate restrictions. These restrictions were not only costly to skirt, they meant competition was held in check to some degree. Their removal heightened competition.
During the 1970s, and before, in-state and interstate branching was limited or prohibited by state and federal laws. But improvements in information technology created an environment in which branching restrictions were untenable. Under pressure from banks, which perceived growing cost advantages from operating large branch networks, and in line with the general trend toward deregulation in many industries, many states dropped in-state restrictions in the mid-1980s and passed laws allowing regional interstate banking. Ultimately federal legislation — the 1994 Riegle-Neal Act — allowed nationwide interstate branching.
The elimination of branching restrictions and the resulting branch expansion leaves today’s consumer with much wider choices, and opens local banking markets to competition from a much wider array of banks. Banks can now operate largely without regard to state boundaries and have responded to this freedom by expanding the number of branches at a much faster rate than the rate of population growth.
While the removal of branching restrictions has certainly produced a more competitive banking environment, it has also enabled the extraordinary consolidation in the banking industry we are all familiar with, and the transition has at times been troublesome for customers — especially small businesses accustomed to dealing with locally owned and managed banks. In some cases small business owners, once able to borrow with little more than a handshake, are now required to submit extensive financial statements for evaluation at the bank’s corporate headquarters.
American markets, however, have a wonderful way of adjusting effectively to these kinds of developments. Consequently, as consolidation has proceeded, new local banks have been formed, and existing community banks have expanded to meet the demand for specialized small business loans. Moreover, the largest banks have endeavored to appeal to small business borrowers by granting local loan officers substantial autonomy to make loan decisions and tailor products to meet local conditions.
Consolidation, of course, though beneficial to both banks and customers in many respects, creates new and tougher responsibilities for bank supervisors. Guarding against excessive risk-taking, the major goal of bank supervision, is far more difficult when supervisors face large complex banking organizations — and large organizations now account for an expanded share of all banking assets due to consolidation. Because today’s financial marketplace features instruments that are quite complex, and the largest banks are heavily involved in trading these instruments, the opportunity for banks to take on risks they may not completely understand is significant. When traditional lending accounted for most of banks’ income the danger was far lower. The increased complexity forces supervisory agencies to recruit more highly skilled, and therefore more highly paid, examiners.
Beyond the removal of Regulation Q ceilings and branching restrictions, during the 1980s banks perceived opportunities to leverage their capabilities honed in banking markets by entering complementary financial businesses such as investment banking and insurance. They found ways to enter these businesses via limited openings in the Glass-Steagall and Bank Holding Company Acts. The Glass-Steagall and Bank Holding Company Act prohibitions were finally dismantled by the Gramm-Leach-Bliley Act in 1999.
Through the 1980s and 1990s banks gained much new competitive flexibility as Regulation Q ceilings, branching restrictions, and prohibitions on securities and insurance activities in banking companies eased. The dismantling of all these barriers has meant rapid change in the business of banking, and created a more efficient and competitive industry to the great benefit of bank customers. Because of the important customer benefits, few would wish to return to the highly regulated banking industry of the past. In fact, a restoration of strict price, geographic, and product regulations would be impossible in a world with so many alternative providers of banking products. At the same time, however, the flexibility to compete confronts banks with both greater pressures and greater opportunities to exploit the financial safety net.
Competition naturally creates incentives to make the most of profit opportunities wherever they exist, including opportunities that exploit features of the financial safety net for profit. These efforts may benefit individual banks; they have the potential, however, to weaken our overall financial system and put taxpayers at risk.
For example, an individual bank may determine that it can take on a little extra risk in its lending or trading activities, and hence increase revenues, without causing its deposit insurance premium to rise. In a highly competitive market, in which customers must be paid a competitive rate of interest and shareholders demand strong returns quarter after quarter, a bank that identifies such a profit opportunity is likely to grab it. While expanding risk would normally cause a firm’s creditors to demand an offsetting increase in debt payments, insured depositors are not particularly concerned about a bank’s risk profile and typically do not demand higher interest payments. We are all well aware of the moral hazard in this situation.
But if one bank profits from such an expansion of risk, others will follow suit. Eventually, the advantage for individual banks will be competed away, flowing instead to bank customers, and banks will gain little from the effort. The initial increase in bank revenues from adding riskier activities quickly dissipates as interest rates on loans or fees from trading activities are driven down. Meanwhile, the banking industry as a whole will have made loans or engaged in trades providing a return that is insufficient to balance the risk, which will reduce financial market efficiency. At the same time, taxpayers will be exposed to risks for which they are not adequately compensated by deposit insurance premiums. Greater competitive freedom and the safety net will have, so to speak, collided.
Banking institutions can exploit the safety net in a number of ways. The actions of troubled S&Ls during the 1980s, of course, immediately come to mind in this regard. At the same time that inflation-induced financial management challenges were arising in the thrift industry, all depositories, including troubled S&Ls, were freed of Regulation Q ceilings on deposit interest rates. Further, S&Ls were granted enlarged lending and investment powers. These S&Ls responded by using their newly won freedom to offer market interest rates to compete — and compete aggressively — for insured deposit funding. At the same time, they engaged in forms of lending with which they were largely unfamiliar, and purchased risky investment securities.
In hindsight it appears that some S&L managers were engaging in fraud. Still, many were simply responding to the incentives of a competitive marketplace juxtaposed with a financial safety net. Managers of troubled S&Ls knew that if they did not undertake aggressive action their institutions were likely to fail. Consequently, some took assertive and in some cases excessive risks, betting on their one chance for recovery. The point here, of course, is that without deposit insurance, such risk taking almost certainly would have been impossible. Uninsured creditors are typically unwilling to advance new funds to seriously troubled firms since they know the firms may employ them in risky investments — at least without tight controls. Insured depositors, however, have no compelling reason to care.
Ultimately this exploitation of the safety net cost consumers more than $100 billion. Perhaps equally importantly, S&L activities almost certainly distorted financial markets by funneling funds to unworthy business projects.
The S&L debacle provides an example of how more robust banking competition in the presence of a financial safety net can produce concrete and measurable costs, at least in terms of losses borne by taxpayers. Other exposures impose costs that are more difficult to measure. One example involves the ongoing consolidation in the banking industry, which — again — is a product of deregulation and increased competition in banking markets. Continuing consolidation may well expand the set of banking assets in institutions that are believed to be too big to fail. Indeed, the prospect of becoming too big to fail could reinforce incentives for further consolidation. Certainly, there are strong efficiency justifications for consolidation, but the potential role of the safety net should not be ignored.
Most banking analysts would agree that depositors and creditors of the largest banks are more likely to be protected in the event of financial troubles than their counterparts in small banks. It is hard to imagine the systemic risk provisions of FDICIA applying in practice to small banks, yet the provisions conceivably could be applied if one of the largest banks became troubled. As a result, large banks are likely to have a funding advantage, other things equal, over smaller banks, and consolidation could be inappropriately encouraged for this reason. In a competitive market in which banks compete aggressively for funding and loans, bank managers who do not exploit this competitive advantage are likely to be ousted by managers who will.
Here, then, is another example of a potential reduction in the efficiency of the financial marketplace produced by the juxtaposition of the government safety net and competitive pressures. Further, the distortion may have increased with consolidation, since more banking industry assets are now held by the very largest banks: consolidation has stretched the implicit government safety net under a greater percentage of the nation’s banking resources.
While banking has been deregulated along a number of dimensions, one remaining stricture on the powers of banking companies is the wall between banking and commerce: in the U.S. banks may not conduct nonfinancial activities nor may they affiliate with firms that do. Over the last 10 years a number of observers — and some members of Congress — have proposed dropping this restriction.
The preceding discussion, however, points to a danger beyond the risks usually associated with the proposal. When a commercial firm that is affiliated with a bank suffers losses, there are circumstances where it may be possible to reduce the holding company’s losses by shifting them to the bank, and potentially on to the FDIC. One would normally expect such shifts to be prevented by the bank’s creditors. But because a large portion of the typical bank’s creditors are insured, shifts are more likely. Since investors would likely recognize this advantage of affiliation with banks, firms so affiliated could well have a funding cost advantage, which — again — could interfere with efficient resource allocation. The point here is that the affiliations would be driven by yet another intensification of competition affecting banks.
What is the appropriate supervisory response to intense competition in the presence of a safety net? How do supervisors achieve a balance — fostering the benefits of competition while preventing exploitation of the safety net? As I have already noted, few policymakers advocate a return to tight regulatory controls and severely restricted competition. Part of the answer is already in place, at least since the passage of FDICIA: careful monitoring of banks and prompt resolution of those that become troubled. Such a response to the emerging S&L problems in the 1980s would almost certainly have reduced taxpayer losses. Additionally, supervisors are working to strengthen and modernize capital requirements through changes proposed in Basel II. And supervisors and legislators have been appropriately cautious about loosening the restraints on combining banking and commerce.
The risks that arise, however, because of the possibility that a very large bank might become troubled some day, remain in place. The fear of wider financial difficulties if supervisors quickly close such a bank and impose losses on creditors and uninsured depositors could lead supervisors to defer decisive, but arguably necessary, action. Economists have developed a number of promising proposals aimed at increasing the likelihood that losses might be imposed on uninsured depositors and creditors, which of course would reduce moral hazard. One prominent example is the coinsurance plan advanced by my colleague Gary Stern, president of the Minneapolis Fed, which would have supervisors provide some protection to uninsured stakeholders, but — with certainty — less than the full amount of their losses. A related idea involves requiring banks to increase their reliance on subordinated debt, the holders of which seem less likely to be bailed out in a crisis. While these proposals could weaken the perception that supervisors would make uninsured depositors and creditors whole, their effectiveness would ultimately depend on the credibility of the supervisors’ pre-commitment. In the absence of this credibility, the behavior of uninsured depositors and creditors is not likely to be influenced by the implementation of these proposals.
So how do supervisors establish credibility with uninsured depositors and creditors? It seems to me that the solution here is the one the Federal Reserve employed in the 1980s in the monetary policy arena to overcome inflation. In the 1970s the Fed allowed inflation rates to rise for extended periods before acting — often not very effectively — to contain it. Beginning in 1979, however, the Fed began to give greater weight to achieving price stability. The Fed brought the inflation rate down over a period of years and then kept it down. In doing so it built and eventually established its inflation-fighting credibility and convinced financial markets that it would not allow inflation to re-emerge.
Similarly, promptly resolving large, troubled banks and imposing costs on uninsured creditors, even at the risk of some short-term financial disruption, is in my view the only means of eliminating the market’s perception that large banks will receive special treatment should they become troubled. Here too, it may be necessary to build credibility over time, in this case through a series of actions that respect the principle of imposing costs on uninsured creditors to encourage market discipline. Supervisors have few opportunities to prove themselves — thankfully — since serious financial troubles at large banks are infrequent, or at least have been for the last decade. Situations could arise, however, where supervisors are asked to intervene on behalf of a troubled financial institution that is not among the largest, but nevertheless large enough to raise concerns of financial disruption. Imposing costs on uninsured creditors in these cases could be useful steps in building full supervisor credibility, and these opportunities obviously should be taken advantage of when they arise. In any case, until supervisors demonstrate their willingness to impose at least some losses on uninsured creditors, financial market outcomes will continue to be distorted to some degree.
Of course, supervisory efforts to close troubled banks are subject to the will of Congress. Legislative action played a prominent role in the S&L crisis of the 1980s, and had the overall effect of delaying the closing of troubled S&Ls and increasing the ultimate cost to taxpayers. The passage of FDICIA, however, indicated clearly — and encouragingly — Congress’ intent to strengthen supervisor credibility in the future.
In sum, over the last 30 years regulatory restraints on competition in banking have been reduced in a number of ways; I have discussed several. In response, the industry has become more efficient, which has benefited consumers and businesses significantly. Consolidation resulting from the increased competition has also led to efficiency gains as banks have responded to the declining costs of managing large enterprises. Yet as banks have been freed to compete more aggressively, pressure has increased to exploit the opportunities for profit that the financial safety net provides. Exploitation of the safety net reduces efficiency and can impose large costs on taxpayers. Congress reduced this risk when it passed FDICIA, so that the widespread abuse of the safety net that occurred during the S&L debacle is unlikely to be repeated. Still, the Act’s systemic risk provision leaves open the possibility of special treatment of the largest banking institutions and therefore an opportunity for these banks to exploit the net. In my view, confronting this continuing “too big to fail” challenge — an unintended consequence of deregulation and increased competition — should be a high priority on the banking agenda.