
What economic principles should policymakers in other countries have learned from the S&L mess? Appropriate incentives for government supervisors are critical for avoidance of financial crises - savings and loan
Edward J. KaneFor individual institutions' insolvencies to trigger a national financial crisis, cumulative losses across an important industry sector must exceed the safety-net support that creditors expect the government to provide. This happens when official supervisors acquiesce to pressure of threatened institutions to engage in risky lending in an effort to stay afloat. Such regulatory risk-taking takes the form of insurance against failure and concealment of the magnitude of the risks to which the insured institutions and their insuring governmental institutions are exposed. The source of such regulatory dereliction is rooted in conflicting incentives for regulators that must be resolved if future crises are to be avoided. The paper describes and analyzes the U.S. savings and loan crisis and financial crises in developing countries to illustrate these points.
**********
An important contributor to the depth of most modern financial crises is the degree to which healthy market discipline was displaced by inefficiently managed systems of government supervision. Taxpayers and public servants around the world need to understand that the extent of a financial system's exposure to bad luck, aggressive management, and looting is an endogenous consequence of the economic principles public policies embody.
Economic insolvency strikes an individual financial enterprise when it suffers losses that destroy its capacity to repay depositors and other creditors without outside assistance. Destructive losses are rooted in poor or dishonest management, bad luck, defective information systems, and superior competitors.
For individual economic insolvencies to trigger a national crisis, cumulative losses across an important industry sector must exceed the safety-net support that creditors expect the government to provide. Entry pressure from more-efficient and better-capitalized competitors can create such losses if implicit and explicit governmental guarantees degenerate into a mechanism for retarding the recapitalization of insolvent deposit-institution competitors. Although no two financial crises unfold in exactly the same way, events are driven by the interaction of (a) risky lending and funding strategies that lead to the economic insolvency of individual institutions and (b) risky regulatory strategies that government supervisors and regulators use to handle individual insolvencies when they develop.
The major lesson of the U.S. savings and loans (S&L) mess is how dangerously regulatory and bank risk-taking can reinforce each other. For individual insolvencies to persist for years on end requires that--at some level of government--officials sell insolvent institutions protection against failure and conspire at least implicitly with internal and external auditors to conceal gaping holes in individual balance sheets. As long as the cover-up succeeds, the lending policies of troubled institutions escape the ordinary weight of depositor discipline.
Around the world, the valuation and itemization principles that deposit-institution accountants and regulators use to measure banking profits and net worth contain options that make it possible for large opportunity losses to be hidden from public view. Until and unless challenged by economic analysis, using these options can generate phantom and nonrecurring profits that overstate profits and net worths for years on end. Cooked books and earnings projections based upon them resemble the digital readouts from a scale rigged by a dishonest butcher. With a show of irrelevant precision, authorities can systematically and repeatedly mismeasure the obligations that deposit insurance is putting on the taxpayers' bill.
Accounting and regulatory dereliction is rooted in incentive conflict experienced by a country's designated watchdog institutions. The solution to incentive conflict lies in reworking the watchdogs' incentives in the social contracts that are breaking down. The conflict at issue is the tradeoff between regulators' and accountants' social missions and the personal and bureaucratic costs of resisting client pressure for relief. Ironically, the accounting and regulatory strategies that ruined the Federal Savings and Loan Insurance Corporation (FSLIC) and the U.S. S&L industry were of the industry's own making. Even more ironically, these discredited strategies closely resemble the policies that multinational firms and the IMF have implicitly urged on crisis countries in Asia and Latin America (e.g., Fischer, 2001).
Guaranteeing the debt of insolvent institutions and covering up the loss exposures this creates for a country's taxpayers is costly in three ways. First, by allowing important institutions to operate in an insolvent condition, authorities leave poorly performing assets and franchises in the hands of managers whose lending and funding incentives are distorted by capital weaknesses. Because the downside of future returns belongs to the guarantor, insolvent firms are tempted to invest the savings entrusted to them in lottery-like projects that combine a negative present value with a small chance of a very large payoff. Second, until the coverup begins to unravel, accounting disinformation insulates the guidance and forbearance decisions that government officials are making from financial and political review. Finally, any cover-up is likely to be accompanied by microeconomically inefficient pricing and entry restrictions intended to protect the markets of troubled firms from close competitors. However, because o f their inefficiency, these restrictions are apt to boomerang against the industry in the long run.
Industry-welcomed restraints on stronger U.S. deposit-institution competitors ultimately helped less-regulated outside competitors--such as foreign banks, money-market mutual funds, and brokerage firms offering cash management accounts--to snatch market share from the industry the restraints were supposed to help. While troubled S&Ls implored government officials to wall off intra-industry access to their customer base behind deposit-rate ceilings and geographic barriers, differently chartered institutions devised substitute instruments that used emerging electronic technologies to innovate through and around the industry's regulatory defenses.
Effective long-run regulatory performance requires improved accountability for policy mistakes, and accountability begins with accurate information. Throughout the S&L mess, authorities showed a propensity for blocking flows of information that threatened to harm their individual and collective reputations.
Incentive reform requires a serious effort to increase the costs and reduce the benefits of accounting coverup. To increase the timeliness and accuracy of information that managers of insured institutions, managers of deposit insurance funds, and incumbent politicians supply to taxpayers, improved economic and political incentives are needed in government service. The more a country's political and cultural environment tolerates de facto corruption, the more useful it would be to offer deferred compensation to the chief executives of banks and deposit-insurance enterprises and to tie retiring officials' right to draw down this compensation either to the absence of crisis during the first five years after their departure or (if information systems permit) to a market-value measure of the change in the insurer's net loss exposure observed during their term in office (Kane, 2002).
Accounting Cover-up in the S&L Mess
In the 1960s and 1970s, the core activity of U.S. thrift institutions was to make long-term mortgage loans financed with short-term deposits. Maintaining a short-funded portfolio exposed these institutions' economic income and net worth to losses whenever interest rates rose. The secular rise in interest rates and interest-rate volatility experienced between 1965 and 1982 generated unbooked losses on S&L mortgages that devastated the economic value of industry income and net worth.
FSLIC furnished dividend-free risk capital and supplied enough of it to fill in the holes in troubled firms' balance sheets for over thirty years. During this interval, red ink flowing through S&L income and balance sheets seeped into the accounts of FSLIC and, through FSLIC, onto U.S. taxpayers.
Although S&L and government accountants refused to formally recognize these losses as they were accruing, the damage being done could not be hidden completely. Across an institution's portfolio, the impact of interest-rate movements could be estimated with a reasonable degree of accuracy by appraisal techniques. For example, estimates summarized by Kane (1989) and Brumbaugh (1988) clarify that, at least from 1971 on, the S&L industry could not expect to repay its deposit liabilities from its own resources. The survival of insolvent S&Ls depended on the black magic of government guarantees rather than on the earning power of their assets. What kept insolvent S&Ls from being closed down by depositor runs was the willingness of the FSLIC to promise depositors credible protection against loss. Their perverse life-in-death existence may be likened instructively to that exhibited by the hordes of marauding zombies featured in George Romero horror movies.
When individual insolvencies spread and deepen in any country, the long-run problem for a deposit-insurance enterprise lies in keeping its guarantees credible to deposit-institution customers. A private guarantor would do this by identifying and forcing the recapitalization of troubled institutions as soon as they weaken. Tolerating widespread insolvency plants and fertilizes the seeds of a deeper crisis. A government insurer can avoid a crisis as long as it keeps its loss exposure small enough that depositors can reasonably rely on the faith and credit of the national Treasury to explicitly or implicitly augment the insurer's resources. However, on average, risk-taking incentives at zombie firms tend to make the insurer's loss exposure grow over time. Crises develop when and as the market value of the implicit government debt embodied in the Treasury's support begins to swamp the incremental tax capacity a nation needs to service it.
A private guarantor would recognize and seek to counter an insolvent client's interest in fabricating profit and net worth. Part of the policy scandal embodied in the S&L mess was authorities' repeated willingness to creatively extend the accounting leeway that S&Ls and FSLIC could use to further slow down reportable deterioration in FSLIC's balance-sheet position. Table 1 shows that until 1986 official estimates of FSLIC's net reserves remained reassuringly positive. Table 2 illustrates the increasing magnitude of the economic losses that authorities were covering up. Using market-value estimates of the enterprise-contributed net worth of every FSLIC-insured institution existing or closed in fiscal years 1985-1989, the table measures the aggregate loss exposure that bureaucrats at FSLIC managed to keep off their books until the cover-up unraveled.
If government guarantees had not been supplied on favorable terms, private creditors would have forced insolvent S&Ls to recapitalize themselves (perhaps by transferring ownership in whole or in part to large creditors) or else see themselves merged or liquidated out of existence. To keep accrued losses from registering on FSLIC's books, regulators had to help insolvent S&Ls to hide losses and to resist forms of exit that would have revealed the size of their capital shortage. Because a guarantor assumes the deep downside risk of private creditors' exposure to loss, economic theory dictates that a conscientious deposit insurer should exercise market-mimicking disciplines. In not staking for itself an explicit claim to the future profits of zombie firms, FSLIC encouraged private owners and managers to take poorly structured, longshot gambles. For zombie S&Ls, the beauty of these gambles was that FSLIC took the downside risk and permitted S&L owners and managers to lay claim to much of the upside potential.
Opportunities for borrowers to over-leverage themselves could not have burgeoned unless insolvent lenders believed that they could shift their own expanding loss exposure to FSLIC. Loss exposures could be expanded because neither insolvent lenders nor their government regulators had strong incentives to fully evaluate and disclose the risks being taken.
In medicine, the word crisis describes the point in the course of a disease at which a decisive change occurs. Often a crisis may be averted by seizing one or more "golden moments" during which improved medical care can cure the disease relatively simply.
In retrospect, one can see that FSLIC passed up two golden moments for capping its losses. First, if, instead of nullifying market pressures for exit and recapitalization during the late 1960s, FSLIC had forced insolvent S&Ls to recapitalize, merge, or liquidate themselves, the S&L industry could never have remained shortfunded and undercapitalized enough to load such large losses onto FSLIC in later years. Second, even if this opportunity had been missed, subsequent losses could have been reduced if recapitalization had been sought when disinflation sharply lowered interest rates in 1982-1983. In fact, the Federal Deposit Insurance Corporation appropriately seized this second window of opportunity to discipline the roughly 500 short-funded savings banks that it insured.
Sadly, by the time that interest rates turned down in mid-1982, S&L trade associations had successfully lobbied congress for new ways to take and hide risk. Supplementing their traditional concentration in long-term home mortgage assets, many thrifts began to load up with riskier loans aimed at financing residential and commercial real estate development and holdings of raw land. If in 1982-1983 insolvent S&Ls had been forced to recapitalize or else, loans made under this new regulatory regime would have been better structured and would have therefore generated fewer losses. Such a recapitalization would have spared the nation a costly spate of overbuilding. Insolvent S&Ls would not have spent so many years pouring funds into an overheated U.S. real estate market in an industry-approved and regulator-authorized effort to help insolvent S&Ls "grow out of their weakness."
Unbooked Fiscal Deficits as a Source of Financial Fragility (1)
Prior to 1989, three strategic elements characterized public policies toward S&Ls. Although these strategies violate economic principles for the efficient production of regulatory services, the same three elements are featured in the banking policies of almost every country in the world today:
1. Politically directed subsidies to a politically favored class of bank borrowers. The policy framework either requires or rewards banks for making credit available to designated classes of borrowers at a subsidized interest rate;
2. Subsidies to bank risk-taking. The policy framework commits government officials to providing on subsidized terms either explicit or conjectural guarantees to holders of bank liabilities;
3. Defective monitoring and control of the subsidies. The contracting and reporting framework for government officials fails to make them directly accountable for controlling the size of either subsidy.
Rent-seeking theory explains why short-horizon authorities would allow banks to extract wealth from taxpayers and would require loan officers to transfer some or all of that wealth to politically favored borrowers. The third element in the strategy explains what prevents taxpayers from seeing the implicit expenditures generated by the first two elements and from disciplining inappropriate transfers in timely fashion through political action or parliamentary review. Imposing civil and criminal penalties on officials who can be shown after the fact to have willfully provided less than their best estimate of their enterprise's economic value would reduce the benefits of forbearance. It would create an enforceable obligation for regulators to report truthfully to taxpayers and watchdog institutions the size of the dual subsidies. Passing this information through the government budget would make authorities accountable for explaining whether and how taxpayer benefits generated by the subsidies justify the costs th at they impose on taxpayers.
Without side payments from the rent-seeking sectors, it would be unlikely that a growing flow of subsidies could prove incentive-compatible for top government officials even for short periods. To enlist high-ranking regulators permanently into the benefit-redistribution game, two further conditions must hold. First, taxpayers must be prevented from assessing by indirect means the magnitude of the costs they face in funding the subsidies. Second, regulators themselves must receive suitably laundered incentive compensation from banks and borrowers. The compensation offered must be sufficient to balance exposure to legal penalties and the risk of damage to the reputations of policymakers and the regulatory bureaus they head if, during their watch on the bridge, the system for covertly financing the subsidy were to break down.
A banking-policy regime that greatly subsidizes risk-taking may be portrayed as an accident waiting to happen. A banking crisis occurs when a sufficient amount of bad luck hits a banking system whose managers have made their institutions vulnerable to this amount and type of bad luck. The S&L mess teaches us that the odds of experiencing a bureaucratic breakdown in a country's inter-sectoral cost-shifting process may be modeled as an evolutionary process in which the odds of breakdown increase as the size of unbooked government guarantees grows.
The larger accumulated opportunity-cost losses become, the larger the off-balance-sheet debt with which fiscal authorities are saddling future taxpayers. What we may call a "silent run" begins not when an institution becomes a zombie, but when the accumulated unbooked fiscal deficit from the government's loss exposure in zombie banks begins to scare large-denomination depositors. As more and more depositors and investors rationally begin to doubt whether officials can or will continue to support the guarantees, the silent run on a country's banking system gathers steam.
Doubts about a government's willingness and capacity to make taxpayers absorb the unfunded cost of guaranteeing the country's zombie banks are a function of its tax capacity. The triggering condition is that the value of aggregate guarantees soars so far above dedicated reserves that taxpayer resistance is expected to develop. This political resistance threatens the survivability of the incumbent government and promises to undermine its ability to raise the funds needed to pay the bill in full. We describe runs by sophisticated large depositors as silent because pressure on a troubled bank from savvy depositors generates far less adverse publicity than a line of panicked small depositors does when a bank is experiencing a conventional run.
A silent run signals an end to the pre-crisis period because it generates observable increases in each zombie bank's funding costs. In developing countries, a zombie bank's first line of defense against a silent run is usually to arrange loans from relatively well-informed foreign banks. Like the sophisticated depositors that zombie bankers manage to retain, foreign banks demand higher interest rates and appropriate collateralization for their claims. The net outflows of domestic deposits that zombie banks experience are financed by a combination of selected asset sales and high-rate new debt. In consciously deciding to finance a silent run, foreign banks may feel confident that (as in Mexico in 1994) they can successfully lobby the IMF, their host government, and their home governments to protect them against defaults by host-country banks. Foreign banks may also find it advantageous to speculate against the currency in offshore derivatives markets. Table 3 reports the amounts and types of external assistanc e received by seven crisis countries in recent years.
Unless and until bank regulators take steps to increase the credibility of their guarantee system (e.g., by establishing a substantial line of credit with the International Monetary Fund), a silent run on a nation's banking system tends to escalate. This is because zombie banks' asset sales and funding-cost increases make the fragility of the zombies' condition visible to less-sophisticated observers by causing an inescapable deterioration in the accounting values of income and net worth. When a zombie bank sells assets at market value, its unhooked losses on subsidized loans become a larger proportion of its footings. Similarly, the more liabilities that a zombie bank rolls over at increased interest rates, the more severely its accounting and economic profit is going to be squeezed.
A silent run increases pressure on regulators to acknowledge that zombie banks are benefiting from government guarantees and that stronger banks and ordinary taxpayers will be asked to pay for the bill. As the pressure builds, it progressively undermines the willingness of taxpayers and stronger banks to tolerate the regulatory status quo. The transfer of benefits to insolvent institutions from taxpayers and viable banks becomes progressively greater the longer a silent run proceeds. Regulatory efforts to retard the exit of inefficient and insolvent deposit institutions lower the profit margins that strong banks can earn on borrowed funds and push their prospective costs for funding the government's guarantee services above the value of the guarantees that they themselves receive.
This theory of crisis may be contrasted with that of authors such as Chang and Velasco (1998) who locate the trigger for financial crisis directly in a growing mismatch in the maturity of a country's international assets and liabilities. Our theory predicts that such an imbalance in maturity develops as an endogenous consequence of insolvency-driven silent runs. A shortening of the maturity of capital inflows is triggered by foreign lenders' increasing concern for being able to unwind the positions they establish in economically insolvent host-country banks. But for troubled banks to receive new funding, their government's guarantees must remain credible domestically.
In country after country, officials have actively encouraged loss-causing patterns of credit allocation and compounded the damage from credit losses by not resolving individual-bank insolvencies until their economic capital had deteriorated disastrously. When the cover-up dissolves, domestic (and sometimes foreign) taxpayers are billed to bail out banks, depositors, and deposit-insurance funds. Caprio and Klingebiel (1999) report that taxpayers' bills for making good on implicit and explicit guarantees has typically run between one and ten percent of GDP. The size of these bills underscores the real costs of allowing the corrupted risk-taking preferences of high government officials to shape the flow of aggregate investment.
Government Exit Resistance
Taxpayers (including strong financial institutions) are likely to be forced eventually to finance deposit insurance losses. Efficiency requires that taxpayers be able to observe and control politically the depth and breadth of deposit-institution insolvencies as they are developing. Regulatory tolerance of go-for-broke risk-taking by insolvent institutions undermines the stability of a country's financial system and allows institutional losses to cumulate relentlessly. In the United States and abroad, regulators' tolerance for risk-taking is negotiated behind closed doors in an unacknowledged and corrupt market for political clout.
Improvements in financial technology and increasing price volatility for financial instruments have steadily transformed the equilibrium market structure of the financial-services industry in and across countries. Starting in the mid-1960s, the U.S. deposit-insurance system began to counteract natural market pressure on failing thrift institutions to voluntarily recapitalize themselves or exit through merger or liquidation. This exit resistance was the root of FSLIC's aggregate losses. Individual client losses were caused by bad economic luck, poor management, insider crime, structural weaknesses in risk-management controls, and the slower-acting effects of well-meaning regulatory interference. However, in trying to sustain the inefficient market structure they inherited, authorities transformed government deposit insurance in the United States, which had enjoyed thirty years of initial success, into a system for nurturing inefficient and unsound firms and perversely rewarding socially imprudent investments. Not forcing insolvent institutions to promptly resolve their insolvencies rewarded unsound banking practices and effected an unintended, badly structured, but (happily) temporary nationalization of a large segment of the nation's financial assets and institutions.
After a long period of expansion, in the mid-1990s Asian and Latin American banking industries faced parallel pressure for domestic-bank exits. In many countries, foreign and nontraditional financial firms had to introduce themselves in circumventive ways. The interlopers did host-country business by making creative use of substitute products, substitute organizational forms, or substitute offshore locations. As in the S&L mess, in most countries, a new entrant's ability to use differently regulated substitute opportunities was facilitated by longstanding and burdensome restrictions on how local deposit institutions could compete domestically.
As in the S&L mess, authorities in these countries were reluctant to encourage the prompt recapitalization of banks that were weakened by outside competition or to estimate the size and publish the opportunity cost of the risk capital that was required from taxpayers to protect these banks. Politically and administratively, it was much easier in the short run to use loopholes in bank and government accounting principles to conceal the extent of industry weakness from public view and to suppress information that might generate political pressure for a different and stricter course of action.
In public-policy discussions, lobbyists euphemistically describe as "regulatory forbearance" those insolvency-management policies that are more lenient than would be expected from informed taxpayer "principals." What makes forbearance strategies attractive is: (1) the campaign donations and other forms of monetary tribute that government officials can collect in exchange and (2) the long period of time during which government and trade-association spokespersons can credibly hide the extent of insolvency. Political deal-making is further assisted by the understanding that the flawed accounting records they certify will make it difficult later to credibly pin the consequences of inappropriate forbearance decisions on the particular officials who conceived and executed them.
Disguised side payments, the lack of reliable measures of an insurer's true condition, and the absence of audit trails for forbearance decisions or their consequences encourage officials to delay recapitalization pressure and to gamble on making a "clean getaway" either to a longer term in office or to a high-paying job in the private sector. By blaming officials disproportionately for whatever problems manage to surface while they are in office and by not nailing officials for the forecastable future damage they create when they adopt short-sighted supervisory strategies, the press and voting public reinforce authorities' propensity to gamble inefficiently with taxpayer money.
The Regulatory-Gambling Theory of Financial Crisis
The seeds of the S&L mess lie in defective incentives for measuring and controlling the taxpayer loss exposures that politicians and top regulatory officials create (Kane, 1989; Barth, 1991; White, 1991). These incentive defects engender conflicts with bureaucratic and personal goals that, in tough times and in tough cases, tempt government officials around the world not to enforce the underwriting standards, coverage limitations, and takeover rights that constitute taxpayers' best theoretical defenses against cumulative deposit-insurance losses.
Applying the theory of principal-agent conflict (Jensen and Meckling, 1976) to safety-net policies can explain allegedly inadvertent policy failures in the pricing and administration of government deposit insurance as calculated risk-taking behavior. Favoring the interests of a nation's decapitalized institutions can serve politicians' and regulators' interests at the expense of society in general. Forbearance may be expected to keep an incumbent politician's and regulator's watch on the bridge less turbulent, to preserve these officials' reputations, to improve opportunities for reelection or postgovernment employment, and to generate a flow of implicit or explicit side payments. An acid test by which to distinguish an innocent mistake from a self-interested "calculated gamble" is the immediacy and sincerity of a perpetrator's regret. An error is regretted simply because it is wrong; a calculated gamble is regretted only if and because it fails.
Incumbent politicians and bureaucrats have short time horizons and narrow career and reputational interests that frequently diverge from those of taxpayers. What was and is missing from deposit-insurance arrangements in most countries is timely accountability for the opportunity losses that tolerating insolvencies imposes on taxpayers. In a representative democracy, once the loss exposure of a government deposit-insurance corporation outruns its budgeted resources, divergences in taxpayers' and regulators' interests make it rational for opportunistic authorities to abuse their discretion by covering up evidence of insolvencies at the institutions they supervise and by postponing painful loss-control activity to their successors' watch on the bridge.
Bankers want government guarantees as a competitive advantage. Politicians see government guarantees as a way to avoid being embarrassed by a wave of bank failures during their limited time on the bridge and--in some cases--as ways to directly or indirectly enrich themselves by corruptly selling options for delaying failure. Because taxpayers do not offer incentive payments and because regulatory officials' reputational and career interests leave them more directly answerable to politicians and bankers than to taxpayers, government regulators are pulled more strongly in practice toward subsidizing deposit-institution operations by avoiding failures than they are toward minimizing the long-run costs of taxpayer loss exposures. Although highly ethical individual regulators may routinely reject these temptations, the main lesson of the S&L mess is that it is fatuous for society to depend on a regulatory framework whose successful operation demands repeated acts of selflessness by its top managers. Although top g overnment officials are explicitly screened for public spiritedness, temptations posed by defects in regulatory incentives become increasingly hard to resist once a deposit-insurance enterprise develops a capital shortage.
In financial markets around the world, technological and political forces have forced decisions to deregulate entry on reluctant politicians and regulators. But markets have had a harder time forcing elected politicians and top regulatory officials to deregulate exit. Officials have a short-run reputational interest in retarding the exit of economically insolvent and inefficient firms when these happen to fall within their traditional client base. The most reliable external control on the costs this principal-agent conflict passes on to a country's taxpayers is the degree of accountability its public servants feel.
Kane (2000a) shows that, in Asian crisis countries, Japanese banks (who have themselves been in continual crisis since the early 1990s) amassed the biggest pre-crisis positions, and during crisis months they beat the strongest retreat. The continued insolvency of major Japanese banks meant that banking policies and conditions in Japan created incentives for Japanese bankers to book extraordinarily high-risk loans at home and abroad (Kane, 2000a and 2000b). Even at year-end 1998, the exposure of Japanese banks in Indonesia, South Korea, and Thailand remained high.
The expansion of foreign lending by insolvent Japanese banks was bound to squeeze the profit margins of host-country banks. Host-country profit margins and economic net-worth were also steadily undermined by domestic political pressure for banks to make subsidized loans to politically selected economic sectors. To restore industry profit margins to a sustainable level, exits had to occur. Crisis became a political mechanism for some insolvent institutions to finally be closed or absorbed into stronger enterprises.
Summary Implications
Economic analysis supports the view that incentive incompatibilities inherent in representative democracy make opportunistic government officials a source of financial instability. They control their reporting frameworks, and they can generate personal and bureaucratic benefits in exchange for adopting suboptimal strategies of cover-up and forbearance. Officials are perennially tempted to distort information flows about the quality of their performance in the short run and to repeatedly delay market-structure adjustments that would serve taxpayers' long-run interests.
It is dangerous for taxpayers not to contractually counterbalance officials' exposure to undisclosed side payments and lobbying pressure that might inappropriately persuade them to give government resources to crippled institutions. In crisis countries everywhere, longstanding systems for subsidizing inefficient loans to favored individuals imposed unbooked losses on their banking systems. Banking policies have been messy: marked by scandal, short-lived administrations, delays in making important decisions, and lack of transparency in decision-making processes.
Nevertheless, the messy policies have lasted for years. The messes turned into banking and currency crises only when doubts began to surface about authorities' willingness and ability to support the growing liabilities of their economically insolvent banking system. The S&L mess teaches us to view a regulation-induced banking crisis as the surfacing of tensions caused by the continuing efforts of zombie institutions to use the safety net to force the rest of society to pick up zombies' unpaid bill for making bad loans. In the S&L mess, pressure to resolve the mess was triggered by silent runs that reflected a growing concern that taxpayers might resist paying the full value of conjectural government guarantees.
The wave of banking and currency crises that has splashed through Asia and Latin America now threatens Eastern Europe. The wave is propelled by two trends. First, advances in information and communications technology are increasingly globalizing previously disconnected local banking markets and are challenging supporting political markets for government subsidies. Second, the globalization of markets for banking and guarantee services is making it increasingly less costly for domestic corporations and wealthy investors to mount silent runs on a country's insolvent banks.
When banking markets are globalized, services that provide regulatory benefits to bank customers are made available from foreign as well as domestic suppliers. The more access customers have to foreign suppliers, the more easily the struggle for net regulatory benefits in one country can spill outside its national boundaries to involve foreign banks and their home-country suppliers of financial regulation.
Inadequate constraint on the pursuit of self-interest by government officials is the root cause of large taxpayer losses, and it continues to threaten countries with inappropriate supervision today. Financial deregulation did not cause the U.S. S&L mess. Nor did it cause recent Asian and Latin American crises. Financial deregulation may be defined as an unambiguous relaxation of the rules of financial-services competition for all players. Modern crises are caused predominantly by corrupt or opportunistic de-supervision of the capital positions and risk exposures of insolvent and inefficient financial-services firms.
Public-service incentives need to be reworked to make it less attractive for authorities to help troubled deposit institutions to resist healthy exit pressure. Using data covering sixty-one countries in the years 1980-1997, Demirguc-Kunt and Detragiache (2000) show that deposit insurance perversely contributes to banking fragility in countries where institutional controls on incentive conflict are weak. Ideally, in every country managerial markets for the services of current and former government officials ought to reward rather than punish officials who protect taxpayer interests faithfully at the expense of having their reputations hammered by regulatory clients and their political allies. But to play this role, the press must be empowered to offer these labor markets much better information about the risks that banks and regulators take and when they take them.
TABLE 1
OFFICIAL ESTIMATES OF FSLIC RESERVES, 1960-1986
Total reserves Percentage of value
Year end ($ million) of accounts insured
1960 381 0.62
1965 1,537 1.35
1970 2,903 2.05
1975 4,120 1.48
1980 6,462 1.28
1985 4,600 0.54
1986 -6,300 -0.71
Source: Kane (1989, p. 9).
TABLE 2
ESTIMATES OF TAXPAYERS' UNBOOKED LOSS EXPOSURE IN FSLIC, 1985-1989 (IN $
BILLION)
Sept. 30, 1985 86.4
Sept. 30, 1986 122.5
Sept. 30, 1987 106.8
Sept. 30, 1988 161.3
Aug. 9, 1989 161.1
Source: Kane (1993).
Note. In these years FSLIC's fiscal year ran from September to
September. August 9, 1989, was the last day on which FSLIC officially
existed. Even these estimates understate the size of FSLIC's losses
because they neglect the implicit financing cost of carrying these
losses.
TABLE 3
ASSISTANCE OFFERED CRISIS COUNTRIES BY THE INTERNATIONAL COMMUNITY
(IN BILLIONS OF U.S. $)
Commitments (a)
Country IMF Multilateral (b) Bilateral
Brazil 18.1 9.0 14.5
Indonesia 11.2 10.0 21.1
korea 21.1 14.2 58.4
Mexico 17.7 0.0 31.3 (d)
Philippines (c) 1.6 0.0 0.0
Russian Federation (e) 15.1 6.0 1.5
Thailand 4.0 2.7 10.5
Total 88.8 41.9 102.0
Country Total
Brazil 41.6
Indonesia 42.3
korea 58.4
Mexico 49.0
Philippines (c) 1.6
Russian Federation (e) 22.6
Thailand 17.2
Total 232.7
Source: Garcia (1998, p. 27) and Christiansen (2002, p. 134)
(a) The rescue packages for each country cover resources made available
for time periods specific to each case.
(b) World Bank and Regional Development Bank.
(c) As of year end 1998.
(d) Includes $10 billion credit line from BIS.
(e) Through end of 1999.
(1.) This section draws heavily on Kane (2000a)
REFERENCES
Barth, James. 1991. The Great Savings and Loan Debacle. Washington: The AEI Press.
Brumbaugh, R. Dan, Jr. 1988. Thrifts under Siege. Cambridge, MA: Ballinger Publishing Co.
Caprio, Gerard, and Daniela Klingebiel. 1999. "Episodes of Systemic and Borderline Financial Crisis." Washington: The World Bank. Financial Sector Strategy and Policy Development.
Chang, R. and A. Velasco. 1998. "The Asian Liquidity Crisis." Cambridge, MA: National Bureau of Economic Research. Working Paper No. 6796.
Christiansen, Hans. 2001. "Moral Hazard and International Financial Crises in the 1990s." Financial Market Trends. No. 78. March. Pp. 115-139.
Demirguc-Kunt, Asli, and Enrica Detragiache. 2000. "Does Deposit Insurance Increase Banking System Stability?: An Empirical Investigation." Journal of Monetary Economics. Forthcoming.
Fischer, Stanley. 2001. "Financial Sector Crisis Management." Seminar on Policy Challenges for the Financial Sector in the Context of Globalization. Sponsored by the World Bank, IMF, and Board of Governors of the Federal Reserve System. Washington, D.C. June 14.
Garcia, Gillian. 1998. "The East Asian Financial Crisis." In G. Kaufman, Ed. Bank Crises: Causes, Analysis and Prevention, Research in Financial Services: Private and Public Policy. Stamford, CT: JAI Press. 10. Pp. 21-32.
Jensen, Michael, and William Meckling. 1976. "Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure." Journal of Financial Economics. 3. Pp. 305-360.
Kane, Edward J. 1989. The S&L Insurance Mess: How Did It Happen? Washington: The Urhan Institute Press.
_____. 1993. "What Lessons Should Japan Learn from the U.S. Deposit Insurance Mess?" Journal of the Japanese and International Economies. 7. Pp. 329-355.
_____. 2000a. "Capital Movements, Banking Insolvency, and Silent Runs in the Asian Financial Crisis." Pacific-Basin Finance Journal. 8. Pp. 153-175.
_____. 2000b. "The Dialectical Role of Information and Disinformation in Regulation-Induced Banking Crises." Pacific-Basin Finance Journal. 8. Pp. 285-308.
_____. 2002. "Using Deferred Compensation to Strengthen the Ethics of Financial Regulation." Journal of Banking and Finance. Forthcoming.
White, Lawrence J. 1991. The S&L Debacle: Public Policy Lessons for Bank and Thrift Regulation. New York: Oxford University Press.
Edward Kane, a professor in finance at Boston College, is the recipient of numerous awards for distinguished teaching and scholarship. He is a research associate of the National Bureau of Economic Research and a former Guggenheim fellow. A banking agency consultant and expert witness in financial-services litigation, his specialty is financial and ethical research. He has published three books, writes regularly for professional journals, and serves on seven editorial boards. He received his BS from Georgetown University and his Ph.D. from the Massachusetts Institute of Technology.
This paper is adapted from a manuscript prepared for the Milken Institute/UCLA research roundtable, "What Can An Examination of S&Ls Reveal About Financial Institutions, Markets and Regulation?' in Los Angeles, January 25, 2002 and is forthcoming in a volume edited by James Barth, Susanne Trimbath and Glenn Yago, as part of the Milken Institute Series On Financial Innovation and Economic Growth. James Moser provided valuable comments on an earlier draft. Much of the material was presented by the author in his Adam Smith Award address at the NABE Annual Meeting on September 30, 2002.
COPYRIGHT 2003 The National Association of Business Economists
COPYRIGHT 2003 Gale Group