PUBLISHED BY CITIZENS AGAINST GOVERNMENT WASTE, 12/92
The mounting problems of the commercial banking industry continue to cripple the U.S. economy and hold American taxpayers hostage. Unwise lending policies over the last decade — largely fueled by open-ended deposit insurance subsidies and antiquated regulatory distortions — resulted in first a squandering of scarce capital and then a rigid tightening of credit. The end product was a weakened banking system, a stagnant economy, a bankrupt deposit insurance fund, and an ominous load of debts to be picked up by U.S. taxpayers.
Although this deepening crisis developed over a number of years, the political need to “address” it reached a peak in 1991. Throughout that year, the Bank Insurance Fund (BIF), which insures deposits in commercial banks and savings banks, began to run out of money as projected losses in failed banks climbed steadily. On October 23, 1991, the Federal Deposit Insurance Corporation (FDIC), which administers the Fund, estimated that it would need to spend at least $32 billion, and perhaps as much as $44 billion, through 1993 to protect depositors in failed banks. Despite a series of steep hikes in insurance premiums assessed against banks in the last few years (nearly tripling since 1989), the FDIC forecast a growing deficit in BIF that would reach at least $9.6 billion in 1992 and increase to a range of $18.3 billion to $28.9 billion in 1993 (depending upon whether future economic conditions deteriorated further).
After wrestling most of 1991 over the terms of comprehensive legislation proposed by the Bush Administration to overhaul the banking industry, Congress deadlocked on most reform issues but finally agreed to a more narrow bill on November 27. The Federal Deposit Insurance Corporation Improvement Act of 1991 provided a $30 billion line of credit from taxpayers to the FDIC and additional borrowing authority to finance assets seized in failed banks. Although the crippled banking industry is supposed to eventually repay any such money borrowed from the Treasury through adjustments in future insurance premiums, skeptical taxpayers are fearful that they are witnessing an instant replay of the still-unfolding disaster within the savings and loan (S&L) industry. Estimates of the total cost to taxpayers in financing the bailout of insolvent S&Ls and their bankrupt insurance fund that was launched in 1988 remain imprecise but have climbed well past the $200 billion mark.
The imminent exposure of taxpayers to mounting losses in the insurance fund designed to protect bank depositors reflects the troubled economics of the banking industry. Particularly within the last decade, bankers have been left crippled by a combination of aggressive competition from nonbank financial services providers, increasingly costly and antiquated regulatory restrictions, and the misincentives of a deposit insurance system that burdens strong institutions and props up weak ones. The damage has been reflected in the industry’s declining profitability, shrinking market share, inadequate capital levels, soaring losses on troubled loans, and rising failure rates.
Both the banking industry and the complex system of government regulation and deposit insurance within which it operates are broken and in need of serious repair. A recent upturn in the industry’s outlook does not appear to be sustainable over the long haul. The current round of “record” earnings reflects an unusually steep, upward-sloping yield curve that cannot last much longer. When bankers run out of one-time gains on securities sales and begin to absorb the downside of interest rate risk, their industry is likely to return to its decade-long pattern of inadequate profitability. The fundamental forces of intensified competition from securities markets and other nonbank financial services providers, plus stiffer capital standards and regulatory compliance costs, will continue to require painful adjustments within the banking industry.
This paper will assess the current problems of the U.S. banking system, analyze how they reached this point, weigh the strengths and weaknesses of the Bush Administration’s proposals to deal with them, review the congressional debate over banking reform during 1991, examine the merits of the final legislation ultimately approved, take a brief look at early experience under the FDIC Improvement Act of 1991, and outline a future agenda for more fundamental reform that focuses on greater market discipline and less taxpayer-subsidized risk-taking in banking.
Digging Into A Deep Hole
Sizing Up the Balance Sheet for Bankers
By any measure, the U.S. banking industry is floundering and has been steadily losing ground over the last decade in particular. Of greatest concern is the fact that bank failure rates have risen steeply. While only 198 federally-insured banks failed from 1942 to 1980 (with never more than 20 bank failures within a single year), the FDIC has been faced with the staggering total of 1256 bank failures from 1981 to 1990. Bank failures reached a peak of 221 in 1988 and averaged 200 a year from 1986 to 1990.
Although officially-declared bank failures were down for 1991, the figure of 127 banks that were “resolved” (either closed, forced to be sold to other banks, or given financial assistance) by the FDIC reflected the agency’s shortage of cash reserves. On November 24, 1991, the late William Taylor, the then-new chairman of the FDIC, acknowledged that federal regulators had been forced to move more slowly in seizing weak banks, particularly larger ones, because of the depletion of the Bank Insurance Fund. A better indication of the growing caseload facing regulators was revealed by the growth in the total value of assets on the problem bank list, which rose to a record-high $486.9 billion by the end of September, 1991, up from $404.2 billion the previous March. By the end of 1991, the asset value total for problem banks had rocketed upward again — to $611.1 billion. It clearly appeared that the FDIC was finding a number of larger banks in increasing distress, but had to await congressional approval of expanded borrowing authority to obtain sufficient resources to deal with them.
Large numbers of explicit bank failures, of course, stem from a broader problem of poor performance by U.S. banks as a whole. Both primary measures of banks’ profitability — return on assets (ROA) and return on equity (ROE) — have declined significantly. During 1989 and 1990, for example, commercial banks’ average ROA was 0.49 percent, down from its average of 0.77 percent during the 1970s and 0.69 percent from 1980-1984. Banks’ average ROE fared poorly as well, dropping from averages of 12.1 percent during the 1970s and 11.7 percent from 1980-1984 to 7.7 percent in 1989 and 1990.
At the heart of this poor earnings performance were record levels of losses on soured loans. The loan loss charge-off rate (the percentage of loans, in dollar value, advanced to borrowers that are later written off and removed from banks’ balance sheets as uncollectible losses) more than tripled over the last decade. In 1990, U.S. banks charged off an all-time record $29 billion in bad loans. By the end of that year, banks held a record level of nonperforming loans that totaled 2.9 percent of all commercial banking assets. In 1991, the rates of loan loss charge-offs and nonperforming loans rose even higher than the 1990 levels.
How We Got Here
The historical roots of the current crisis in banking trace back to the Great Depression. Much of the fundamental structure of modern banking regulation stems from the political judgment of the 1930s that “excessive competition” among financial institutions led to a wave of contagious bank failures. Even though the unprecedented banking crisis of the late 1920s and early 1930s was essentially due to the combined effect of fatally flawed macroeconomic policy (particularly inaction by the Federal Reserve as “lender of last resort”) and the undiversified fragility of the largely unit banking system of the time, the Glass-Steagell Banking Act of 1933 imposed for decades the view that banking markets needed to be kept “stable” and free from “unfair” competition.
On the asset side, this philosophy was translated into limits on bank activities (to deter conflicts of interest) and geographic restrictions (to prevent the dominance of large, nationwide banking firms). On the liability side, the Depression view of banking meant deposit insurance protection to prevent sudden withdrawals of funds by anxious depositors (bank runs) and controls against interest rate bidding wars.
Glass-Steagall divided up financial product markets. It restricted the range of permissible underwriting activities for banks and prohibited their affiliation with securities firms.
The Banking Act amended the McFadden Act of 1927 to effectively set state boundary lines as the limit on geographic expansion through bank branching, while deferring to each state’s judgment regarding the latter’s own particular branching structure.
The legislation completed its set of cartel-like restrictions on product lines, market entry, and pricing by imposing price controls on interest payments to bank depositors. It prohibited such payments on demand deposits and authorized regulators to set rates on time deposits and savings accounts.
At the same time, the Banking Act offered new protection to depositors against losses from bank failures by implementing a threshold level ($2500 per depositor) of “deposit insurance” guarantees.
Later legislation (the Bank Holding Company Act of 1956 and the 1970 Amendments to it) reinforced the above limits on product and geographic expansion for bank companies and also effectively precluded commercial firms from acquiring controlling interests in banks.
The Depression-era regulatory structure for banks created a precarious balancing act involving both regulatory protections and restrictions for banks.
Legal restrictions that separated the geographic markets in which individual banks could operate were intended to limit the competitive pressure they faced. But they also hindered banks’ ability to benefit from geographic diversification that would reduce susceptibility to regional economic downturns.
Product restrictions (and limits subsequently imposed on who could own banking companies) were initially designed to assign bankers an exclusive franchise in activities “closely related to banking” while securities firms handled underwriting of debt and equity products, and insurance companies underwrote and sold insurance. Although such limits may have been imposed in the name of preventing conflicts of interest, financial concentration, and/or unsafe competition, they also prevented banks from diversifying their profit-seeking activities across product lines and from reducing costs through more efficient production of a full line of financial services.
Price controls on the interest payments that banks (and thrifts) could offer to attract deposits further dampened price competition between depository institutions and helped lower the cost of funding for the banking industry. As long as banks enjoyed a virtual monopoly in providing payments transactions services, inflation was low, and marketplace interest rates for alternative savings vehicles did not rise significantly above the deposit interest rate caps set by the Federal Reserve Board under Regulation Q, such price controls provided an artificial deposit subsidy to banks and allowed them to remain relatively more conservative in the credit risk that they took. In later years, however, the combination of rigid interest rate caps and inflationary pressures would encourage the development of innovative financial products by nonbank competitors that would offer more attractive terms to savers and dry up sources of low-cost funding for the banking industry.
In short, many of the regulatory “protections” provided to banks in the depressed economic landscape of the 1930s operated as a double-edged sword that, by also restricting banks’ ability to diversify and compete, created long-term destabilizing forces within the banking industry.
The safety net guarantees provided by deposit insurance could be seen in part as an offsetting subsidy that would compensate banks for these regulatory handicaps. Even as deposit insurance helped banks attract funds and cover losses, however, it also provided an incentive for banks to increase their expected returns by selecting riskier asset portfolios. This “moral hazard” dimension of deposit insurance (the incentive for those insured against risk to adopt riskier behavior) in turn strengthened the case for tighter regulatory controls on banks’ risk-taking opportunities.
Thus, the deposit insurance system and the bank regulatory system developed a mutually reinforcing “quid pro quo” relationship with each other. An extensive safety net of insurance and regulatory subsidies would compensate for the destabilizing effects of legal restrictions on the range of geographic and product diversification permitted for banks. In return, an extensive system of regulation, supervision and examination (in addition to the aforementioned restrictions) was imposed on banks to control the undesirable incentives set in motion to incur risk and maximize profits.
This overall regulatory regime of market allocation and competitive restraint proved paternalistically benign for U.S. banks during the post-World War II period of low inflation, stable interest rates, and steady economic growth. In those days, federal deposit insurance seemed to provide a relatively riskless and inexpensive guarantee against “irrational” runs on banks that enjoyed protected markets, earned steady profits, and rarely failed.
Nevertheless, the fundamental design of FDIC-administered insurance was flawed from the outset. It underpriced risk by imposing the same flat-rate premiums on all banks regardless of their balance sheet composition. This operated as a cross-subsidy within the banking industry that rewarded weak banks at the expense of strong ones.
Like most seemingly inexpensive government-provided “insurance” programs, the FDIC fund was structured to hold inadequate financial reserves to cover its potential long-term losses. The Banking Act of 1935 limited the FDIC’s assessment powers and set insurance premiums below the level that would reflect historical loss experience. Since the 1950 Deposit Insurance Act also required the FDIC to rebate “excess” net assessment income, insurance premium increases tended to lag the Fund’s long-term exposure.
The deposit insurance system also grew in importance to banks over time as a subsidy in attracting funds, leveraging capital, and covering losses. In its first year of operation, the FDIC insured 45 percent of all bank deposits. (The initial ceiling of $2500 per depositor was soon raised to $5000 per insured account.) The average bank’s ratio of equity capital (“net worth”) to total assets — representing both a cushion against unexpected losses and bank owners’ stake in managing their enterprises wisely — was 15 percent at that time. By the 1980s, when the FDIC insured over 70 percent of all deposits (hitting a high of 77 percent in 1987) and official coverage had been raised to $100,000 per insured account, these bank capital ratios averaged only about 6 percent. With the assurance that the federal government would protect their money from loss in whichever bank they chose, depositors increasingly relied less on the strength of the capital base in the particular bank that held their funds. Thus, banks could at least in part substitute expanding deposit insurance guarantees for their own capital. Over time, this would eventually lead to an ironic phenomenon — an increasingly undercapitalized banking industry developing an overreliance on an underfunded deposit insurance fund.
In addition, the scope of deposit insurance protection repeatedly strained against official coverage limits. Although the FDIC was initially authorized to use only deposit payoffs in failed bank situations, the Banking Act of 1935 allowed the agency to facilitate mergers of failing insured banks with stronger institutions. By the 1940s, the FDIC had transformed this tool into purchase and assumption (P&A) arrangements that effectively extended 100 percent protection to most uninsured depositors as well.
Such expanded coverage drew little protest during these early years of relatively infrequent bank failures and modest losses in resolving troubled institutions. But after incurring losses of over 30 percent of insured deposits in the four bank failures of 1950, the FDIC promised that it would apply an informal “cost test” that would supposedly limit complete rescues of bank creditors through P&As to instances where they were cheaper than straight payoffs of only insured depositors. Since P&As were considered less disruptive of local banking services and cheaper than liquidation in most cases (by capturing the so-called ongoing “franchise” value of failing banks that are sold to other depository institutions), the FDIC soon found that it could easily manipulate the loosely-applied cost test whenever it needed to get around it. By the mid-1960s, blanket protection for uninsured depositors through P&As was again the dominant FDIC resolution practice. Regulatory interpretations of different “rights and capacities” by which deposits are owned provided an additional means of expanding insurance coverage.
As the deposit insurance safety net expanded, it tilted the bank regulatory system’s “apparent” political tradeoff between short-term stability and long-term economic efficiency increasingly in the direction of the former. Although the longstanding design flaws within FDIC-administered deposit insurance (underpriced risk, inadequate reserves, informal 100 percent coverage) always offered a latent subsidy to banks operating on thin capital levels, the potential benefit could only be fully captured and exploited by banks willing to take on greater risks in their asset portfolios.
In the 1980s, this “moral hazard” of deposit insurance finally broke loose as old regulatory limits on risk-taking and protections from competitive pressure lost their force. During the previous decade, the twin forces of double-digit inflation and technological innovation had begun to erode both the benefits and effectiveness of government-created restrictions on the delivery of financial services. As the financial industry became less segmented and banks faced vigorous new rivals offering innovative financial products, onetime legal protections for banks turned into competitive handicaps.
The increasing burden of regulation on banks throughout the 1970s encouraged the entry of unregulated providers of financial services and drove business outside of the traditional banking industry. Inflationary pressure made the interest rate ceilings of Regulation Q more binding on both banks and thrift institutions, while helping fuel the development of money market mutual funds. Since such funds operated unencumbered by deposit rate caps and reserve requirements, they could bid aggressively for depositors’ funds and grew spectacularly during the double-digit inflation years of the late 1970s. Money market funds became even more attractive by introducing limited transactions (check-writing) services to investors.
At the same time, securities firms developed “cash management accounts.” These financial products combined the sale of bank certificates of deposit with other asset allocation services and were offered to customers through extensive brokerage office networks.
Many thrift institutions gained regulatory approval to offer another innovation — negotiable order of withdrawal (NOW) accounts — which, in effect, allowed S&Ls to offer certain kinds of interest-bearing transactions accounts and add to the growing competition for depositors.
By the end of the 1970s, banks had in effect lost their traditional monopoly over transaction account services and found their low-cost-of-funds subsidy under siege by competitors. Banks could no longer earn sizeable profits simply by underpaying many of their best customers (the elderly, corporations, affluent consumers) for their deposits and then lending the money out to borrowers at higher market rates. In the face of high inflation and new competitive alternatives, “[those] contributing the lion’s share of the [deposit] subsidy became more sophisticated and began leaving the banking system.”
Forced to deal with the competitive threat posed by the nonbank sponsors of money market funds, Congress phased out rate controls in the Depository Institutions Deregulation and Monetary Control Act of 1980 and the Garn-St Germain Act of 1982. Although banks to some degree stabilized their share of deposit funding, the days of easy financing through below-market interest rate payments were over.
As banks entered the 1980s, they were also being squeezed on the asset side of their balance sheets. Technological improvements in information processing and telecommunications lowered the cost of credit assessment, which facilitated both entry of new nonbank competitors into financial markets and the rapid evolution of asset securitization.
Banks faced increasing numbers of innovative nonbank rivals in markets for credit cards and auto loans. Many commercial companies (including a number of diversified financial conglomerates) exploited a loophole around the Bank Holding Company Act’s restrictions on bank ownership to form so-called “nonbank banks” — limited-service financial institutions covered by deposit insurance that either accepted demand deposits or made commercial loans, but not both.
Banks’ best customers — high-quality corporate borrowers — discovered that they could largely bypass banks and reduce their borrowing costs by issuing their own commercial paper to eager investors.
Growing secondary markets in residential mortgages and consumer loans opened up opportunities for new investors, increased competition, reduced net interest spreads, and ate further away at banks’ traditional role as financial intermediaries.
In addition, foreign banks made increasing inroads within U.S. financial markets, and thrift institutions were unleashed by the Garn-St Germain Act to compete more vigorously with banks.
The delayed effect of decades of once-protective overregulation hit banks from both directions. In one way or another, most of the new competitors confronting U.S. banks operated with the comparative advantage of being less burdened by regulatory limits on the kind of assets they could hold, the markets in which they could operate, the rates they could pay to depositors, the ratio of capital they must hold against their assets, the amount they could lend to a single borrower, and/or the portion of their deposits they must keep as noninterest-earning reserves in the Federal Reserve system. While nonbank firms thrived by working around banking law restrictions on product offerings and geographic expansion, banks still faced the same legal barriers in their own efforts to diversify into other financial markets.
The banking industry struggled to counterattack on several fronts, with mixed success throughout the 1980s.
Geographic liberalization to meet the competition of nonbank competitors not bound by the McFadden Act and the Douglas amendment to the Bank Holding Company Act advanced more steadily than product line liberalization. By the end of the 1980s, dozens of states had expanded intrastate branching, entered into interstate regional compacts, and even opened themselves up to nationwide entry through holding company acquisitions. Automated teller networks, brokered deposits, and interstate acquisitions of failed banks and thrifts further eroded the remaining force of territorial restrictions. But the legacy of McFadden/Douglas still prevented the development of truly nationwide banks, concentrated loan risk exposure in local and regional markets, hampered the attraction of core retail deposits, and generally kept banks from following their customers except through less efficient and more costly methods.
Throughout the 1980s, however, the banking industry focused its lobbying campaign for congressional relief from regulation in terms of gaining greater asset powers (primarily in the securities field) to meet both domestic and international competition. The congressional debate over legislation to expand bank powers was dominated by endless conflict over how the financial market “pie” would be divided up among the respective interests of the banking, securities, insurance, and real estate industries. Brokering past such gridlock, in effect, necessitated an elusive supermajority. In fact, the debate shifted in the latter part of the decade from expansion of bank powers to holding off bankers’ perceived encroachments (through regulatory rulings) upon the turf of the other three financial sectors.
Though stalemated on the congressional front, banks made more headway with federal regulators and state legislators, who gradually broadened permissible activities for banks throughout the 1980s. Nevertheless, the ad hoc nature of regulatory liberalization by the Federal Reserve Board and the Comptroller of the Currency evolved in a piecemeal fashion full of legal uncertainties and repeated court challenges. Entry into nontraditional activities was further burdened by the inefficiencies of revenue limitations and costly “firewall” requirements.
Although a number of larger banks repeatedly tried to innovate new financial products, expand “off-balance sheet” activities that generate fee income, and generally stretch the boundaries of their regulatory restraints, the banking industry as a whole was whipsawed by the multiple forms of new competition (money market mutual funds, commercial paper, asset securitization, nonbank banks). Commercial banks’ share of U.S. financial market assets shrank from 51 percent in 1950 to 31 percent in 1989.
As longtime regulatory barriers became less effective in deflecting competition and risk on the asset side, the deposit insurance system grew in importance to many banks as a subsidy in attracting funds and covering losses on the liability side. With increased competition meaning lower spreads on traditional loans and the flight of high-quality customers to other financial outlets, weaker banks sought to restore their former profits by drawing upon their last regulatory asset — deposit insurance — to raise funds to underwrite higher-yielding credit risks. The combined effect of deposit interest rate decontrol, higher official deposit insurance coverage ceilings, and growing brokered deposit operations that redistributed insured funds around the country was to extend an open invitation for troubled banks to try growing their way out of trouble by bidding up the cost of funds and taking on high-yield, high-risk asset gambles.
As the deposit insurance subsidy became more valuable to weak banks, it grew more expensive for stronger depositories. The latter felt, on one hand, the pressure of matching higher deposit rates or losing their own funding base. On the other hand, they faced the burden of paying the rising bill for deposit insurance at the same flat rate of assessment on domestic deposits as their risk-taking colleagues.
The higher costs facing the FDIC insurance fund encouraged the agency to briefly experiment during 1983 and 1984 with moving away from complete protection of uninsured bank creditors. Under so-called modified payoffs, the FDIC advanced to uninsured depositors and unsecured creditors partial payments of what the agency expected to recover later on the failed institution’s assets. This limited effort to restore a measure of market discipline through uninsured depositors/creditors, yet provide timely liquidity, was ultimately undermined by the FDIC’s decision in 1984 to fully protect all general creditors and all largely- uninsured depositors of the $34 billion Continental Illinois National Bank. The Continental bailout openly affirmed the Too Big To Fail (TBTF) doctrine for large banks and made it clear that uninsured depositors would be completely protected in future failures. During the rest of the decade, the FDIC protected over 99 percent of the value of uninsured deposits in failed banks.