«Volume Title: Politics and Economics in the Eighties Volume Author/Editor: Alberto Alesina and Geoffrey Carliner, editors Volume Publisher: ...»
With the defeat of the St Germain amendment, HR 27 was adopted by the House, 402 to 6. There were no other controversial amendments to the bill reported by the committee. The Senate had also approved its committee’s bill (S 790) with only minor changes. This lack of controversy on the floor provides additional evidence for our thesis that there was not a significant constituency for confronting the full magnitude of the thrift problem or to curtailrather than to extend-forbearance. Even a small group of dissenters can require roll call votes. In this way, the dissenters can force their opponents to go on record against a position, while the dissenters signify their concern on the issue. The virtual absence of roll call votes on the thrift issue in 1986-87 is like Sherlock Holmes’s dog that did not bark in the night. It is mute testimony to the fact that members of Congress did not believe there was an audience to whom it was worth sending stronger signals about the thrift problem.41 6.4.4 Compromise in Conference: The Competitive Equality Banking Act As the House and Senate conference committees prepared to meet to resolve differences between the two bills, administration spokesmen indicated that the president was seriously considering a veto of any bill that did not raise the recapitalization limit or-more important-relax the restraints on nonbank banks. As to raising the FSLIC limit, Proxmire “told an industry group...
By the end of July, as conferees were meeting, the General Accounting Office reported that FSLIC was $6 billion in the red and was confronting future claims up to $50 billion. In an agreement with the administration, leaders of the conference committee agreed to raise the FSLIC borrowing limit to $10.8 billion, with not more than $3.75 billion to be raised in any one year.
The forbearance provisions were left intact. In early August, both chambers passed the conference report (the House by 382 to 12 and the Senate by 96 to 2).42The president signed the Competitive Equality Banking Act on 10 August 1987.
The final legislation reinforced continued forbearance by allowing thrifts in farm and oil-patch states to continue to use the lenient regulatory accounting practices adopted in 1982. Thrifts in these areas (and other areas deemed economically depressed) would also be allowed to stay open with a.5% capitalPolitical Foundations of the Thrift Debacle asset ratio, instead of the 3% required by existing law. The act also explicitly reaffirmed the commitment that the “full faith and credit” of the U.S. government stands behind FSLIC. Given the limited amount of new FSLIC funding relative to the magnitude of the thrift insolvency problem, the act ensured that many failing thrifts would have considerable more time to gamble for resurrection. The crisis would grow.
6.5 The Thrift Debacle and Beyond The policy of forbearance did not, of course, lead to the resurrection of the zombie thrifts. Instead, as many economists had warned, the losses incurred by thrifts gambling for resurrection continued to escalate. The net income of the thrift industry for 1988 was - $12.0 billion, down from - $7.8 billion in
1987. Barth, Bartholomew, and Bradley (1989) report that the estimated cost of the 205 thrift resolutions begun in 1988 is $31.8 billion. Over half of these thrifts had been GAAP-insolvent for more than three years before they were closed or merged. At year-end 1988, there were 364 thrifts that were insolvent but still open. Many of these thrifts had been insolvent for a long time-some for as long as 10 years (Barth, Bartholomew, and Labich 1990).
Congress did not return to the thrift problem in 1988. By the time the woes of FSLIC reappeared on the legislative agenda after the 1988 elections (in which they played little or no role), the costs of delay and forbearance were increasingly evident to nearly everyone. The 1989 “bailout” (Financial Institutions Reform, Recovery, and Enforcement Act, or FIRREA) was enacted in the context of predictions that the liabilities facing FSLIC were expected to exceed $200 billion.43The FIRREA and its legislative history are beyond the scope of this paper. But its enactment was necessitated by the legislative failures of the previous five years.
While it is now widely recognized that a policy of forbearance under the existing structure of deposit insurance has been a prime contributor to the escalation of the thrift debacle, the political foundations for this behavior are not as widely appreciated. This paper has pointed to these foundations as being rooted in the logic of the connections between elected officials and regulatory policy. By this view, regulatory agencies are viewed not as autonomous decision makers, but as actors closely tied to the political system.
Constituency pressures work on regulators through the (often implicit) connections with politicians.
The legislative response to the problems of the thrift industry in the 1980s is an excellent example of how this process works. In summary, we note the
1. The political process provides at least tacit support for regulatory policy when the policy is consistent with the preferences o active constituent groups.
f This characterized policy toward thrifts until the late 1970s.
2. Signijicant changes in the economic environment typically lead to conThomas Romer and Barry R. Weingast stituency pressures for regulatory change. This change may be initiated by the regulatory agency, but to be sustainable, it must meet congressional and presidential approval. The 1982 legislative response to the erosion of thrift profitability-deregulation coupled with forbearance for sick thrifts-can be seen in this light.
3. Regulatory change that is resisted by active constituencies and not supported by other active constituencies rarely succeeds. In 1985 the FHLBB appeared to be deviating from its earlier course. The 1986 stalemate in refinancing FSLIC and the 1987 legislation worked-through delay, direct intervention, and explicit limitation on FSLIC resources-to rein in the FHLBB.
The president, because he can command national attention, is the primary political actor with the power to counter narrow constituency pressure. In the case of the thrifts, the president opted for a stance that did not oppose the congressional tendency toward forbearance.
4. The emergence of constituencies for whom an issue becomes more salient can alter legislative policy preferences. As the scale of insolvencies grew, depositor concern over FSLIC solvency also mounted. By late 1988, the policy preferences of the insolvent thrifts (in favor of continued forbearance) were more strongly opposed by concerns about the viability of deposit insurance guarantees. As our empirical results on the St Germain amendment suggest, this would lead to an increase in support for larger levels of FSLIC capitalization. Moreover, once the required funding exceeded an amount that could be covered mostly by assessments on the industry (around $15420 billion over five years), healthy thrifts would no longer oppose additional amounts. The financing costs of such increments would come from general revenues; deposit guarantees financed in this way would benefit healthy thrifts.
6.5.1 Information and Policy Choice It is sobering to note that the political behavior surrounding the thrift debacle is absolutely ordinary. Through 1988, congressmen behaved as they do in ordinary circumstances: paying solicitous attention to active, wellorganized interests, provided that the readily apparent costs to their other constituents are not noticeably high.
It is an intriguing question whether congressmen would have acted differently had they “really known” in 1986 that delay, forbearance, and intervention would lead to a $400 billion (or more) debacle. If by “really known,” we mean that there was a widely shared consensus held by broad constituencies, then the answer to the question would probably be yes. But this was not the situation. It is no doubt correct to say that any economist who thought seriously about the situation would have seen that the combination of forbearance and deposit insurance was a recipe for disaster. Some economists actually said so, within the regulatory agency and outside it.
But the more widespread belief, reinforced by constituency pressures, was problem would abate rather than explode. To a that-as in 1982-84-the Political Foundations of the Thrift Debacle large extent, of course, this perception was created by the very process of the forbearance that politically endorsed policies made possible. Accounting procedures and regulatory reporting effectively minimized the magnitude of the losses. This put most congressmen in the position of “not knowing” the consequences of forbearance. The lack of an active constituency against forbearance (together with knowledge that agencies often cry “wolf”) allowed congressmen to act in standard ways; namely, to intervene in the regulatory process on behalf of their constituents. Thus, even in the face of the GAO predictions that FSLIC was facing a $50 billion problem in 1987, nothing near this level of recapitalization was ever on the political agenda in that year. Nor was there any consideration of restructuring the guarantees under deposit insurance.
6.5.2 Beyond the Thrift Debacle Given the nature of incentives facing politicians, it is extremely difficult for Congress to deal at a sufficiently early state with emerging policy problems that may turn into catastrophes if left unattended. Unless they can claim credit for actions to stem the crisis today, congressmen face great difficultiesin bucking the current set of interest group forces. Perversely, even when congressmen know that policy crises are emerging, they may have to wait for the crisis to occur before they can take and be rewarded for remedial actions. Part of the foundation for this tendency is that individual legislators are not seen by their voters as playing a role in the problem. They are not penalized for letting the problem grow.
These considerations are particularly relevant because many aspects of the thrift debacle appear now to be replicated in several other financial problems that loom on the congressional horizon-the solvency of commercial banks, farm loan guarantee programs, and various government-guaranteed pension systems.” If the wrong lessons are drawn from the thnft debacle, similar crises may reappear in new settings.
It is tempting to explain the thrift debacle as a story about greed, fraud, and criminal behavior. Surely these are part of the story. It is also tempting to argue that this was a case of regulators in bed with the industry they were supposed to oversee. This certainly also played a role. But underlying these factors is the essential component: Congress sanctioned regulatory forbearance and actively intervened when regulators sought a new, more restrictive path.
In order to argue that the next potential crisis would unfold differently, one would need to demonstrate that structure or incentives have changed. It is possible that, in commercial banking, for example, healthy firms will behave differently than such firms did in the thrift industry. They may perceive that early action to deal effectively with insolvencies is in the long-run interest of the rest of the industry. But it is also possible that, in each case, the regulated interests will remain major constituents of politicians and will argue against forceful r e g ~ l a t o r sWill politicians be more wary? The answer to this deThomas Romer and Barry R. Weingast pends on the degree to which currently diffuse interests-generally, taxpayers as a group-see their representatives as having played a major role in the crisis. By moving the focus to shifty operators in the industry or by blaming incompetent or biased regulators, the political source of the crisis will be obscured.
1. We will sacrifice some precision and use the terms “savings and loan” and “thrift” interchangeably to mean federally insured thrift institutions (which include both S&Ls and some mutual savings banks).
2. The Wall Street Journal reported on 6 April 1990 that Congressional Budget Office and General Accounting Office projections of spending through the 1990s will be between $300 billion and $350 billion, “before factoring in increased net losses from an unexpectedly greater number of [thrift] insolvencies” (emphasis in original).
The article quotes estimates by close observers of the industry that the likely 10-year cost will exceed $400 billion.
3. By 1981, market-value net worth of federally insured thrifts had fallen to
- 17.3% of total assets (Brumbaugh 1988, 50, table 2-7).
4. In 1982 nearly 10 percent of FSLIC-insured thrifts were insolvent by GAAP standards, more than in any prior year (Brumbaugh 1988, 37, fig. 2-1).
5. Prior to the Garn-St Germain Act, the Federal Home Loan Bank Board (FHLBB) had eased some accounting rules to help ailing thrifts, by lengthening the list of items that could be excluded as liabilities and those included as assets, in computing net worth. Under Garn-St Germain, “well-managed” thrifts with net worth between 0.5% and 3% of assets could issue “net worth certificates” that could be exchanged for FSLIC promissory notes and counted as assets. In this way a thrift institution could convert what was, in effect, a liability into an asset. Moreover, once FSLIC agreed to buy such certificates from an institution, it was committed to continue buying them as long as the institution was deemed to be “well-managed.’’
6. A typical report, carried in the 13 August 1983 issue of National Journal, proclaimed that Garn-St Germain had “rescued” the thrift industry and that thrift executives were optimistic about their new options.
7. Kane (1989b) provides a good overview of the first and third of these hypotheses.
Explanations that lean heavily on fraud and skulduggery are presented in a spate of “inside story” books: e.g., Adams (1989), Pizzo, Fricker, and Muolo (1989), Pilzer and Dietz (1989). Movie versions cannot be far behind.